**Chemical Traits of Fermented Alfalfa Brown Juice: Its Implications on Physiological, Biochemical, Anatomical, and Growth Parameters of Celosia**

**Nóra Bákonyi 1, Szilvia Kisvarga 2, Döme Barna 1, Ibolya O. Tóth 1, Hassan El-Ramady 1,3, Neama Abdalla 1,4, Szilvia Kovács 1, Margaréta Rozbach 5, Csaba Fehér 5, Nevien Elhawat 1,6, Tarek Alshaal 1,3,\* and Miklós Gábor Fári <sup>1</sup>**


Received: 3 December 2019; Accepted: 3 February 2020; Published: 7 February 2020

**Abstract:** Brown juice is a byproduct of fractionated green biomass during leaf protein isolation. It represents approximately 45%–50% of the total pressed fresh biomass. Disposal of brown juice is a serious issue in leaf protein production due to its high biological oxygen demand and carbohydrates content. The current study aimed to find a possible potential use of brown juice. Therefore, chemical and biochemical properties of brown juice—derived from alfalfa green biomass—were determined before and after fermentation by lactic acid bacteria. Additionally, the growth stimulation potential of fermented brown juice on plumed cockscomb (*Celosia argantea* var. plumose 'Arrabona') plants were tested. Celosia seedlings were sprayed at different rates of fermented brown juice (i.e., 0.5%, 1%, 2.5%, 5%, and 10%) and tap water was applied as control. The results revealed that lactic acid bacteria successfully enhanced the stabilization of brown juice via reducing sugars content and increasing organic acids content. After fermentation, contents of glucose monomers were 15 times lower; while concentrations of lactic and acetic acids increased by 7- and 10-fold, respectively. This caused a reduction in the pH of fermented brown juice by 13.9%. Treating Celosia plants at lower rates of fermented brown juice (up to 1.0%) significantly induced their growth dynamics and antioxidant capacity. Higher values of vegetative parameters were measured in treated plants compared to control. The brown juice treatments caused significant changes in histological parameters as well. The activity of catalase and peroxidase increased in plants that received fermented brown juice especially at low rates. Moreover, an increase in water-soluble protein and phenol was measured in different tissues of plants sprayed with fermented brown juice. Malondialdehyde content was lowered in treated plants compared to control. Fermented brown juice at high rates slightly reduced the amount of photosynthetic pigments; however, this reduction was not reported for low rates of fermented brown juice. These results surely illustrate the potential use of fermented alfalfa brown juice as a growth stimulator for crops particularly at rates below 2.5%.

**Keywords:** deproteinized leaf juice; fermentation; lactic acid bacteria; plant nutrition; antioxidant capacity; ornamental plants

#### **1. Introduction**

Due to the continuous growth in the global population (7.2 billion) and malnutrition, the global demand for the protein will increase in the next years [1]. The lack of protein supply has existed as a health problem for many years and is considered as one of the main types of malnutrition in developing and developed countries [2]. Over the next decades, a dramatic increase in the global protein demand is expected and overall protein consumption is predicted to nearly double by 2050. These rapid changes will create serious and accelerated pressure on land and water resources and their scarcity [3]. To meet the increased protein demand there are several approaches to introduce novel protein sources or alternatives [4,5]. The extraction of proteins from forage crops such as alfalfa, clover or grass is a potential process for the production of leaf protein concentrates (LPC), which can be utilized as feed or food but also hydrolyzed into amino acids for the cosmetics or pharma industries [6]. Alfalfa or lucerne plant is well known as the king of forage. It is a perennial flowering plant belonging to the legume family Fabaceae. This plant has several advantages including high-quality leaf protein (50%–60%), strong adaptability, high nutritional value, good taste, wide distribution, and stable productivity [7]. It can also yield crude protein 2-, 3-, and 4-fold higher than peas, soybean, and wheat, respectively [8]. Therefore, alfalfa nowadays is considered as the most promising crop for LPC. Isolation of leaf protein in form of LPC aims to extract solid or insoluble proteins (i.e., the protein of mitochondria, chloroplasts, nucleoprotein, and cell wall) and soluble proteins (i.e., the soluble fraction of mitochondrial proteins, chloroplast matrix, and cytoplasm proteins). Therefore, the thermal treatment of green juice obtained by pressing the fresh biomass is needed to coagulate these types of proteins. During coagulation of leaf protein, a brown liquid byproduct is produced, and it is referred to as "brown juice". One kilogram of fresh alfalfa biomass can produce up to 500 g of brown juice [9]. These large amounts of brown juice are rich in protein and phenols as well as micronutrients. Plant phenolic compounds are known to be able to modulate important physiological routes like signal transduction and transcriptional regulation. Phenolic compounds in brown juice associated with auxin bioregulators [10] prove that the disposal of these amounts of valuable brown juice is a waste. Disposal is high in its costs and will waste the nutritional value of this byproduct, which would be easily adaptable to the circular economy concept; a technology that generates no further waste by utilizing all the produced renewable resources [11–13]. The main product, the leaf protein produced by coagulation, is widely studied [14]; however, the brown liquid, also known as whey or brown juice [15], has limited literature especially in the case of the plant nutrition aspect. Brown juice is mentioned in some articles as DPJ (Deproteinized Plant Juice) [16] or deproteinized leaf extracts or leaf juice, deproteinized whey [17] as a byproduct of plant protein-producing technologies. DPJ can be applied for several purposes; for instance, as a fertilizer for plants, milk for calves, excellent fodder for cattle and rabbits, medium for microbial growth, and also for in vitro rhizogenesis [18–21]. The dry matter and protein content of brown juice range from 13% to 15% and 16%–20%, respectively, whereas the cellulose content is 25%–30% [18]. The alfalfa brown juice has a dry matter content of 4%–8% which is influenced by the species, varieties, weather conditions, phenophase, methods of harvest, and processing.

Several microorganisms like lactic acid bacteria (LAB) are useful, having advantageous features, and can be found in a range of locations from soil and natural water, to the surface of plants up to the human intestinal tract [22]. These microorganisms have been applied for decades in the fermentation processes of raw materials because of their beneficial effects. It has been validated that ferments containing lactic acid bacteria (or other PGPB—plant growth-promoting bacteria) (isolated from different sources) have plant growth-promoting properties [23]. Lactic acid bacteria containing ferments were proven to be effective biofertilizers, biocontrol agents, and biostimulants because they promote plant health, growth, and resilience as they improve nutrient availability [24], however, the functional roles of these bacteria in the phytomicrobiome have not been discovered yet [25]. *Celosia* genus is native to tropical America and Africa. *Celosia argentea* is a food crop in West Africa as well as a medicinal plant in China and India with considerable pharmacological properties [26]. Among 13 green leafy vegetables, *Celosia argentea* was one of the few that had exceptionally high iron (13.5 mg 100 g<sup>−</sup>1), calcium (188 mg 100 g−1), sodium (240.6 mg 100 g−1), ascorbic acid (26 mg 100 g−1), and β-carotene (4.42 mg 100 g−1) content. The edible portion of *Celosia argentea* was found to be 55 g 100 g−<sup>1</sup> fresh weight which was one of the highest, while its moisture and protein content was found to be 87.6 and 3.2 g 100 g−1, respectively [27]. Plumosa Group of *Celosia argentea* is an attractive ornamental plant characterized by a wide range in size and color of flowers. Plumosa cultivars can grow from dwarf to tall. The inflorescence of narrow pyramidal, plume-like, is consistent with tiny, vivaciously colored (e.g., orange, red, purple, yellow) flowers.

This research aimed to enhance the stability of stored brown juice through fermentation by lactic acid bacteria; assess physiochemical traits of alfalfa brown juice before and after fermentation; determine whether the different fermented brown juice concentrations have any impact on the formation of the stem's anatomy; and evaluate the potential of fermented brown juice as a growth stimulator using *Celosia argantea* var. *plumosa* as a model plant.

#### **2. Materials and Methods**

#### *2.1. Brown Juice Production and Its Characteristics*

#### 2.1.1. Source of Alfalfa Biomass

A field experiment of alfalfa (*Medicago sativa* L. var. Hunor-40) was carried out during 2017 and 2018, under the GINOP (2.2.1-15-2017-00051) project labeled Proteomill [28], at the experimental farm of Tedej Zrt., Hajdúnánás, Hungary. The seeds were sown on chernozem soil at the rate of 25 kg ha<sup>−</sup>1. All recommended agronomic practices such as irrigation, weed control, and fertilization were done. The alfalfa fresh biomass was used as a source for brown juice. The first cut of alfalfa plants was carried out in the middle of May 2018 directly before the flowering stage since at this time protein in alfalfa biomass is at its highest content. Plants were harvested early morning and directly transferred in special boxes to the laboratory to avoid the degradation of protein by protease enzyme.

#### 2.1.2. Extraction of Brown Juice

Alfalfa fresh biomass was fractionated into the fiber, leaf protein concentrate (LPC), and deproteinized plant juice (DPJ, brown juice) as follows: fresh biomass was pressed and pulped mechanically using Angel Juicer (5500, Angel Ltd., Praha, Czech Republic) into fiber and green juice fractions. Later, the green juice was thermally treated at 80 ◦C in order to coagulate mainly the chloroplastic and cytoplasmic proteins. After thermal coagulation, the mixture was left at room temperature for approximately 10 min, then the coagulant was separated from brown juice using moistened 100% natural unbleached cotton cloth filter (pore size = 10 microns).

#### 2.1.3. Fermentation of Brown Juice

Fermentation of brown juice was necessary to increase the stability of brown juice and its storage period because fresh brown juice rapidly spoils due to high sugar and protein content. After cooling, the brown juice was transferred into a 20-L container and inoculated with AdiSil LG-100 Perfect (Fides Agro, Šardice, Czech Republic) containing heterofermentative lactic acid bacterial cultures (10<sup>11</sup> CFU g<sup>−</sup>1, *Pediococcus acidilactici, Lactobacillus paracasei, Lactobacillus plantarum*) at the rate of 0.01 g L<sup>−</sup>1. The inoculated samples were kept at 35 ◦C for 48 h.

#### 2.1.4. Determination of Lactic Acid Bacteria

The qualitative measurement of lactic acid bacteria in the fermented brown juice was determined at the end of the fermentation process by methylene blue test [29]. Briefly, 1 mL methylene blue reagent was added to 10 mL fermented brown juice, and then the samples were incubated at 37 ◦C for 48 h. The time needed for the disappearing of blue color is an indication of lactic acid bacteria density in the solution.

#### 2.1.5. Chemical Properties of Brown Juice

The pH of brown juice was measured by pH-meter (Mettler Toledo S20 Seven Easy, Switzerland). Electrical conductivity (EC) was determined using EC-meter (Thermo Scientific, Orion Model 209A<sup>+</sup> type, Germany). Degree Brix was recorded manually by a refractometer (RBR32-ATC, Germany). The content of macro- and micro-elements in brown juice before and after fermentation was measured using HNO3-H2O2 wet digestion method as described by Kovács et al. [30]. Briefly, 1 g lyophilized brown juice was weighed into a Kjeldahl digestion tube, then 10 mL HNO3 (99%, VWR International, USA) was added. The mixture was placed on the heater at 100 ◦C for 45 min; after cooling 5 mL H2O2 (30%, Sigma-Aldrich, St. Louis, MO, USA) was added for complete oxidation of organic materials and samples were kept on the heater for additional 45 min at 120 ◦C. After cooling the sample volume was brought to 50 mL using distilled water and then filtered using MN 640 W filter paper. The elemental content of brown juice was measured by ICP-OES spectrometer (Perkin Elmer made OPTIMA 3300 DV, Pittsboro, NC, USA).

Total phenol content in brown juice was determined spectrophotometerically using Ultrospec spectrophotometer (2100 pro, Amersham BioSciences, Amersham, United Kingdom) as previously described by Boór and Bélafiné Bakó [31]. Determination of total N content was carried out by Kjeldahl method [32] (Sparks et al., 1996). Concentrations of glucose and organic acids were determined by HPLC using BioRad (Hercules, CA, USA) Aminex HPX-87H (300 × 7.8 mm) column at 65 ◦C, and a refractive index detector. The eluent was 5 mmol L−<sup>1</sup> H2SO4 at a flow rate of 0.5 mL min−1. The injection volume was 40 μL. Concentrations of fructose, xylose, and arabinose were determined by HPLC using Phenomenex (Torrance, CA, USA) Rezex RPM-Monosaccharide Pb+<sup>2</sup> (300 <sup>×</sup> 7.8 mm) column at 80 ◦C, and a refractive index detector. The eluent was ultrapure (milli-Q) water at a flow rate of 0.5 mL min−1. The injection volume was 40 μL. Total sugars include monomer sugars and sugar oligomers solubilized. Monomer sugar concentrations were determined by HPLC after a sample preparation of 5 min boiling followed by centrifugation (5000 rpm) to eliminate residual proteins. To determine the oligomer sugar content of the samples, weak acid hydrolysis was performed. The samples were mixed with 8 w/w%H2SO4 at a volume ratio of 1:1 and treated at 120 ◦C in the autoclave for 15 min to decompose sugar oligomers into monomers, which were determined by HPLC.

#### *2.2. Celosia Experiment*

This experiment was carried out to assess the potential use of brown juice as a plant growth stimulator. In the present study, Celosia (*Celosia argantea* var. *plumosa* 'Arrabona') was used as a model plant for examining physiological, biochemical, and anatomical responses to fermented brown juice in our department and the National Agricultural Research and Innovation Center (NARIC, Budapest, Hungary). Celosia seeds were obtained from NARIC.

#### 2.2.1. Experimental Design

A greenhouse pot experiment was carried out at the NARIC. The experimental layout was the Randomized Complete Block design (RCB) with 15 replicates. A polyethylene pot (7 × 7 × 8 cm) was filled with potting soil for horticultural crops (Klassman-Deilmann TS 3 FINE type, Geeste, Germany). The physical and chemical properties of potting soil are structure fine, pH (H2O) 6, N 140 mg L−1, P (P2O5) 100 mg L−1, K (K2O) 180 mg L−1, Mg 100 mg L−1, S 150 mg L−1. Seeds of Celosia were

sown in nursery substrate on 16th July 2018 and 4 days later germinated seeds were fertilized using different rates of brown juice. After two weeks, identical and healthy seedlings were transferred to the pots. Fermented brown juice was applied as a foliar application at rates of 0.5%, 1.0%, 2.5%, 5%, and 10%. The final application volume was 250 mL and equally shared among all replicates of the same treatment. The control plants were sprayed with tap water. Brown juice was applied once a week from starting the experiment on 16th July until 14th August, then we applied brown juice twice a week (Tuesdays and Fridays) until the end of the experiment on 11th September. At the end of the experiment, the following vegetative parameters were measured: root and stem length, root and stem volume, root and stem fresh and dry mass, and the number of leaves.

#### 2.2.2. Determination of Water-Soluble Protein and Antioxidant Enzymes

Water-soluble protein fraction of lyophilized root, stem, and leaf tissues was determined using Coomassie Brilliant Blue G-250 according to Bradford [33] in triplicate with bovine serum albumin as standard. Briefly, 20 mg plant tissue was ground into homogenate in the mortar with quartz sand, then transferred into a volumetric flask, and then suspended in 100 mL distilled water to extract water-soluble protein fraction. The solution was centrifuged at 3000 rpm for 5 min. The supernatant was used for the assay of water-soluble protein content using UV-160A spectrophotometer (Shimadzu, Japan) at 595 nm. Peroxidase activity was determined in lyophilized roots, stems, and leaves of Celosia plants according to Roxas et al. [34]. Briefly, 100 mg plant tissue was macerated in 4 mL of phosphate buffer 0.01 M (pH 6.0). The homogenate was centrifuged at 13,000 rpm for 10 min to collect the supernatant. The supernatant was used to measure peroxidase activity using UV-160A spectrophotometer (Shimadzu, Japan) at 460 nm for 1 min. The unit of peroxidase activity was defined with the increase of one unit of absorbance per mL−<sup>1</sup> min−<sup>1</sup> g−<sup>1</sup> of dry matter. Catalase (CAT) activity in lyophilized Celosia leaves was measured by following the decomposition of hydrogen peroxide at 240 nm according to Woodbury et al. [35]. The reaction included 0.2 mL supernatant, 1.5 mL phosphate buffer (pH 7.8, 0.2 M), and 1 mL distilled water. The colorimetric determination of CAT was conducted by the model UV-160A spectrophotometer (Shimadzu, Japan) at 240 nm. The biochemical reaction was initiated by adding 0.3 mL 0.1 M H2O2. The activity of CAT was expressed as μmol H2O2 consumed/mg protein/min.

#### 2.2.3. Malondialdehyde Measurement

The malondialdehyde (MDA) content was determined from roots, stems, and leaves of Celosia plants by the method of Zhang and Huang [36]. Briefly, 100 mg lyophilized sample was homogenized in 1 mL 0.1% (*w*/*v*) TCA solution using cold mortar and pestle. The homogenates were centrifuged at 10,000× *g* for 10 min. Then, 4 mL of 0.5% thiobarbituric acid (TBA) in 20% TCA solution was added into 1 mL of supernatant and incubated at 96 ◦C for 30 min. The tubes were cooled by transferring into an ice bath. The absorbance of the supernatant was recorded at 532 nm. The standard curve was generated from MDA standard. The concentration of MDA of samples was calculated from the absorbance knowing calibration curve.

#### 2.2.4. Photosynthetic Pigment

The photosynthetic pigment content of Celosia leaves was measured spectrophotomertically based on methods described by Porra et al. [37]. For the sample preparation, the leaf disc was cut and the chlorophyll content was extracted by N'N dimethyl-formamide overnight. The absorbance was measured by spectrophotometer (Amersham Biosciences Ultrospec 2100 Pro UV/Visible) on 663 and 645 nm wavelengths and from these data, the chl a, b, a + b, and a/b ratio were calculated.

#### 2.2.5. Histology

We used three specimens per treatment for the stem's histological examination. Each plant was cut into smaller pieces and the third internodes (from beneath) fixed separately in a mixture of glycerin:alcohol:water (1:1:1) for a week. Then, several cross-sections were prepared using blades, after clarification, they were stained with Toluidin-blue. All analyses were performed under a light microscope (Zeiss Axioscope 2+; Zeiss International, Oberkochen, Ostalbkreis, Germany) with a compatible camera, and the Scope Photo software (Scopetek, München, Germany) was used for processing the images. For the measurement, we used at least 15 different cross-sections per internodes. The measured parameters were thick at the epidermis, primary cortex, pith, primary and secondary vascular tissue.

#### *2.3. Statistical Analysis*

Before the ANOVA test, Levene's Test for Equality of Variances was performed. The Levene's test for different variables at the six treatments of brown juice (i.e., 0%, 0.5%, 1%, 2.5%, 5%, and 10%) was negative, *p* < 0.05, and then the variances showed homogeneity. Results of the experiments were subjected to one-way (for fresh and dry weight, chlorophyll pigments, protein, MDA, POD, and catalase) and two-way (for root and shoot lengths, root and shoot volumes, and number of leaves) ANOVA by 'SigmaPlot 12.0' software and the means were compared by Duncan's Multiple Range Test [38] at *p* < 0.05.

#### **3. Results**

#### *3.1. Characteristics of Brown Juice*

#### 3.1.1. Chemical Traits of Brown Juice

The fermentation of brown juice significantly changed its chemical properties (Table 1). Inoculation of fresh brown juice by lactic acid bacteria under anaerobic conditions caused a 13.9% reduction in pH. The degree Brix slightly increased after fermentation as it changed from 7.03 to 7.20. Total phenolic content dropped down after fermentation by almost 33.4%. Moreover, EC of fermented brown juice was 25.2% lower than fresh brown juice. Additionally, the density of brown color, that brown juice has, was reduced as its absorbance at 430 nm was diminished by 35.9%.



Notes: † Degree Brix = water-soluble sugar content (one degree Brix means 1 g of sucrose in 100 mL aqueous solution); ‡ nd = not detected; sample size (*n* = 6); *H*-samples run on Aminex HPX 87 H column; *Pb*-samples run on Aminex HPX 87 Pb column.

#### 3.1.2. Contents of Sugars and Organic Acids in the Brown Juice

Furthermore, the effect of lactic acid bacteria was not only reflected in the chemical characteristics of brown juice but also was noticed in sugars content. Interestingly, contents of monomer and oligomer forms of glucose, xylose, arabinose, and fructose reduced after fermentation, except arabinose monomer which was below the detected limit in fresh brown juice and became 0.1 g L−<sup>1</sup> after fermentation; also, no fructose oligomer was detected either in fresh or fermented brown juice samples (Table 1). The highest decrease was found for glucose monomer as it lowered by 16 times in fermented brown juice compared to fresh brown juice. Fructose monomer, also, was four times lower in fermented brown juice, while arabinose oligomer recorded a decrease of 57.3% (Table 1). In contrast to sugars content, organic acids such as acetic, lactic, and propionic acids were considerably increased after fermentation by lactic acid bacteria. The content of lactic acid was 10-fold higher in fermented brown juice, as the highest recorded increase for any measured organic acid, while acetic acid content changed by seven times higher. Propionic acid content was below the detected limit in fresh brown juice; however, after fermentation it increased, recording 1.2 g L−<sup>1</sup> (Table 1).

#### 3.1.3. Macro- and Microelements Content of Brown Juice

Content of macro-and microelements of brown juice meaningfully changed due to fermentation by lactic acid bacteria (Table 2). Fermentation of brown juice resulted in a substantial reduction in the concentration of N, P, K, and S by 11%, 32%, 38%, and 21%, respectively. Otherwise, the contents of other elements displayed in Table 2 were found to be considerably higher after treating brown juice with lactic acid bacteria under anaerobic conditions. Interestingly, concentrations of Ca, Mg, Mo, Sr, and Ba were increased by 55%, 63%, 36%, 54%, and 109%, respectively. Furthermore, Na, Mn, Fe, Zn, B, and Al contents were 14.5-, 2.0-, 11.0-, 2.5-, 1.5-, and 5.7-fold higher, respectively, in fermented brown juice than fresh brown juice. No Cu was detected in brown juice either fresh or fermented.


**Table 2.** Content of macro- and microelements (mg L<sup>−</sup>1) in alfalfa brown juice before and after fermentation.

Notes: † Standard deviation; ‡ not detected; sample size (*n* = 6).

#### *3.2. Fermented Brown Juice as A Growth Stimulator*

The possible utilization of fermented brown juice as a growth stimulator was evaluated. Celosia seedlings were treated with different doses of fermented brown juice through foliar application.

#### 3.2.1. Growth Dynamic of Celosia

Spraying of Celosia seedlings with fermented brown juice significantly induced the development of stems (Figure 1). The application of brown juice at low concentrations had better effects on plant growth than higher concentrations. Spraying Celosia plants with 0.5% of fermented brown juice resulted in the tallest stem (26.0 cm); however, higher concentrations drastically diminished stem length. For instance, at the rate of 10% fermented brown juice stem length was 15.6 cm (Figure 1A). The root system of Celosia plants responded to fermented brown juice differently to the shoot part. All rates of brown juice resulted in very similar lengths of root systems except the rate of 10% which caused a significant reduction in root systems (16.9 cm). However, the tallest root system was found in control plants sprayed with tap water (Figure 1A).

**ZĂƚĞƐŽĨĨĞƌŵĞŶƚĞĚĂůĨĂůĨĂďƌŽǁŶũƵŝĐĞ ;йͿ**

**Figure 1.** Length (**A**) and volume (**B**) of root and stem systems of Celosia plants fertilized at different rates of fermented alfalfa brown juice applied as a foliar application. Sample size (*n* = 6). Different letters above the same columns show significant differences at the level of *p* < 0.05.

Although, length of shoot and root systems is considered as a good indicator for plant growth and its response to the newly added fertilizers and/or stimulators, alone it does not precisely describe the real status of plant health. Therefore, to have a comprehensive description of the shoot and root systems, their volumes should be also measured. This is very essential particularly to describe the root system and its architecture as shoot parts respond to growth conditions in a proportional way. Concerning stem volume, similar findings as for its length were reported. At lower rate of fermented brown juice (0.5%) the highest volume of stem (6.0 cm3) was measured while increasing the rate of fermented brown juice gradually and significantly declined the stem volume and lowest volume (2.3 cm3) was measured for plants sprayed at 10% fermented brown juice (Figure 1B). Results of root volume presented in Figure 1B displayed that although control plants had the tallest root length, its volume was the lowest among all the treatments. This means that control plants had long roots but unbranched ones with few lateral roots. All treated Celosia plants with fermented brown juice showed higher root volume compared to control plants. The highest root volume was noticed at plants sprayed with 5% of fermented brown juice. Additionally, results show that higher rates of fermented brown juice, i.e., 5% and 10% resulted in higher measured root volumes (Figure 1B).

Fresh mass of different Celosia tissues (roots, stems, and leaves) significantly responded to spraying the plants with different rates of fermented brown juice as shown in Figure 2. Fresh mass of roots, stems, and leaves of all plant parts was higher for plants treated with fermented brown juice compared to control plants sprayed with tap water. The highest fresh mass of roots, stems, and leaves was 4.30, 8.87, and 8.49 g plant<sup>−</sup>1, respectively, that measured at rates of 2.5%, 0.5%, and 5% fermented brown juice, respectively (Figure 2A). Control plants displayed the lowest dry mass of roots, stems, and the number of leaves 0.17, 0.31, and 0.55 g plant<sup>−</sup>1, respectively; while sprayed plants with 2.5% fermented brown juice showed the highest determined dry mass 0.44, 0.64, and 0.86 g plant<sup>−</sup>1, for roots, stems, and leaves, respectively (Figure 2B). All rates of fermented brown juice, except 10%, significantly increased the number of leaves per plant (Figure 2C). Applying fermented brown juice at the rate of 10% significantly decreased the number of leaves not only compared to other fermented brown juice rates but also control plants. The highest number of leaves per plant was 18 and was counted for plants treated with 1% fermented brown juice. However, the differences between treatments of 0.5%, 1%, 2.5%, and 5% of fermented brown juice were not significant.

**Figure 2.** *Cont.*

**Figure 2.** Fresh (**A**) and dry (**B**) masses and the number of leaves (**C**) of different plant tissues (roots, stems, and leaves) of Celosia plants sprayed at different rates of fermented alfalfa brown juice. Sample size (*n* = 6). Different letters above the same columns show significant differences at the level of *p* < 0.05.

#### 3.2.2. Antioxidant Capacity of Celosia Plants Treated with Fermented Brown Juice

Spraying Celosia plants with fermented brown juice significantly induced the activity of catalase (CAT) enzyme in the leaves (Figure 3A). All treated plants had higher activities of CAT enzyme compared to control plants (sprayed with tap water). However, increasing the rate of applied brown juice gradually reduced the CAT activity up to 5%, but this reduction was still higher than the control. Treated Celosia plants at the rate of 10% achieved the highest CAT activity among all treatments (0.290 μmoL H2O2 consumed mg−<sup>1</sup> protein min<sup>−</sup>1).

Different Celosia plant tissues (i.e., root, stem, and leaf) showed a significant response of peroxidase enzyme activity (POD) to added fermented brown juice (Figure 3B). Higher rates of fermented brown juice above 1.0% resulted in higher POD activity in the root system than both lower rates and control plants. The root POD activity in treatments of 2.5%, 5%, and 10% of fermented brown juice was higher than lower rates and the control; however, no statistically significant differences were calculated among these treatments. Interestingly, applying fermented brown juice at the rate of 1% resulted in the lowest determined activity of POD in the root system among all treatments including the control plants. The activity of POD in the stem tissue of Celosia plants was totally in contrast to POD activity in the root system (Figure 3B). The high rates of fermented brown juice above 1% showed lower POD activity in the stem than low rates (i.e., 0.5% and 1%) and control plants. The lowest POD activity in stems

was noticed when plants were sprayed at 10% fermented brown juice, while the highest measured POD activity in the stem was found for plants that received 0.5% fermented brown juice (Figure 3B). Except for treatments of 2.5% and 5% fermented brown juice, all other treatments including control plants showed similar POD activity in leaf tissue without significant differences. The highest leaf POD activity was measured in the leaves of treated plants with 2.5% fermented brown juice, while at the rate of 5% fermented brown juice the lowest leaf POD activity was determined (Figure 3B).

Malondialdehyde (MDA) content in different tissues of Celosia plants was measured as a marker for the degree of lipid peroxidation of unsaturated fatty acids due to oxidative stress. In the root system of Celosia plants, the highest measured value of MDA content was denoted in control plants. All treated plants with fermented brown juice had lower root MDA content than control plants. However, the response of treated plants with fermented brown juice hesitated as no clear trend was seen. The lowest applied rate 1% fermented brown juice showed the lowest root MDA content, while the highest root MDA content was measured in the root system of plants sprayed with 2.5% fermented brown juice (Figure 3C). In contrast to the root system, stem MDA content was found to increase as the rate of fermented brown juice increased up to 5% then reduced at the rate 10% recording the lowest MDA content in stem tissue among all treated plants with fermented brown juice. However, the lowest MDA content in the stem was displayed in control plants. Leaves of control plants showed higher MDA content than plants that received different rates of fermented brown juice. No significant differences were found in the MDA content of leaves of plants sprayed at the rates of 1%, 2.5%, and 5% fermented brown juice (Figure 3C). However, the lowest leaf MDA content was determined in the leaves of plants treated with 0.5% fermented brown juice.

**Figure 3.** *Cont.*

**Figure 3.** The activity of catalase (**A**) and peroxidase (**B**) and malondialdehyde content (**C**) in different plant tissues (roots, stems, and leaves) of Celosia plants sprayed at different rates of fermented alfalfa brown juice. Sample size (*n* = 6). Different letters above the same columns show significant differences at the level of *p* < 0.05.

#### 3.2.3. Phenolic, Protein, and Photosynthetic Pigments Contents

The results of water-soluble phenol content are depicted in Figure 4A. Different plant tissues of Celosia plants possessed different water-soluble phenol contents as the root system showed the lowest content, while the highest water-soluble phenol content was measured in leaves. The addition of fermented brown juice as a foliar application to Celosia plants significantly affected the water-soluble phenol content in the root system. The highest water-soluble phenol content (46.7 μg g−1) was measured in the root system of plants that received 0.5% fermented brown juice, while, when plants were allowed to grow in the presence of 10% fermented brown juice, the water-soluble phenol content was 12.8 μg g−<sup>1</sup> (Figure 4A). The root system of the control plant displayed 18.6 μg g−<sup>1</sup> water-soluble phenol content. In stem tissues, water-soluble phenol content in plants treated with fermented brown juice showed lower water-soluble phenol content than control plants. Increasing the rate of fermented brown juice up to 2.5% gradually increased the content of water-soluble phenol in stem tissues, then a linear increase was recorded when rates of fermented brown juice were increased up to 10%. The highest stem water-soluble phenol content (56.8 μg g<sup>−</sup>1) was measured for control plants (Figure 4A). Except for treatment of 0.5% fermented brown juice, all fermented brown juice rates showed higher water-soluble phenol content in leaf tissues. The highest water-soluble phenol content (μg g-1) was measured in leaves of plants that received 1% fermented brown juice; then a gradual decrease was noticed with increasing the rate of fermented brown juice up to 10% (Figure 4A).

The content of water-soluble protein was higher in leaf tissue followed by the root system, while the lowest content was denoted in stem tissue. Significantly, the application of fermented brown juice improved water-soluble protein content in the root system. Treatments of 1%, 2.5%, and 10% fermented brown juice displayed higher water-soluble protein content than control and treatments of 0.5% and 5% (Figure 4B). The lowest water-soluble protein content (2.18 mg g<sup>−</sup>1) was measured in the root system of plants treated with 0.5% of fermented brown juice. Similar results were found in stem tissue of Celosia plants, where all treated plants with fermented brown juice had higher water-soluble protein content than control except treatment of 0.5% fermented brown juice. Although the content of water-soluble protein in leaves was higher than measured in the root system, the trend in which roots and leaves responded to spraying with fermented brown juice was almost the same. Leaf water-soluble protein contents in plants of treatments of 0.5% and 5% were the lowest among all treatments including the control. Other fermented brown juice rates enhanced the water-soluble protein content in leaf tissue over the control plants (Figure 4B).

