**Silicon Modulates the Production and Composition of Phenols in Barley under Aluminum Stress**

### **Isis Vega 1,2, Cornelia Rumpel 3, Antonieta Ruíz 4, María de la Luz Mora 2,4, Daniel F. Calderini <sup>5</sup> and Paula Cartes 2,4,\***


Received: 7 July 2020; Accepted: 31 July 2020; Published: 5 August 2020

**Abstract:** Silicon (Si) exerts beneficial effects in mitigating aluminum (Al) toxicity in different plant species. These include attenuating oxidative damage and improving structural strengthening as a result of the increased production of secondary metabolites such as phenols. The aim of this research was to evaluate the effect of Si on phenol production and composition in two barley cultivars under Al stress. Our conceptual approach included a hydroponic experiment with an Al-tolerant (Sebastian) and an Al-sensitive (Scarlett) barley cultivar treated with two Al doses (0 or 0.2 mM of Al) and two Si doses (0 or 2 mM) for 21 days. Chemical, biochemical and growth parameters were assayed after harvest. Our results indicated that the Al and Si concentration decreased in both cultivars when Al and Si were added in combination. Silicon increased the antioxidant activity and soluble phenol concentration, but reduced lipid peroxidation irrespective of the Al dose. Both barley cultivars showed changes in culm creep rate, flavonoids and flavones concentration, lignin accumulation and altered lignin composition in Si and Al treatments. We concluded that Si fertilization could increase the resistance of barley to Al toxicity by regulating the metabolism of phenolic compounds with antioxidant and structural functions.

**Keywords:** aluminum toxicity; antioxidant; barley; lignin; phenols; silicon

### **1. Introduction**

Silicon (Si) is a beneficial element that improves the growth, development and yield of plants subjected to different stresses [1–3]. However, the beneficial effects of Si depend on the capacity of plants to take up Si from the growth media, and transport it to the plant tissues [4]. To date, numerous studies have indicated that Si uptake and accumulation in plants are modulated by different influx and efflux transporters [5–8]. Most of the Si taken up by plants is deposited in the cell walls, where it increases mechanical strength [9,10].

In recent years, it has been suggested that Si can alter the secondary metabolism of plants [2,11–14]. In this regard, Si appears to stimulate the production of phenols in plants subjected to salinity, drought, temperature stress, UV radiation, cadmium, chromium [2], manganese [15], aluminum [16,17],

nickel [11], soil acidity [18] and biotic [12,19,20] stresses. There is some evidence showing that the positive effects of Si on phenol metabolism in plants growing in stressful environments is due to (i) the regulation of the gene expression or activity of key enzymes in the phenylpropanoid pathway [21,22], (ii) the enhancement of total phenol production [17,23], and/or (iii) the formation of complexes involving lignin and carbohydrates [24,25] or Si-polyphenol in the cell wall [15]. However, there is still no information about the impact of Si on the production and composition of phenolic compounds with either antioxidant capacity or structural action.

Barley is one of the most cultivated cereals around the world due to the high nutritional value of its grains, which provide complex carbohydrates, proteins, minerals, fiber and antioxidants, including phenols [26]. The main phenolic compounds in barley grains belong to the group of flavonoids (cyanidin-3-glucoside, petunidin-3-glucoside, delphinidin-3-glucoside) and phenolic acids (ferulic acid, *p*-coumaric acid, vanillic acid, sinapic acid), which benefit human health by reducing the risk of various diseases such as cancer and coronary heart diseases [27,28]. However, in acid soils, barley growth is limited due to its high sensitivity to Al3+, which reduces both the yield and quality of the grains. In this context, some reports have demonstrated an improvement in the Al tolerance of barley following Si addition [17,29,30].

For various other plant species, it has been suggested that Si attenuates Al phytotoxicity by means of (i) increasing the pH in the growth media, (ii) the formation of aluminosilicate complexes, (iii) enhancement of the chlorophyll and carotenoids content in plant tissues (iv) the stimulation of antioxidant enzyme activities and production of antioxidant compounds, and (vi) the exudation of phenolic compounds with Al chelation ability by plant roots [7,8,18,31–33].