Significant differences were noticed in a few cases among treatments for chlorophyll pigments content (Figure 4C). Content of *chl a* was reduced gradually with increasing the rate of applied fermented brown juice. Application of fermented brown juice at low rates (i.e., 0.5%) significantly improves the *chl a* content recording the highest value among all other treatments but was similar to control plants. On the other hand, the *chl b* content was found to respond negatively to increasing the rate of applied fermented brown juice as a gradual significant reduction was noticed. Content of total *chl a* + *b* displayed a similar tendency as it slightly decreased with increasing the rate of fermented brown juice. The low rate of fermented brown juice showed a slightly higher content than the control, but this increase was not significant (Figure 4C). Except treatment of 0.5% fermented brown juice, carotenoids content in all treatments including the control showed similar values as no significant differences were statistically measured. The lowest carotenoids content was determined in leaves of plants that received 0.5% fermented brown juice (Figure 4C).

**Figure 4.** *Cont.*

**Figure 4.** Water-soluble phenol (**A**), water-soluble protein (**B**), and chlorophyll pigments contents (**C**) in different plant tissues of Celosia plants sprayed at different rates of fermented alfalfa brown juice. Sample size (*n* = 6). Different letters above the same columns show significant differences at the level of *p* < 0.05.

#### 3.2.4. Anatomical Features of Celosia Stem after Brown Juice Application

Regarding the cross-sections, 10–15 cm from the apex were analyzed, and the tissue structure was representative of an older Celosia's stem anatomy, with successive cambia [39,40]. Stems were covered by the epidermis (single row); beneath its primer cortex containing angular collenchyma (four to six cells thick) was visible. In the pith primary vascular bundles were located surrounded by a cylinder of anomalous cambium. Secondary and primary vascular tissues were separated by the conjunctive tissue [41]. Both the conjunctive tissue and the inner part of the pith were composed of parenchymatous cells (Figure 5).

There is no fundamental difference in the tissue structure in connection with the treatment, but there are significant differences in the thickness of the tissues, which support the differences that are visible to the naked eye too (e.g., thicker, stiffer stem). All levels of concentration caused a reduction in the thickness of the epidermis, while it was the 1% treatment that caused a reduction to a greater extent. The thickness of the primary cortex reinforced with angular collenchyma was decreased by most treatments, except the 0.5% and 10% treatments, where statistically verified thickening was observed. The proportion of pith involved vascular tissues increased for all treatments. The more concentrated brown juice treatments resulted in significantly thicker primary vascular tissues, except the 1% and the 10% treatments. Growing of the secondary vascular tissue was the highest at the 0.5% treatment, where the new successive cambium formed almost entirely closed xylem and phloem, significantly contributing to the strength of the stem (Table 3).

**Table 3.** Impact of different concentrations of fermented brown juice on stem tissue of *Celosia argentea* var. *plumosa (*μm*)* (mean ± SD, *n* = 45).


Notes: Different letters in each column indicate statistically significant differences (*p* < 0.05).

**Figure 5.** Anatomical sections of *Celosia argentea* var. *plumosa* stem. *ep* epidermis, *co* cortex, *ang* angular collenchyma, *pi* pith, *sc* successive cylinder (*svt* secondary vascular tissue), *ca* cambium, *cjt* conjunctive tissue, *pvb* primary vascular bundle after spraying Celosia plants with different rates of fermented brown juice (i.e., control, 0.5%, 1%, 2.5%, 5%, and 10%). *Scale bar is* 200 μm.

To sum up, it can be stated that the brown juice treatments (applied as foliar) influence the proportions of the Celosia stem's tissue. As a result of the treatments, the thinning of the epidermis and the intense growth of the vascular tissues (especially the secondary vascular tissue) can be projected. The growth rate of secondary tissues within the pith is the highest at 0.5% treatment.

#### **4. Discussion**

Recently, isolation of protein from plant green leaves has gained increasing attention as an attempt to bridge the gap between protein production and demand due to the dramatic increase in population and the increase in living standards. Brown juice (referred to as deproteinized plant juice or DPJ as well) is a byproduct generated during the coagulation of soluble protein in green juice through thermal treatment. Brown juice has gained less attention than LPC and press cake. It represents nearly 50% of pressed and pulped fresh biomass [9]. Therefore, these huge amounts could be an obstacle facing the acceleration of this approach and its acceptance by both politicians and the public. Disposal of alfalfa brown juice is a serious issue in LPC production due to its high biological oxygen demand (BOD) and carbohydrates content [42]. Due to its richness in free amino acids, peptides, soluble sugars, vitamins, and many macro- and microelements, it can be directed towards animal feeding and production of many chemicals [43]. Additionally, it can be used as a fertilizer, growth stimulator and/or growth medium for microorganisms [42]. Although some pieces of literature have been reporting the possible utilization of brown juice as a ruminant feed [44], few studies have been focusing on brown juice as a fertilizer [42,45].

During our recent experiments on LPC production from alfalfa biomass, it became clear that the storage of the brown juice at room conditions leads to fast spoiling. Therefore, we had to store it below 4 ◦C. This may be due to its high carbohydrate content, which represents a suitable environment for bacteria to grow [42]. Therefore, converting sugars into organic acids and subsequent decrease in the pH of brown juice through fermentation using lactic acid bacteria seemed to be an ideal solution since fermented brown juice is stable and this facilitates its handling.

Lactic acid bacteria have long been known for their role in the fermentation of carbohydrates. Consequently, it has wide applications in medicine and food processing. Nowadays, lactic acid bacteria have been found to play an important role in agriculture, bioenergy production, and bioremediation of the environment [46]. In the present study we, firstly, aimed to stabilize alfalfa brown juice through reducing its water-soluble sugars content and pH using lactic acid bacteria. Accordingly, sugars content in brown juice was reduced after fermentation, because lactic acid bacteria use sugars as energy and carbon sources [47]. As shown in Table 1, most of the sugars in brown juice were found to be below the quantification limits after fermentation indicating that lactic acid bacteria consumed them. Comparable results have been previously presented by many researchers [48,49]. On the other hand, organic acids (e.g., lactic, acetic, and propionic acids) were increased in fermented brown juice causing a subsequent reduction in pH (Table 1). Novik, et al. [46] reported lactic acid as the main acid produced after fermentation of water-soluble sugars such as glucose and fructose either monomer or oligomer by lactic acid bacteria. Similar findings were cited by Bautista-Trujillo et al. [48], who observed a decrease in pH of maize silage after inoculation by lactic acid bacteria due to the increase in the production of organic acids, mainly lactic and acetic acids. Moreover, they reported an increase of 46.3% in lactic acid content. However, lactic acid content was found to increase by 8-fold after fermentation compared to unfermented brown juice (Table 1). This high increase in lactic acid content may be attributed to the initial low pH of fresh brown juice (4.54), which helps to hydrolyze the oligo- and polysaccharides, therefore they become available for lactic acid bacteria [48]. Additionally, another possible reason for high lactic acid content could be attributed to the high Mn content in brown juice. Cheng et al. [49] stated that applying Mn to Jerusalem artichoke juice enhanced the lactic acid production by lactic acid bacteria up to 12 g L−1. Moreover, Dimitrovski et al. [50] stated that the fermentation of Jerusalem artichoke tuber juice by lactic acid bacteria reduced its pH from 6.5 to 4.7 after 30 h. In the current study, at the end of the fermentation process, the pH of brown juice was 3.91. Lactic acid bacteria significantly reduced the absorbance of brown juice by 35.9%. This result was supported by that previously cited by Kwaw et al. [51]. They studied the effect of different strains of lactic acid bacteria on colorimetric properties of mulberry juice, reporting a 6.9% reduction in the color. They referred this reduction to the increase in content of the monomeric anthocyanin. Although an increase in total phenolic content has been previously reported for fermented mulberry juice [51] and pomegranate juice [52], our results displayed a decrease of 33.4% after inoculation of brown juice by lactic acid bacteria. Except N, P, K, and S other macro- and microelements were higher in fermented alfalfa brown juice (Table 2). The reduction in concentrations of N, P, K, and S could be attributed to the fact that they are essential elements for the growth of lactic acid bacteria. However, similar findings were described

by Kim [53], who stated that fermented kale juice had higher elemental composition than unfermented juice. Moreover, he cited significant differences between kale juice fermented by different lactic acid bacterial strains. The increase in the concentration of microelements, in particular, may be due to the increase of brown juice acidity. Although, the concentration of macronutrients (e.g., N, P, and K) was reduced after fermentation, the content of macro- and microelements is still high, and this makes the fermented brown juice a potential growth stimulator. On the whole, these results are supported by earlier findings of Ream et al. [45]. They reported that alfalfa brown juice contains a relatively high content of N and K, in addition to small amounts of P, Ca, Mg, and other microelements.

In the present study, fermented alfalfa brown juice as a growth stimulator was evaluated using the Celosia plant as a model. Brown juice was applied at different rates by foliar application. The foliar application of fermented brown juice showed a significant potential on the development of Celosia seedlings in comparison with control. Noticeably, increasing the application rate of brown juice sprayed on Celosia seedlings led to a considerable reduction in shoot parts, particularly the stem length. From a horticultural point of view, this seems to be a good result since the target is the flower not the vegetative growth of Celosia. Application of brown juice at low rates such as 0.5% and 1.0% enhanced the growth and resulted in high values of stem length, the volume of stem and root, fresh masses of stem and root, and number of leaves. Shorter but more branched root systems were observed when Celosia seedlings were sprayed with brown juice (Figure 2). This phenomenon is supported by data of length and the volume of roots (Figure 1). The beneficial effect of alfalfa brown juice could be attributed to its high content of macro- (i.e., N, P, K, Ca, and Mg) and microelements (i.e., S, Mn, Fe, Cu, Zn, and Mo); all in phyto-available forms (Table 2). Similarly, Ream et al. [45] observed that using brown juice as a fertilizer added at an annual rate of 1.25 cm induced the growth and yield of alfalfa, corn, and bromegrass; while, at the higher rate (2.5 cm) a reduction in yield and plant damage were noticed in all crops. However, they referred to the damage in plant growth caused by high rates of brown juice to unknown reasons; moreover, they considered it a not serious problem since the added amount of brown juice can be controlled. These results are supported by findings of Reddy et al. [42], who earlier stated that the application of alfalfa brown juice at low rates enhanced germination and growth of cowpea, mung bean, and groundnut; while high rates inhibited the germination and reduced the plant growth. They reported that alfalfa brown juice can be used as a fertilizer if it would be added at a lower level than 10%. Additionally, they could not give a reason for such damaging effects of high rates of brown juice, except what previously was mentioned by Pirie [54], who stated that alfalfa brown juice contains some phytotoxic organic compounds.

In our experiment, the reduction in plant growth of treated Celosia plants at high rates of fermented brown juice can be explained by high EC value and low pH of brown juice solutions. Increasing the rates of brown juice caused a gradual increase in EC and a decrease in pH as shown in Table 4. At treatment of 0.5% brown juice, EC (dS m−1) and pH were 0.12 and 4.21, respectively; while, at the highest applied rate 10% they were 1.99 and 4.38, respectively. Low pH is not favorable for the development of plants; it reduces photosynthesis due to a reduction in stomatal conductance [55]. This might explain why we found diminished growth of Celosia plants treated at high rates of brown juice. However, in the current study, the lowest pH was 4.38 when Celosia plants received 10% fermented brown juice. Long et al. [56] had earlier reported a reduction in citrus growth below pH 4, while higher pH did not inhibit the growth and seedlings reached their maximum growth at pH 5. The reduction in growth may be attributed to H+-toxicity which damages leaves. Absorption of nutrients applied as foliar application depends on the pH of the solution. Extreme pH (below 2 and above 12) was reported to burn the leaves. Moreover, some elements prefer different pH values for their optimum absorption by plant leaves.

Antioxidant enzymes such as CAT and POD are among the most important antioxidant enzymes which play a vital role in scavenging reactive oxygen species generated in cells due to different biotic and abiotic stresses [57]. Thus, enhancing the activity of these enzymes is considered an important step in improving the plants' tolerance to different kinds of stress [58,59]. The results abstracted from this research showed that the application of alfalfa brown juice after fermentation by lactic acid bacteria significantly increased the activity of CAT and POD in different Celosia tissues. However, the low rates of brown juice seemed to be more effective than higher ones, as a reduction in the activities was noticed. These results are confirmed by results of MDA, as treated plants with fermented brown juice had lower MDA content than control plants (untreated plants) regardless of the type of plant tissues. These results demonstrate that fermented brown juice can potentially be exploited as a growth stimulator, particularly at low rates. Besides, fermented brown juice had a significant effect on water-soluble phenol and protein contents, as they were higher in treated plants in comparison to control ones. The high rates of brown juice were found to reduce the photosynthetic pigment content. This could be attributed to low pH at high rates of fermented brown juice. This result was in accordance with that cited by Solati et al. [60], who reported a decrease in chlorophyll content due to low pH.

Brown juice could be very useful as a soil fertilizer/conditioner particularly in alkaline soils and/or sandy soil due to its rich composition in macro- and microelements and sugars. These could induce the microbial growth in soil increasing soil fertility. Additionally, sugars play an important role in soil stabilization through maintaining soil aggregation, which subsequently leads to better water holding capacity [42,61]. While delivering deeper insights into the possible use of alfalfa brown juice as a growth stimulator and trying to precisely determine the most effective rate and application method, there are many issues that should be addressed in the future [62]. Results, undoubtedly, suggest that brown juice tolerance can be plant species dependent; therefore, more studies on different plant species at different rates of brown juice are crucially needed. Additionally, phytotoxicity of brown juice should be the focus of future studies.


**Table 4.** pH and electrical conductivity (EC, dS m<sup>−</sup>1) values of alfalfa brown juice solutions before and after fermentation at the beginning of the experiment.

Notes: † Standard deviation.

#### **5. Conclusions**

The present study highlights the possible use of alfalfa brown juice as a growth stimulator. Brown juice is a serious problem in LPC production, where its disposal represents a threat to the environment due to its high content of water-soluble sugars as well as macro- and microelements. Fermentation of brown juice using lactic acid bacteria significantly improved its nutritional value and stability, because these bacteria—as our data showed—produce a significant amount of organic acids i.e., lactic, acetic, and propionic acids through their metabolism making the nutrients more available and the pH of row material (brown juice) lower, thereby stabilizing it. Most water-soluble sugars were under the detectable level after fermentation as the bacteria used them as carbon source. Moreover, the concentration of nutrients increased—showing the effect of bacteria for nutrient availability—after fermentation except N, P, K, and S showed a slight decrease. In this study, treating *Celosia argentea*, a valuable ornamental species with significant food and medical uses, with low rates of fermented brown juice through foliar application significantly improved the growth, as all of the vegetative parameters such as stem and root length, shoot and root volume, fresh mass of stem and root, and the number of leaves increased. The brown juice treatments in low (0.5%) concentration caused positive changes in histological parameters, in the growth rate of secondary tissues. Additionally, fermented brown juice showed a considerable impact on the antioxidant capacity of Celosia plants, as CAT and

POD activities increased while MDA content decreased. Moreover, both water-soluble phenol and protein were found to increase in treated plants with fermented brown juice compared to the control showing the beneficial effect of lactic acid fermentation and chemical properties of brown juice. These results conclude and state the potential use of fermented alfalfa brown juice as a sustainable growth stimulator for crops with a particular interest in horticultural crops. Our data regarding the chemical and microbiological properties of brown juice and the effects (listed plant responses) it triggered confirm the scientific investigations where plant growth-promoting properties of lactic acid bacteria contribute greatly to the maintenance of the health of plants (also strengthening disease resistance) and by consuming these plants, they are also beneficial in human digestive processes. It should be noted that the sample size number was modest in our study that is why it was difficult to draw strong far-going conclusions, however preliminary conclusions support the fact that phytomicrobiome engineering can be a promising strategy for sustainable agriculture, but the data available is limited to understand properly these complex symbiotic relationships. Therefore, examination of fermented alfalfa brown juice's effect on physiological, biochemical, and anatomical parameters of other horticultural and agricultural crops is in progress.

**Author Contributions:** Conceptualization, N.B., S.K. and M.G.F.; methodology, D.B., N.E. and I.O.T.; software, D.B.; validation, D.B. and I.O.T.; formal analysis, N.A. and S.K.; investigation, M.R. and C.F.; resources, H.E.-R. data curation, T.A.; writing—original draft preparation, N.B. and D.B.; writing—review and editing, N.B., T.A., H.E.-R. and D.B.; visualization, S.K.; supervision, M.G.F.; project administration, N.B.; funding acquisition, N.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The research was financed by the Higher Education Institutional Excellence Programme (NKFIH-1150-6/2019) of the Ministry of Innovation and Technology in Hungary, within the framework of the Biotechnology thematic programme of the University of Debrecen."; "Complex Rural Economic and Sustainable Development, Elaboration of its Service Networks in the Carpathian Basin (Project ID: EFOP-3.6.2-16-2017-00001, Hungary)" research project. Also, support was given by Tempus Public Foundation (TPF), Hungary and Széchenyi 2020 under the GINOP-2.2.1-15-2017-00051.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

### **Nitrogen Use and Uptake E**ffi**ciency and Crop Performance of Baby Spinach (***Spinacia oleracea* **L.) and Lamb's Lettuce (***Valerianella locusta* **L.) Grown under Variable Sub-Optimal N Regimes Combined with Plant-Based Biostimulant Application**

**Ida Di Mola 1,\*, Eugenio Cozzolino 2, Lucia Ottaiano 1, Sabrina Nocerino 1, Youssef Rouphael 1, Giuseppe Colla 3, Christophe El-Nakhel <sup>1</sup> and Mauro Mori <sup>1</sup>**


Received: 14 January 2020; Accepted: 12 February 2020; Published: 15 February 2020

**Abstract:** An optimized nitrogen (N) fertilization may have a positive effect on leafy vegetables by increasing growth, yield and nutrient content of plants. Nevertheless, crop performance must be coupled with an increase in Nitrogen Use Efficiency (NUE) in order to limit external N inputs and to avoid N surpluses associated with environmental and health problems. The aim of the current study was to assess the effects of a legume-derived plant hydrolysates (LDPH; Trainer®) and N fertilization levels (0, 2.25 and 4.5 g N m−<sup>2</sup> for spinach and 0, 2.5 and 5.0gNm−<sup>2</sup> for lamb's lettuce; N0%, N50%, N100%, respectively) on agronomical, biochemical, qualitative responses and NUE of these two important greenhouse leafy vegetables. Spinach and lamb's lettuce were sprayed four times during the growing period (at a concentration of 4 mL L−<sup>1</sup> of LDPH). In baby spinach, the LDPH application elicited a significant increase at the three levels of N fertilization: +16.8%, +14.2%, and 39.4% at 0, 2.25 and 4.5 g N m<sup>−</sup>2, respectively. Interestingly, in lamb's lettuce, the N50% plants treated with LDPH reached similar values of marketable yield in comparison to treated and non-treated plants under N100% conditions. The presumed mechanism involved in the enhancement of yield response in the two leafy greens could be associated to a better activity of the photosystem II (higher SPAD index), biochemical (higher content of chlorophyll a, b and total) and leaf nitrate status. The foliar application of LDPH produced a major fortification in lipophilic and hydrophilic antioxidant activities (+11.6 and 6.3% for spinach and lamb's lettuce, respectively). The biostimulant application also improved N-use efficiency and N-uptake efficiency compared to untreated plants: +17.8% and +18.8%, and +50% and +73.3%, for spinach and lamb's lettuce, respectively.

**Keywords:** N fertilization; nitrogen use efficiency; antioxidant activity; leaf quality; protein hydrolysate; *Spinacia oleracea* L.; sustainable agriculture; *Valerianella locusta* L.

#### **1. Introduction**

Chemical fertilizers, especially nitrogen (N), are basically the main input for boosting yield and concomitantly one of the most expensive inputs in terms of economics and environment. Many crops require high amounts of this element to maximize yield [1], but N fertilization requires a particular care

because it is involved in many environmental and health risks [2]. The main environmental impacts of N can be summarized in the contamination of surface and groundwater resources and greenhouse gases emissions [3,4]. The effects on human health strongly depend on the accumulation of nitrate in edible plant tissue; when nitrate is reduced to nitrite in human body it can cause methemoglobinemia, which is dangerous to children [5–7]. Moreover, nitrite can also react with several chemical compounds (amines and amides), producing N-nitrous compounds, known as probably carcinogenic to humans [7–9].

On the other hand, it is certainly necessary to adapt the correct management of N fertilization through a balanced application of the elements in order to reach the right dose, nevertheless by choosing the convenient chemical form and application time. Moreover, another possible perspective is to raise the nitrogen use efficiency (NUE) that is linked to the capacity of plants to uptake nutrients, nevertheless to their systems of transport, storage and mobilization and to the N loss into the environment [10]. NUE is expressed as the harvestable yield per the amount of available N in the soil or per N supply [11–13].

In recent years, the approach to improve NUE, passed through biotechnology and plant breeding strategies, but currently it is necessary to evaluate alternative means, which are environmentally friendly, such as the use of plant biostimulants. These products can be used to complement fertilizers in order to reduce the inputs and increase the NUE [14]. They act in several ways: on plant growth, physiology, carbon and nitrogen metabolism, productivity, product quality and tolerance to abiotic stress [14]. Moreover, some studies found that plant biostimulants, particularly commercial legume-derived proteins, have a great potential to reduce nitrate accumulation in the leaves of some green leafy vegetables [15]. It is of a major result because these crops have the genetic predisposition to greatly accumulate nitrate in their leaves [16]. It is known that the different crops ability of nitrate accumulation can depend on different localization and activity of nitrate reductase (NR) [17,18], but also on unbalanced relationship between nitrate uptake and NR activity, as well as the different capacity of uptake, translocation and accumulation of plants [16]. Moreover, this behavior is worsened by specific environmental conditions, where nitrate accumulation increases at low solar radiation [19–21].

In addition, the cultivation in protected environment causes a similar effect, because the plastic film cover reduces the solar radiation transmission. Likewise, the photoperiod and growing period affect nitrate accumulation; in fact, both conditions are matched to conditions of low solar radiation.

Green leafy vegetables play a key part in the economical market of many countries, both in the Mediterranean area and Nord-Europe, because they are widely used in ready-to-eat salads. In addition to typical leafy greens such as lettuce and rocket, also spinach and lamb's lettuce are largely spreading. Italy is a leading country in the European production of green leafy vegetables destined for the ready-to-eat market, with more than 150 kilotons harvested per year in protected conditions [22,23]. Among these crops, spinach is the less-efficient in terms of N uptake and use [24], requiring high rates of fertilization to grow well and reach higher leaf quality (dark green leaves) [25].Instead, lamb's lettuce is still under-studied, and its behavior regarding NUE under different N regimes is unknown.

Previous studies regarding vegetable crops including leafy greens have documented that the application of plant biostimulants triggers several molecular and physiological processes, accompanied by improvement in growth, yield, quality, NUE and tolerance to abiotic stress [22,26–37]. The capacity of biostimulants to improve NUE is the utmost reason for which they are spreading in the market, considering their economic and environmental motives [38]. However, relatively few researches regarding biostimulants effects on plants grown under sub-optimal N conditions are available [33,35,39–41], especially about green leafy vegetables. The reduction of N inputs in leafy vegetables is very important, both for containing the phenomenon of nitrate accumulation in leaves and for reducing the economic and environmental impacts of fertilization. Di Mola et al. [42] reported that the foliar application of different biostimulants (in particular seaweed extract and legume-derived hydrolysate protein) on greenhouse baby lettuce boosted plant growth, mainly in sub-optimal N fertilization. Furthermore, in baby rocket cultivated under greenhouse conditions, Di Mola et al. [43]

found that the application of plant-based biostimulants boosted the marketable yield at low N levels compared to the control.

The aim of this study was to assess the effect of foliar application of legume-derived protein hydrolysates on N demand and uptake efficiency of two important leafy greens. Therefore, two experiments were carried out for evaluating the possible beneficial effects of a plant-derived protein hydrolysates applied on greenhouse spinach and lamb's lettuce grown under variables N conditions, in order to depict its influence on NUE, yield and leaf quality.

#### **2. Materials and Methods**

#### *2.1. Experimental Setting, Leafy Vegetables Tested and Cultural Practices*

Two consecutive experiments were carried out in a plastic tunnel during winter 2018/2019 and spring 2019 seasons at the experimental site "Gussone Park" of the Department of Agricultural Sciences (40◦48.870 N; 14◦20.821 E; 70 m a.s.l.), University of Naples Federico II, Italy. The two tested crops were cultivated in large pots (diameter 0.70 m and height 0.60 m) filled with sandy soil, with the following physical and chemical proprieties: pH 7.4, 2.5% organic matter, 0.9 g kg−<sup>1</sup> total N (Kjeldhal method), 252.6 mg kg−<sup>1</sup> P2O5 and 490.9 mg kg−<sup>1</sup> of K2O.

For the first experiment (Winter 2018/19), baby spinach (*Spinacia oleracea* L. cv. Platypus RZ F1, Rijk Zwaan, Bologna, Italy), a widely spread cultivar in Italy with dark green leaves, was sown on January 17th (1000 seeds per square meter) and harvested on March 12th. While for the second experiment (Spring 2019), lamb's lettuce (*Valerianella locusta* L. cv. Princess HM CLAUSE, Torino, Italy) was sown on March 26th (1200 seeds per square meter)—this cultivar is characterized by deep green leaves and a high adaptability to different growing seasons—and harvested in five different dates from May 10th till the 25th, upon reaching the marketable size according the different treatments. The germination time was 8 and 10 days after sowing and the plant densities after germination were 900 and 1100, for spinach and lamb's lettuce, respectively. For both crops, there were no differences between the treatments in terms of plant density.

Considering the chemical composition of soil, no phosphorus or potassium was given to either crop; while N was added as ammonium nitrate (34%) in a single application 27 and 20 days after the sowing, for spinach and lamb's lettuce, respectively. Water was not a limiting factor; the crop evapotranspiration was calculated with the Hargreaves method and the deficit was fully restored by sprinkler irrigation.

#### *2.2. Experimental Design, Nitrogen Fertilization and Biostimulant Application*

A factorial combination of three nitrogen fertilization levels and two biostimulant applications (treated and non-treated control) distributed in a randomized complete-block design were adopted for both experiments. Each treatment was replicated three times accounting a total of 18 pots (3 N levels × 2 biostimulant applications × 3 replicates).

The optimal nitrogen dose was calculated based on the balance method that considers all inputs and outputs. For the first experiment (spinach) N levels were: optimal dose (N100%) <sup>−</sup>4.5 g m−2, sub-optimal dose (N50%) <sup>−</sup>2.25 g m−<sup>2</sup> and no fertilization (N0%). While for the second experiment (lamb's lettuce) N levels were: optimal dose (N100%) <sup>−</sup>5.0 g m<sup>−</sup>2, sub-optimal dose (N50%) <sup>−</sup>2.5 g m−<sup>2</sup> and no fertilization (N0%).

The plant-based biostimulant used for both green leafy vegetables was a legume-derived protein hydrolysates, promoted as Trainer ® by Italpollina S.p.A. The legume-derived PH biostimulant obtained through enzymatic hydrolysis contains 75% of free amino acids and peptides, 22% of carbohydrates and 3% of mineral nutrients. The detailed aminogram of the product along with the phenolics, flavonoids, and elemental composition were reported in detail by Rouphael et al. [22]. For both crops, the treated plants were sprayed four times at 21, 27, 33 and 39 days after sowing, at a concentration of 4 mL L<sup>−</sup>1. Untreated control spinach and lamb's lettuce plants were only sprayed with water. Each pot was

sprayed with a solution volume of 38.5 mL (=1000 L ha<sup>−</sup>1) corresponding to a biostimulant application rate of 0.000154 mL per pot (=4 L of biostimulant per ha).

#### *2.3. Marketable Yield and Sampling*

In both experiments, the whole area of all the pots at harvest was cut and leaves were weighed in order to measure the marketable fresh yield. In addition, a representative sub-sample of each replicate was dried in a forced air oven at 70 ◦C and then weighed in order to determine dry weight and then to calculate leaf dry matter content and subsequently used for N content determination (total N and nitrate) by chemical analysis. For qualitative analysis, fresh samples were also collected from each replicate and conserved at −80 ◦C.

#### *2.4. Nitrogen Determination, N-use E*ffi*ciency and Uptake E*ffi*ciency*

The Kjeldahl method [44] was used to determine the concentration of N in dried leaves samples that were mineralized with sulfuric acid, while nitrate content was determined using the Foss FIAstar 5000 continuous flow Analyzer (FOSS analytical Denmark).

Nitrogen use efficiency (NUE) was calculated by dividing yield by N application dose plus the available N in the soil and expressed as ton per kg. In addition, N uptake efficiency was determined as the ratio between N content in the leaves and N application dose and it was expressed as kg kg<sup>−</sup>1.

#### *2.5. Leaf Quality: Antioxidant Activity and Compounds, Chlorophyll Content and SPAD Index*

Lipophilic (LAA) and hydrophilic (HAA) antioxidant activities were determined using the protocols of Re et al. [45] and Fogliano et al. [46], respectively. The two extract fractions, lipophilic and hydrophilic, were measured by the means of a Hach DR 2000 spectrophotometer at 734 and 505 nm, respectively.

The Kampfenkel et al. [47] method was used to determine ascorbic acid spectrophotometrically. A wavelength of 525 nm was set in order to measure the absorbance of the extract. Total phenols were also assessed spectrophotometrically, and the absorbance solution was detected at 765 nm, based on the Singleton et al. method [48].

Leaves chlorophyll content was measured spectrophotometrically: the first step was the extraction of fresh material by ammoniacal acetone as described by Wellburn [49], then the absorbance of solutions was measured at 662 and 647 for chlorophyll a and b, respectively.

The soil plant analysis development (SPAD) index was measured at harvest, on 15 leaves by replicate, using a portable SPAD-502 chlorophyll meter.

#### *2.6. Statistical Processing*

In both experiments, a two-way ANOVA was conducted using the SPSS 21 software package. Duncan's Multiple Range Test (DMRT; significance level 0.05) was adopted for mean comparisons on each of the independent measured variables.

#### **3. Results**

#### *3.1. Marketable Yield and SPAD Index*

The effects of both tested factors (N fertilization rates and biostimulant application) on marketable fresh yield and SPAD index were reported in Figure 1A,B and Figure 2A,B, where the relevant F and P values and the degrees of freedom are reported in Table 1.


**Table 1.** Analysis of variance of marketable fresh yield and SPAD index of spinach and lamb's lettuce (Figure 1A,B and Figure 2A,B).

In particular, the marketable yield of baby spinach was positively influenced by N fertilization, but it was further boosted by biostimulant application (Figure 1A). The LDPH application elicited a significant increase at all the levels of N: +16.8%, +14.2%, and 39.4% at 0, 2.25 and 4.5 g N per square meter, respectively.

As with baby spinach, the marketable yield of lamb's lettuce increased with higher N dose and it was positively affected by biostimulant foliar application (Figure 1B). However, no significant difference was recorded between LDPH-treated and non-treated control plants at the higher N fertilization level (N100%). Interestingly, the N50% plants treated with LDPH reached significantly similar values of marketable yield in comparison to treated and non-treated plants under N100% conditions.

**Figure 1.** Marketable yield of baby spinach (**A**) and lamb's lettuce plants (**B**) as affected by nitrogen (N) fertilization levels (0, 2.25 and 4.5 g N m−<sup>2</sup> and 0, 2.5 and 5.0 g N m<sup>−</sup>2; N0%, N50%, N100%, respectively) and biostimulant application (non-treated control and LDPH: Legume-derived protein hydrolysates). Different letters indicate significant differences according to the DMR test (*p* < 0.05). Vertical bars indicate ± standard error of means.

The SPAD index statistically increased with the higher availability of N and also with the foliar application of LDPH in both baby spinach (Figure 2A) and lamb's lettuce (Figure 2B). The average increase of the SPAD index of fertilized and sprayed spinach plants was 8.6% compared to fertilized unsprayed plants. At 0 g N per square meter, the SPAD index of spinach plants treated with LDPH was +7% compared to untreated N0% plants. Finally, for lamb's lettuce the SPAD index increases due to biostimulant application were less marked: +5.2% and +2.9% for fertilized (N50% and N100% plants) and non-fertilized plants (N0%).

**Figure 2.** The SPAD index of baby spinach (**A**) and lamb's lettuce plants (**B**) as affected by nitrogen (N) fertilization levels (0, 2.25 and 4.5 g N m−<sup>2</sup> and 0, 2.5 and 5.0 g N m<sup>−</sup>2; N0%, N50%, N100%, respectively) and biostimulant application (non-treated control and LDPH: Legume-derived protein hydrolysates). Different letters indicate significant differences according to the DMR test (*p* < 0.05). Vertical bars indicate ± standard error of means.

#### *3.2. N-Use and Uptake E*ffi*ciency*

The results regarding the two efficiency parameters: N use efficiency and N uptake efficiency in baby spinach and lamb's lettuce are presented in Table 2. In both leafy vegetables, significant effects were noted on N use efficiency with both N and biostimulant treatments, but not the N × B interaction, whereas N uptake efficiency was only affected by foliar biostimulant application (Table 2).


**Table 2.** Nitrogen use and uptake efficiency of baby spinach and lamb's lettuce plants as affected by nitrogen (N) fertilization levels (0, 2.25 and 4.5 g N m−<sup>2</sup> and 0, 2.5 and 5.0 g N m<sup>−</sup>2; N0%, N50%, N100%, respectively) and biostimulant applications (control and LDPH: Legume-derived protein hydrolysates).

NS, \*, \*\* Non-significant or significant at *p* ≤ 0.05 and 0.01. Different letters within each column indicate significant differences according to Duncan's test (*p* ≤ 0.05). The numbers in parenthesis are the data of 90% confidence interval.

When averaged over the N treatments, the baby spinach plants sprayed with the plant-based biostimulant showed a 17.8% and 50.0% increase compared to untreated plants, for N-use efficiency and N-uptake efficiency, respectively (Table 2). Moreover, irrespective of biostimulant application,

the N0% and N50% plants had the highest values of NUE (Table 2). The trends of the two efficiency parameters in lamb's lettuce were similar to those of spinach but were always higher. In particular, the NUEs of unfertilized plants were significantly higher than in N50% and N100% plants around +69% and +59.5%, respectively (Table 2). When averaged over the N treatments, the foliar application of LDPH improved N-use efficiency and N-uptake efficiency compared to untreated plants, by 18.8% and 73.3% respectively (Table 2).

#### *3.3. Total Chlorophyll, Chlorophyll a and b and Nitrate content*

In spinach, the N fertilization levels statistically affected the content of total chlorophyll, chlorophyll a and b, as well as nitrate content in leaves. This latter was the only parameter also affected by the biostimulant application (Table 3). Particularly, chlorophyll (a, b, and total) content increased when N dose was raised. The two treatments N50% and N100% were not significantly different, but N100% was significantly higher than N0% (+10%, +37.3%, and +20.3% respectively).


**Table 3.** Chlorophyll a and b, total chlorophyll and nitrate content of baby spinach plants as affected by nitrogen (N) fertilization levels (0, 2.25 and 4.5 g N m<sup>−</sup>2; N0%, N50%, N100%, respectively) and biostimulant application (control and LDPH: Legume-derived protein hydrolysates).

NS, \*, \*\* Non-significant or significant at *p* ≤ 0.05 and 0.01. Different letters within each column indicate significant differences according to Duncan's test (*p* ≤ 0.05). The numbers in parenthesis are the data of 90% confidence interval.

As expected, our results demonstrated that increasing N fertilization from 0 to 5.0 g m−<sup>2</sup> elicited a significant linear increase in nitrate content compared to non-fertilized plants. Particularly, at N100%, the nitrate content in LDPH-treated plants exceeded the limits imposed by the European Regulation No. 1258/2011 for the commercialization of fresh spinach (3500 mg kg−<sup>1</sup> on fresh weight basis) as determined by the cultivation practices, growing conditions and latitude (Table 3).

In lamb's lettuce, all parameters were affected by both factors, but not by their interaction (Table 4). The chlorophyll a, b and total content increased with increasing N level; N100% had the highest values and was statistically different from the other two treatments: +19%, +24.7% and +21% over the mean value of N0% and N100%, for chlorophyll a, b and total chlorophyll, respectively. Moreover, the increases due to biostimulant applications were 26.6%, 44.0% and 32.3% for chlorophyll a, b and total chlorophyll, respectively.


**Table 4.** Chlorophyll a and b, total chlorophyll and nitrate content of lamb's lettuce plants as affected by nitrogen (N) fertilization levels (0, 2.5 and 5.0 g N m<sup>−</sup>2; N0%, N50%, N100%, respectively) and biostimulant applications (control and LDPH: Legume-derived protein hydrolysates).

NS, \*\* Non-significant or significant at *p* ≤ 0.05 and 0.01. Different letters within each column indicate significant differences according to Duncan's test (*p* ≤ 0.05). The numbers in parenthesis are the data of 90% confidence interval.

Furthermore, in lamb's lettuce nitrate content in leaves increased when nitrogen dose was raised, but without significant differences between N50% and N100%, and it was higher in the plants sprayed with biostimulants compared to untreated plants. For this crop, the European Community has not fixed any threshold, but if we consider the limit imposed for fresh spinach, only the value of biostimulant-sprayed plants overcame it.

#### *3.4. Leaf Quality: Antioxidant Activity and Compounds*

In spinach, LAA and the content of total phenols and ascorbic acid (AsA) were significantly affected by N fertilization, while the biostimulant application influenced only LAA. HAA was neither affected by N fertilization treatments nor by biostimulant application (Table 5). Irrespective of biostimulant application, LAA, total phenols, and AsA were significantly higher in N0% plants in comparison to N100% plants, around 3.9%, 29.8%, and 41.8% respectively. Interestingly, when averaged over N treatments, the foliar application of LDPH boosted LAA compared to untreated plants by 11.6% (Table 5).

In lamb's lettuce, all the measured leaf quality parameters (LAA, HAA, total phenols, and AsA) were significantly affected by N fertilization levels, while only HAA was affected by the biostimulant application (Table 6). Regarding LAA, total phenols, and AsA, the trends were similar to those observed for spinach; where the values of N0% plants were higher (+8.3%, +23.3%, and +26.9%, respectively) compared to N100% plants. Instead, HAA had an opposite trend: it was higher in fertilized plants (+18.5% compared to unfertilized plants) and it was also higher in the plants sprayed with biostimulant (+6.3%).



NS, \*, \*\* Non-significant or significant at *p* ≤ 0.05 and 0.01. Different letters within each column indicate significant differences according to Duncan's test (*p* ≤ 0.05). The numbers in parenthesis are the data of 90% confidence interval.

**Table 6.** Lipophilic (LAA) and hydrophilic (HAA) antioxidant activity, total phenols and ascorbic acid (AsA) of lamb's lettuce plants as affected by nitrogen (N) fertilization levels (0, 2.5 and 5.0gNm<sup>−</sup>2; N0%, N50%, N100%, respectively) and biostimulant applications (control and LDPH: Legume-derived protein hydrolysates).


NS, \*, \*\* Non-significant or significant at *p* ≤ 0.05 and 0.01. Different letters within each column indicate significant differences according to Duncan's test (*p* ≤ 0.05). The numbers in parenthesis are the data of 90% confidence interval.

#### **4. Discussion**

In order to increase the supply of food produced on the available arable land—since the global population will reach 10 billion by 2050—growers must boost the yield of their produce, through the massive use of technical means, in particular N fertilization. Nowadays, it is impossible to adopt an agriculture that is not sustainable and environmentally friendly. Therefore, the objective of boosting crop productivity must occur through the reduction of N fertilizers, but also through the improvement of nitrogen use efficiency (NUE), that assures reasonable yield and a profit margin for farmers [50].

Several researches have highlighted that plant-based biostimulants have a triggering effect on growth and yield, but they are also capable of improving the NUE in consideration of both economic and environmental motives [38,51]. The plant-based biostimulant used in this test was Trainer®, a legume-derived protein hydrolysate (containing free amino acids and signaling molecules such as small soluble peptides), for which previous researches have already demonstrated its ability to boost crops' resources use efficiency (RUE) [15,52]—especially N uptake and assimilation [39]—as well as productivity [6,32] and quality [53,54]. Our results highlighted the ability of LDPH to enhance yield of both baby spinach and lamb's lettuce (+24.6% and +13.5% for plant sprayed with Trainer® compared to control plants, respectively), which is in line with Carillo et al.'s [35] findings on spinach, and Di Mola et al. [42,43] on other two important leafy greens (lettuce and baby leaf rocket) cultivated under variable N regimes. The positive effects of the foliar application of LDPH, irrespective of the N fertilization treatments, were more pronounced in spinach than in lamb's lettuce, demonstrating a species-specific response [15,55], especially that the same commercial plant-based biostimulant was used. The different responses between the two leafy vegetables species could be attributed to the different leaf permeability and cuticle morphology as well as the stomatal aperture and thus the efficacy of the plant biostimulant [38]. Therefore, our results highlight, that further study is warranted to assess the physiological and molecular mechanisms behind the biostimulant action and to investigate the specificity of species dependent responses in impacting leaf characteristics and consequently interacting with the different bioactive compounds of plant biostimulants. Interestingly, in our study the marketable fresh yield of LDPH-treated spinach and lamb's lettuce grown under N50% was similar to those grown under N100% (especially the non-treated plants). A number of biochemical and physiological aspects may have contributed to this result, including (i) a higher chlorophyll content (a, b and total) and SPAD index in biostimulant-treated than in non-treated plants, and (ii) improved leaf status in terms of nitrate content, triggering a more efficient translocation of assimilates to potential photosynthetic sinks, thus boosting plant growth and yield [35,42,43]. Moreover, several authors attributed the stimulation action and the increased N assimilation in response to LDPH application to multiple mechanisms of action involving (i) the hormones-like activities (i.e., auxin and giberrellins-like activities), (ii) the increase in the activity of the key enzymes glutamine synthetase and nitrate reductase, and (iii) the upregulation of specific genes responsible in N assimilation and pigment synthesis [27,33,56–58].

Similar to the effect N fertilization on agronomic performance, our findings highlighted the higher NUE of baby spinach and lamb's lettuce, even without N fertilization. The current results are in agreement with the findings of several researches such as Abdelraouf [59], Canali et al., 2011 [60], and Zhang et al. [61], which in spinach observed a linear decrease in NUE when N dose increased. Moreover, our findings about N uptake are in line to the results of Canali et al. [60], which observed that this parameter was not affected by variable nitrogen regimes.

Interestingly, our findings also indicated that foliar application of LDPH can be considered an efficient tool to reduce N additional inputs to the cropping system, hence cutting down the production costs for farmers and N surpluses into the environment [62]. Mainly because the LDPH-treated baby spinach and lamb's lettuce plants exhibited both higher NUE and higher N-uptake efficiency, irrespective of the N fertilization levels. The positive effect of foliar application of LDPH on the two N efficiency parameters can be attributed to the improvement of root architecture (i.e., more vigorous root apparatus) which is related to an overall increase in nutrient accessibility caused by its power to boost the capacity of absorption, translocation and assimilation of macro and micro minerals, especially when N is limiting plant growth [27,56,63]. This phenomenon associated to the PH-induced remolding of root advocating N uptake and translocation was described by Colla et al. [64], as "nutrient acquisition response". The stimulation of root system architecture—in particular the increase in root hair density and length—was observed previously by several authors on a wide range of agronomic and horticultural species such as corn, sunflower, tomato, eggplant, lettuce and *Brassica* genus [27,57,64,65].

Although the application of fertilizers (nitrogen, phosphorus and potassium) generally increases the crop yield; alternatively, the excessive application of synthetic fertilizers—especially N—can result in undesirable nutritional quality changes such as a decrease in some bioactive compounds (phenols and vitamin C) and soluble sugars [66]. This was the case in the current study, whereby baby spinach and lamb's lettuce cultivated under N100% negatively modulated the synthesis and accumulation of antioxidant molecules such as total phenols and ascorbic acid along with low antioxidant activity. Similar trends were reported recently by Di Mola et al. [42,43] on baby lettuce and rocket grown under optimal and supra-optimal N regimes.

Concerning the effect of LDPH application on the quality of the two tested leafy greens, some findings demonstrated that the application of protein hydrolysates-based biostimulant was able to modify plant primary and secondary metabolism [15,55], leading to the synthesis and accumulation of phytochemicals with health-promoting properties. This was the case in the current greenhouse experiment, since baby spinach and lamb's lettuce plants treated with the commercial protein hydrolysates Trainer® positively modulated both the lypophilic and hydrophilic antioxidant capacity, which are considered important traits in evaluating the quality of food including leafy vegetables [23]. However, the foliar LDPH application did not affect the concentration of total phenols and ascorbic acid in both leafy vegetables. A variable effect of three commercial plant biostimulants containing mainly free amino acids (Aminovert, Megafol and Veramin) was also observed on the chemical composition, phenolic profile and bioactive properties of two greenhouse spinach cultivars [67]. Therefore, future research should focus on designing the ideotype plant biostimulants and identifying the best species × biostimulant × fertilization (N) combination(s) for the production of healthy and nutrient-dense leafy vegetables.

#### **5. Conclusions**

Sustainable agriculture is the greatest challenge of our century, and plant-based biostimulants represent an efficient and concrete possibility to reach this objective by maintaining high production and improving the NUE of leafy greens with several economic, nutritional and environmental benefits. The positive effects of the LDPH biostimulant were manifested in terms of marketable fresh yield in baby spinach, irrespective of N fertilization treatments and at low N rates (N0% and N50%) in lamb's lettuce. Such benefits were likely derived from the signaling molecules (such as small peptides) as a result of augmented leaf nitrate content, SPAD index and pigments synthesis. These stimulation actions of the LDPH application were more pronounced under sub-optimal (N0% and N50%) than under optimal (N100%) N regimes. Interestingly, foliar LDPH application in both tested leafy vegetables boosted NUE and N uptake efficiency, which is fundamental for both economic and environmental reasons. Our results also demonstrated that the foliar application of LDPH can promote the antioxidant capacity which is important for the human diet and may constitute an added value for both growers and consumers. Overall, our findings suggest that the application of protein hydrolysates can be a sustainable practice in intensive greenhouse cropping systems to enhance crop productivity and NUE under both optimal and sub-optimal (low-input conditions) N regimes.

**Author Contributions:** Conceptualization, M.M. methodology, I.D.M., S.N. and E.C. software, L.O., I.D.M. and C.E.-N. validation, M.M., I.D.M., E.C. and L.O. formal analysis, C.E.-N. and L.O. investigation, I.D.M. resources, M.M. data curation, I.D.M. and L.O. writing—original draft preparation, I.D.M. writing-review and editing, Y.R., C.E.-N. and G.C. visualization, L.O., I.D.M. and C.E.-N. supervision, Y.R., G.C. and M.M. project administration, M.M. funding acquisition, M.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Brief Report* **Exploratory Study on the Foliar Incorporation and Stability of Isotopically Labeled Amino Acids Applied to Turfgrass**

**Rachel M. McCoy 1,2, George W. Meyer 1,2, David Rhodes 1, George C. Murray 3, Thomas G. Sors <sup>4</sup> and Joshua R. Widhalm 1,2,\***


Received: 13 February 2020; Accepted: 3 March 2020; Published: 5 March 2020

**Abstract:** There is increasing interest in the use of amino acid-based biostimulant products due to their reported abilities to improve a number of quality characteristics in a variety of specialty crops. However, when it comes to the foliar application of amino acids to turfgrass, there are still many basic questions about their uptake forms and incorporation into cellular metabolism. In this study, we shed light on the fate of amino acids exogenously applied to turfgrass foliage through a series of time-course, isotopic-labeling studies in creeping bentgrass (*Agrostis stolonifera* L.) leaves. Using both 15N-labeled and 15N,13C double-labeled L-glutamate applied exogenously to creeping bentgrass foliage, we measured the uptake of glutamate and its integration into γ-aminobutyric acid (GABA) and L-proline, two amino acids with known roles in plant stress adaptation. Our results demonstrate that glutamate is rapidly absorbed into creeping bentgrass foliage and that it is utilized to produce GABA and proline. Based on the labeling patterns observed in the endogenous pools of glutamate/glutamine, GABA, and the proline from applied glutamate-[13C5 15N1], we can further conclude that glutamate is predominantly taken up intact and that mineralization into other forms of nitrogen is a minor fate. Taken together, the collective findings of this study provide evidence that amino acids exogenously applied to turfgrass foliage can be rapidly absorbed, and serve as stable sources of precursor molecules to be integrated into the metabolism of the plant.

**Keywords:** biostimulants; amino acids; isotopic labeling; turfgrass

#### **1. Introduction**

The use of biostimulants to promote quality traits in specialty crops has gone up over the last decade [1]. With an estimated annual growth rate of more than 10% each year, the projected market for biostimulants is estimated to be at \$4.9 billion by 2025 [2]. Biostimulants is a broad term referring to extracts, lysates, purified natural compounds or microorganisms that are applied to crops in small amounts to enhance aspects like health, resiliency, and/or vigor [1,3] but whose primary role is not to fertilize or protect against pathogens [4].

Amino acids and small peptide-based biostimulants have received increased attention for their positive effects on plant performance [5]. Whilst externally applied amino acids are poorly taken up by roots because of competition with soil microbes, foliar application has the potential to improve availability due to reduced competition [6]. As a result, amino acids are emerging in many foliarly applied products marketed to golf course superintendents and sports turf managers with claims of enhanced growth, greening, and increased resistance to stress. Despite the substantial sales of such products from a variety of companies in the turfgrass market, there have been limited studies on the uptake by and the fate of amino acids in turfgrass foliage. Using 15N-labeled glycine, L-glutamate, and L-proline, it was previously demonstrated that the nitrogen from these applied amino acids was absorbed into creeping bentgrass (*Agrostis stolonifera* L.) foliage to similar degrees as other nitrogen fertilizer forms [7]. Assuming that no mineralization into other transportable forms of nitrogen occurred on the leaf's surface, this suggests that amino acids can be directly taken up through bentgrass foliage. This now raises questions about the metabolic fate of exogenously applied amino acids once inside the plant. The objective of this study was to determine the uptake form, stability, and incorporation of amino acids exogenously applied into metabolism in turfgrass foliage. To accomplish this, we conducted a series of exploratory tracer studies using 15N- and 15N,13C-labeled glutamate, applied exogenously to bentgrass foliage, and we measured their integration into endogenous amino acid pools and derived metabolites.

#### **2. Materials and Methods**

#### *2.1. Plant Growth Conditions, General Experimental Procedures, and Reagents*

Turfgrass used in this experiment was PennTrio Bentgrass (Tee-2-Green Corporation, Hubbard, OR, USA) which is a creeping bentgrass (*Agrostis stolonifera* L.) mix that contains equal parts Penncross, Penneagle, and Pennlinks. Turfgrass was grown in a controlled environment in 8" pots at 23–24 ◦C with an average humidity of 45% under daylight spectrum fluorescent lighting with 12-h days. Plants were watered weekly and fertilized once at germination with a fertilizer containing 12% nitrogen, 6% phosphorus (P2O5), 6% potassium (K2O), and micronutrients boron, copper, iron, manganese, and zinc. Unlabeled amino acid standards were purchased from Sigma-Aldrich (St. Louis, MO, USA). Stable isotopes were purchased from Cambridge Isotopes (Tewksbury, MA, USA). All other reagents were purchased from Fisher Scientific (Pittsburgh, PA, USA). For gas chromatography-mass spectrometry (GC-MS) experiments, an Agilent 7890B GC (Agilent Technologies, Santa Clara, CA, USA) connected with an Agilent 5966A mass spectrometer (Agilent Technologies, Santa Clara, CA, USA) were used. All analyses were done using Agilent Chemstation software.

#### *2.2. Stable Isotope Labeling of Turfgrass*

Labeling was conducted by spraying the foliage of the potted plants with a mixture of each stable-isotopically labeled amino acid in water at a rate of 804 L per hectare, at the concentrations indicated below. The first trial used 10 mM glutamate-[15N1] (Cambridge Isotopes) with sampling at 0, 1, 4, 8, 24, and 48 h post application. In the second trial, 4 mM glutamate-[13C5- 15N1] was applied and sampling occurred at 0, 0.25, 0.5, 1, 4, 8, 24, 48, and 72 h post application. At each timepoint, the aboveground tissue was cut and rinsed to remove any residue and then the leaves were transferred directly into methanol to quench metabolism, and stored at 4 ◦C until extraction.

#### *2.3. Extraction and Quantification of Amino Acids*

Amino acids were extracted according to a protocol adapted from Rhodes et al. [8] using a ration of 10 mL methanol for every 500 mg of creeping bentgrass leaves. Extracts were spiked with 25 μL of 10 mM α-aminobutyrate, vortexed well, and then incubated in the dark at 4 ◦C for 2 d to extract metabolites. Next, for every 10 mL methanol, 5 mL chloroform and 6 mL water were added and incubated for 1 h at room temperature to allow phase separation. The aqueous phase was collected and evaporated to dryness under N2 gas using a Techne sample concentrator. The dried aqueous phase was resuspended in 1 mL of water and applied to a Dowex-50-H<sup>+</sup> 200 mesh column. The column was washed with 7 mL water, and amino acids were eluted from the column with 6 mL of 6 M NH4OH and dried. Amino acids were derivatized for GC-MS analysis, as described previously [9], with 1 μL of each derivatized sample being analyzed by GC-MS on an Agilent 19091s-433 HP-5MS capillary column (30 m × 0.25 mm; film thickness 0.25 μm) as described previously [8]. Labeling percentage was calculated by dividing the intensity of the shifted molecular ion by the sum of the shifted and unshifted ion and corrected for natural isotope abundance. See Supplementary Tables S1 and S2 for masses analyzed for each labeled and unlabeled amino acid.

#### **3. Results and Discussion**

#### *3.1. Nitrogen from Foliar Applied Glutamate is Incorporated into Proline and* γ*-aminobutyric acid (GABA)*

To investigate whether amino acids are absorbed by turfgrass leaves and incorporated into cellular metabolism, we measured time course labeling in the endogenous pools of glutamate and some major glutamate-derived amino acids from glutamate-[15N1] applied to the foliage of creeping bentgrass. In addition to serving as the precursor for the synthesis of chlorophylls and proteins, glutamate functions as a hub metabolite in plant amino acid metabolism (Figure 1). Glutamate is a substrate for producing L-glutamine from ammonia; it serves as the primary α-amino donor for aminotransferases involved in synthesizing multiple amino acids, and its carbon skeleton and amino group are directly incorporated into L-arginine, L-proline, and γ-aminobutyric acid (GABA) [10]. The accumulation of GABA, a non-proteinogenic amino acid found ubiquitously in plants, functions in adaptive responses to mitigate plant stress, including defense against drought and insect herbivory [11]. The overproduction of proline was also demonstrated to be a metabolic response involved in plant stress tolerance. Proline functions as an osmolyte to maintain cell turgor, stabilizes membranes to prevent electrolyte leakage, and helps prevent oxidative bursts by lowering the concentrations of reactive oxygen species [12]. Therefore, because enhanced resiliency to environmental stresses underlies one of the major purported benefits of amino acid-based biostimulant products, we focused on labeling in GABA and proline from glutamate-[15N1] exogenously applied to creeping bentgrass foliage.

**Figure 1.** Glutamate occupies a central position in plant amino acid metabolism. The arrows indicate the multiple fates for the carbon backbone and/or amino group of glutamate in plant cells. The arrow thickness approximately correlates with relative flux toward each metabolite. The arrows labeled with a question mark (aspartate to β-alanine (aspartate decarboxylase), GABA to alanine (GABA: pyruvate aminotransferase), and glutamate to ammonium (glutamate dehydrogenase)) denote metabolic fates that are controversial.

To ensure that the endogenous precursor pool was labeled highly enough to detect possible labeling in GABA and proline, we first examined labeling in glutamate by looking at the glutamate/glutamine pool. Note that in the current sample preparation protocol, glutamine is converted into glutamate during derivatization, so the two amino acids are quantified together by GC-MS as glutamate. Within 1 h of foliar application with glutamate-[15N1], the glutamate/glutamine pool was labeled by 60% and remained constant over the 48-h experiment (Figure 2). The pool of GABA, which is formed via the irreversible decarboxylation of glutamate in plant cytoplasm by glutamate decarboxylase (GDC) [10], was labeled by 29% within 1 h of glutamate-[15N1] application, increased to over 40% labeled 4 h post application, and then remained relatively constantly labeled for the duration of the experiment (Figure 2). The rapid incorporation of glutamate into GABA is consistent with the observation that the expression of the gene encoding GDC in rice roots increased nearly 10-fold in response to exogenous application of glutamate [13]. Labeling in proline, whose biosynthesis from glutamate can take place in chloroplasts or cytoplasm [14], was in comparison expectedly delayed (Figure 2). The proline pool was labeled by 12% 4 h after application with glutamate-[15N1], increased to 23% labeled by 8 h, and then remained constant until 48 h. Taken together with the fact that glutamate must be present in cytoplasm to produce GABA and in the cytoplasm or chloroplast to synthesize proline, these data are consistent with not only glutamate-[15N1] being absorbed into the foliage of creeping bentgrass, but also with it being taken up by cells where it can be utilized to produce metabolites with well-established roles in plant stress adaptation.

**Figure 2.** Time course of percent labeling of glutamate/glutamine-[15N1], GABA-[15N1], and proline-[15N1], from 10 mM glutamate-[15N1] applied to the foliage of creeping bentgrass (*Agrostis stolonifera* L.).

#### *3.2. The Carbon Skeleton from Foliar Applied Glutamate is also Incorporated into Proline and GABA*

In a previous labeling study by Stiegler et al. [7], it was found that the uptake of nitrogen from glycine, glutamate, and proline into creeping bentgrass foliage is equal to or less than that of nitrogen from urea. Thus, it is possible that glutamate-[15N1] applied to creeping bentgrass foliage in the current study was mineralized on the leaf surface and that ammonia-[15N] was absorbed and then re-assimilated into glutamine/glutamate (Figure 1) before being used to synthesize GABA-[15N1] and proline-[15N1] (Figure 2). To definitively determine whether glutamate-[15N1] was taken up intact or mineralized before absorption, we performed the same time course labeling experiment with glutamate-[13C5 15N1]. By using double-labeled glutamate, in which the nitrogen and all carbon atoms are labeled, it is possible to differentiate between the uptake of mineralized ammonia-[15N] and the intact amino acid.

Similarly to what was observed with glutamate-[15N1] (Figure 2), applied glutamate-[13C5 15N1] rapidly labeled the glutamate/glutamine pool (Figure 3A). The predominant form detected was the fully intact form, glutamate/glutamine-[13C5 15N1], which represented approximately 55% of the total pool and remained relatively constant for the duration of the experiment. The second most abundantly labeled form detected was glutamate/glutamine-[13C5]. It was found to represent approximately 10% of the total pool and then attenuated to nearly 0% by 24 h after application. This form would a priori derive from the metabolism of glutamate-[13C5 15N1] to α-ketoglutarate-[13C5] that is transaminated back to glutamate-[13C5] with an unlabeled nitrogen. The least abundant form detected was glutamate/glutamine-[15N1], representing less than 3% of the total pool by 1 h post application and rapidly decreasing thereafter. This form likely results from the labeled nitrogen of absorbed glutamate-[13C5 15N1] being used to transaminate an unlabeled α-ketoglutarate to produce glutamate-[15N1]. This form could also originate if applied glutamate-[13C5 15N1] was mineralized on the leaf surface to produce ammonia-[15N] that was absorbed and then re-assimilated back into amino acid metabolism to produce glutamate-[15N1] (Figure 1). Regardless of how it was formed, because glutamate-[15N1] accounted for such a small fraction of the total glutamate/glutamine pool compared to the 13C-labeled forms, this suggests that the intact amino acid was the predominant form absorbed by turfgrass foliage.

Next, we examined whether the glutamate-[13C5 15N1] applied to creeping bentgrass foliage labeled GABA and proline like what was observed with glutamate-[15N1] (Figure 2). Peak labeling in GABA occurred 1 h after application with glutamate-[13C5 15N1], though labeling was already detectable at 15 min (Figure 3B). Unlike the first experiment (Figure 2), there was a decrease in labeled GABA pools following the initial peak (Figure 3B). This likely reflects the fact that less glutamate-[13C5 15N1] was administered. Previous work in rice by Kan et al. [13] showed that the expression of the gene encoding GDC displays a sensitive dosage-dependent induction in response to glutamate. The subsequent decline and increase in GABA labeling is likely related to the incorporation of GABA into the GABA shunt, a bypass pathway in which the GABA produced in the cytoplasm is imported into the mitochondria, where it is converted to succinate that can enter the tricarboxylic acid (TCA) cycle. The GABA shunt is the major source of succinate in foliage during the day (reviewed in Michaeli and Fromm, 2015 [15]).

The most abundantly labeled form of GABA detected was GABA-[13C4 15N1], which represented approximately 9% of the total pool. This isotopic form likely originated from decarboxylation of glutamate-[13C5 15N1], the predominant labeled form found in the glutamate pool (Figure 3A). The other isotopic forms of GABA detected after 1 h, GABA-[13C4] and GABA-[15N1], represented 7.5% and 3.3% of the total pool, respectively (Figure 3B). Because GABA-[15N1] is a priori synthesized from glutamate-[15N1], the observation that GABA-[15N1] was the least abundant labeled form present is consistent with glutamate-[15N1] being the minor form in the glutamate pool (Figure 3A). Along the same lines, proline-[13C5 15N1] and proline-[13C5] were more abundant than proline-[15N1]; however, like in the previous experiment (Figure 2), labeling was delayed, peaking at 24 h post application with glutamate-[13C5 15N1] (Figure 3C). Thus, in all cases, the double-labeled 13C and 15N isotopes and the single-labeled 13C isotopes of glutamate, GABA, and proline were more abundant than the single-labeled 15N isotopic forms (Figure 3A–C). These data imply that intact amino acids are taken up by turfgrass foliage rather than being mineralized to other transportable forms of nitrogen. The data also indicate that once inside the plant, exogenously applied amino acids are imported into cells where they can be rapidly and directly incorporated into metabolism.

597

**Figure 3.** Time course of percent labeling of single- and double-labeled isotopic forms of glutamate/glutamine (**A**), GABA (**B**) and proline (**C**) from 4 mM glutamate-[13C5 15N1] applied to foliage of creeping bentgrass (*Agrostis stolonifera* L.).

#### **4. Conclusions**

In this exploratory study, we investigated questions about the uptake forms and the incorporation of exogenously applied amino acids on turfgrass foliage. Through time course labeling studies with glutamate-[15N1] and glutamate-[13C5 15N1], we demonstrated that glutamate is rapidly absorbed intact into creeping bentgrass leaves and directly utilized as a precursor to synthesize GABA and proline, two well-studied glutamate-derived metabolites with roles in plant stress adaptation. Our results also provide evidence that the mineralization of glutamate into other nitrogen forms is likely a minor fate of

the amino acids applied to the foliage, though future work measuring the formation and foliar uptake of other nitrogen forms should be performed to independently investigate this question. Furthermore, the labeling in the endogenous pools of glutamate/glutamine remained stable for 72 h, the latest point measured in this study. Taken together, the collective findings of our work suggest that amino acids applied to turfgrass foliage, like those in some specialty turf care products, can be rapidly absorbed and serve as stable sources of precursor molecules to be integrated into the metabolism of the plant.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/3/358/s1, Table S1: Fragments of each labeled and unlabeled amino acid from glutamate-[15N1] as analyzed by GC-MS; Table S2: Fragments of each labeled and unlabeled amino acid from glutamate-[13C5 15N1] as analyzed by GC-MS.

**Author Contributions:** Conceptualization, D.R., T.G.S, G.C.M., and J.R.W.; methodology, D.R., T.G.S, G.C.M., and J.R.W.; formal analysis, R.M.M., G.W.M., D.R., and J.R.W.; investigation, R.M.M., G.W.M., and D.R.; resources, G.C.M and J.R.W.; writing—original draft preparation, R.M.M., G.W.M., D.R., G.C.M., and J.R.W.; writing—review and editing, R.M.M., G.W.M., D.R., G.C.M., T.G.S., and J.R.W.; visualization, R.M.M., G.W.M., D.R., and J.R.W.; supervision, D.R., G.C.M., and J.R.W.; project administration, D.R., G.C.M., T.G.S., and J.R.W.; funding acquisition, D.R., G.C.M., and J.R.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded in part by EnP Investments, LLC (Mendota, IL USA), start-up funds from Purdue University to J.R.W., and by the USDA National Institute of Food and Agriculture Hatch Project number 177845.

**Acknowledgments:** We thank Kyle Ladenburger of EnP Investments, LLC for his assistance in performing amino acid applications and collecting tissue.