Nevertheless, to the best of our knowledge only a few reports have described the effects of Si on the phenolic metabolism of barley subjected to Al stress. Moreover, the influence of Si on the phenolic metabolism of barley cultivars with contrasting Al tolerance has rarely been studied [17]. Therefore, the general objective of this research was to evaluate the effect of Si on phenol production and composition in tolerant and sensitive barley cultivars under Al stress. To address this objective, we carried out a hydroponic experiment with both types of cultivars grown under Al toxicity with and without Si addition and investigated the barley growth and the antioxidant as well as the structural phenol composition. We hypothesized that Si addition would improve barley's resistance to Al stress due to enhanced phenol production.

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

### *2.1. Plant Material and Growth Conditions*

Seeds of the barley cultivars Sebastian (Al-tolerant) and Scarlet (Al-sensitive) were germinated (10 days) on filter paper moistened with deionized water. After germination, 48 seedlings of each barley cultivar were transferred to plastic containers filled with 8 L of nutrient solution [34], and grown for 15 days under controlled conditions. Thereafter, Al and Si were applied in combination according to the following treatments: −Al/−Si (0 mM Al and 0 mM Si; control), +Al/−Si (0.2 mM Al and 0 mM Si), −Al/+Si (0 mM Al and 2 mM Si), +Al/+Si (0.2 mM Al and 2 mM Si). These treatments were selected from our previous kinetic study concerning the effect of Si on barley under Al stress [17]. The nutrient solutions were replaced every 5 days, and the pH was adjusted (HCl or NaOH) daily to 4.5. For the experiment, barley cultivars were arranged in a factorial design with three replicates per treatment. Plants were harvested 21 days after the initiation of the experimental treatments and subjected to chemical and biochemical analyses.

### *2.2. Plant Growth Traits and Chemical Analyses*

### 2.2.1. Growth Traits

Plant tissues (shoots and roots) were dried (65 ◦C for 48 h) to determine the dry weight (DW). Barley growth was determined by measuring the length of the longest root and the shoot of 10 plants randomly selected from each plastic container.

### 2.2.2. Aluminum and Si Concentration in Barley

The Al concentration in barley tissue (shoots and roots) was determined with the method described by Sadzawka et al. [35]. Briefly, dried samples were heated at 500 ◦C for 8 h, and treated with 2 M hydrochloric acid. The Al concentration was quantified by flame atomic absorption spectrophotometry (FAAS) at 324.7 nm. For Si concentration, dried shoots and roots (0.1 g) were digested with 5 mL of nitric acid (HNO3) at 70 ◦C for 5 h. Thereafter, 1 mL of hydrofluoric acid (HF, 40%) and 10 mL of deionized water were added, and left overnight. The next day, the solutions were treated with 5 mL boric acid (H3BO3, 2% w/v), and the solution was made up to 25 mL by adding distilled water. Silicon concentration of the digested samples was determined by FAAS at 251.6 nm as described in Pavlovic et al. [36].

### *2.3. Biochemical Analyses*

### 2.3.1. Total Soluble Phenols in Plants

Total soluble phenols were determined in root and shoot samples according to the Slinkard and Singleton method [37] using Folin–Ciocalteu reagent. The standard curve was calculated using chlorogenic acid as standard, and the absorbance was measured spectrophotometrically at 765 nm.

### 2.3.2. Identification and Quantification of Phenolic Compounds in Barley

Barley roots and shoots (0.1 g) were milled with liquid nitrogen and macerated in methanol as described by Slinkard and Singleton [37]. Phenolic compounds were determined by high performance liquid chromatography with a diode array detector (HPLC-DAD) using a Shimadzu HPLC system (Tokyo, Japan) with a LC-20AT quaternary pump, a DGU-20A5R degassing unit, a CTO-20A oven, a SIL-20a automatic injector and an SPD-M20A UV-Vis diode spectrophotometer. Data were analyzed using Lab solutions (Shimadzu, Duisburg, Germany) for DAD analysis. Identification was performed by LC-MSD Trap VL, model G2445C VL with electrospray ionization (ESI-MS/MS) detectors (Agilent, Waldbronn, Germany); control and data analyses were carried out by the Agilent ChemStation (version B.01.03) data processing station and Agilent LC-MS Trap Software (version 1.3, Santa Clara, CA, USA). The chromatographic separation method (HPLC-DAD) for the determination of phenolic compounds used a Kromasil ClassicShell-2.5-C18 (4.6 × 100 mm, 2.5 μm) column and a C18 precolumn (Novapak; Waters, Milford, MA, USA; 22 × 3.9 mm, 4 μm) as reported by Santander et al. [38]. The samples were injected using water:acetonitrile:formic acid (92:3:5 v/v/v) and water:acetonitrile:formic acid (45:50:5 v/v/v) as A and B mobile phases, respectively, with an elution gradient between 6% and 50% B over 30 min at 0.55 mL min−<sup>1</sup> and 40 ◦C. Quantification was carried out by external calibration using chlorogenic acid for the roots and apigenin for the shoots as standards at 320 nm.