**Conflicts of Interest:** G.C.M. is the president of EnP Investments, LLC (manufacturer of the Foliar-Pak brand, Mendota, IL USA) where he is also the chief formulator and inventor. He was involved in the design of the study and collection of the samples, and decision to publish, but was not involved in the analyses or interpretation of data. The remaining authors declare no competing financial or non-financial interests.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

### **The Use of a Plant-Based Biostimulant Improves Plant Performances and Fruit Quality in Tomato Plants Grown at Elevated Temperatures**

**Silvana Francesca 1, Carmen Arena 2, Bruno Hay Mele 3, Carlo Schettini 4, Patrizia Ambrosino 5, Amalia Barone <sup>1</sup> and Maria Manuela Rigano 1,\***


Received: 21 February 2020; Accepted: 3 March 2020; Published: 6 March 2020

**Abstract:** Abiotic stresses can cause a substantial decline in fruit quality due to negative impacts on plant growth, physiology and reproduction. The objective of this study was to verify if the use of a biostimulant based on plant and yeast extracts, rich in amino acids and that contains microelements (boron, zinc and manganese) can ensure good crop yield and quality in tomato plants grown at elevated temperatures (up to 42 ◦C). We investigated physiological responses of four different tomato landraces that were cultivated under plastic tunnel and treated with the biostimulant CycoFlow. The application of the biostimulant stimulated growth (plants up to 48.5% taller) and number of fruits (up to 105.3%). In plants treated with the biostimulant, antioxidants contents were higher compared to non-treated plants, both in leaves and in fruits. In particular, the content of ascorbic acid increased after treatments with CycoFlow. For almost all the traits studied, the effect of the biostimulant depended on the genotype it was applied on. Altogether, the use of the biostimulant on tomato plants led to better plant performances at elevated temperatures, that could be attributed also to a stronger antioxidant defence system, and to a better fruit nutritional quality.

**Keywords:** antioxidants; biostimulant; tomato; fruit quality; abiotic stress

#### **1. Introduction**

Tomato (*Solanum lycopersicum* L.) is one of the most consumed vegetables worldwide also owing to the development of products such as soups, juices, purees, and sauces [1]. Tomato is an essential component of the Mediterranean diet and of other traditional diets. However, heat can negatively affect vegetative and reproductive growth phases in tomato resulting in up to 70% harvest losses [2,3]. Indeed, in tomato, when temperatures exceed 35 ◦C different physiological functions result adversely affected including seed germination, seedling and vegetative growth, flowering and fruit set and ripening [3]. High temperature stress leads also to inhibition of chlorophyll biosynthesis and of photosystem II activity [4]. Indeed, photosynthesis is one of the processes most affected by elevated temperatures [5].

Considering the importance of this crop, the development of new management practices to enhance tolerance to abiotic stresses, including heat stress, could contribute to global food production. The use of biostimulants is proposed as an innovative solution to address the novel challenge to improve the sustainability of agricultural systems and reduce the use of chemical fertilizers [6,7]. The most accepted and complete definition of a biostimulant is the one from Du Jardin that defines

a plant biostimulant as "any substance or microorganism that applied to plants, regardless of its nutrients content, is able to enhance nutrition efficiency and also abiotic stress tolerance and quality traits" [8]. Du Jardin allocated the biostimulants into eight classes: humic substances, complex organic materials, beneficial chemical elements, inorganic salts, seaweed extracts, chitin and chitosan derivatives, anti-transpirant and free amino acids and considered other N-containing substances with microorganism a potential ninth category. The mechanisms activated in plants by the different biostimulants are still not known as they can act directly on plant metabolism and physiology or indirectly on soil conditions [9]. The effects of biostimulants compounds include stimulation of enzyme activities of glycolysis, Krebs cycle, nitrate assimilation, and of hormonal activities [10]. It has been also demonstrated that biostimulants application is able to enhance tolerance to different abiotic stresses, such as drought [11,12], salinity [7,13,14], and thermal stresses [15]. For example, it has been demonstrated that applications of algal extracts are able to promote tolerance to drought, salinity, and heat, while extracts rich in amino acids can help increasing tolerance to thermal stresses [16,17]. Lettuce plants (*Lactuca sativa*) treated with a mixture derived from enzymatic hydrolysis of proteins and subjected to cold showed higher fresh weights and better stomatal conductance compared to non-treated plants [18]. In another work, perennial ryegrass (*Lolium perenne* L.) treated with hydrolyzed amino acids had improved photosynthetic efficiency compared to non-treated plants at high temperatures (36 ◦C) [15]. In general, the application of amino acids was found to exert positive effects on plant growth due to their use for the biosynthesis of a large number of non-protein nitrogenous compounds (pigments, vitamins, coenzymes, purine, and pyrimidine bases). Therefore, amino acids applications could directly influence the physiological activity in plant growth and yield also under abiotic stress [19]. Protein hydrolysates can also improve soil respiration, microbial biomass and activity and impact on plant nutrition by forming complexes and chelates between amino acids and soil nutrients [20].

To improve the tolerance to high temperatures the use of biostimulants has been previously investigated, even if it is presently unclear to what extent these compounds are able to improve the physiological performances of tomato plants under elevated temperatures [7]. We hypothesize that the use of an amino acid-based biostimulant could stimulate natural processes to enhance plant performances also at elevated temperatures. Indeed, the use of protein hydrolysates could directly stimulate carbon and nitrogen metabolism and indirectly enhance nutrient availability, nutrient uptake and nutrient use-efficiency in plants [21]. To verify this hypothesis, we used a novel plant-based biostimulant named CycoFlow and we performed physiological and biochemical analyses on four different tomato landraces grown at elevated temperatures and treated or not with this biostimulant. We reasoned that treatments with CycoFlow could facilitate stress adaptation because of its putative cytokinin-like action and its high concentration of glycine betaine known to mitigate the effect of heat stress [7,22]. Considering climate changes and the expected rise of temperatures in the next few years, to understand the contribution of biostimulants to ensure good plant performances at high temperatures may become increasingly important.

#### **2. Materials and Methods**

#### *2.1. Plant Growth, Experimental Design, and Treatments*

One-month-old tomato seedlings (landraces E17, E36, E107, PDVIT, described in Table 1) were transplanted in May 2018 under walk-in plastic tunnel (22 <sup>×</sup> 8 m2) in Battipaglia in the Campania Region in Southern Italy (40◦57'68"N 14◦95'97"E). The tunnel was covered in polyethylene sheet and was open on both sides. Microclimatic conditions and temperatures were not regulated but were recorded during the growing season. All four genotypes have an indeterminate growth habit. The genotype E17 is characterized by large fruits (200–500 g), the genotype E107 is characterized by medium-sized fruits (70–100 g) and the E36 and the PDVIT genotypes are characterized by small cherry fruits (Table 1). Only the mature fruits of the E107 genotype are yellow while the fruits of the other genotypes are red. Tomato plants were grown following the standard cultural practices of the

area. The experimental design consisted of a completely randomized design with three replicates *per* treatment and ten plant *per* each biological replication. There were two different groups: one control, which did not receive any biostimulant, and one that was treated with the biostimulant. The biostimulant CycoFlow (Agriges, Benevento, Italy) was produced by mixing sugar cane molasses with yeast extract obtained by autolysis of previously grown *Saccharomyces cerevisiae* yeasts. It is rich in high free amino acids, peptides, nucleotides, B-vitamins, trace elements, and other growth factors. Its chemical composition contains total nitrogen of 4.5% and organic carbon of 19.5%. The aminogram of the Biostimulant Cyco Flow is reported in Supplementary Table S1. The product contains also Boron (0.2%), Manganese (1%) and Zinc (1.2%). The biostimulant has a pH of 5.0, a density of 1200 kg/m3 and an EC value of 15.0 dS/m. The Biostimulant, in liquid formulation, was initially applied directly to the soil (400 mL *per* plant) at the moment of transplanting, and thereafter every 15 days, until the end of the cultivation cycle for a total of four total applications. CycoFlow was applied by fertigation at a final concentration of 3 g/L. The control and the treatment groups received the same amount of water. No fertilizer has been applied. During the whole growing period climatic data (Figure S1) were recorded using the weather station VantagePro2 from Davis Instrument Corp. At the end of the cultivation cycle, plants were harvested and separated into leaves, stems, roots and fully ripe fruits. Plant height, numbers of leaves *per* plant, fresh weight of biomass, total number of fruits, weight of fruit and final yield were recorded. Dry biomass (in grams) was determined by drying plant tissues to constant weight in a forced-air-oven at 80 ◦C for 72 hours. Measurements were done on three randomly selected plants *per* each biological replication *per* genotypes for each treatment.


**Table 1.** Details of the tomato genotypes used in this study.

#### *2.2. Pollen Viability*

Pollen viability was analyzed using five flowers *per* plant sampled from three different plants *per* replicate. In the laboratory, pollen grains were spread on microscope slides. One droplet of DAB solution (SIGMA) was added on each pollen sample; slides were gently warmed with a gas lighter and mounted with a cover slip [23]. Scoring was made using an LEITZ Laborlux12 microscope.

#### *2.3. Ascorbic Acid Quantification*

Reduced ascorbic acid (AsA) and total ascorbic acid (AsA + dehydroascorbate − DHA) measurements were carried out by using a colorimetric method [24] with modifications reported by Rigano et al. [25,26]. Briefly, 500 mg of frozen powder from tomato fruits or leaves were extracted with 300 μL of ice cold 6% trichloroacetic acid (TCA) and the mixture was then incubated for 15 min on ice and centrifuged at 14,000 rpm for 20 min. For reduced AsA evaluation, to 20 μL of supernatant were added 20 μL of 0.4 M phosphate buffer (pH 7.4), 10 μL of double distilled (dd) H2O and 80 μL of color reagent solution. This solution was prepared by mixing solution A (31% (*w*/*v*) H3PO4, 4.6% (*w*/*v*) TCA and 0.6% (*w*/*v*) FeCl3) with solution B (4% (*w*/*v*) 2,2 -Dipyridyl). For total AsA, to 20 μL of sample, 20 μL of 5 mM dithiotreitol in 0.4 M phosphate buffer (pH 7.4) were added and the mixture was incubated for 20 min at 37 ◦C. Ten microliters of N-ethyl maleimide (NEM; 0.5% (*w*/*v*) in water) were added and left for 1 min at room temperature. Eighty microliters of color reagent were added as previously described for reduced AsA. Both the final mixtures were incubated at 37 ◦C for 40 min and measured at 525 nm by using a Nano Photometer TM (Implen, Munich, Germany). Three separated biological replicates for each sample and three technical assays for each biological repetition were measured. The concentration was expressed in mg/100 g of fresh weight (FW).

#### *2.4. Total Carotenoids and Chlorophylls Content*

The evaluation of total carotenoids and chlorophylls was carried out according to the method reported by Wellburn [27] and by Zouari et al. [28] as modified by Rigano et al. [2]. To obtain the lipophilic extract, 0.25 grams of sample were extracted with 24 mL of acetone/hexane (40/60, *v*/*v*). The mixture was centrifuged at 15,000 rpm for 5 min at 4 ◦C. Supernatants were collected and stored at −20 ◦C until analyses. For carotenoids and chlorophylls a and b levels determination, absorbance of lipophilic extracts was read at 470, 663, and 645 nm, respectively. For lycopene and β-carotene levels absorbance was read at 505 and 453 nm, respectively. Results were converted into mg/100 g FW. Three separated biological replicates for each sample and three technical assays for each biological repetition were measured.

#### *2.5. Antioxidant Activity Determination*

Hydrophilic antioxidant activity (HAA) was evaluated in the water-soluble fraction, obtained by adding to 2 g of frozen powder 25 mL of 80% methanol, using the ferric reducing/antioxidant power (FRAP) method [29] with slight modifications. The FRAP assay was carried out by adding in a vial 2.5 mL of acetate buffer at pH 3.6, 0.25 mL of TPTZ solution (10 mM) in 40 mM HCl, 0.25 mL of FeCl3·6H2O solution (12 mM), and 150 μL of methanolic extract. The mixture was incubated for 30 min in the dark, and then readings of the colored products (ferrous tripyridyltriazine complex) were taken at 593 nm using a spectrophotometer. Results were expressed as micromoles of Trolox equivalents (TE) per 100 g FW. Lipophilic antioxidant activity (LAA) determination was carried out according to the 2,2 -azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) method, using the lipophilic extract obtained as described in the previous paragraph [30]. The ABTS assay was based on the reduction of the ABTS•+ radical action by the antioxidants present in the sample. A solution constituted by 7.4 mM ABTS•+ (5 mL) mixed with 140 mM K2S2O8 (88 μL) was prepared and stabilized for 12 h. This mixture was then diluted by mixing ABTS•+ solution with ethanol (1:88) to obtain an absorbance of 0.70 ± 0.10 unit at 734 nm using a spectrophotometer. Methanolic extracts (100 μL) were allowed to react with 1 mL of diluted ABTS•+ solution for 2.5 min, and then the absorbance was taken at 734 nm using a spectrophotometer. All biological replicates of samples were analyzed in triplicate. Results were expressed as micromoles of TE per 100 g FW.

#### *2.6. Fluorescence Emission Measurements*

Fluorescence emission measurements were performed on five replicates *per* each treatment, coming from five different plants. A portable FluorPen FP100max fluorometer, equipped with a light sensor (Photon System Instruments, Brno, Czech) was used for measurements, following the procedure reported in Figlioli et al. [31]. The ground fluorescence signal, Fo, was induced on 40 dark adapted leaves, by a blue LED internal light of about 1–2 μmol m−<sup>2</sup> s<sup>−</sup>1. The maximal fluorescence level in the dark, Fm, was induced by a 1s saturating light pulse of 3000 μmol m−<sup>2</sup> s<sup>−</sup>1. The maximum quantum efficiency of PSII photochemistry, Fv/Fm, was calculated as (Fm − Fo)/Fm, according to Kitajima and Butler [32].

#### *2.7. Leaf Functional Traits Determination*

Fully expanded leaves, without apparent damages, were collected to determine the functional leaf traits following Arena et al. [33]. Leaf area (LA) was measured by the program Image J 1.45 (Image Analysis Software) and expressed in *per* square centimeter, specific leaf area (SLA) was measured as the ratio of leaf area to leaf dry mass and expressed as square centimeter *per* gram dry weight (DW). For dry mass determination, leaves were dried at 70 ◦C for 48 h. Leaf dry matter content (LDMC) was measured as the oven-dry mass of a leaf divided by its water-saturated fresh mass and expressed as gram *per* gram of water saturated leaf mass (WSLM). Relative water content in leaves (RWC) was calculated by dividing the amount of water in the fresh leaf tissue by the water in the leaf tissue after rehydration multiplied by 100 [34].

#### *2.8. Statistical Analysis*

Data were subjected to analysis of variance using a two-way ANOVA. To separate means within each parameter, the Tukey-HSD's test was performed. Differences at *p* < 0.05 were considered to be significant. ANOVA was performed by using SPSS (Statistical Package for Social Sciences) Package 6, version 23.0. To explore the overall data, we used the R environment for statistical computing and graphics R Core Team (2018). We first selected variables of interest for each genotype, treatment and plant part (4 × 2 × 2) then calculated the arithmetic mean (n = 3), and finally used the scale function to center the data around the mean and scale it using the standard deviation. The transformed data were visualized using a heatmap (heatmap function). To aid interpretation of the data, we also performed an SVD-based Principal Component Analysis over the multivariate matrix (function prcomp in base R) after normalization.

#### **3. Results**

#### *3.1. Phenotypic and Physiological Analyses*

In this study four different tomato genotypes were transplanted under a plastic walk-in tunnel with a delay of one month compared to the usual transplanting period (tomato plants in the South of Italy are usually transplanted in April), thus imposing a high-temperature condition during flowering and fruit setting. Indeed, the maximum temperature of 32 ◦C during the day, which represents a critical threshold in the sensitive stages of reproductive development, was frequently exceeded in this trial [3] (Figure S1). The four different tomato landraces were treated with a plant-based biostimulant named CycoFlow. According to ANOVA analyses, the treatment with the biostimulant increased the height of genotypes E107 and PDVIT by 48.5% and 30.1%, respectively (Supplementary Table S2). Generally, the number of leaves was lower in the biostimulant treated group compared to the control, independently of the genotype it was applied to (no significant interaction G X T). For the fresh biomass parameter, in PDVIT the treatment with CycoFlow increased the above ground fresh biomass by 68.4% (Figure 1a). Genotypes E17 and E36 showed, instead, lower values in treated plants compared to non-treated ones (−53.8% and −21.1%, respectively). A slightly higher pollen viability was also observed in the genotypes treated with the biostimulant compared to the respective controls (Figure 1b). In particular, in the genotype E107 the treatment with the biostimulant increased pollen viability by 125%. Generally, the treatment with the biostimulant increased the number of fruits, independently of the genotype (no significant interaction G X T). In particular, the treatment with the biostimulant increased the number of fruits in the genotype PDVIT by 105.3% (Figure 1c). The medium fruit weight was significantly affected only by the factor genotype (Supplementary Table S2). Generally, the final yield (kg *per* plant) showed a tendency to be higher in all the samples from the treated genotypes, even though these differences were not significant (Figure 1d). Interestingly, the yield was significantly affected only by the factor treatment (Supplementary Table S2).

**Figure 1.** Effect of CycoFlow on: (**a**) Fresh weight (FW) biomass, (**b**) pollen viability, (**c**) fruit number, and (**d**) final yield in four tomato genotypes. Values are mean ± SE. Different letters indicate significant differences based on Tukey-HSD test (*p* ≤ 0.05).

The treatment with the biostimulant CycoFlow also increased the maximal PSII photochemical efficiency (Fv/Fm) in the E107 and PDVIT genotypes (Figure 2). The monitoring of leaf functional traits evidenced that biostimulant application did not affect these traits significantly (Supplementary Table S3).

**Figure 2.** Maximal photochemical efficiency (Fv/Fm) in leaves of four tomato genotypes. Data are mean ± SE (n = 5). Different letters indicate significant differences based on Tukey-HSD test (*p* ≤ 0.05).

#### *3.2. Leaf and Fruit Antioxidant Content*

The main interaction effects of the biostimulant Cyco Flow on the content of antioxidants in leaves from treated and non-treated plants is reported in Table 2.

**Table 2.** Analyses of variance and mean comparison for reduced and total ascorbic acid (AsA), total phenols, carotenoids, chlorophylls a and b and total lipophilic and hydrophilic antioxidant activities (LAA and HAA, respectively) in leaves of different tomato cultivars treated with the biostimulant CycoFlow applied by fertirrigation four times. Means ± SD within rows and columns followed by the different letter are significantly different based on Tukey-HSD test (*p* ≤ 0.05).


G = genotype; T = treatment; \* = p ≤ 0.05; \*\* = *p* ≤ 0.01; \*\*\* = *p* ≤ 0.001.

For the hydrophilic antioxidants, the treatment with the biostimulant increased the content of reduced AsA in the genotypes E17 and E36 and of total AsA in the leaves of the genotypes E36 and PDVIT. In particular, in the genotype E107 a 60.8% higher content of total AsA was registered in leaves treated with the biostimulant. As for the content of phenolic compounds, two genotypes (E17 and E107) showed lower contents of total phenols in the leaf after treatment with the biostimulant. In particular, in the E17 genotype a 41.6% decrease in the treated compared to the non- treated samples was demonstrated. Only in the PDVIT genotype the treatment with the biostimulant increased phenols content. It has been reported that phenolics compounds are the most important contributors to HAA [35]. Accordingly, in the leaves of the treated plants, HAA was lower in E17 compared to the respective non-treated control. For the lipophilic antioxidants, the treatment with the biostimulant increased the content of carotenoids in the genotypes E36 and E107 and the content of chlorophylls a and b only in the genotype E36. Particularly, the E36 genotype showed a 15.8% higher content of carotenoids in the treated leaves compared to the non-treated one, and 17.35% and 48% higher levels of chlorophyll a and b, respectively. The treatment with the biostimulant also increased total lipophilic antioxidant activities in E107 and surprisingly also in PDVIT, suggesting that other compounds outside of carotenoids contributed to this parameter.

In Table 3 is reported the content of hydrophilic antioxidants determined in red ripe fruit from genotypes treated or non-treated with the biostimulant CycoFlow. In general, the content of hydrophilic antioxidants in the fruits was higher in almost all the genotypes treated with biostimulants compared to the non-treated ones. The treatment with the biostimulant increased the content of reduced AsA independently of the genotype it was applied on (not significant interaction G X T). The content of reduced AsA was 28.7%–58.7% higher in fruits from treated genotypes compared to non-treated genotypes. Moreover, a content 112.8% higher of total AsA was registered in fruits from PDVIT treated with the biostimulant compared to the respective non-treated control. Contrary to what seen in the leaf, the content of total phenols in berries of treated E17 and E36 genotypes was higher compared to the non-treated control. In particular, in the E17 genotype 72.8% higher values were registered. Moreover, a significantly higher antioxidant activity HAA was demonstrated in fruits from E36 plants treated with CycoFlow, according to ANOVA analyses. Assessing the content of lipophilic antioxidants, the treatment with the biostimulant had no effects on the content of carotenoids and chlorophylls but only on the total lipophilic antioxidant activity, as reported in Supplementary Table S4. In particular, LAA was higher in fruits from the treated genotypes E17, E36, and E107.

**Table 3.** Analyses of variance and mean comparison for reduced and total ascorbic acid (AsA), total phenols, hydrophilic antioxidant activities (HAA) in fruits of different tomato cultivars treated with the biostimulant CycoFlow applied by fertirrigation four times. Means ± SD within rows and columns followed by the different letter are significantly different based on Tukey-HSD test (*p* ≤ 0.05).


G = genotype; T = treatment; \* = *p* ≤ 0.05; \*\* = *p* ≤ 0.01; \*\*\* = *p* ≤ 0.001; ns = not significant.

#### *3.3. Heat Map Analysis*

A heat map providing the morphological, biochemical, and physiological changes in leaves and fruits of four different tomato genotypes in response to the addition of one biostimulant is displayed in Figure 3. With regard to leaves, the heat-map identified two main clusters which divided the analyzed samples differently (Figure 3, panel a). The first cluster separated the control genotypes E107 and E17 from the other genotypes and respective treated samples, the second cluster associated the treated genotypes E107, E17, and PDVIT in a sub-group and control PDVIT and E36 genotypes in another sub-group (Figure 3). Our data indicate that biostimulant application was the main clustering factor for E107, E17, and PDVIT genotypes, on the basis of differences in some leaf traits, Fv/Fm, phenols, yield and HAA, suggesting that the biostimulant utilization produces significant effect on many metabolites. The heat map built on tomato fruits clearly separated the treated PDVIT genotype from all others, in particular for number of fruits and reduced AsA (Figure 3, panel b), indicating this genotype as the most responsive to biostimulant application for fruit characteristics. A remarkable separation was also evident for control E107 and E36 compared to treated genotypes, grouped in two sub-clusters on the basis of pigments (chlorophylls and carotenoids) and LAA. A PCA analyses was also performed (Supplementary Figure S2). The PCA output further showed an evident separation between the treated and the non- treated genotypes.

**Figure 3.** Cluster heat map analysis summarizing the behavior of the different tomato genotypes E36, E17, E107, PDVIT treated or non-treated with the biostimulant CycoFlow in leaf (panel **a**) and in fruit (panel **b**). The heat map was generated using the R environment for statistical computing and graphics (https://www.R-project.org/online) program package with Euclidean distance as the similarity measure and hierarchical clustering with complete linkage.

#### **4. Discussion**

In this paper four different tomato landraces were grown at elevated temperatures under a plastic walk-in tunnel and were treated or not with a plant-based biostimulant named CycoFlow. The higher height demonstrated in the majority of the tomato plants treated with CycoFlow compared to non-treated plants is in agreement with previous studies on different plant species and biostimulants [36–40]. Probably, the presence of signaling molecules in the biostimulant, such as free amino acids, promoted endogenous phytohormonal biosynthesis thus stimulating growth and also fruit setting [41]. Indeed, several authors demonstrated that the application of plant-based biostimulants exhibited cytokinin-like activity promoting cell division [42]. Moreover, cytokinins mitigate stresses induced by free radicals by direct scavenging and also by preventing ROS formation inhibiting xanthine oxidation [39]. Also, the treatment with CycoFlow overall increased the number of fruits, as previously demonstrated also in tomatoes treated with other biostimulants [10,36,39–41]. For example, Rouphael et al. [41] demonstrated that application of a protein hydrolysate in tomato increased in one cultivar the fruit mean weight and in another cultivar the number of fruits. In this study, in the genotype E107, the higher number of fruits observed was also linked to a higher pollen vitality observed after CycoFlow treatment. This result could be due to a combination of multiple effects. While the cytokinin-like activity could have favored cell division, the high level of proline present in the biostimulant, an amino acid whose natural content in the flower organs is ten times higher than that in the leaves, may have played an important role [31]. Indeed, it is known that also the amino acid proline promotes the translocation of nutrients towards developing flowers (sink) [43]. The positive effects of biostimulants based on amino acid on growth and yield is also due to the fact that the amino acids present in plant-based biostimulants stimulate plant defenses, participate in the synthesis of organic compounds (such as amines, purines, pyrimidines, vitamins) and affect the uptake of macro and micronutrients [37]. The CycoFlow effects observed in this

study on yield and yield components are even more remarkable considering the elevated temperatures (up to 43 ◦C) reached under the plastic walk-in tunnel in Battipaglia. Indeed, this temperature normally impairs fertilization and reduces pollen viability [10]. It can be hypothesized that the presence of glycine betaine in the CycoFlow may have enhanced the tolerance of tomato plants to elevated temperatures. Indeed, it has been previously demonstrated that during tomato germination glycine betaine applied exogenously improved tolerance to high temperatures and enhanced the expression of heat shock genes [44]. At elevated temperatures, the glycine betaine compound may have also a crucial role in the repair of photodamaged PSII, in maintaining the activity of Rubisco and in alleviating the inhibition of gas exchanges [22]. Accordingly, a higher maximal photochemical efficiency was observed in the genotypes E107 and PDVIT treated with the biostimulant. These results are consistent with other papers, which demonstrated that applications of plant- and animal-based biostimulants are able to enhance photosynthetic rates and ensure a higher carbon assimilation efficiency [45,46]. For example, under drought stress conditions, Arabidopsis plants treated with an *Ascophyllum nodosum*-extract maintained a better photosynthetic performance compared to non-treated plants during the dehydration period, showing a higher capacity to dissipate thermally the excess of energy in the PSII reaction centers [47]. These results were linked to the fact that pre-treatments with the Ascophyllum-extracts induced partial stomatal closures and also modifications of the expression levels of genes involved in ABA-responsive and antioxidant system pathways [47]. Accordingly, our data indicate that biostimulant treatment induced the activation of the antioxidant defense system, as demonstrated by the higher content of reduced and total AsA in treated leaves. Although the precise reasons for these increases are not explained, it is known that biostimulants components, including glycine betaine, can promote the activity of specific enzymes involved in antioxidant homeostasis [22,41,48]. The ability to maintain an optimal chlorophyll content during heat stress is another key heat tolerance trait in tomato [49]. Interestingly, herein we observed higher contents of carotenoids and chlorophylls in two genotypes (E36 and E107) treated with the biostimulant compared to the non-treated samples. The higher chlorophylls content detected in these genotypes could be linked to limited chlorophyll degradation and leaf senescence [9]. In particular, this could be the case for the genotype E107 that demonstrated a higher maximal photochemical efficiency after treatment with the biostimulant.

The biostimulant-mediated effects on photosynthesis and secondary metabolism could also enhance fruit quality [10]. Indeed, one interesting finding of this study is the positive effect of the biostimulant CycoFlow on the quality of the tomato fruits. In general, the content of hydrophilic antioxidants in the fruits, including AsA, was higher in almost all the genotypes treated with biostimulants compared to the non-treated ones. Higher content of reduced AsA was observed in all the genotypes and of total AsA in the genotypes E17 and PDVIT. This result confirms data previously obtained in other studies that demonstrated an increase in AsA content in tomato, in kiwi fruits and in peppers after the application of plant-based biostimulants [36,41]. Contrary to what seen in the leaf, the content of total phenols in berries of treated E17 and E36 genotypes was higher compared to the non-treated control. Moreover, a significantly higher antioxidant activity HAA was demonstrated in fruits from E36 plants treated with CycoFlow. These results are in agreement with results previously obtained in other crops (soybean seeds, common bean, tomato, corn), even if the reported effects depended on the type of biostimulants, their concentrations and the number of applications [37]. Assessing the content of lipophilic antioxidants, the treatment with the biostimulant had no effects on the content of carotenoids and chlorophylls but only on the total lipophilic antioxidant activity. Similar results were obtained by Chehade et al. [36] in tomato. On the contrary, Rouphael et al. [41] demonstrated that in tomato foliar applications of a legume-derived protein hydrolysate had an effect also on lycopene content. Also, Colla et al. [10] demonstrated that foliar applications of protein hydrolysate, plant and seaweed extract affected lycopene content in greenhouse tomato. In the future, foliar application of CycoFlow will be also tested in order to verify if the results obtained in this study are also linked to the used application regimen.

Altogether, the genotypic factors remain decisive in the response obtained in the different tomato lines to the biostimulant. Indeed, for almost all the traits considered the effect of the biostimulant depended on the cultivar it was applied to, as seen by the interaction between the effect of the biostimulant and cultivars in most of the studied parameters. These variations can be explained by the differences in the genetic background between the different cultivars that were used in this study [33]. Indeed, the four genotypes here tested differed in terms of fruit shape and size and also in terms of fruit color (e.g., fruit of E107 is yellow). The geographical origin is also different with the E107 genotype coming from Spain and the other coming from Italy. These further highlight the fact that one biostimulant should be tested on a certain number of cultivars in order to assess its mechanisms of action.

#### **5. Conclusions**

In this paper we investigated the effects of the application of one plant-based biostimulant named CycoFlow on the nutritional quality and yield of tomatoes grown in walk-in tunnel under elevated temperatures. The application of the CycoFlow biostimulant had a clear effect on plant growth and final crop quality. Indeed, CycoFlow application had a significant effect on the content of hydrophilic antioxidants in both tomato leaves and fruits. In particular, the content of AsA increased after treatments with CycoFlow. Herein, the biostimulant application improved plant performances and fruit quality mostly in the genotypes E107 and PDVIT. In particular, in the genotype PDVIT application with CycoFlow determined a higher plant height, a higher number of fruits, a higher pollen vitality, a higher photochemical efficiency, a higher accumulation of AsA and a higher antioxidant activity. Additional studies are now planned in order to investigate if different applications regimen, such as foliar application, can also influence the observed effects.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/3/363/s1, Table S1: Amino acid composition expressed in g/100 g of the biostimulant CycoFlow, Table S2: Analyses of variance and mean comparison for height, number of leaves, fresh weight (FW) and dry weight (DW) biomass, number of fruits, medium fruit weight, yields, and pollen viability (%) per plants of different tomato cultivars treated with the biostimulant CycoFlow applied by fertirrigation four times. Means ± SD within rows and columns followed by the different letter are significantly different based on Tukey-HSD test (*p* ≤ 0.05). Table S3: Analyses of variance and mean comparison for maximal PSII photochemical efficiency (Fv/Fm), leaf area (LA), specific leaf area (SLA), leaf dry matter content (LDMC) and relative water content (RWC) *per* plants of different tomato cultivars treated with the biostimulant CycoFlow applied by fertirrigation four times. Means ± SD within rows and columns followed by the different letter are significantly different based on based on Tukey-HSD test (*p* ≤ 0.05). Table S4: Analyses of variance and mean comparison for total lipophilic antioxidant activities (LAA), carotenoids, chlorophylls a and b (Chl A and Chl B, respectively) content in fruit of different tomato cultivars treated with the biostimulant CycoFlow applied by fertirrigation four times. Means ± SD within rows and columns followed by the different letter are significantly different based on Tukey-HSD test (*p* ≤ 0.05). Figure S1: Maximum temperatures recorded in the experimental field located in Battipaglia during the day from May to August 2018. Figure S2: Principal component analysis (PCA) of phenotypic and physiological traits in tomato plants treated or not with the biostimulant CycoFlow. The treated genotypes are indicated by the letter T after the name.

**Author Contributions:** Conceptualization: S.F., A.B. and M.M.R.; Data Curation: S.F., M.M.R., C.A., B.H.M.; Funding acquisition: A.B.; Writing—original draft: S.F., M.M.R., C.A., A.B.; Writing—review and editing: S.F., M.M.R., C.A., A.B., P.A., C.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors have received funding from the European Union's Horizon 2020 research and innovation program through the TomGEM project under grant agreement No 679796. The APC was also funded by the TomGEM project under grant agreement No. 679796.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

1. Del Giudice, R.; Petruk, G.; Raiola, A.; Barone, A.; Monti, D.M.; Rigano, M.M. Carotenoids in Fresh and Processed Tomato (*Solanum lycopersicum*) Fruits Protect Cells from Oxidative Stress Injury. *J. Sci. Food Agric.* **2016**, *97*, 1616–1623. [CrossRef] [PubMed]


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

## **Changes in Assimilation Area and Chlorophyll Content of Very Early Potato (***Solanum tuberosum* **L.) Cultivars as Influenced by Biostimulants**

#### **Wanda Wadas 1,\* and Tomasz Dziugieł <sup>2</sup>**


Received: 7 February 2020; Accepted: 10 March 2020; Published: 12 March 2020

**Abstract:** This paper analyses the effects of foliar application of the seaweed extracts Bio-algeen S90 (*Ascophyllum nodosum*) and Kelpak SL (*Ecklonia maxima*), as well as the humic and fulvic acids ini HumiPlant (leonardite extract), on the assimilation area and chlorophyll content of very early potato cultivars ('Denar', 'Lord', Miłek'). The field experiment was carried out in central-eastern Poland over three growing seasons, using Luvisol. The biostimulants were applied according to the manufacturers' recommendations. The use of biostimulants resulted in enlargement of the assimilation area, but had no effect on the specific leaf area (SLA) or chlorophyll content (Soil Plant Analysis Development (SPAD) value). The assimilation area was larger, on average, by 0.0505 m2 and leaf area index (LAI) was higher by 0.30 compared with the plants from the control group without a biostimulant. The SLA and SPAD depend on the cultivar and weather conditions, or nitrogen and magnesium content in soil, to a greater extent. The biostimulants enhanced abiotic stress tolerance and increased marketable tuber yield (diameter above 30 mm) 75 days after planting (the end of June), on average by 2.15 <sup>t</sup>·ha<sup>−</sup>1. Bio-algeen S90 and Keplak SL produced better results in a warm and very wet growing season, whereas HumiPlant produced better results in a year with lower air temperature and with drought periods during potato growth. No correlations were found between the tuber yield and assimilation area or between the tuber yield and SPAD value, although a significant negative correlation was found between the tuber yield and SLA.

**Keywords:** seaweed extract; humic acids; leaf area index (LAI); specific leaf area (SLA); Soil Plant Analysis Development (SPAD) index; tuber yield

#### **1. Introduction**

In recent years, the growth and productivity of crop plants have been greatly influenced by abiotic stresses. Periods of high temperature and drought are becoming more frequent in regions with extensively crop production, such as Central Europe, South-Central Asia, south-eastern South America and the south-eastern United States [1]. Under climate change conditions, biostimulants play an important role in sustainable crop production. These natural products (seaweed extracts, humic substances, hydrolysed proteins, and amino acids containing products or microorganism) contain a bioactive substance which enhances nutrition efficiency, abiotic stress tolerance, and/or crop quality traits, regardless of its nutrients content [2–5]. In recent years, the use of seaweed extracts and humic substances as plant growth stimulants has been increasing. Seaweed extracts and humic acids can promote plant growth, enhance abiotic stress tolerance as well as increase nutrient use efficiency [6–10].

Many plant growth-stimulating compounds (auxins, cytokinins, gibberellins, betaines, polysaccharides, polyamines, abscisic acids, brassinosteroids, and minerals) have been identified from

seaweed. The chemical composition of seaweed extracts depends on the algae species and on the method of extraction. Brown algae (*Phaeophyta*) are most commonly used for the manufacture of extracts used as biostimulants of plant growth, including *Ascophyllum nodosum* and *Ecklonia maxima* [7,8,11]. An increase in leaf area and chlorophyll content are common plant responses to seaweed extract treatment. Cytokinins present in the seaweed extracts stimulate cell division, resulting in enlarged leaf area, and also stimulate chlorophyll biosynthesis, whereas betaines slow chlorophyll degradation and delay leaf senescence [7,8]. *Ascophyllum nodosum* extracts applied on foliage or to soil caused an increase in the leaf chlorophyll content of French bean, tomato, barley, maize, wheat, pepper, and strawberry [8,11,12]. A one-year study carried out in Iraq showed an increase in chlorophyll content in potato following the application of brown seaweed *Sargassum* extracts [13]. Foliar application of seaweed extracts *Ascophyllum nodosum* and *Ecklonia maxima* increased potato yield [14–16]. Biostimulants based on seaweed extracts improved plant growth and yield of wheat, barley, maize, potato, tomato, pepper, onion, and carrot [7,8,10,11].

The biological activity of humic substances depends on their source, chemical structure, and concentration. Humic substances may influence both respiration and photosynthesis. One of the effects of humic substances applied to growing plants was an increase in chlorophyll content, which can affect photosynthesis [17]. Leonardite is the most common commercial source of humic substances. Leonardite humic acids stimulate melon and soybean growth and chlorophyll synthesis [6]. A one-year study carried out in Iraq showed an increase in chlorophyll content in potato following the application of humic and fulvic acids in HumiMax [13]. A one-year study carried out in Egypt showed that the application of humic acid under water stress conditions enhanced the leaf chlorophyll content of very early potato cultivars [18]. Application of humic substances originating from leonardite increased potato yield and nutrient uptake [19]. In most experiments, foliar or soil application of humic and fulvic acids increased potato yield [13,20,21], but one study showed no clear effect of humic and fulvic acids on the potato yield [22]. Humic and fulvic acids improved plant growth and yield quality of wheat, maize, tomato, pepper and cucumber [2,22–24]. The effect of humic acids on plant growth depends of their source and concentration, and on the date and method (foliar or soil) of application, as well as the plant species and environmental conditions [9,17].

There is a relationship between leaf chlorophyll content and Soil Plant Analysis Development (SPAD) index [25]. Leaf SPAD values is related to nutrient plant status, especially nitrogen [26,27]. There was a relationship found between SPAD value and potato yield. A higher SPAD does not always guarantee a higher potato yield [28–31]. Plant-based biostimulants increased SPAD index and marketable yield of tomato and rocket [32–34].

To date, few studies have been focused on the effect of seaweed extract and humic acid application in early crop potato culture. The aim of the study was to determine the effect of foliar application of brown seaweed extracts and humic acids on the asssimilation area and chlorophyll content of very early potato cultivars. In the current study, it was hypothesised that seaweed extracts and humic acids could contribute to increasing assimilation area and chlorophyll content and, as a result, increase the early crop potato yield. The assumption was also made that the response to the application of these biostimulants depends on the cultivar and environmental conditions.

#### **2. Materials and Methods**

#### *2.1. Experimental Site and Season*

The study was carried out in central-eastern Poland (52◦03 N, 22◦33 E), over three growing season 2012–2014, on Luvisol with a low total nitrogen content, a high content of available phosphorus, a medium-to-high content of potassium and a low-to-medium content of magnesium, with an acidic-to-slightly-acid reaction. Spring triticale was grown as a potato forecrop. Farmyard manure was applied in autumn, at rate of 25 t·ha−1, and mineral fertilizers were applied at rates of 80 kg N

(ammonium nitrate), 35 kg P (superphosphate) and 100 kg K (potassium sulphate) per hectare in spring. Potato cultivation was carried out according to common agronomical practice.

The thermal and moisture conditions during the potato growth period were different (Table 1). The mean air temperatures were above or similar to the long-term average. In 2012, total precipitation was similar and, in 2013 and 2014, above the long-term average, although it was unevenly distributed during the potato growth period. The most favourable hydrothermal conditions for early crop potato culture were in the warm and moderately wet growing season of 2012. The next year, 2013 was warm and with heavy rainfall, whereas 2014 was cool with heavy rainfall after plant emergence and a drought in the period of tuber growth.


**Table 1.** Hydrothermal conditions during potato growing period.

Hydrothermal index value: up to 0.4 extremely dry; 0.41–0.7 very dry; 0.71–1.0 dry; 1.01–1.3 rather dry; 1.31–1.6 optimal; 1.61–2 rather humid; 2.01–2.5 humid; 2.51–3 very humid; >3 extremely humid [35].

#### *2.2. Plant Material and Experimental Design*

The field experiment was established in a split-plot design with three replications. The experimental factors were: (1) plant biostimulant; and (2) cultivar. The potato plants were treated with three biostimulants: Bio-algeen S90 and Keplak SL containing seaweed extracts, and HumiPlant based on humic and fulvic acids. Bio-algeen S90 is an extract from *Ascophyllum nodosum* which contains amino acids, vitamins, alginic acids and other active components of seaweeds, as well as macronutrients (N, P, K, Ca, Mg) and micronutrients (B, Fe, Cu, Mn, Zn, Se, Co). Kelpak SL is an extract from *Ecklonia maxima* containing auxin (11 mg·dm–3) and cytokinin (0.031 mg·dm–3). HumiPlant is an extract from leonardite which contains humic acid (12%) and fulvic acid (6%) as well as macronutrients (K, Ca, Mg, S) and micronutrients (Fe, Mn, B, Mo, Zn, Cu). The biostimulants were applied according to the manufacturers' recommendations: Bio-algeen S90–2 dm3·ha−<sup>1</sup> at the beginning of leaf development stage (BBCH 10–11) and 2 dm3·ha−<sup>1</sup> two weeks after the first treatment, Kelpak SL–2 dm3·ha−<sup>1</sup> at the leaf development stage (BBCH 14–16) and 2 dm3·ha−<sup>1</sup> two weeks after the first treatment, HumiPlant–2 dm3·ha−<sup>1</sup> at the leaf development stage (BBCH 14–16) and 2 dm3·ha−<sup>1</sup> one week after the first treatment. Potato plants sprayed with water were used as a control without a biostimulant.

The most popular very early potato cultivars (Denar, Lord and Miłek) in the research area were grown. In successive years, 6-weeks pre-sprouted seed potatoes were planted on April 12, April 18 and April 7 with a row spacing of 0.25 m and 0.675 m between rows. The plots were six rows wide and 4 m long (96 plants per plot). Potatoes were harvested 75 days after planting (the end of June).

#### *2.3. Determination of Assimilation Area, Chlorophyll Content and Tuber Yield*

At the tuber formation stage (BBCH 41–43), the assimilation area, leaf area index (LAI), specific leaf area (SLA), and chlorophyll content (SPAD value) were determined. The measurements were made on four successive randomized plants per plot. The assimilation area was measured by the weight method [36]. SLA was calculated as the ratio of assimilation area/weight of leaves [37].

The chlorophyll content was estimated with non-destructive methods using a portable SPAD-502 chlorophyll meter (Minolta, Osaka, Japan). The measurements were made on the youngest fully expanded leaf, i.e., the fourth or fifth leaf from the top.

The total and marketable tuber yield were determined. The marketable tuber yield constituted tubers with a transverse diameter above 30 mm, excluding cracked and deformed tubers. The marketable tuber yield was determined on the basis of the total tuber yield of ten successive plants per plot using a hand calibrator with a square hole.

#### *2.4. Statistical Analysis*

The results of the study were analysed statistically with an analysis of variance (ANOVA) for the split-pot design. The significance of differences between the compared averages was verified using Tukey's test at the significance level *p* ≤ 0.05.

#### **3. Results**

#### *3.1. Assimilation Area*

The effect of biostimulants on the assimilation area depended on the weather conditions during potato growth (Table 2). In the year with the highest air temperature and heavy rainfall after plant emergence (2013), the greatest enlargement of the assimilation area was caused by Kelpak SL, whereas in the year with the lowest air temperature and heavy rainfall after plant emergence (2014), the greatest enlargement of assimilation area was caused by Bio-algeen S90. The assimilation areas were larger, on average, by 0.0624 m2 (11.5%) and 0.0941 m<sup>2</sup> (10%) respectively, and the leaf area index (LAI) was higher by 0.37 and 0.56 compared with the plants from the control group without a biostimulant. Regardless of the biostimulant applied, the assimilation area was largest in the year with the highest air temperature and moderate rainfall at the end of May (Table 3).



Means within columns followed by the same letters do not differ significantly at *p* ≤ 0.05.

The potato cultivars tested showed different responses to the biostimulants applied (Table 2). The type of biostimulant had a greatest effect on the assimilation area of the 'Lord' cultivar. The greatest enlargement of the assimilation area of 'Lord' was caused by Bio-algen S90. Following the application of this biostimulant, the assimilation area of 'Lord' was larger, on average, by 0.1583 m<sup>2</sup> (24.5%) and the LAI value was higher by 0.94 compared with the plants from the control without biostimulant. The differences were highest in the year with a low air temperature and heavy rainfall after the plant emergence (2014). Despite the biostimulant applied, the assimilation area was higher for 'Miłek' than for 'Denar' and 'Lord' (Table 3).



Means within columns followed by the same letters do not differ significantly at *p* ≤ 0.05.

Only in a warm and moderately wet growing season (2012), following application of Bio-algeen S90 and HumiPlant, was the specific leaf area (SLA) higher, on average, by 0.29 m2·kg−<sup>1</sup> compared with the plants from the control group without biostimulant (Table 2). With the use of Kelpak SL, the difference was smaller and not statistically confirmed. The SLA depended to a greater extent on the weather conditions during potato growth. Irrespective of the treatment (with or without biostimulant), the SLA was highest in the year with the highest air temperature and heavy rainfall after plant emergence (Table 3). The type of biostimulant and cultivar interaction effect on SLA was not statistically confirmed (Table 2). Regardless of the treatment, the SLA values of the potato tested cultivars were similar (Table 3).

#### *3.2. Chlorophyll Content (SPAD Value)*

The biostimulants used in the experiment had no significant effect on the chlorophyll content in leaves (Figure 1). The SPAD value depended to a greater extent on the cultivar and weather or soil conditions during potato growth. Irrespective of the treatment (with or without biostimulant), the SPAD values were higher for 'Denar' and 'Lord' than 'Miłek'. The SPAD was highest in the warm and wet growing season (2013) and, at the same time, the highest content of total nitrogen and available magnesium in soil (Figure 2).

**Figure 1.** Effect of plant biostimulants on chlorophyll content (Soil Plant Analysis Development (SPAD) value); average of the three year tests on three cultivars. Means followed by the same letters do not differ significantly at *p* ≤ 0.05.

**Figure 2.** Chlorophyll content (SPAD value) in relation to potato growing season (**a**) and cultivar (**b**). Means followed by the same letters do not differ significantly at *p* ≤ 0.05.

#### *3.3. Relationship between Tuber Yield, Assimilation Area and Chlorophyll Content (SPAD Value)*

The biostimulants used in the experiment had no effect on the weight of leaves [38], but caused enlargement of the assimilation area (Table 4). Over the three years of the study, the assimilation area was larger, on average, by 0.0505 m2 (7%) and the LAI was higher by 0.30 compared with the plants from the control group without a biostimulant. The biostimulants had no significant effect on the SLA and SPAD (Figure 1).

**Table 4.** Effect of plant biostimulants on assimilation area; average of the three year tests on three cultivars.


Means within columns followed by the same letters do not differ significantly at *p* ≤ 0.05.

The biostimulants used in the experiment had a significant effect on the tuber yield [38]. The yield-increasing effects of biostimulants were comparable (Table 5). In the three years of the study, the total tuber yield was higher, on average, by 2.64 t·ha−<sup>1</sup> (7.7%) and marketable tuber yield (diameter above 30 mm) by 2.15 t·ha−<sup>1</sup> (6.5%). The yield-increasing effect of biostimulants depended on weather conditions during the potato growing season. Bio-algeen S90 and Kelpak SL caused the highest increase in tuber yield in the warm and very wet growing season (2013), and HumiPlant in the year with a low air temperature and a drought in the period of tuber growth (2014).

The tuber yield was not significantly correlated with the weight and assimilation leaf area or LAI (Table 6). A significant negative correlation was found between the marketable tuber yield and SLA. No significant correlation was found between the marketable tuber yield and SPAD value.



Means within columns followed by the same letters do not differ significantly at *p* ≤ 0.05.

**Table 6.** Correlation coefficient between tuber yield and assimilation area and SPAD.


\* significant at *p* ≤ 0.05.

#### *3.4. E*ff*ect of Experimental Factors on Assimilation Area, Chlorophyll Content and Tuber Yield*

The effect of the experimental factors and their interactions on potato assimilation area and chlorophyll content (SPAD value) are presented in Table 7.

**Table 7.** Effect of experimental factors on assimilation area, chlorophyll content (SPAD value) and tuber yield.


\* significant at *p* ≤ 0.05, \*\* significant at *p* ≤ 0.01, ns—non-significant.

#### **4. Discussion**

In sustainable crop production, biostimulants play an important role in improving plant growth and crop quality. Assimilation area and chlorophyll content are important parameters of assessment plant growth. The biostimulants used in the experiment caused enlargement of assimilation area, but had no effect on the chlorophyll content (SPAD value) in leaves of very early potato cultivars. SPAD value depended on the cultivar and weather or soil conditions to a greater extent. The effect of

foliar application of seaweed extracts on potato assimilation area was comparable to humic and fulvic acids. In the three years of the study, following biostimulant application, the average leaf area index (LAI) was 4.64, being higher by 0.30 compared to the average for the untreated control group. Potato cultivars showed different responses to the applied biostimulants. Studies have shown the highest light absorption efficiency values at the LAI value of 3, which corresponded to maximum ground cover. If potato LAI exceeds 3, the intercepted photosynthetically active radiation value changes very little [39,40]. According to Howlader and Hoque [41], irrespective of potato cultivars, LAI increased progressively over time, reaching a peak at 60 days after planting and thereafter declining. The rate of assimilation area expansion showed the interaction between genotype and environment and varied by year [42], which was confirmed in the present study. The effect of seaweed extracts on potato assimilation area depended on the weather conditions after plant emergence. In the year with the highest air temperature and heavy rainfall after plant emergence, the assimilation area was larger after the application of Kelpak SL (*Ecklonia maxima*), whereas in the year with the lowest air temperature and with heavy rainfall after plant emergence, the assimilation area was larger after the application of Bio-algeen S90 (*Ascophyllum nodosum*). Potato plants are very sensitive to heat stress. In general, heat stress increases plant height, reduces leaf size, increases leaf chlorophyll content, and severely reduces tuber mass [43]. Kelpak SL contains auxins and cytokinins in a ratio of 350/1. Exogenous auxin plays an important role in plant stress resistance. The action of auxin depends on its concentration, the light conditions and carbohydrate content in the plant [44]. Exogenous cytokinins also play an important role in plant adaptation to environmental stresses [45]. Cytokinins present in the seaweed extracts stimulate cell division, resulting in enlarged leaf area [7,8], which was confirmed in the present study.

The leaf area index describes the growth of lowland fields, whereas the growth of individual plants is characterized by the specific leaf area (SLA). Biostimulants caused enlargement of the assimilation area, but had no effect on the SLA. The SLA for potato depends on the cultivar and growth stage, and temperature [42], which was confirmed in the present study. Early foliar expansion of potato is associated with a strong increase in SLA [41].

Foliar or soil application of *Ascophyllum nodosum* extracts caused an increase in the chlorophyll content of some agriculture (barley, wheat, maize) and horticulture (French bean, tomato, pepper, strawberry) plants [8,11,12], which was not confirmed in the present study. A study carried out in Egypt showed that the application of humic acid under water stress conditions enhanced the chlorophyll content of very early potato 'Spunta' grown on sandy soil [18], which was not confirmed in the present study with very early potato cultivars grown on loamy soil (Luvisol). A one-year study carried out in Iraq showed that foliar application of humic and fulvic acids caused an increase in the chlorophyll content of medium-early potato cultivar [13]. The effect of humic acids depends on their source and concentration, and on the date and method of application, as well as the plant species and cultivar [9]. The increase in chlorophyll alone does not necessarily result in higher yields [17,26].

The biostimulants used in the experiment enhanced tolerance to abiotic stress and improved crop quality. In the three years of the study, the marketable tuber yield (diameter above 30 mm) was higher, on average, by 2.15 t·ha<sup>−</sup>1. Bio-algeen S90 and Keplak SL containing seaweed extracts produced better results in a warm and very wet growing season, whereas HumiPlant based on humic and fulvic acids produced better results in a year with lower air temperature and with drought periods during potato growth.

A correlation between the tuber yield and assimilation area was not found. Li et al. [46] found a significant positive correlation between LAI and tuber yield, which suggests that the enlargement of leaf area could enhance the export of photosynthetic products and cause an increase in tuber yield. According to Ascione et al. [47], the tuber growth rate is only slightly correlated with LAI, and still less so with SLA, which was not confirmed in the present study. A significant negative correlation was found between the total and marketable (diameter above 30 mm) tuber yield and SLA.

No correlation was found between the tuber yield of three very early potato cultivars and SPAD value measured on the fourth or fifth leaf from the top at the tuber formation stage (BBCH 41-43), which suggest that the biostimulants used in the experiment had no effect on the plant nitrogen status. Bărăscu et al. [30] found a significant negative correlation between SPAD measured on the fourth and fifth leaves from the top and the tuber weight of two mid-early potato cultivars, which could have been associated with oxidative stress [29]. SPAD index as an indicator of crop nitrogen status may be used for the prediction of the potato yield, however a higher SPAD does not always guarantee a higher tuber yield [26,28,31]. SPAD value is a useful indicator for selecting the high yield cultivars in the early period, however, no single threshold leaf SPAD value can be used for all potato cultivars. The SPAD value can predict the level of tuber yield if the value is calibrated for a particular potato cultivar [28,31]. Establishing threshold SPAD value is quite difficult due to the influence of climate and technical factors. SPAD values can be affected by leaf age and position, as well as, time of the day [26,27]. As a rule SPAD measurements are carried out on the third-fifth leaf from the top. Recently it was demonstrated that there is a significance difference in SPAD values between the upper and lower leaves among potato cultivars. It was shown that cultivar affects the SPAD values of the fourth and eighth leaf, but does not affect SPAD value of the fourth-eighth leaves and the difference between SPAD of the fourth and eighth leaf. Therefore the SPAD values of the fourth-eighth leaves could be applied as a general index of nitrogen status across different potato cultivars [27].

#### **5. Conclusions**

In conclusion, the foliar application of seaweed extracts *Ascophyllum nodosum* (Bio-algeen S90) and *Ecklonia maxima* (Kelpak SL), as well as humic and fulvic acids from leonardite (HumiPlant), resulted in enlargement of the assimilation area of very early potato cultivars, but had no effect on the SLA or chlorophyll content (SPAD value). The assimilation area was larger, on average, by 0.0505 m<sup>2</sup> (7%), and LAI was higher by 0.30 compared with the plants from the control group without a biostimulant. The SLA and SPAD depend on the cultivar and weather conditions, or nitrogen and magnesium content, in soil to a greater extent. These biostimulants enhanced abiotic stress tolerance and increased marketable tuber yield (diameter above 30 mm) 75 days after planting (the end of June), on average, by 2.15 t·ha<sup>−</sup>1. Bio-algeen S90 and Keplak SL containing seaweed extracts produced better results in a warm and very wet growing season, whereas HumiPlant based on humic and fulvic acids produced better results in a year with lower air temperature and with drought periods during potato growth. No correlation was found between the tuber yield and assimilation area or between the tuber yield and SPAD value, although a significant negative correlation was found between the tuber yield and SLA.

**Author Contributions:** Conceptualization, W.W. and T.D.; methodology and formal analysis, W.W.; investigation, W.W. and T.D.; writing—original draft preparation, W.W. and T.D.; writing—review and editing, W.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was financed from the science grant granted by the Polish Ministry of Science and Higher Education, research theme number 218/05/S.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **E**ff**ect of** *Pterocladia capillacea* **Seaweed Extracts on Growth Parameters and Biochemical Constituents of Jew's Mallow**

**Mohamed Ashour 1, Ahmed A. El-Shafei 2,3,\*, Hanan M. Khairy 4, Doaa Y. Abd-Elkader 5, Mohamed A. Mattar 2,6,\*, Abed Alataway 3,\* and Shimaa M. Hassan <sup>5</sup>**


Received: 13 February 2020; Accepted: 15 March 2020; Published: 19 March 2020

**Abstract:** We performed field experiments to evaluate the influence of two extraction treatments, seaweed (*Pterocladia capillacea* S.G. Gmelin) water extraction (WE) and ultrasound-assisted water extraction (USWE) at three concentrations (5%, 10%, and 15%), as well as control NPK traditional mineral fertilizer on the growth, yield, minerals, and antioxidants of Jew's Mallow (*Corchorus olitorius* L.) during the two seasons of 2016 and 2017 in Egypt. Plant height, number of leaves, and fresh weight of WE10 treatment were the highest (*p* < 0.05) as 59.67 cm, 10.67 and 2.41 kg m−<sup>2</sup> in 2016, respectively, and 57.33 cm, 11.00 and 2.32 kg m−<sup>2</sup> in 2017, respectively. WE10 and USWE5 treatments produced the highest dry matter (17.07%) in 2016 and (16.97%) in 2017, respectively. WE10 plants had an increased water productivity of 41.2% relative to control plants in both seasons. The highest chlorophyll 'a' was recorded after the WE10 treatment in 2016 and 2017 (17.79 μg g−<sup>1</sup> and 17.84 μg g<sup>−</sup>1, respectively). The highest levels of total antioxidant capacity, total phenolics, and total flavonoids were also recorded after the WE10 treatment. Application of WE10 boosted growth, yield, minerals, and antioxidants of Jew's Mallow. The CROPWAT model was used to estimate the evapotranspiration, irrigation water requirements, and yield response to irrigation scheduling. Our data showed a yield reduction in the initial growth stage if a limited amount of water was provided. Therefore, irrigation water should be provided during the most important stages of crop development with the choice of effective irrigation practices to avoid water losses, as this helps to maximize yield.

**Keywords:** seaweed extract; ultrasound-assisted water; foliar spray; *Pterocladia capillacea*; bio-fertilizer; growth parameters; antioxidants; Jew's Mallow; CROPWAT model

#### **1. Introduction**

Vegetables and their products contain non-enzymatic antioxidants and micro-nutrients that stabilize free radicals and in turn, increase the capacity of the plant to fight against pathogens that may affect humans and animals [1–4]. For example, the antioxidant compounds in vegetables and fruits could help prevent oxidative stress, diabetes, neurodegenerative disorders, cardiovascular disease, and cancer [5,6]. Jute (*Corchorus olitorius* L.), or Jew's mallow, belongs to the *Tiliaceae* family. *C. olitorius* thought to have originated from South China, from where it was introduced to India and Pakistan. However, a wild variety has been discovered in many areas in India, China, Australia, and Africa, particularly in Southeastern Nigeria. Jute leafy vegetable is commonly used in the preparation of soup [7]. The young shoot tips can be consumed raw or cooked and contain elevated concentrations of protein and vitamin C [8,9]. Jute is generally suggested for pregnant and nursing women because it is thought to be rich in iron [10].

Chemical fertilizers have been used in large quantities to compensate the nutrients deficiency in the soil. It has been observed that this use affects soil, plants, and human health. Their potential carcinogenicity and toxicity have been demonstrated, particularly after the reduction of nitrate to nitrite, or just reacting with amines and/or amides in the formation of N-nitroso compounds, N-nitrosamines, and other nitrogen compounds with high levels of nitrate [11]. Screening of native algal species must be considered to achieve a successful commercial and biotechnological potential of native algal species [12]. Seaweeds are the most promising plants from marine ecosystems and are used as a source of food and medicine. The coast of Egypt has a wide range of wild seaweed available throughout the year, even the Mediterranean coast [13] or the Red Sea coast [14]. Along the Egyptian Mediterranean coast, especially near Alexandria, red algae (*Pterocladia capillacea*; Rhodophyta) are the most dominant native seaweeds. Khairy and El-Shafay [13] studied the seasonal variations (spring, summer and autumn 2010) of biochemical composition of *P. capillacea* collected from Abu Qir Bay, Mediterranean Coast of Alexandria, Egypt. In 2010 spring season, *P. capillacea* achieved the highest significant protein (23.72%) and lipid (2.71%), while in 2010 summer season, *P. capillacea* achieved the highest significant carbohydrate (50.96%), ash (15.81%), and moisture (10.19%). Total fatty acids (248–515 μg/g), total saturated fatty acids (189–360 μg/g), total mono-unsaturated fatty acids (29–77 μg/g), total poly-unsaturated fatty acids (30–78 μg/g), total amino acids (2836–3924 μg/g), total essential amino acids (1136–1445 μg/g), and total non-essential amino acids (1700–2445 μg/g) of seasonally collected *P. capillacea* species were observed. Moreover, Khairy and El-Sheikh [15] observed the mineral composition and antioxidant activities of *P. capillacea* species collected seasonally (spring, Summer and autumn 2010) from Abu-Qir Bay, Mediterranean coast of Alexandria, and they concluded that this species is a rich in carotenoids, phenolic compounds, DPPH free radicals and minerals, therefore, this species can be used as potential source of health food in human diets and may be of use to food industry. In general, marine algae are a rich source of protein, lipids, carbohydrates, polysaccharides, minerals, antioxidants, and other bioactive compounds that can serve in multiple biological activities related to different industries [15,16].

Seaweed extracts can be used as fertilizer for flowering plants, vegetables, and grain crops [17–19]. Furthermore, they have been marketed as fertilizer additives, which are better than other fertilizers [20,21]. Using such extracts (bio-fertilizers) in cultivation many protect the soil and improve crop quality. Therefore, applying them to seeds or adding them to the soil stimulates plant growth [22]. Liquid extracts obtained from seaweeds have gained popularity as foliar sprays for many crops. These extracts contain cytokines, growth promoting hormones, elements, vitamins, and amino acids [23,24]. Some unknown bioactive component in seaweed acts to illicit the plant's own production of plant hormones through internal metabolic pathways [25].

Booth [20] reported that the efficacy of seaweeds as extracts was due to the presence of several metabolites and trace elements. The green seaweed *Enteromorpha* has a high potential for commercial exploitation because of its abundant and varied chemical composition, quality, and concentration of basic nutrients [26]. *Enteromorpha* sp. contains 28 times more calcium than spinach, 26 times more than nopal, and 13 times more than quelite [27]. *Ulva lactuca* and *Enteromorpha intestinalis* are used as seaweed liquid

extract for many crops [21,28]. Rama Rao [29] reported good yields of *Zizyphus rugosa* fruits, when leaves sprayed with seaweed liquid extracts obtained from *Sargassum*. Seaweed extracts are now available commercially as Maxicrop (Sea-Born), Algifert (Marinure), Goemar GA14, Kelpak 66, Seaspray, Seasol, Cytex, and Seacrop. It has been reported that seaweed extracts are better than other extracts [21,23].

Traditional methods employed for extracting bioactive compounds are time consuming and have low extraction efficiencies. To overcome these disadvantages, novel technologies for extraction of bioactive compounds from marine algae have been investigated including the use of microwaves [30], enzymes [31], and super-critical fluids [32]. Recently, ultrasonic technologies have been used to enhance the extraction efficiencies of bioactive compounds (total phenolics, fucose, and uronic acid) from brown seaweed *Ascophyllum nodosum* [33–35] and starch from microalgae *Chlamydomonas fasciata* Ettl NIES-437 [36]. Moreover, ultrasonic assisted extraction was utilized in various industrial fields including phenolic compounds from citrus peel [37], lycopene from tomatoes [38], and anthocyanins from raspberries [39]. Ultrasound-assisted extraction is a simple and employed to improve extraction of bioactive compounds from seaweed [34–36,40].

Full irrigation should be practiced to maximize water productivity [41]. Weekly skipping of irrigation during seed filling may substantially reduce seed yield and water productivity. Skipping during seed germination may be a viable option when water is scarce and land is not limiting. Economic evaluation will provide guidance to policy makers at basin scales for formulating improved and efficient water management plans under all varying weather conditions. CROPWAT is a software for irrigation planning and management [42,43]. Its main functions are: To calculate reference evapotranspiration (ETo), crop water requirements, and crop irrigation requirements, which may be used to develop irrigation schedules under multiple management conditions and water supply schemes, to estimate rain-fed production and drought effects, and to evaluate the efficiency of irrigation practices. The CROPWAT model has been validated in previous studies for estimating the dynamics main components of soil water balance [44–47]. We undertook this study to investigate the effect of *P. capillacea* seaweed liquid extracts on the growth, yield, minerals, and antioxidants of Jew's Mallow (*C. olitorius* L.).

#### **2. Materials and Methods**

#### *2.1. Seaweed*

#### 2.1.1. Sampling

*P. capillacea* seaweed was collected in spring 2016 from the submerged rocky site near Boughaz El-Maadya, Abu-Qir Bay, Alexandria (31.3000◦ N and 30.1667◦ E) in Egypt. Harvested samples were transferred to the Microalgae and Invertebrates Aquaculture Laboratories, National Institute of Oceanography and Fisheries (NIIOF), Alexandria, Egypt. Epiphytes were removed from samples, and the seaweed samples were cleaned, washed, and air-dried in shadow. Dried seaweed samples were powdered and stored at room temperature in plastic bag for further analysis and utilization.

#### 2.1.2. Biochemical Composition

Protein, lipid, carbohydrates, and ash of identified seaweed *P. capillacea*, collected in spring 2016, were determined. Total proteins were extracted according to Rauch [48] and determined according to Hartree [49]. Total carbohydrates were extracted according to Myklestad and Haug [50] and determined according to Dubois et al. [51]. Total lipid was calculated according to Bligh and Dyer [52]. Fatty acids and amino acids were extracted and estimated as described by El-Shenody et al. [14].

#### 2.1.3. Seaweed Liquid Extracts Preparation

In this study, seaweed crude liquid extracts of *P. capillacea* were prepared using two extraction methods: three treatments using water extract (WE) and the other three treatments using Ultrasound-Assisted Water Extraction (USWE), as shown in Figure 1. For WE, 100 g seaweed powder was soaked in 1 L distilled water in a 60 ◦C water bath for 60 min (extraction phase I). The residual filtrate was filtered and soaked in 1 L distilled water (1:10, w/v) in a 60 ◦C water bath for 60 min (extraction phase II) and this process was repeated a third time (extraction phase III). Each extraction phase was filtered through Whatman No. 3 filter paper and the supernatants of the three phases (I, II, and III) were combined to the final WE volume of 3 L and stored at −20 ◦C. For USWE, the three extraction phases were prepared as described above for WE, but after each phase the mixture was subjected to ultrasonication. The USWE extraction was performed at 60 ◦C for 5 min and 99% amplitude of 20 kHz; these conditions were adjusted and stable for all three extraction phases. After three USWE extraction phases, the supernatants were combined to a final volume of 3 L, and stored at −20 ◦C. The final combined supernatants, both WE and USWE, were considered to be a 100% crude extract that was utilized as seaweed foliar spray.

**Figure 1.** Procedures for water extraction (WE) and ultrasound-assisted water extraction (USWE) of pre-treated *Pterocladia capillacea*.

#### *2.2. Experimental Design*

The field experiment with Jew's Mallow (*C. olitorius* cv. Balady) was conducted for two successive growing seasons (2016–2017) at Abeis Experimental Farm, Alexandria University, Alexandria (31.2001◦ N and 29.9187◦ E) in Egypt. Before sowing, soil samples were collected (0–30 cm depth) to determine physical and chemical properties following Page [53] (Table 1). Climatic data, such as maximum and minimum air temperature (Tmax and Tmin), relative humidity (RH), wind speed (u2), and rainfall (P), were collected at a meteorological station near the experimental field location (Figure 2) to calculate daily ETo using the Penman–Monteith FAO-56 equation [54]. Evapotranspiration was estimated during growth using crop coefficient (Kc) values [54] multiplied by ETo. The experimental area of 220.5 m<sup>2</sup> was divided into three replicate blocks separated by 2-m buffer zones. Each block consisted of seven plots including one traditional fertilizer and six seaweed extract treatments. Each plot covered an area of 10.5 m<sup>2</sup> (3 <sup>×</sup> 3.5 m). A randomized complete block design (RCBD) was used. Commercial seeds were sown on March 20, 2016 and March 22, 2017 at the rate of 28 kg ha−<sup>1</sup> [55]. The site was irrigated five times during the first 20 days after sowing to allow germination and establishment before the application of extract treatments. After that, irrigation was carried out every six to seven days for all treatments. The first dose (0.5 m<sup>3</sup> ha−1) of growth fertilizer or seaweed extract was applied 10 days

after sowing (DAS), the second one (0.75 m<sup>3</sup> ha<sup>−</sup>1) was applied 18 DAS, and the third dose (1 m3 ha<sup>−</sup>1) was adding 26 DAS. Harvesting included two cuttings at 45 and 70 DAS.


**Table 1.** Soil physical and chemical properties.

**Figure 2.** Daily climate parameters in the 2016 and 2017 experimental periods during the Jew's Mallow growing season. (**a**) daily maximum and minimum air temperature (Tmax and Tmin, ◦C), (**b**) relative humidity (RH, %) and wind speed (km h<sup>−</sup>1), and (**c**) reference evapotranspiration (ETo, mm) and rainfall (mm).

#### *2.3. Treatments*

The following seven treatments were used: mineral NPK fertilizer (control); water extracted seaweed at 5%, 10%, and 15% (WE5, WE10, and WE15); and ultrasound-assisted water extraction seaweed at 5%, 10%, and 15% (USWE5, USWE10, and USWE15).

NPK fertilization was carried out according to the recommendations for commercial production of Jew's Mallow plant. The NPK treatment dose consisted of ammonium nitrate NH4NO3 (33%N) at the rate of 300 kg ha−1, calcium superphosphate (Ca(H2PO4)2.H2O (15% P2O5); 525 kg ha−1); and potassium sulphate (K2SO4 (48% K2O); 125 kg ha<sup>−</sup>1). Nitrogen fertilizer was applied thrice at 7, 15, and 21 DAS. Phosphorus fertilizer was mixed during soil preparation. Potassium fertilizer was applied at 15 DAS.

#### *2.4. Measurements*

#### 2.4.1. Agronomic and Physiological

Plants were harvested (cut) twice, at 45 DAS and 70 DAS, from the center of each plot (treatment) per season to determine leaf and stem fresh weight (kg m<sup>−</sup>2). Five plants were randomly chosen from each plot to measure plant height and; number of leaves. Ratios between dry leaf weight and fresh leaf weight were determined after drying 70 ◦C in a forced-air oven until reaching a constant weight. Water productivity (WP, kg m<sup>−</sup>3) was used to evaluate treatments, calculated by dividing total fresh weight (kg m<sup>−</sup>2) at harvest by the amount of water applied (supplemental irrigation plus rainfall, m3 m<sup>−</sup>2) to the crop. Chlorophyll 'a', Chlorophyll 'b', and total carotene (μg g<sup>−</sup>1) as described by Dere, et al. [56].

#### 2.4.2. Nutrient Contents

Plant nutrient content (N, P, and K) was analyzed and expressed as percentage on leaf dry weight basis. Total N and P contents were determined calorimetrically using a spectrophotometer at 662 and 650 nm, following the methods of Evenhuis [57]. K was quantified by atomic absorption spectrometry as described by Cottenie, et al. [58].

#### 2.4.3. Antioxidant Activities

Antioxidant activities of crude extracts of WE, USWE and Jew's Mallow were observed. Free radical scavenging activity against DPPH (2,2-diphenyl-1-picrylhydrazy) was determined as described by Suresh Kumar, et al. [59]. The total antioxidant content (TAC; mg g−1) was determined with a Phosphomolybdate assay using ascorbic acid as the standard [60]. The total phenolic content (TPC; mg g<sup>−</sup>1) was determined by using the Folin–Ciocalteu method as modified by Suresh Kumar, et al. [59]. Total flavonoid content (TVC; μg g−1) was determined according to the method of Chang, et al. [61] with Quercetin as the standard.

#### *2.5. CROPWAT Model*

Before Jew's Mallow cultivation, the water application depth and irrigation timing intervals were calculated using the CLIMWAT 2.0 and CROPWAT models. CLIMWAT 2.0 is climatic software [62] presenting the monthly agro-climatic data of over 5000 stations worldwide, including the Alexandria-Nouzha agroclimatic station, which was the nearest to the experimental site (4 km). The CROPWAT model was used for calculation of crop water requirements and the development of irrigation schedules using the option to irrigate at critical depletion and refill soil to field capacity. Irrigation times and the amounts were estimated based on the efficiency of the basin irrigation system and applied for both growth seasons (Figure 3). At the end of each season, the CROPWAT model with the options of user defined application depth and irrigation at user defined intervals were used to evaluate the irrigation schedule. The input data for the CROPWAT version 8.0 model [63] required the following data:


The output of CROPWAT model consists of daily root zone depletion (*Dr,i*, Equation (1)), deep percolation (*DPi*), actual water use by crop (*ETc*)*actual*, efficiency of the irrigation schedule (*EIS*, Equation (2)), deficiency of the irrigation schedule (*DIS*, Equation (3)) and yield reduction (*YR*, Equation (4)) were collected and analyzed using the following equation:

$$D\_{r,i} = D\_{r,i-1} + (ET\_{c,i})\_{\text{actual}} - P\_i - I\_i + RO\_i + DP\_i \tag{1}$$

where *Dr,i*, and *Dr*,*i*−<sup>1</sup> are at days *i* and *i*−1, *Pi* is total rainfall over day *i*, *Ii* is net irrigation on day *i*, *ROi* is water loss by runoff from the soil surface on day *i*, in our study the *RO* is equal to zero, and *DPi* is water loss by deep percolation on day *i*.

$$EIS = \frac{\sum (I\_i - DP\_i)}{\sum I\_i} \times 100\tag{2}$$

$$DIS = \frac{\text{Sesaonal } (ET\_c)\_{\text{potential}} - \text{Sesaonal } (ET\_c)\_{\text{actual}}}{\text{Sesaonal } (ET\_c)\_{\text{potential}}} \times 100\tag{3}$$

$$Y\_R = K\_y \left( 1 - \frac{(ET\_c)\_{\text{actual}}}{(ET\_c)\_{\text{potential}}} \right) \tag{4}$$

**Figure 3.** Amounts of irrigation water (mm) applied during the Jew's Mallow growing season.

#### *2.6. Statistical Analyses*

Analysis of variance (ANOVA) with RCBD was performed on data obtained from both growing seasons (2016 and 2017) using the IBM SPSS Version 23 software to determine the significance of differences among treatments. Standard errors (SE) were presented for the mean of data from both growing seasons. Differences among means of replicates were measured using Duncan method at *p* ≤ 0.05 [64].

#### **3. Results**

#### *3.1. Biochemical Composition Seaweed P. Capillacea*

Nutritional compositions of the red seaweed species *P. capillacea* collected during spring season of 2016 were investigated. Lipid, protein, carbohydrate, and ash percentages, based on dry weight, were 2.46%, 18.47%, 51.36%, and 13.71%, respectively. Moreover, total fatty acids, total saturated fatty acids, total mono-unsaturated fatty acids, total poly-unsaturated fatty acids, total amino acids, total essential amino acids, and total non-essential amino acids were 247.6, 188.6, 29.1, 29.9, 2836, 1136.3, and 1700 μg g<sup>−</sup>1, respectively. The antioxidant activities result of WE and USWE crude extracts observed that no significant differences (*p* < 0.05) were found in total antioxidant content (22.48 and 22.33 mg g<sup>−</sup>1) and total phenolic content (17.79 and 16.85 mg g−1) in WE and USWE, respectively, while USWE achieved a significant difference (*p* < 0.05) in total flavonoid content (45.68 μg g−1) and total carotene (2.03 μg g<sup>−</sup>1) in comparing to WE (34.77 μg g−<sup>1</sup> and 1.29 μg g−1, respectively).

#### *3.2. Agronomic Traits*

Table 2 shows the significant differences in in plant height (*p* < 0.05). WE10 treatment produced the tallest plants, followed by WE5 in both seasons. In WE10 and WE5 treated plants, height increased by 39.8% and 28.1%, respectively, compared with mineral fertilizer treated plants (control treatment) in 2016 while plant height increased by 28.3% and 23.2% in 2015. The USWE10 treatment had the smallest effect on height in both years, with increases of 4.6% and 0.7% in 2016 and 2017, respectively, relative to control. Statistically significant differences (*p* < 0.05) were found between treatments for leaf number, where WE5 and WE10 treatments had the highest values in both seasons.


**Table 2.** Effects of water extract (WE) and ultrasound-assisted Water Extraction (USWE) on growth characteristics of Jew's Mallow during 2016 and 2017 growing seasons.

Control: NPK fertilization; WE5, WE10, and WE15: water extracted seaweed at 5%, 10%, and 15%, respectively; USWE5, USWE10, and USWE15: ultrasound-assisted water extraction seaweed at 5%, 10%, and 15%, respectively. Data are means ± SE. Different superscript letters in each column indicate significant differences (*p* ≤ 0.05).

There was a significant difference (*p* < 0.05) in fresh weight and dry matter between the treatments in 2016 and 2017. Fresh weight was the highest in WE10 treated plants in both seasons, followed by the USWE10 treatment. For WE10 and USWE10 treatments, the fresh weight increased by 47% and 35.4% in 2016, respectively; then 40.6% and 30.9% in 2017, compared to control treatment. Across all treatments, WE15 and USWE15 treatments reduced the fresh weight in both seasons. There were significant differences (*p* < 0.05) between bio- and mineral fertilizer-treated plants dry matter in 2016 and 2017 (Table 2). The highest dry matter value was recorded in plants that received the WE10 and USWE5 treatments, which was 23% and 19.9%higher, respectively, than control in 2016.

#### *3.3. Water Productivity*

Water requirement varied from 3.8 to 9.1 and 3.5 to 8.5 mm day−<sup>1</sup> from the early stage to the peak demand period (mid-season) for 2016 and 2017, respectively. Water productivity (WP) values determined for treatments in 2016 and 2017 are shown in Figure 4. In both seasons, there were significant differences (*p* < 0.05) between WP values. The highest WP values were recorded with the WE10 treatment 41.2% higher than control. Among extract treatments, USWE15 and WE15 had the lowest WP in 2016 and 2017, which was lower by 23% and 20.7% than WE10, respectively.

#### *3.4. Physiological Traits*

Water extraction treatments had the highest content of chlorophyll 'a' in both seasons (average, 17.49 μg g<sup>−</sup>1), while the lowest chlorophyll 'a' content was observed with the control treatment (average, 9.4 μg g−1, Table 3). WE10 and WE15 treated plants showed significant increases in chlorophyll 'b' content compared to control treatment in both seasons. The chlorophyll 'b' content for the WE10 and WE15 treated plants (average, 13 μg g−<sup>1</sup> and 13.3 μg g<sup>−</sup>1, respectively) was two-fold higher than the content in the control treatment in 2016 and 2017. Conversely, USWE5 and USWE10 application resulted in the lowest chlorophyll 'b' content in both growing seasons, 16.8% and 26.8% lower than control, respectively. The lowest carotene content was achieved by control treatment in 2016 and 2017 (2.9 and 2.8 μg g−1, respectively; Table 3). The highest carotene content in 2016 and 2017 was measured with the USWE10 treatment (71.2% and 72% higher than control, respectively), followed by the USWE15 treatment (53.7% and 54.3% higher than control, respectively).

**Figure 4.** Water productivity (kg m<sup>−</sup>3) of Jew's Mallow as a function of water extract (WE) and ultrasound-assisted water extract (USWE) treatments in the 2016 and 2017 growing seasons. Different letters (a, b, etc.) above bars indicate a significant difference among treatments in each season.



Control: NPK fertilization; WE5, WE10, and WE15: water extracted seaweed at 5%, 10%, and 15%, respectively; USWE5, USWE10, and USWE15: ultrasound-assisted water extraction seaweed at 5%, 10%, and 15%, respectively. Data are means ± SE. Different superscript letters in each column indicate significant differences (*p* ≤ 0.05).

#### *3.5. N, P, and K*

Nutrient content (i.e., N, P, and K) of Jew's Mallow plant treated with different seaweed extracts in comparison to control treatment are presented in Table 4. Control treatment had the highest N content in 2016 and 2017 (1.78% and 1.71%, respectively), while WE10 had the lowest N content (1.20% and 1.33%), respectively. USWE10 treatment had the highest P content in 2016 and 2017 (0.74% and 0.77%, respectively). The highest K content (1.90%) was achieved by WE15 treatment in both seasons.

**Table 4.** Effects of water extract (WE) and ultrasound-assisted Water Extraction (USWE) on N, P, and K content in Jew's Mallow during the 2016 and 2017 growing seasons.


Control: NPK fertilization; WE5, WE10, and WE15: water extracted seaweed at 5%, 10%, and 15%, respectively; USWE5, USWE10, and USWE15: ultrasound-assisted water extraction seaweed at 5%, 10%, and 15%, respectively. Data are means ± SE. Different superscript letters in each column indicate significant differences (*p* ≤ 0.05).

#### *3.6. Antioxidant Activity*

The highest DPPH percentage was achieved by WE10 in 2016 and 2017 (40.78% and 40.74%, respectively). The lowest DPPH percentage was recorded in USWE15 in 2016 and 2017 (8.75% and 8.74%, respectively; Figure 5). The highest TAC was recorded in WE10 in both seasons (43.97 and 44.22 mg g<sup>−</sup>1, respectively), followed by the USWE10 treatment ((35.69 and 36.38 mg g−1; Table 5). The lowest TAC was recorded with control (26.30 mg g<sup>−</sup>1) in 2017 and USWE15 treatment (26.00 mg g−1) in 2018. In both seasons, the highest significant TPC was obtained with WE10 treatment (116.28 and 115.81 mg g<sup>−</sup>1, respectively), while the lowest TPC was obtained with the USWE15 treatment (49.62 and 49.61 mg g<sup>−</sup>1, respectively). Although USWE5 had a higher TVC value, significant differences between extracts treatments were not observed, except for USWE15 treatment.

**Figure 5.** DPPH inhibition (% inhibition) of Jew's Mallow with water extract (WE) and ultrasound-assisted water extract (USWE) in the 2016 and 2017 growing seasons. Different letters (a, b, etc.) above bars indicate a significant difference among treatments in each season.

**Table 5.** Effects of water extract (WE) and ultrasound-assisted Water Extraction (USWE) on total antioxidant content (TAC), total phenolic content (TPC), and total flavonoid content (TVC) of Jew's Mallow during the 2016 and 2017 growing seasons.


Control: NPK fertilization; WE5, WE10, and WE15: water extracted seaweed at 5%, 10%, and 15%, respectively; USWE5, USWE10, and USWE15: ultrasound-assisted water extraction seaweed at 5%, 10%, and 15%, respectively. Data are means ± SE. Different superscript letters in each column indicate significant differences (*p* ≤ 0.05).

#### *3.7. CROPWAT Model*

Figure 6 shows the depletion curve before and after each irrigation event during the 2016 and 2017 growth seasons. The highest values of depletion were 52 and 48 mm in the mid-season for each year, respectively. The depletion values were between those of field capacity and readily available moisture except for the initial 20 days of both seasons. Thus, there was no water stress. In the initial stage, there was a maximum *DP* of 28 mm at the first irrigation event in both seasons and after that it decreased to 2.5 and 8.5 mm, respectively, in the 2016 and 2017 growing seasons. The effective rainfall means modeled for 2016 and 2017 were 9.4 and 1 mm, respectively leaving deficits of 387 and 401.7 mm to be made up from irrigation. Thus, effective rainfall showed an ineffective pattern across the growth stages. By applying the basin irrigation system, the application water efficiencies were 78% and 79%, respectively, in 2016 and 2017. Hence, the values of *EIS* were 95.1% and 98.6%, respectively, in 2016 and 2017. The values of (*ETc*)*actual* were 393.2 and 399.8 mm, respectively in 2016 and 2017; the values of *DIS* were 0.8% and 0.7%. On the other hand, there were no yield reductions across the growth stages, with the maximum values of 3.7% and 3.1% in the initial stage, respectively in 2016 and 2017.

**Figure 6.** Soil water depletion during the 2016 and 2017 growing seasons.

#### **4. Discussion**

Marine algae are considered very important bioindicator for the marine ecosystem [13–15]. Many studies have reported that the constituents, diversity, and communities of marine algae are affected by variations in environmental parameters and nutrient limitation [65–70]. During the last few years, the attention on scientific and commercial interest to biotechnological applications of algae as a sustainable source and global commercial for aquaculture [71–75], biofuel [12,76], extracts [77,78], food supplement, pharmaceuticals, and cosmetics were increased [12].

In current study, data of biochemical composition (protein, lipid and carbohydrate) of *P. capillacea* showed that the large component is carbohydrate (51.36%), followed by protein (21.49%) and lipid (2.06%). The presented data may be act as an indicator for related bioactive secondary metabolites of *P. capillacea* liquid extract. However, our data is in the same line of the results presented by Khairy and El-Shafay [13] who found that, during spring season of 2010, the highest component is carbohydrate (50.49%), followed by protein (23.72%) and lipid (2.71%). Many authors reported that the biochemical constituents of marine algae are affected by variations in environmental conditions and nutrient availability [65–70].

The nutrient contents of seaweed *P. capillacea* used in current study were investigated previously by Khairy and El-Sheikh [14], at the same collected location of current study too, who observed that mineral were potassium (50.9 mg/100g), calcium (68.4 mg/100g), magnesium (22.1 mg/100g), cupper (0.5 mg/100g), ferrous (18.37 mg/100g), and zinc (0.19 mg/100g). In current study, although the applied seaweed extract is a rich source of nutrient, it not characterized as a nutrient fertilizer because of many consecrations like its constituent of bioactive compound which act as a plant growth promoting. Interestingly, the *P. capillacea* seaweed species is reported as a potential source for human healthy food because its constituent of bioactive compounds *like* protein, lipid, carbohydrate, fatty acids (saturated, mono-unsaturated and poly-unsaturated fatty acids), amino acids (essential and non-essential), carotenoids, phenolic compounds, and DPPH [13,14]. Hence, *P. capillacea*, collected from the same study location, is reported as a rich source of alkaloids, flavonoids, steroids, terpenoids, phlobatannins and many other phytochemicals and secondary metabolites [79].

Moreover, *P. capillacea* as a red alga is characterized as a rich source of different phytohormones [40,80,81]. It well known that some unknown bioactive component in seaweed acts to illicit the plant's own production of plant hormones through internal metabolic pathways [25]. Seaweeds and its extracts are becoming of increasing importance because of their bioactive compounds and their potential application in different industries. Liquid seaweed extract is commonly used as commercial agricultural biostimulants because of many considerations.

In current study, to enhance the efficiency of seaweed liquid extract, we evaluate two extraction methods; (1) water (WE); and (2) ultrasound-assisted water extraction (USWE). The effect of different seaweed extracts as a foliar spray on quantity (growth and yield) and quality (minerals and antioxidants activity) of Jew's Mallow (*C. olitorius* L.), comparing to NPK traditional fertilizers were observed. In general, Jew's Mallow (*C. olitorius* L.) treated with liquid seaweed extract (either WE or USWE) achieved the highest significant quantity (yield) and quality (antioxidant activity, P %, and K %), comparing to NPK traditional fertilizers, which only achieved the highest significant N %. Jew's Mallow (*C. olitorius* L.) treated with WE10 and USWE10 were achieved the highest significant yield (fresh weight), and P %. The highest significant Chlorophyll a and b; total antioxidant activity and total phenolic compounds were achieved by WE10, while the highest significant carotene and total flavonoid compounds were achieved by USWE10. In general, in the present study, it was observed that the seaweed liquid extract prepared from *P. capillacea* presented to Jew's Mallow gave better results in all aspects of growth to yield when compared to NPK traditional fertilizers. Using ultrasound-assisted water extraction (USWE) method was significantly improved the total flavonoid and carotene content in *P. capillacea* USWE crude extract, which is positively reflected on these compounds of Jew's Mallow (*C. olitorious* L.), when comparing to WE. Carotenes are indispensable to plants and act as precursors for the biosynthesis of phytohormones and strigolactones, improve the plant development and responses to unstable environmental, and serve as a source of pro-vitamin A [82].

In the present study, WE10-treated plants showed the best response in plant height and leaf number. Similarly, Stephenson [40] reported that seaweed liquid extract prepared from *Ascophyllum* and *Laminaria* accelerated maize growth. Blunden and Wildgoose [83] reported a marked increase in lateral root development in potato plants as a result of treatment with seaweed extract. Similar results were obtained with *Padina* biofertilizer, which induced maximum growth in *Cajanus cajan* [84]. Thirumaran, et al. [85] reported similar findings 20% seaweed liquid extract from brown algae *Rosenvingea intricate* had an increased growth of *Cyamopsis tetragonoloba*. Similarly, Whapham, et al. [86] observed that the application of seaweed *Ascophyllum nodosum* liquid extract increased the chlorophyll content in cucumber cotyledons, tomato, and guar plants [83].

Seaweed liquid extracts can be an effective way to some crop plants to increase both the nutrient content of the soil and crop yield. Hence, seaweeds play a vital role in agriculture, where irrational use of chemical fertilizer and pesticides is a cause of concern. Extensive regional trials with the product are needed to determine the environmental limitations of biological activity and to monitor the survival and dispersal of the inoculate [87]. Hence, use of modern agriculture in conjunction with traditional farming practices is the sustainable solution for the future. The expansion of nature source of other manures, seaweed extract application will be useful in enriching the production in the place of costly chemical fertilizer. The use of seaweed liquid extracts helps to avoid environmental pollution by high doses of chemical fertilizer. The beneficial effects of seaweed extract on terrestrial plants are improving the overall growth, yield and the ability to with stand adverse environmental conditions [88].

From the outputs of the CROPWAT model for 2016 and 2017 growing seasons, it appeared that additional irrigation was required to meet the daily crop water requirements as rainfall had minor effects or none. This high irrigation requirement may be attributed to the low rainfall during the growing seasons. Our data indicate that irrigation is crucial in the initial growth stage of Jew's Mallow due to high *DP* caused by basin irrigation system. To avoid yield reductions in Jew's Mallow cultivation, large quantities of water should be applied during the initial stage. In areas where water is a restricting factor in crop production, a well-designed irrigation schedule can improve water productivity when full irrigation is not plausible. However, a certain yield reduction should be expected due to the relationship between ETc and yield of some crops [44,89–91].

#### **5. Conclusions**

Seaweeds are one of the most important marine resources for food, industrial raw materials, therapeutic and botanical applications. In current study, to enhance the efficiency of seaweed liquid extract, we evaluate two extraction methods; (1) water (WE); and (2) ultrasound-assisted water extraction (USWE). The effect of different seaweed extracts as a foliar spray on quantity (growth and yield) and quality (minerals and antioxidants activity) of Jew's Mallow *C. olitorious* L., comparing to NPK traditional fertilizers were observed. The present study observed that the seaweed liquid extract prepared from *P. capillacea* (either WE or USWE) presented to Jew's Mallow *C. olitorious* L. gave better results in all aspects of quantity and quality when compared to NPK traditional fertilizers. No significant differences of quantity (yield) of *C. olitorious* L. treated with WE10 and USWE10. Water extraction (WE) method improves the Chlorophyll 'a' and 'b'; total antioxidant activity and total phenolic compounds of Jew's Mallow *C. olitorious* L. While, using ultrasound-assisted water extraction (USWE) method improves the carotene and total flavonoid compounds of *P. capillacea* USWE crude extract which positively reflected on the contents of these compounds in Jew's Mallow *C. olitorious* L., when comparing to WE. Carotenes are indispensable to the plants and act as precursors for the biosynthesis of phytohormones and strigolactones, improve the plant development and responses to unstable environmental, and serve as a source of pro-vitamin A. Thus, USWE is an attractive novel technology enhancing the efficiency of seaweed liquid extract on Jew's Mallow. The CROPWAT model has shown that an adequate amount of water is vital, especially during the initial growth stage of Jew's Mallow, but also in other stages. Therefore, it is important to adopt efficient irrigation practices to maximize yields while reducing adverse effects on water resources.

**Author Contributions:** M.A., conceptualization, funding acquisition, investigation, methodology, writing—original draft, project administration; A.A.E.-S., formal analysis, investigation, data curation, methodology, writing—original draft, writing—review and editing; H.M.K., visualization, methodology, writing—original draft; D.Y.A.-E., supervision, visualization, methodology, writing—original draft; M.A.M., project administration, methodology, writing—review and editing, writing—original draft; A.A., Funding acquisition, project administration, resources, supervision, writing—original draft; S.M.H., methodology, investigation, supervision, project administration, writing—original draft. All authors have read and agree to the published version of the manuscript.

**Funding:** This work was supported by the Vice Deanship of Research Chairs at King Saud University.

**Acknowledgments:** This project was financially supported by the Vice Deanship of Research Chairs at King Saud University.

**Conflicts of Interest:** The authors declare no conflicts of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

### **Appraisal of Biodegradable Mulching Films and Vegetal-Derived Biostimulant Application as Eco-Sustainable Practices for Enhancing Lettuce Crop Performance and Nutritive Value**

**Eugenio Cozzolino 1,**†**, Maria Giordano 2,**†**, Nunzio Fiorentino 2, Christophe El-Nakhel 2, Antonio Pannico 2, Ida Di Mola 2, Mauro Mori 2, Marios C. Kyriacou 3, Giuseppe Colla <sup>4</sup> and Youssef Rouphael 2,\***


Received: 6 March 2020; Accepted: 19 March 2020; Published: 20 March 2020

**Abstract:** Scientists, extensions specialists, and growers are seeking sustainable agricultural practices that are able to cope with these objectives in order to ensure global food security and minimize environmental damage. The use of mulching films and plant biostimulants in agriculture seems to be a valid solution for tackling these rising concerns. A greenhouse experiment was conducted in order to elucidate the morpho-physiological and nutritive characteristics of lettuce (*Lactuca sativa* L.) in response to foliar application of a tropical plant extract (PE) biostimulant and the use of plastic mulches. Two biodegradable mulch treatments (Mater-Bi® 1 and Mater-Bi® 2) were compared to black polyethylene (LDPE) and bare soil. Biodegradable mulch film Mater-Bi® 1 produced a comparable marketable fresh yield to the commercial standard polyethylene (LDPE), whereas Mater-Bi® 2 exhibited the highest crop productivity. When averaged over biostimulant application, lettuce plants grown with biodegradable film Mater-Bi® 2 exhibited superior quality traits in terms of K, Ca, total ascorbic acid, and carotenoids content. The combination of film mulching (LDPE, Mater-Bi® 1 or Mater-Bi® 2) with the tropical plant extract biostimulant exhibited a positive and significant synergistic effect (+30%) on yield. The PE-biostimulant induced higher values of SPAD index and total chlorophyll content when compared to untreated greenhouse lettuce. The mineral content of leaf tissues was greater by 10% and 17% (for P and Ca, respectively) when compared to the untreated lettuce (no PE application). Nitrate content was significantly reduced by 23% in greenhouse lettuce plants receiving PE as compared to the untreated control. The positive effect of Mater-Bi® 2 film on the ascorbic acid content has also been highlighted when combined with the biostimulant application, where a major amplification of total ascorbic acid (+168%) was recorded in comparison to the untreated lettuce. Overall, our work can assist leafy vegetables growers in adopting good agricultural practices, such as biodegradable plastic mulches and vegetal-derived biostimulants, to improve the sustainability of greenhouse production.

**Keywords:** eco-friendly practices; *Lactuca sativa* L.; total ascorbic acid; tropical plant extract; Mater-Bi®; nitrate; mineral composition; SPAD index; functional quality

#### **1. Introduction**

A widespread agricultural practice across the world consists of covering the soil around plants with plastic films. The introduction of this technique in agriculture dates back to the 1970s, and its success is still linked to multiple benefits. In fact, plastic films can: (i) increase soil temperature and keep it constant throughout the first 20–30 cm layer, so that plants' roots develop faster [1,2]; (ii) reduce soil evapotranspiration and preserve moisture; (iii) prevent soil erosion and excessive leaching of nutrients from plants' rhizosphere; and, (iv) improve the performance of plants in a quantitative and qualitative manner [1,3–5]. In addition, mulching films suppress weeds growth, protect crops against pests and various diseases, and reduce the use of pesticides and herbicides. Based on their color (black, clear or white), they absorb and/or reflect sunlight, differently varying soil temperature, thus affecting crop growth and productivity [5]. Plastic films are widely used for growing vegetables under both open-field and greenhouse conditions [6]. Moreover, these films are mainly made by low-density polyethylene (LDPE) [3], having a strong resistance and high durability, even though, like all petroleum products, they are non-compostable and non-biodegradable. The presence of LDPE residues in the soil beyond the duration of a crop cycle is associated to soil contamination with phthalate and phthalic acid esters due to thermal degradation [7]. Therefore, farmers must manually or mechanically collect from the field and recycle or dispose them to comply with the legislative directives of each country. Unfortunately, the frequent illegal burning of plastic mulches by farmers is becoming a common practice, with the aim of reducing production costs by avoiding disposal expenses, which results in a consequent emission of toxic and harmful substances for humans and the environment [1,3]. In such a way, plastic mulching films increase plastic wastes that are used in agriculture, such as pipes and fittings; agricultural packaging, such as bags, liners, and containers [3]. Therefore, there is an urgent need to use compostable and biodegradable materials in modern agriculture. Nowadays, research is projected towards the creation of films made of biopolymers, such as starch, polylactic acid, and cellulose. These materials are derived from renewable resources, such as corn, potato, and rice [1,6]. Their degradation is in compliance with the European laws and Italian ones (UNI 10785, 1999) on biodegradability (EN 13432, 2000). In fact, these materials are entirely degraded by soil microorganisms and they are mineralized in carbon dioxide and methane, water, and biomass, without the production of toxic substances. Any biodegradable material is designed to disappear within the soil in 5–6 months after the end of the crop [2].

Efficient management of natural resources, such as water and soil, is needed in a scenario where the world population is growing, and agriculture must meet an increasing food demand. On the other hand, the use of plant biostimulants in agriculture has been recognized during the last two decades as an efficient tool to boost yield under optimal and sub-optimal conditions, to improve quality as well as increase nutrient uptake and use efficiency of field and horticultural crops [8–11]. Under the new European Union Regulation 2019/1009, plant biostimulants are specified based on their agronomical effects on crops (i.e., claims), and they include humic substances, protein hydrolysates, algae and plant extracts, inorganic compounds (e.g., silicon), growth-promoting bacteria, and mycorrhizal fungi. Many recent studies on vegetal-based biostimulants have shown to increase the tolerance of crops to abiotic stress (extreme temperature, drought, and salinity), and improve the quality of the produce, in terms of organoleptic and nutraceutical characteristics [11]. They have also contributed to the reduction of unwanted substances content, such as nitrates and heavy metals, in crops [12]. Among these, plant extracts that mainly contain signaling molecules (i.e., small peptides and free amino acids) can influence both primary and secondary metabolism in plants, by stimulating glycolysis enzymes' activity, Krebs' cycle, and nitrates' assimilation [13,14]. Moreover, it has been shown that vegetal-derived plant biostimulants effects involve the size modifications of roots by increasing the length and the number of root hairs, as well as the intake of both macroelements and microelements, leading to better crop performance and the nutritive value of the final produce [8,13,14].

Lettuce (*Lactuca sativa* L.) belongs to the *Asteraceae* family and it is one of the most intensively produced leafy vegetables being widespread all over the world. It is valued for its organoleptic

properties and is considered an important source for health-promoting metabolites (carotenoids, chlorophylls, macro and trace elements, phenolics, and vitamins), which are crucial in human nutrition [15,16]. Lettuce has a high water and low fat content, which makes it ideal for dietary plans [15]. Italy dedicates vast areas to lettuce production, and has a broad market, which places it as a European leader in this sector [10]. More importantly, production systems and agronomic practices are pre-harvest factors that can determine the quantitative and qualitative variations in lettuce bioactive compounds [17].

On the basis of the above-mentioned considerations, the aim of our work was to combine two eco-sustainable agricultural practices, such as the use of biodegradable films and plant-based biostimulant (tropical plant extract), and test their effect on the morpho-physiological performance, mineral composition, and nutritive value of greenhouse lettuce plants. The films used were two biodegradable mulching films, namely Mater-Bi® with different composition, which effect was compared with that of a polyethylene film and bare soil. The findings of the study will elucidate the biostimulant × mulch interaction to select the best combination (s) able to improve crop performance and nutritive value of this important leafy vegetable. We also believe that these results will be of great interest for horticulturists, extension specialists, and scientists.

#### **2. Materials and Methods**

#### *2.1. Greenhouse Growth Conditions, Treatments and Experimental Design*

The experimental test was implemented in a protected environment made of an unheated greenhouse, which was located at the experimental farm of the Department of Agriculture, University of Naples Federico II, Portici—Naples (lat. 40◦49 N; long 14◦20 E, 37 a.s.l). The main physical and chemical characteristics of the soil at the experimental site were: sandy loam texture (74% sand, 20% silt, 6% clay), electrical conductivity of 0.5 dS m<sup>−</sup>1, neutral pH-7.0, total nitrogen (N) of 0.12%, and organic matter of 1.20% (w/w). The nitrate N, ammoniacal N, Olsen phosphorus, and exchangeable potassium were 105, 12, 40, and 936 mg kg<sup>−</sup>1, respectively. The butterhead lettuce F1 hybrid SINTIA RZ (42–160; Rijk Zwaan, Der Lier, The Netherlands) was used in this test. This lettuce is very resistant to tip burn and bolting and it is characterized by bright green leaves. SINTIA RZ was selected as the most representative commercial cultivar that was used in Italy during the autumn and winter growing seasons under protected environment. On 16 September 2017 three mulching films (M) were installed, two black biodegradable films, namely Mater-Bi® PC 17 N1 (15 μm thick, commercial; Novamont S.p.A, Novara, Italy) and Mater-Bi® PC 17 N2 (15 μm thick, experimental; Novamont S.p.A, Novara, Italy), and one traditional black low-density polyethylene (LDPE) plastic film (50 μm thick, Idroland s.r.l., Bari, Italy). The compositions of the two biodegradable films are composed of thermoplastic starch and copolyester. The two Mater-Bi® mulching films differ in the presence of Masterbatch (PC 17 N2), a solid additive that is used for imparting color or other properties to plastics, with innovative characteristics to improve the color of mulches with low impact on the original polymer. The soil additive is a concentrated mixture of pigments that was made through a heating process and it includes a carrier resin (e.g., wax) that is cut into granules and then added to plastics.

The greenhouse consisted of a galvanized steel frame with plastic covering material, two non-automated side openings, and a mechanized roof opening. The total greenhouse surface corresponded to an area of 162 square meters (27 m × 6 m). The soil was prepared with low energy inputs consisting of a manual grubbing-up of weeds and then a shallow hoeing (20–25 cm) to allow for a leveling of the soil in a single pass. Water was not a limiting factor, the crop evapotranspiration was calculated with the Hargreaves method, and the water deficit was fully restored by using a drip irrigation system. The irrigation system consisted of a main polyethylene pipeline with a diameter of 32 mm with a low operating pressure of 2 atm, while a series of semi-compensating dripping wings (16 mm diameter and 10 cm interpolation) were laterally attached.

The lettuce seedlings were transplanted in the greenhouse on September 25th on raised furrows. On each furrow, the lettuce seedlings were arranged in double rows, at a plant density of 12.3 plants per m2. The antiperonosporic protection was performed with Metalaxil seven days after transplantation in order to limit the development of fungal pathogens.

Figure 1 presents the trend of minimum, maximum, and mean daily air temperature inside the greenhouse during the cropping cycle. The soil temperature measurements (minimum, maximum, and mean temperature) were also recorded with microchip sensors (0.5 ◦C sensitivity) that were placed at 10 cm depth. All of the measurements were collected on a data logger (Davis Vantage Pro*2*, CA, USA). Nitrogen fertilization was applied by fertigation with ammonium nitrate at eight and 16 days after transplantation (DAT). Half of the plots were treated with Auxym® (Italpollina USA Inc., Anderson, IN, USA) product in order to assess the action of the biostimulant (B). This biostimulant is obtained from fermented tropical plant biomass. It contains phytohormones, amino acids, vitamins, phytochelatins, and enzymes. Auxym® contains as well micro and macroelements (g/kg): N 8.3, P 4.0, K 25.0, Ca 0.9, Mg 1.2, Fe 6.6, Mn 6.4, B 4.4, Zn 0.4, and Cu 0.2 [18]. The biostimulant was applied -at a concentration of 2 mL per liter of water- on plants by a sprayer shoulder pump and application took place five times at seven days' intervals starting 10 DAT (i.e., foliar application).

**Figure 1.** Daily maximum, mean, and minimum values of air temperature recorded inside the greenhouse during the growing period of lettuce.

The experimental scheme provided a two factors factorial combination that resulted in eight treatments in which the factors were mulching (M; three mulching films and bare soil) and biostimulant application (B; control treatment and foliar application of biostimulant). Each treatment was replicated three times and all of the treatments were organized in a randomized complete-block design, resulting in a total of 24 experimental plots.

#### *2.2. Growth Analysis, Yield, Harvest and Quality Analysis Sampling*

The harvest was manually carried out by cutting the plants at the crown area, just when commercial weight was attained. For each replicate, a total of 15 representative plants were collected. Each plant was first weighed as a whole (leaves and stem) in order to determine the total yield, while the commercial yield was estimated after separation and weighing of leaves. In both cases, yield was expressed in g plant<sup>−</sup>1. Finally, the leaves were counted and the leaf area (cm2 plant<sup>−</sup>1) was determined using a LiCor 3100C leaf area meter (LI-COR Biosciences, Lincoln, NE, USA). Five fresh plants from each treatment were randomly sampled, and then stored at −80 ◦C until the determination of bioactive compounds content.

#### *2.3. Soil Plant Analysis Development (SPAD) Index and Color Measurements*

Fifteen SPAD index measurements were performed by a chlorophyll meter (Minolta SPAD-502, Tokyo, Japan) and averaged to a single value on five fully expanded lettuce leaves per replicate. Leaf color (space parameters L\*, a\* and b\*) was recorded with a Minolta chroma meter (CR-300, Minolta Camera Co. Ltd., Tokyo, Japan), on the center of the upper leaf surface with special care to avoid the central vein.

#### *2.4. Total Chlorophyll and Carotenoid Content Determination*

On 1 g of fresh leaf samples, the total chlorophylls and carotenoids content was determined following the Lichtenhaler and Buschman [19] method. Fresh sample was extracted in pure acetone, for 15 min in the dark. Subsequently, the absorbance of the extracted solutions was measured at 662, 645, and 470 nm, while using a Hach DR 2000 spectrophotometer (Hach Co., Loveland, CO, USA).

#### *2.5. Dry Matter, Nitrate and Macromineral Content Analysis*

After the determination of fresh yield, the leaves were put to a ventilated stove at a temperature of 65 ◦C for 72 h until constant weight for dry weight determination. The dry mater content was expressed as percentage (%). Mineral analysis was carried out in 250 mg of dry ground leaves (IKA, MF 10.1, Staufen, Germany), which were sieved and diluted in 50 mL of ultrapure water (Milli-Q, Merck Millipore, Darmstadt, Germany). A syringe with a 0.45 μm pore filter (Phenomenex, Torrance, CA, USA) was used to inject each sample into an ion chromatography (ICS-3000, Dionex, Sunnyvale, CA, USA). For macrocations determination, an IonPac CG12A (4 × 250 mm) guard column and IonPac CS12A (4 × 250 mm) analytical column were used. While, for macroanions determination, an IonPac AG11-HC (4 × 50 mm) guard column and IonPac AS11-HC analytical column (4 × 250 mm) were used. All of the macrominerals were expressed on a dry weight (dw) basis (g kg<sup>−</sup>1), while the nitrate content was expressed as mg kg−<sup>1</sup> fw based on the respective leaf sample dry matter content.

#### *2.6. Hydrophilic Antioxidant Activity Determination*

In order to measure the hydrophilic antioxidant activity (HAA), 200 mg of lyophilized sample were extracted twice with distilled water, following the *N*,*N*-dimethyl-p-phenylenediamine (DMPD) method [20]. An aliquot of 20 μL of extract was combined with 2 mL of DMPD + solution. The bleaching of solution was proportional to the amount of antioxidant compounds concentration. The reduction in absorbance, as measured by UV Vis spectrophotometry at 505 nm, allows for determining the antioxidant activity. For this purpose, an ascorbate external standard calibration curve was used.

#### *2.7. Total Phenols and Total Ascorbic Acid Content Determination*

The total phenols content was assessed with the Folin–Ciocalteau procedure [21]. 250 mg of lyophilized sample were extracted with 10 mL of methanol/water (60:40 *v*/*v*). After an incubation of 90 min., the absorption was measured at 765 nm while employing a UV-Vis spectrophotometer. The results were calculated using an external gallic acid calibration curve (Sigma Aldrich Inc., St. Louis, MO, USA). Total ascorbic acid content was measured according to the method of Kampfenkel et al. [22], and it was quantified by a spectrophotometer at 525 nm against an external ascorbate standard calibration curve.

#### *2.8. Statistical Analysis*

The normal distribution of the data was verified through the Shapiro–Wilk's and Kolmogorov–Smirnov's procedures. All of the data were subjected to Two-way ANOVA using SPSS 20 software package. For mulching factor, the treatment means were confronted utilizing Duncan's Multiple Range Test that was performed at *p* ≤ 0.05, while, for the biostimulant effect, the means were compared using the *t*-test.

#### **3. Results**

#### *3.1. Soil Temperature Trends*

The minimum, maximum and mean soil temperatures under the three tested mulches were influenced by the composition of the utilized mulching material (Figure 2). The differences between the minimum and mean soil temperatures between LDPE and the two biodegradable mulches (Mater-Bi® 1 and Mater-Bi® 2) were notable during the first 15–20 days after transplanting, whereas the differences became narrowerer towards the end of the growing cycle (Figure 2). Concerning the maximum soil temperatures, the Mater-Bi® 1 film had similar maximum soil temperature values to LDPE and slightly higher ones than Mater-Bi® 2. However, the soil minimum, maximum, and mean temperatures trends that were recorded in bare soil were regularly lower than those reached among the three tested mulch materials (Figure 2). The soil temperature trends were similar under the three mulching films in all cases, since they had the highest values at the beginning of the crop cycle and underwent a gradual decrease afterwards, especially towards the end of the growing period. The average minimum soil temperature varied between 13.2–23.9 ◦C in LDPE, 12.7–22.2 ◦C in Mater-Bi® 1, 12.7–22.2 ◦C in Mater-Bi® 2, and 10.6–20.8 ◦C in bare soil. Finally, the average maximum soil temperature fluctuated between 15.1–30.0 ◦C in LDPE, 15.1–29.5 ◦C in Mater-Bi® 1, 14.7–28.3 ◦C in Mater-Bi® 2, and 12.0–25.2 ◦C in bare soil.

#### *3.2. Yield and Biometric Parameters*

The combination of LDPE or biodegradable mulching materials (Mater-Bi® 1 or Mater-Bi® 2) with the PE-based biostimulant positively affected the total and marketable yields of greenhouse lettuce when compared to the untreated plants, although the beneficial effect of biostimulant application was not apparent for bare soil treatment (Table 1). According to the average effect of the mulching films, a tendency to higher total yield values was recorded for Mater-Bi® 2 (319.5 g plant−1), with a 22% increase as compared to bare soil (261.3 g plant−1), even though no significant differences were recorded between the three mulching treatments. However, this trend became apparent for marketable yield with significantly higher values for Mater-Bi® 2 when compared to Mater-Bi® 1 or LDPE and especially to bare soil (Table 1). The positive effect of PE-treated lettuce plants that were cultivated under the three mulching materials (LDPE, Mater-Bi® 1, or Mater-Bi® 2) was mainly attributed to an increment in the total leaf area and not to an increase in the plant leaf number based on the M × B interaction (Table 1). Moreover, the effect of PE foliar application, when averaged over all mulching treatments, was shown to affect leaf number, which was higher by 10% in PE-treated than in untreated greenhouse lettuce plants. Finally, our findings demonstrated that lettuce plants that were grown under LDPE or Mater-Bi® 2 elicited a significant increment in the number of leaves confronted to the bare soil treatment, whereas the plants cultivated under Mater-Bi® 1 exhibited intermediate values (Table 1).

**Figure 2.** Daily minimum (**A**), maximum (**B**), and mean (**C**) values of soil temperature recorded at a depth of 10 cm in bare soil, LDPE, Mater-Bi® 1 and Mater-Bi® 2 mulches.


**Table 1.** Mean comparison and analysis of variance for total and marketable yield, leaf number, and total leaf area of untreated and biostimulant-treated greenhouse lettuce grown under low-density polyethylene (LDPE) mulch and biodegradable (Mater-Bi® 1 and Mater-Bi® 2) mulching materials in relation to bare soil.

NS, \*, \*\*, \*\*\* Non-significant or significant at *p* ≤ 0.05, 0.01, and 0.001, respectively. Different letters in the same column indicate significant differences according to DMR test (*p* = 0.05). Means of biostimulant effect are compared according to Student's *t*-test (*p* = 0.05). All data are expressed as mean ± SE.

#### *3.3. SPAD index, Chlorophyll Content and Colorimetric Indices*

The non-destructive (SPAD index) and destructive measurement of chlorophylls content were significantly affected by mulching materials and biostimulant applications, with no significant effects from the M × B interaction (Table 2). The PE-based biostimulant provoked greater values of SPAD index and chlorophyll content (+6% and 30%, respectively) in comparison to the untreated control, irrespective of the mulching materials (Table 2). Moreover, when averaged over biostimulant applications, the total chlorophyll content was enhanced by 33% in mulched lettuce plants (avg. 51.3 mg 100 g−<sup>1</sup> fw) when compared to bare soil (avg. 38.6 mg 100 g−<sup>1</sup> fw), with no significant differences being observed among the three mulching materials (Table 2).

Concerning the Hunter color parameters, the ANOVA highlighted no significant M x B interaction for all of the examined color parameters (Table 2). In general, neither mulching nor biostimulant application had a significant effect on leaf yellowness (+b\*; avg. 33.4) of greenhouse lettuce. Moreover, the use of LDPE as a mulching material resulted in greater lightness (i.e., lowest L\* values) of greenhouse lettuce leaves (Table 2). Finally, the foliar application of PE-based biostimulant provoked greater values of brightness and greenness, in comparison to the untreated control, irrespective of the mulching materials (Table 2).


**Table 2.** Mean comparison and analysis of variance for Soil Plant Analysis Development (SPAD) index, total chlorophyll content, and Hunter color parameters L\* (brightness), a\* (−a\* = green) and b\* (+b\* = yellow) of untreated and biostimulant-treated greenhouse lettuce grown under LDPE mulch and biodegradable (Mater-Bi® 1 and Mater-Bi® 2) mulching materials in relation to bare soil.

NS, \*, \*\*\* Non-significant or significant at *p* ≤ 0.05 or 0.001, respectively. Different letters in the same column indicate significant differences according to DMR test (*p* = 0.05). Means of biostimulant effect are compared according to Student's *t*-test (*p* = 0.05). All data are expressed as mean ± SE.

#### *3.4. Dry Matter Percentage and Leaf Mineral Profile*

The leaf dry matter percentage and nitrate content were significantly influenced by M × B interaction (Table 3). The recorded leaf dry matter percentage across the eight experimental treatments ranged from 3.5 to 4.2%, with the lowest values being recorded in bare soil without biostimulant application (Table 3). The recorded nitrate content across the eight experimental treatments (836–2685 mg kg−<sup>1</sup> fw) was within the limits set by the EU Commission Regulation No 1258/2011 for the commercialization of fresh lettuce (3000–5000 mg kg−<sup>1</sup> fw). Our results also demonstrated that the presence of mulching materials, in particular, the use of Mater-Bi® 2, evoked a significant increment in nitrate content confronted to bare soil in both untreated and biostimulant-treated lettuce plants. Interestingly, the nitrate content was significantly reduced by 23% in greenhouse lettuce plants receiving foliar application with tropical plant extract (1566 mg kg−<sup>1</sup> fw) confronted to the control (2037 mg kg−<sup>1</sup> fw) (Table 3).


#### *Agronomy* **2020** , *10*, 427

654

Neither mulching materials nor PE-application had a significant influence on Mg and S concentrations in greenhouse lettuce leaves (avg. 3.4 and 1.0 g kg−<sup>1</sup> dw, respectively). The concentrations of target macronutrients and sodium in leaf tissues were significantly affected by mulching materials, with the highest values of K, Ca, and Na being recorded in lettuce plants that were grown under Mater-Bi® 2 mulching material (Table 3). Interestingly, PE biostimulant treatment, as averaged over mulching materials (M × B interaction = ns), affected P, Ca, and Na leaf tissues concentrations, which were greater by 10% and 17% (for P and Ca, respectively) and lower by 12% (for Na) when compared to the untreated lettuce (Table 3).

#### *3.5. Antioxidant Activity and Bioactive Compounds*

The hydrophilic antioxidant fraction of lettuce ranged from 5.6 to 7.5 mmol ascorbate eq. 100 g−<sup>1</sup> dw (Table 4). Regardless of mulching materials, the antioxidant capacity in lettuce that was treated with the commercial biostimulant Auxym® was significantly higher (+9%) as compared to the untreated control (Table 4). Neither mulching materials nor PE-application had a significant influence on total phenols content in lettuce leaves (avg. 3.4 mg gallic acid eq. 100 g−<sup>1</sup> dw). Moreover, phytochemicals with antioxidant properties, such as total ascorbic acid and carotenoids, were affected by both the tested factors (mulching materials, biostimulant application, and their combination). When averaged over the biostimulant application, the use of the Mater-Bi® 2 film evoked a significant increase in the biosynthesis and the accumulation of carotenoids (Table 4). The positive effect of Mater-Bi® 2 film on total ascorbic acid content has also been highlighted in the interaction with the biostimulant, where a major increase of total ascorbic acid (+168%) was recorded in comparison to the untreated and PE-treated lettuce grown in bare soil (Table 4).

**Table 4.** Mean comparison and analysis of variance for hydrophilic antioxidant activity, total phenols, total ascorbic acid and carotenoid contents of untreated and biostimulant-treated greenhouse lettuce grown under LDPE mulch and biodegradable (Mater-Bi® 1 and Mater-Bi® 2) mulching materials in relation to bare soil.


NS, \* Non-significant or significant at *p* ≤ 0.05, respectively. Different letters in the same column indicate significant differences according to DMR test (*p* = 0.05). Means of biostimulant effect are compared according to Student's *t*-test (*p* = 0.05). All data are expressed as mean ± SE.

#### **4. Discussion**

The use of biodegradable mulching films and plant-based biostimulants has revolutionized modern agriculture in the last two decades. Nevertheless, no scientific studies have assessed the combinatorial effect of these two agricultural practices on crop performance and nutritional value of an important greenhouse leafy vegetable, such as lettuce. Our findings indicated that biodegradable mulching film Mater-Bi® 1 produced comparable marketable fresh yield to the commercial standard polyethylene (LDPE), while Mater-Bi® 2 exhibited the highest crop productivity. It is well established that plastic mulching films increase soil temperature in comparison to bare ground. This was the case in the current experiment, since the soil minimum, maximum, and mean temperature trends that were recorded in bare soil were always lower by 2.3–3.3 ◦C, 3.5–4.2 ◦C, and 2.8–3.8 ◦C, respectively, than those that were observed among the three tested mulching materials. The differences in fresh yield could be also associated to differences in soil temperatures, when temperature is a limiting factor (autumn–winter growing season; [23]). The results that were recorded in this greenhouse experiment endorse the previous study, where the span of soil temperature under the different mulching materials had a pronounced effect on marketable lettuce yield [24–27]. Our findings concerning the beneficial effect of mulching versus bare soil were also reported in previous studies on open-field and greenhouse vegetables. For instance, melon plants had more fruits and higher fruit mean weight when grown with biodegradable films and LDPE, as compared to bare soil [2]. An increase in marketable yield in the presence of polyethylene and biodegradable (Mater-Bi®) films when compared to bare soil was also observed in pumpkin [24], tomato [1,4,25], strawberry [3,26], garlic chives [5], as well as lettuce [27]. The use of plastic films may have preserved soil moisture and prevented water evaporation and the excessive leaching of nutrients in the rhizosphere [1].

Interestingly, the combination of film mulching (LDPE, Mater-Bi® 1, or Mater-Bi® 2) with the tropical plant extract biostimulant exhibited a positive and important synergistic effect (+30%) on both total and marketable yield. Particularly, the higher marketable production that was observed in greenhouse lettuce plants that were grown under mulching films and treated with PE-biostimulant, was due to an increase in the leaf area and not to the number of leaves per plant. The increase in crop productivity and biometric parameters of lettuce plants grown under protected cultivation has been previously reported in several research studies testing the action of this tropical plant extract biostimulant on leafy and fruit vegetables, such as tomato, jute, wall rocket, and lettuce [18,28,29]. The biostimulant action of the commercial product Auxym® on PE-treated lettuce plants could be associated to the presence of signaling molecules, such as carbohydrates, vitamins, but especially free amino acids and soluble peptides [14,18,30]. The hormone-like activity of plant-derived peptides that are contained in Auxym® has been proposed in many scientific papers, where the foliar application of vegetal-based biostimulants elicited auxin- and gibberellin-like activities and, thus, boosted yield [31,32]. Since many other signaling peptides have been identified in plant cells controlling growth, development, and stress responses of plants [33], it is expected that more signaling-peptide based PE will be developed in the near future. Furthermore, some indirect effects of amino acids can be postulated. The amino acid L-tryptophan is a precursor of indole compounds (thus including auxins), while L-methionine is known as the precursor of ethylene [30]. Finally, these bioactive compounds that are present in the plant-based biostimulants can act on the primary metabolism, increasing the photosynthetic activity of the plants, and it can act as well on root growth, which might increase water and nutrient absorption efficiency, thus resulting in a yield increase [18]. This was the case in the current study, where plants that were grown under plastic mulching films and treated with tropical plant extract were characterized by better physiological and biochemical status. The greater SPAD index and chlorophyll content of lettuce leaves corroborated this, thus confirming the better photosynthetic efficiency that leads to better plant performance. Similar results on the stimulation of the physiological and biochemical status of biostimulant-treated plants were also previously observed in greenhouse tomato [18], spinach [34], lettuce [35], and jute [28].

The leaf appearance in peculiar color is among the visual characteristics of leafy vegetables that steadily govern consumer preference and selection choice [36]. Lettuce green color is directly dependent upon chlorophyll synthesis in leaf tissue. Plant extract-biostimulant application affected lettuce greenness color (−a\*) to the extent it affected chlorophyll content, as observed earlier in a broad span of leafy greens, such as spinach, lamb's lettuce, and baby lettuce [29,37].

A negative aspect in the quality of leafy vegetables is, certainly, the high content of nitrates, as they are involved in the onset of different diseases [38]. Generally, vegetables that belong to the *Brassicaceae, Chenopodiaceae,* and *Asteraceae*families [18] may accumulate nitrates in their leaves. Significant genotypic variations in nitrate accumulation are shown for lettuce [39–41]. The nitrate concentration in plants is closely related to nitrate reductase activities [42]. This reality has prompted the European Commission to regulate the nitrate limits for lettuce. In our experiment, nitrate concentrations for plants cultivated with LDPE films, Mater Bi® 1, and Mater Bi® 2 films (1898–2368 mg kg−<sup>1</sup> fw), were within the set limits for fresh lettuce according to Commission regulation (EU) No 1258/2011 (3000–5000 mg kg−<sup>1</sup> fw) [43]. The PE-biostimulant application decreased nitrates concentration in lettuce leaves by 23% (avg. 1566 mg kg<sup>−</sup>1), as compared to the control (avg. 2037 mg kg−1). This positive effect could be linked to the presence of a high content of free amino acids in the biostimulant product, which, once absorbed by the plant, might exert the inhibition of the nitric ion transporters that are present in the root. On the other hand, the ability of the plant-based biostimulant Auxym® to reduce nitrates accumulation could be associated with the regulation of nitrogen metabolism in plants, which involves the activity of nitrate and nitrite reductase, glutamate synthase, as well as glutamine synthetase [14,18,44]. Various studies confirmed our results, such as that of Bulgari [45] performed on iceberg lettuce, which showed that a biostimulant of vegetal origin enriched with micro-elements (one), kept nitrate levels well under the limit required by the EC. Similar results were also obtained in spinach, on which the effect of amino acid-based biostimulant (Aminoplant) was evaluated [46]. Other studies on corn, soy, and wheat also showed that exogenous amino acids application can significantly reduce nitrate absorption [18].

Scientists recommend that people should consume fruits and vegetables daily, because they satisfy 11%, 35%, 7%, and 24% of the daily intake of P, K, Ca, and Mg, respectively [10]. These macronutrients help against certain diseases, such as blood pressure imbalances, hypertension (K), and osteoporosis (P, Ca, and Mg) [15]. For lettuce, several authors reported a potassium content between 48–72 mg g−1, phosphorus 4–6 mg g−1, magnesium 1.4–2.8 mg g−1, and calcium 4–10 mg−<sup>1</sup> on a dry weight basis [15,16,47]. In our work, the use of biodegradable films influenced the biofortification of macronutrinets in lettuce leaves. In particular, the use of polyethylene film Mater-Bi® 1 increased P content, confronted to the control, whereas lettuce plants that were grown under Mater-Bi® 2 exhibited higher values of K and Ca when compared to the bare soil treatment. Our results match with previous studies on the 'nutrient acquisition response' of plant-based biostimulant application on tomato [18], jute [28], and spinach [29]. In addition to the accumulation of macronutrients in leaf tissues of biostimulant-treated plants, the use of PE reduced sodium concentration in lettuce leaves by 12%, confronted to the control, which is in harmony with Carillo et al. [28] findings. This is a very important aspect, because Na causes hypertension and cardiovascular diseases [48].

Furthermore, lettuce is considered to be a good source of nutraceutical molecules, such as vitamin C and carotenoids [15]. These molecules represent the radical scavenging power that protects plants from the oxidative damage caused by free radicals. In our work, when averaged over biostimulant application, lettuce plants that were grown under Mater-Bi® 2 had the highest total ascorbic acid and carotenoids content. Similarly, Morra et al. [3] recorded a higher antioxidant activity, total polyphenols, and anthocyanins in two strawberry cultivars grown under biodegradable Mater-Bi film as compared to those cultivated with LDPE or in bare ground. These results are also confirmed for melon plants that are grown with biodegradable mulching films [27,49]. The Mater Bi® 2 behavior could be related to the fact that below this film there is a greater evaporation of the soil, which results in a lower accumulation of water in plants. Therefore, this mild condition of stress might trigger the plant to

synthesize defensive molecules [27,49]. More compelling, these secondary metabolites are also crucial to human well-being [50,51].

In our work, the foliar application of PE on greenhouse lettuce also influenced antioxidant activity and health-promoting secondary metabolites, since the antioxidant potential increased by 10% when compared to the untreated control plants. The latter is a notable qualitative functional parameter in leafy vegetables, since it is correlated to the synergetic effect of low-molecular weight biologically active compounds, such as phenolic compounds and carotenoids [50]. Ertani et al. [52] showed an increase in antioxidant activity, lycopene, phenols, and ascorbic acid of *Capsicum chinense* L., in response to the application of plant extract based biostimulants. The synergistic action of Mater-Bi® 2 with tropical plant extract is of significant interest for scientists and nutritionists, because it resulted in the production of superior greenhouse lettuce leaves in terms of vitamin C content (+168% as compared to the control). In fact, as also shown by Carillo et al. [28], signaling compounds that are present in the tropical plant extract Auxym®, like glutamic and aspartic acids are involved in the stimulation of primary and secondary metabolism, thus, leading to a greater synthesis of antioxidant molecules, such as vitamin C [28].

#### **5. Conclusions**

In recent years, horticultural research has focused on improving farming practices in the framework of a more sustainable agricultural, including the use of biodegradable mulching films and vegetal-based plant biostimulants to improve the crop performance and nutritive quality of the produced commodities. Our greenhouse experiment on lettuce confirmed that the use of biodegradable plastic mulching materials, especially Mater-Bi® 2, could be considered as an alternative to LDPE and bare soil cultivation. This biodegradable mulching material increased marketable yield irrespective of the biostimulant application, due to many agronomic benefits, in particular, the better microclimate (minimum and maximum soil temperatures) in the rhizosphere. Our results also demonstrated, that lettuce plants grown under biodegradable film especially Mater-Bi® 2 exhibited superior quality traits in terms of K, Ca, total ascorbic acid, and carotenoids. Interestingly, the foliar application of PE-biostimulant in the presence of mulching materials was able to improve the total and marketable yield and biometric traits. The synergistic effect of mulching with plant-based biostimulant was linked to better physiological and biochemical status (higher SPAD index and chlorophyll content) and a higher nutrient acquisition response (higher P and Ca and lower Na content). The PE-biostimulant treated lettuce had a lower nitrate content and higher antioxidant scavenging capacity than the non-treated control, while the combination of Mater-Bi® 2 and PE-biostimulant resulted in the production of premium greenhouse lettuce leaves in terms of vitamin C content. The outcomes of the current study can encourage leafy vegetables producers to replace LDPE films with biodegradable ones in combination with plant-based biostimulants in order to attain high productivity and reach consumer expectations for high quality produce. In addition, the substitution of plastic mulching with biodegradable ones can significantly tackle the environmental issues that are related to the disposal of mulching materials at the end of the cropping cycles. The absence of dumping costs for farmers could likely offset the higher costs due to biodegradable mulching, favoring the application of biodegradable mulching materials on a wide scale.

**Author Contributions:** Conceptualization, E.C. and Y.R. methodology, E.C. software, C.E.-N., A.P., M.G., I.D.M., and M.M. formal analysis, C.E.-N., A.P., M.G., I.D.M., and M.M. investigation, E.C. resources, Y.R. data curation, C.E.-N., A.P., M.G., I.D.M., and M.M. writing—original draft preparation, M.G. writing-review and editing, M.G., N.F., C.E.-N., M.C.K., G.C. and Y.R. visualization, E.C. and Y.R. supervision, E.C., N.F. and Y.R. project administration, Y.R. funding acquisition, Y.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors are grateful to Vincenzo Esposito for his technical assistance in the greenhouse experiment.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

### **Response of Soil Bacterial Community and Pepper Plant Growth to Application of** *Bacillus thuringiensis* **KNU-07**

**HyungWoo Jo 1,2, Setu Bazie Tagele 1, Huy Quang Pham 1, Min-Chul Kim 1, Seung-Dae Choi 1, Min-Ji Kim 1, Yeong-Jun Park 1, Jerald Conrad Ibal 1, Gun-Seok Park 1,3 and Jae-Ho Shin 1,4,\***


Received: 20 March 2020; Accepted: 9 April 2020; Published: 10 April 2020

**Abstract:** Many *Bacillus* species are among the plant growth-promoting rhizobacteria (PGPR) that promote the growth of many different plant species. This study aimed to investigate the effects of *Bacillus thuringiensis* KNU-07 on the growth of pepper plants and the soil microbiota. We also designed primers specific for the strain KNU-07 to monitor the population in pepper-cultivated soil. Accordingly, a strain-specific primer pair was designed using a database constructed from 16,160 complete bacterial genomes. We employed quantitative PCR (qPCR) to track the abundance of the strain KNU-07 introduced into pepper-cultivated soil using the strain-specific primers. Our study revealed that the strain was found to possess plant growth-promoting (PGP) activities, and it promoted the growth of pepper plants. The soil bacterial community structure due to the application of the PGPR strain was significantly changed after six weeks post-inoculation. In addition, based on qPCR analysis, the population of the introduced strain declined over time. In this study, application of a PGPR strain increased the growth of pepper plants and changed the soil bacterial community structure. The successful results of monitoring of a bacterial strain's population using a single strain-specific primer pair can provide important information about the quantification of bio-inoculants under non-sterile soil conditions.

**Keywords:** *Bacillus thuringiensis*; *Capsicum annuum*; PGPR; microbiome; strain-specific primer; tracking

#### **1. Introduction**

Plant growth-promoting rhizobacteria (PGPR), which are found in the vicinity of crop roots, increase the growth and health of the plants [1]. Considering the growing public concern about the use of chemical fertilizers, there is increasing high demand to use PGPR, such as *Bacillus* spp., *Pseudomonas* spp., *Streptomyces* spp., etc. [2–4]. *Bacillus* is one of the most important genera that provides plants with potent plant growth-promoting effects, and many *Bacillus* species have been successfully used for agricultural purposes as commercial bio-inoculants [5,6]. Some strains of *Bacillus thuringiensis* have been used as a PGPR to improve soil fertility and enhance crop growth [7–9]. In addition, although the effect of PGPR on the indigenous soil bacterial communities and their functional properties has been studied, there is very limited information on the effects of *B. thuringiensis* on bacterial communities in the soil [10,11]. The beneficial effects of *B. thuringiensis* on plants are due to direct and indirect

mechanisms, including nitrogen fixation, siderophore production, plant nutrient solubilization, and plant growth hormone production [12–14]. However, the plant responses are often variable due to inconsistent performance of inoculants under field conditions [15].

The ability of inoculants to survive in the soil is an important factor for their ability to function under field conditions [15,16]. Hence, quantification of inoculants in the soil is helpful to determine the success of PGPR under field conditions [1,6,17]. However, measuring the spatiotemporal dynamics of PGPR in the environment remains challenging [18,19]. Various culture-dependent and culture-independent methods have been employed to track and quantify bio-inoculants in the soil [20,21]. However, culture-dependent methods are only successful under sterile conditions, and less than 1% of the soil microbial diversity is recovered with such methods [21,22]. On the other hand, culture-independent methods, such as reporter nucleic acid-based, gene-based, and immunological methods, are capable of detecting less abundant, slow-growing, and unculturable bacteria [8–10]. However, most culture-independent methods are incapable of monitoring population dynamics at a species level, making it difficult to determine the fate of some strains [23,24].

In this study, we investigated the effects of the PGPR strain KNU-07 (hereafter referred to as KNU-07) on the growth of pepper plants and soil bacterial community composition. More importantly, we developed a strain-specific primer pair and developed a qPCR protocol to track the quantity of *Bacillus thuringiensis* KNU-07 in pepper-cultivated soil in a greenhouse. KNU-07 increased the growth of pepper plants, and the application of KNU-07 significantly changed the soil bacterial community structure after six weeks. The established strain-specific primer was successful in quantifying and monitoring KNU-07 in non-sterile soil conditions.

#### **2. Materials and Methods**

#### *2.1. Bacterial Strains and Growth Conditions*

The genome of the rhizospheric bacterial strain used in this study, KNU-07, was assembled using PacBio RSII and comprised 6,152,737 bp. KNU-07 was cultured in Luria-Bertani (LB) broth and LB agar (Difco Laboratories, Sparks, MD) and incubated for 24 h at 30 ◦C. Bacterial strains used for in vitro PCR are indicated in Table S1.

#### *2.2. Bioinformatics Approach for Designing a Strain-Specific Primer for KNU-07*

#### 2.2.1. Designing a Primer to Target a Unique Sequence of KNU-07

Strain-specific primers were designed using a Python script that was developed in house. The complete genome sequence of KNU-07 was cut into truncated fragments of 500 bp. BLASTN was used to search for each fragment in a custom database constructed by downloading 16,160 complete bacterial genomes from the NCBI Genome Browse section (updated on 27 December 2019). The BLASTN searches were performed with the following parameters: ungapped alignment (-ungapped), no filter query sequence with dust (-dust no), and apply filtering locations as soft masks (-soft\_masking false). Unique fragments containing regions that had no overlap with any complete genome in the NCBI custom database were chosen for further analysis. A primer pair targeting the unique sequence was designed using the web-based tool Primer-BLAST (NCBI Primer-BLAST).

To validate the selected unique fragment, BLASTN was used as mentioned above. The primer pair was checked to ensure homology to the KNU-07 unique sequence region, and the specificity of the primer was validated by performing in silico PCR evaluation using ecoPCR software on the NCBI bacteria complete genome database [25]. The number of mismatches in the binding regions of the target sequence for either forward or reward primers were set to 0 to 2. The targeted PCR product size was set to a minimum of 50 bp and a maximum of 500 bp. For comparison, a universal primer pair, 27F/1492R, which amplifies a region of the 16S rRNA gene in prokaryotes, was used as a positive control (Table 1). Finally, an ecotaxstat script was used to summarize taxonomy information from the in silico PCR products.


**Table 1.** KNU-07-specific primers and universal primers used in this study.

2.2.2. In Vitro Validation of the Strain-Specific Primer Pair

PCR was performed to verify whether the primer pair designed by in silico PCR analysis amplified the unique sequence of KNU-07. In addition, we conducted additional PCR assays to investigate whether the strain-specific primer pair amplified genomic DNA from other bacterial species and environmental samples in vitro (Tables S1 and S2). Each PCR reaction contained 10 ng of template DNA, 0.2 μL of each primer (10 μM), 5 μL of EmeraldAmp GT PCR Master Mixture (Takara Korea Biomedical Inc., Seoul, Korea), and sterile distilled water to a total volume of 20 μL. PCR amplifications were carried out using the following cyclic program: initial denaturation at 95 ◦C for 7 min, followed by 30 cycles of 30 s denaturation at 95 ◦C, 55 ◦C annealing for 30 s, 72 ◦C extension for 30 s, and a final extension at 72 ◦C for 5 min.

#### *2.3. In Vitro Plant Growth-Promoting (PGP) Traits Assay*

An in vitro assay was carried out to evaluate the effect of KNU-07 on plant growth potential. The indole acetic acid (IAA) production potential was evaluated following the method of Gordon and Weber [26]. The concentration of IAA was quantified based on a standard curve of pure IAA (Sigma-Aldrich, St. Louis, MO, USA). IAA identity and its purity were confirmed by gas chromatography–mass spectrometry with a SIM (6890N network GC system, and 5973 network mass selective detector; Agilent, CA, USA). In addition, the potential of the strain to exhibit urease activity [27], siderophore production [28], and phosphate solubilization [29] was determined.

#### *2.4. Greenhouse Experiment*

#### 2.4.1. Plant Material and Bacterial Strain Preparation

Seeds of hot pepper (*Capsicum annuum* cv. CM334) were used in this experiment. KNU-07 was incubated at 30 ◦C for 24 h at 200 rpm. The pellet was collected after centrifugation, washed, and resuspended in sterile distilled water. The bacterial inoculum was adjusted using a sterile distilled water to concentrations of 7.8 <sup>×</sup> 106 cells mL−<sup>1</sup> soil and 7.8 x 108 cells mL<sup>−</sup>1.

#### 2.4.2. In Vivo Assay

The effect of strain KNU-07 on the growth of pepper plants in pots under greenhouse conditions for two months was assessed. Pepper seeds were surface-sterilized with ethanol (70%) for 1 min and soaked in a disinfectant solution (Clorox, distilled water, and 0.05% Triton X-100 in a 3:2:2 ratio (v/v/v)) for 5 min and washed 7–10 times with sterile, deionized, distilled water. Pepper seeds were vernalized for 48 h in a refrigerator at 4 ◦C and germinated by placing the seeds on sterile, wet filter paper in a growth chamber for 7 days at 30 ◦C. The germinated seeds were then sown in plastic trays containing mixed soil. The mixed soil was composed of garden soil and Biosangto-Mix soil (Heung Nong Co., Ltd., Pyeongtaek, Republic of Korea) in a 1:9 ratio (v/v). The pepper seedlings were incubated in a growth chamber (25 ◦C, 65% relative humidity, and cycles of 16 h light and 8 h dark). After two weeks, uniform-sized pepper seedlings having shoots approximately 5 cm in height were each transplanted into a pot containing 300 g of mixed soil. To assess the effect of KNU-07, the soil of some pots was inoculated with 3.85 mL of KNU-07 at one of the following concentrations: 1.0 <sup>×</sup> 105 cells g−<sup>1</sup> soil and 1.0 <sup>×</sup> <sup>10</sup><sup>7</sup> cells g−<sup>1</sup> soil. Application of bio-inoculants at 1.0 <sup>×</sup> 105 cells g−<sup>1</sup> soil is a very common practice in South Korea. Seedlings treated only with sterile distilled water served as the non-inoculated control. The experiment was replicated three times with five plants per replication. After 11 weeks of treatment, plant growth data, including plant shoot length, root length, and total biomass, were recorded.

#### 2.4.3. DNA Extraction from Pure Cultures and Soil Samples

To analyze the soil bacterial community, the soil where *Capsicum annuum* cv. CM334 was growing in each pot was sampled weekly. Very small amounts of soil sample (less than one gram) were taken at five different sites in each pot and pooled into a composite sample per pot. Genomic DNA from the soil and the strain culture was extracted using a Power Soil DNA Isolation Kit (MoBio Laboratories, Carlsbad, CA, USA) according to the manufacturer's protocol. DNA concentrations were determined using a Qubit 2.0 Fluorometer (Thermo Fisher Scientific). For strain-specific PCR assays, KNU-07 genomic DNA was extracted using a Wizard Genomic DNA Purification Kit (Promega Corporation, Madison, WI, USA).

#### 2.4.4. DNA Library Preparation and Amplicon Sequencing

The diversity of the soil bacterial community was assessed by amplifying and analyzing the V4-V5 hypervariable region of 16S rRNA gene using the universal primer pair 515F/ 907R (Table 1). The V4-V5 primer pair was tailored with Ion Torrent PGM adapter and barcode sequences, which are unique to each sample. The PCR reaction (50 μL) contained 1 ng template DNA, 1 μL of each primer, and 25 μL of EmeraldAmp GT PCR Master Mixture (Takara, Japan). The PCR conditions were as follows: initial denaturation at 95 ◦C for 7 min; 10 cycles of denaturation at 95 ◦C for 30 s, annealing at 55 ◦C for 30 s, and extension at 72 ◦C for 30 s; 30 cycles of denaturation at 95 ◦C for 30 s, annealing and extension at 72 ◦C for 45 s; and a final extension at 72 ◦C for 5 min.

Ion Torrent PGM sequencing technology and data analysis were used to sequence the amplified products. The quality of the amplified DNA library was assessed using an Agilent 2100 Bioanalyzer with a High-Sensitivity DNA (HS DNA) Kit (Agilent Technology, Santa Clara, CA, USA). The amplified DNA library was diluted to 6 pM to perform emulsion PCR with Ion Sphere Particles (ISPs) in the Ion OneTouch System II (Thermo Fisher Scientific), followed by enrichment for template-positive Ion Sphere Particles using Dynabeads MyOne Streptavidin C1 beads (Thermo Fisher Scientific, Waltham, MA, USA). Each sample was loaded on barcoded chips (Ion 316 Chip Kit v2 BC). Sequencing was carried out using the Ion Torrent PGM system and an Ion PGM Hi-Q Sequencing Kit (Thermo Fisher Scientific). The Torrent Suite Software, along with Ion Torrent PGM specific pipeline software, was employed to generate sequence reads, trim adapter sequences, filter, and exclude poor signal profile reads. Quality-filtered sequence reads were analyzed using a QIIME package (V1.9.1). Operational taxonomic units (OTUs) having 97% similarity were selected by an average neighbor algorithm and were identified using the sequence database of the National Center for Biotechnology Information (NCBI).

#### 2.4.5. Continuous Tracking of KNU-07 Using qPCR

The abundance of inoculated KNU-07 in pepper-cultivated soil was monitored using qPCR with strain-specific primers (Table 1). The total bacteria in the soil were quantified using qPCR with the universal primer pair (27F/1492R) targeting a 16S rRNA gene (Table 1). Each PCR reaction mixture (10 μL) consisted of 10 ng of DNA from the soil sample, 0.3 μL of each KNU07 specific primer (10 μM), and MG 2X qPCR mix (SYBR green, MGmed). qPCR reactions were performed in triplicate under the following cycling conditions: 95 ◦C for 15 min, followed by 35 cycles of denaturing at 95 ◦C for 30 s, annealing at 55 ◦C for 30 s and extension at 72 ◦C for 30 s. Gel electrophoresis using a CFX Real-Time PCR Detection System (BioRad) was conducted to ensure the appropriate size of the amplified products.

A standard curve based on copy number was used to determine the abundance of KNU-07 and total bacteria in the soil. Briefly, a six-fold serial dilution of amplicons of KNU-07 unique sequence was prepared in triplicate. The copy number of each concentration was calculated based on the amplicon concentration and length. A regression equation was calculated based on the cycle threshold (Ct) value to the known amount of serially diluted copy number of the unique sequence. By using the standard curve, the abundance of KNU-07 was deduced and expressed as the number of genome equivalents. A genome equivalent corresponds to the number of KNU-07 cells. In addition, the abundance of the total bacteria was determined using qPCR with a universal primer pair that amplifies a conserved region of the 16S rRNA genes of multiple bacteria species.

#### *2.5. Nucleotide Accession Numbers*

The complete genome sequence of *B. thuringiensis* KNU-07 was deposited in GenBank under accession number CP016588. The unique DNA sequence of KNU-07, which was used to design the strain-specific primers, is located at the sequence position 1,904,488 bp to 1,904,728 bp. The NGS data of all raw sequence reads were deposited in the NCBI Short Read Archive (SRA) database under accession number SRP243872.

#### *2.6. Statistical Analysis*

The alpha diversity of KNU-07-treated and control samples was analyzed using taxonomic diversity indices, such as the Shannon index, Simpson's index, and the number of observed OTUs. The community diversity difference was analyzed based on principal coordinate analysis (PCoA) using Bray–Curtis distances in QIIME1. A dissimilarity analysis of Bray–Curtis based on permutational multivariate analysis of variance (PERMANOVA, ADONIS function) [30] and an analysis of similarity (ANOSIM function) [31] were conducted to determine the impact of KNU-07 application on the soil bacterial community composition. The abundance of predicted gene function of the soil bacterial community in each experimental sample was determined by the PICRUSt pipeline using an OTU table normalized to the 16S rRNA gene copy number [32]. The data of predicted function were analyzed using the STAMP software package [33]. All data of greenhouse experiments were arranged in a randomized design with at least three replications. Analysis of variance (ANOVA) was performed for plant growth parameters using SAS software version 9.4 [34], and treatment means were separated using post hoc Tukey significant difference (HSD) tests.

#### **3. Results**

#### *3.1. In Silico and in Vitro PCR Verification of KNU-07-Specific Primer Pairs*

The genome of KNU-07 was truncated into 500 bp fragments, and 10,687 fragments were found. Among these, 81 unique windows were identified, and one window (located at 1,904,488 bp to 1,904,728 bp) was selected for designing the primers. A primer targeting a unique sequence of KNU-07 was designed to have 25 bp (Table 1) using the Primer-BLAST tool on the NCBI web site.

For the in silico PCR analysis, complete genomes of 16,160 bacteria comprising 52 phyla, 173 orders, 1130 genera, and 3747 species were used. A universal primer pair, 27F/1492R, targeting the bacterial 16S rRNA gene matched perfectly with 76% of the species tested (no mismatch), 89% of species had one mismatch, and 92% of species had two mismatches. On the other hand, our strain-specific primer pair targeting the unique sequence of KNU-07 had a perfect match to only one genome, that of *B. thuringiensis* KNU-07 (Table 2). Even by increasing the number of mismatches, no bacterial species other than KNU-07 was found to match, indicating that the primer pair was highly specific to KNU-07.


**Table 2.** In silico PCR verification that strain-specific primers target sequences unique to KNU07.

\* Perfect matches with *B. thuringiensis* KNU-07.

To verify whether our designed primer pair specifically detected KNU-07, an in vitro PCR assay was conducted using DNA samples from pure cultures of 28 bacterial strains. The results showed that the primer pair amplified the expected band size of 241 bp from strain KNU-07; however, no visible band was detected with any other bacterial strain, including *Bacillus* spp., other than KNU-07 (Figure S1). Our strain-specific primer can precisely detect and distinguish KNU-07 from other tested *Bacillus* species. Furthermore, the discrimination power of the KNU-07 strain-specific primer was verified using 28 diverse environmental DNA samples. The results confirmed that the strain-specific primer was able to successfully amplify KNU-07 with the expected band size of 241 bp, while no visible band was detected in any environmental sample (Figure S1), demonstrating that the strain-specific primer was selective in detecting KNU-07.

#### *3.2. PGP Activity of KNU-07*

KNU-07 was positive forin vitro PGP activities, including IAA production, siderophore production, phosphate solubilization, and urease activities (Figure S2). In addition, gas chromatography/mass spectrometry experiments revealed that the amount of IAA produced by KNU-07 with and without a tryptophan supplement was 4.886 and 0.167 μg mL−1, respectively. The in vivo effects of KNU-07 inoculated at different concentrations on the growth of pepper plants was determined under non-sterile conditions. After 11 weeks of growth, a significant (p < 0.05) difference was found between bacterized and non-inoculated pepper seedlings (Figure 1). Plants treated with high concentrations of KNU-07 exhibited significant increases in root length (30.7%), shoot length (19.7%), and total dry biomass (30.7%) compared to non-inoculated control plants (Figure 1).

#### *3.3. Response of the Soil Bacterial Community to KNU-07*

The effect of KNU-07 on the composition of the soil bacterial community in pepper-cultivated soil was analyzed by the Ion Torrent PGM platform based on 16S rRNA gene amplicon sequences. In this study, 3146 observed OTUs, 27 phyla, and 408 genera were identified by a BLASTN search against the Green gene database (data not shown). The results revealed that the alpha diversity indices, the Shannon index, and the number of observed OTU increased similarly over time in inoculated and non-inoculated control samples (Figure 2A). The Simpson's index showed that diversity increased with each treatment, except when KNU-07 was applied at the highest concentration (1.0 <sup>×</sup> <sup>10</sup><sup>7</sup> cells g−<sup>1</sup> soil) (Figure 2B). During the first two weeks, Simpson's index was low in soil treated with the highest concentration of KNU-07. However, after three weeks, Simpson's index increased over time (Figure 2C).

**Figure 1.** Effect of KNU-07 inoculation on the growth of pepper plants over 11 weeks post-inoculation in greenhouse conditions. The numerical value of (**A**) root length and shoot length and (**B**) total dry biomass. (**C**) Pictorial view of pepper plants inoculated with the indicated concentrations of *B. thuringiensis* KNU-07. Non-inoculated plants served as control. Mean values having different letters in each of the growth parameters are significantly different (p ≤ 0.05).

**Figure 2.** Taxonomic α-diversity analysis: (**A**) observed operational taxonomy units (OTUs); (**B**) Shannon index; (**C**) Simpson's index. Non-inoculated control was not inoculated with *B. thuringiensis* KNU-07.

Xanthomonadales and Saprospirales were the two most abundant orders in this study regardless of the KNU-07 application, and the abundance of these orders gradually decreased over time (Figure 3). In contrast, the abundance of orders Rhizobiales and Ellin329 increased over time in all samples, including controls. The abundance of Acidobacteriales decreased over time in all samples. The abundance of Bacillales, the order to which KNU-07 belongs, was comparatively high in KNU-07-inoculated soil during the first three weeks, but then it decreased (Figure 3). At genera level, the abundance of *Bacillus* spp. was comparatively high in soil inoculated with a high concentration of KNU-07 (1.0 <sup>×</sup> 107 cells g−<sup>1</sup> soil) (Figure 4). At a higher concentration of KNU-07, although the

abundance of *Bacillus* spp. was decreasing over time, the abundance of *Bacillus* spp. was still higher than the remaining treatments. The abundance of *Bacillus* spp. in the soil inoculated with a lower concentration of KNU-07 (1.0 <sup>×</sup> 105 cells g−<sup>1</sup> soil) and non-inoculated control was comparatively higher in the last three weeks (Figure 4). However, it is important to note that the resolution power of NGS of 16S rRNA coding region is not strong enough to discriminate KNU-07 from indigenous *Bacillus* spp. Hence, we designed a strain-specific primer for KNU-07 to monitor the population dynamics of KNU-07 using qPCR with strain-specific primers.

**Figure 3.** Bubble plot showing the abundance of the bacterial community at an order level based on the 16S rRNA gene in pepper-cultivated soil inoculated with strain KNU-07.

**Figure 4.** Bubble plot showing the abundance of bacterial taxa, where KNU-7 belongs, over eight weeks post-inoculation using 16S rRNA gene sequencing.

The results of the beta diversity analysis based on principal coordinate analysis (PCoA) at an OTU level revealed that the soil bacterial community compositions were separated over time in all treatments including control (Figure 5). More importantly, bacterial community compositions of the soil treated with a high concentration of KNU-07 were separated from non-inoculated control samples in the last three weeks (Figure 5). These test results were similar to non-parametric statistical analyses based on ADONIS and ANOSIM. The analysis confirmed that the beta diversity between KNU-07-bacterized and non-inoculated control samples was significantly (p < 0.05) separated six weeks post-inoculation (Table S3).

**Figure 5.** Principal coordinate analysis of 16S rRNA genes of total bacteria based on the Bray–Curtis similarity index at 97% identity (operational taxonomic unit level) for eight weeks (W1–W8). PCoA1 and PCoA2 explained 57% and 12% of the variance, respectively.

We employed the PICRUSt program to predict the function of the soil bacterial community based on 16S rRNA gene data (Figure 6). The PICRUSt functional analysis showed that pathways related to germination and sporulation were overrepresented before six weeks in samples that received an application of KNU-07 (1.0 <sup>×</sup> 107 cells g−<sup>1</sup> soil) (Figure 6). After six weeks, the pathways that were positively impacted by the application of KNU-07 (1.0 <sup>×</sup> 107 cells g−<sup>1</sup> soil) were energy metabolism and metabolism of cofactors and vitamins (Figure 6).

**Figure 6.** Predicted metabolic function from 16S rRNA gene sequences of soil bacterial community collected from KNU-07 bacterized and non-bacterized samples using PICRUSt and STAMP analysis before and after six weeks inoculation with the indicated concentrations of *B. thuringiensis* KNU-07. (**A**) Energy metabolism, (**B**) metabolism of cofactors and vitamins, (**C**) germination, and (**D**) sporulation. Non-inoculated control (Control). Non-inoculated plants served as controls. Mean values having different letters in each parameter are significantly different (p ≤ 0.05).

#### *3.4. Tracking of KNU-07 Population Using qPCR*

The results of qPCR data showed soil treated with KNU-07 at higher concentrations had the highest abundance of KNU-07 throughout the experiments. As expected, KNU-07 cells were not detected in any non-inoculated control soil at any time (Figure 7). The abundance of KNU-07 decreased over time, regardless of the initial concentration of the KNU-07 inoculum (Figure 7). KNU-07 cells were detected long after inoculation (six weeks) from soils initially inoculated with a high concentration of KNU-07 (1.0 <sup>×</sup> <sup>10</sup><sup>7</sup> cells g−<sup>1</sup> soil). However, KNU-07 cells were detected in soil initially inoculated with a low concentration of KNU-07 (1.0 <sup>×</sup> 105 cells g−<sup>1</sup> soil) only within 3 weeks of inoculation (Figure 7). After eight weeks of inoculation, KNU-07 cells were not detected in any sample. We also investigated the total bacteria population using the 16S rRNA gene to determine whether there was a decrease in the total bacteria population, as was observed for KNU-07. The results of qPCR data showed that the abundance of total bacteria in all samples, including controls, increased slightly over time (Figure 7).

**Figure 7.** Abundance of (**A**) *B. thuringiensis* KNU-07 based on unique sequence copies and (**B**) total bacteria based on 16S rRNA gene copies in the soil over eight weeks post-inoculation using qPCR.

#### **4. Discussion**

In this study, we assessed the effect of KNU-07 on the growth of pepper plants and the soil bacterial community and designed a strain-specific primer pair to track the population dynamics of KNU-07 in soil using a qPCR-based method. Similar to the results of our in vitro assays, IAA production, siderophore production, phosphate solubilization, and urease activity by several strains of *B. thuringiensis* have been previously reported [12–14]. An increase in IAA production in the presence of L-tryptophan may attributed to the nature of the strain to utilize L-tryptophan as a physiological precursor [35]. Our in vivo assays also showed that KNU-07 promoted the growth of pepper plants after inoculation into the soil. Strains of *B. thuringiensis* have been used to promote the growth of plants, and our findings are consistent with these reports [36–38]. Previous studies reported that PGPR strains possessing siderophore production play a great role in helping plants to acquire iron for plant growth [13]. In addition, IAA production, phosphate solubilization, and urease activities play important roles in enhancing nutrient and water uptake and thereby enhance plant growth [12,14].

The change in soil bacterial community structure due to the presence of KNU-07 was less visible before six weeks post-inoculation. However, the community structure was separated after six weeks. Ke et al. [39] reported that the inoculation of soil with *Pseudomonas stutzeri* A1501 significantly changed the indigenous soil bacterial community structure after 2 months of inoculation, and our findings are consistent with this report. Similarly, Wang et al. [40] discussed the significant effect of bio-inoculants on soil microbial communities. In this study, the abundance of the Ellin 329 and Rhizobiales orders were higher in all samples. This may be due to the loss of Acidobacteriales [41,42]. There was also a change in soil microbial community structure over time. In our study, KNU-07-bacterized plants

exhibited superior growth relative to control plants. Plant age has been reported to influence the dynamics of the soil microbiome [43,44], and our findings are consistent with these reports.

Predicting the function of the total bacterial community provides information about its interaction with the surrounding environment [45]. Hence, we employed PICRUSt to predict changes in the function of the soil microbiota due to KNU-07 inoculation. Several metabolic pathways that facilitate growth in plants were overrepresented following application KNU-07 (1.0 <sup>×</sup> 107 cells g−<sup>1</sup> soil). He et al. [46] reported that rhizobacteria inoculation had beneficial effects on the function of the bacterial community. Predicted metabolic functions related to sporulation and germination were significantly affected during the first week after inoculation with KNU-07 at the highest concentration. The elevated abundance of predicted genes related to sporulation and germination might arise from the inoculated KNU-07, which belongs to the Bacillales order. Sporulation is a survival mechanism of *Bacillus* spp. in response to unfavorable environmental conditions [47]. More importantly, after six weeks post-inoculation, KNU-07 pathways related to energy metabolism and the metabolism of cofactors and vitamins were found to be overrepresented. This might give the pepper plants growing in inoculated soil better nutrition and plant growth [39,41,48].

Quantifying the abundance of a microbial inoculant in the soil is one of the best strategies for tracking [20,24]. Tracking helps to investigate the potential of inoculated microbes because PGPR is based on their persistence in the soil. Tracking bio-inoculants in the soil has been performed using different methods, including dilution plating and microscopy [49,50]. However, such methods can be laborious, time-consuming, and limited to sterile conditions [51]. Interestingly, a few recent studies proposed the possibility of tracking bacterial populations in field samples by using strain-specific primers in qPCR-based protocols [24,52]. To the best of our knowledge, this study is the first report of monitoring *B. thuringiensis* abundance in non-sterile soil using a single strain-specific primer pair in a qPCR-based method. The abundance of KNU-07 was relatively stable during the first two weeks post-inoculation and decreased over time regardless of the initial KNU-07 concentration. Coy et al. [53] reported that the population of *Bacillus sphaericus* drastically declined after six weeks post-inoculation. The bacterial population of antagonistic bacteria has also been reported to decline over time [54]. These decreases in the abundance of soil bio-inoculants might be attributed to physical and biological factors found in the soil environment [55]. Another factor that might cause a decrease in the abundance of KNU-07 may be microbial competition [53,56]. In this study, the amplicon sequence data of 16S rRNAs revealed that there was a gradual increase in the abundance of the total bacterial over time.

The design of a strain-specific primer pair and being able to track the strain in the soil by qPCR offers important information about the fate of PGPR under non-sterile soil conditions, which is an important step in registering a microbe as a PGPR product. Nevertheless, further studies are needed to identify ways to increase the survival of KNU-07 under different soil environmental conditions.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/4/551/s1, Figure S1: Effectiveness of the single primer pair for specific detection of *B. thuringiensis* KNU-07. (A) Lane M: Doctor protein 1 kb plus ladder, lane 1: negative control, lane 2: KNU-07, lane 3–30: different bacterial strains samples (Table S1). (B) Lane M: Doctor protein 1 kb plus ladder, lane 1: negative control, lane 2: KNU-07, lane 3–30: soil samples isolated from different locations (Table S2). Primer pair KNU07F/ KNU07R without template KNU-07 DNA served as the negative control, Figure S2. Potential of some *Bacillus* spp. (1–8) and KNU-07 (9) for indole acetic acid production (A), siderophore activity (B), urease activity (C) and phosphatase activity (D). *1* = *Bacillus licheniformis* KACC 10476, *2* = *Bacillus megaterium* KACC 10482, *3* = *Bacillus polymyxa* KACC 10485, *4* = *Bacillus subtilis* KACC 10854, *5* = *Bacillus pumilus* KACC 10917, *6* = *Bacillus macerans* KACC 11233, *7* = *Bacillus amyloliquefaciens* KACC 12067, *8* = *Bacillus velezensis* KACC 14004. Table S1: Bacterial strains used in this study, Table S2: Sources of soil samples used for in vitro PCR assays, Table S3. Statistical analysis of bacterial community structure at an operational taxonomic unit level in the last three weeks.

**Author Contributions:** Conceptualization: H.J. and J.H.S.; methodology: H.J. and H.Q.P; software: H.Q.P; validation: H.J., S.B.T., M.-C.K. and S.C; formal analysis: H.J. and J.-H.S.; investigation: H.J., S.B.T. and J.-H.S.; resources: G.-S.P. and J.-H.S.; data curation: H.J., S.-D.C., M.-J.K. and Y.-J.P.; writing (original draft preparation): H.J., H.Q.P. and S.B.T.; writing (review and editing): S.B.T. and J.-H.S.; visualization: H.J., S.B.T., J.C.I. and J.-H.S.; supervision: J.H.S.; project administration: H.J. and J.-H.S.; funding acquisition: J.-H.S. All authors have read and agreed to the published version of the manuscript.

**Acknowledgments:** This work was supported by the Strategic Initiative for Microbiomes in Agriculture and Food (Grant number 918010-4), Ministry of Agriculture, Food and Rural Affairs, Republic of Korea; and the Cooperative Research Program for Agriculture Science and Technology Development (Project number PJ013383), Rural Development Administration, Republic of Korea.

**Conflicts of Interest:** Author H.J. and G.-S.P. were employed by the company COSMAX BTI Inc. and Atogen Co., Ltd., respectively. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