### 2.3.3. Antioxidant Capacity in Barley Plants

The antioxidant capacity of the roots and shoots was analyzed by the method described by Chinnici et al. [39] using DPPH (2,2-diphenyl-1-picrylhydrazyl) radical, and Trolox as standard. The absorbance of samples was measured in spectrophotometer at 515 nm.

### 2.3.4. Lipid Peroxidation Assay

Lipid peroxidation was assayed on fresh root and shoot samples by following the thiobarbituric acid reactive substances (TBARS) procedure reported by Du and Bramlage [40]. The absorbance of the samples was registered spectrophotometrically at 532, 600 and 440 nm.

### *2.4. Plant Stretching*

The creep rate of culm was measured as an index of plant stretching. The extension of the first 5 cm of culm was measured with a constant load extensometer as described by Perini et al. [41]. Briefly, the fresh culms were scraped with carborundum to break the cuticle and then were placed in hot water for 15 min. Subsequently, the tissue was inserted between two clamps under a constant tension of 10 g per 30 min. The extension was measured through the movement of the upper clamp, detected by an electronic sensor and recorded in a microcomputer. All extension tests reported here were repeated at least three times for each sample.

### *2.5. Lignin Accumulation and Composition in Plants*

To visualize the lignin distribution in the plant tissues, fresh roots and leaf sections were stained with Safranine O, and analyzed by Laser Scanning Confocal Microscopy (CLSM; Olympus FV1000, Arquimed, Tokyo, Japan) at λ emission/excitation of 543/590 nm according to the method described by Sant' Anna et al. [42]. The images were processed using Image Processing software (software FV10-ASW v0.200c; Arquimed).

A quantitative analysis of the total lignin composition (calculated as the sum of monomers vanillyl [V], syringyl [S] and cinnamyl [C]) was carried out by means of the alkaline cupric oxide (CuO) oxidation method proposed by Kögel and Bochter [43]. Briefly, 0.05 g of roots or shoots were oxidized in teflon vials for 2 h under N2. Thereafter, the CuO oxidation products were purified by acidification and solid phase extraction using a C18 inverted column. Samples were derivatized by the addition of BSTFA (N, O-Bis (trimethylsilyl) trifluoroacetamide) before being analyzed by gas chromatography. For the separation and quantification of the monomers (V, S and C), a HP 6890 gas chromatograph (Agilent Technologies, Santa Clara, CA, USA) equipped with a SGE BPX-5 column and a flame ionization detector (GC/FID) was used. Phenylacetic acid was used as an internal standard for quantification.

### *2.6. Data Analysis*

Experimental data were checked for normality by the Shapiro-Wilk test and for homogeneity of variance by the Levene test. Statistical differences of means (95% significance level) were analyzed using two-way (cultivar and treatment) analyses of variance (two-way ANOVA). Post hoc tests were performed with a Tukey-test to determine the explanatory variables independently when the ANOVA detected significant differences. For each data set, the standard deviation (SD) was also determined. In addition, the relationship between two response variables was analyzed through Pearson correlation at a significance level of 5%.

### **3. Results**

### *3.1. Plant Growth and Concentrations of Al and Si*

### 3.1.1. Plant Growth Traits

The interaction between cultivar and treatment had a significant effect on the growth traits of roots, but dry weight (DW) and length of shoots were not significantly affected by the interaction (Table 1). Aluminum toxicity led to a significant reduction in DW and the tissue length of the plants. For the Al-tolerant Sebastian cultivar, the DW of the shoots and roots decreased by about 30% when grown under Al toxicity (Table 1). Greater reductions were recorded for the Al-sensitive Scarlett cultivar

ranging from 36% for shoots to 52% for roots. However, DW was enhanced in Sebastian (roots) and Scarlett (roots and shoots) when Al and Si were supplied together. The root length of Sebastian was greater than that of Scarlett, and both cultivars showed strong diminution of root length under +Al/−Si compared to the control treatment (−Al/−Si; Table 1). In plants growing under Al supply, the addition of 2 mM Si improved root length by 10% (Sebastian) and 17% (Scarlett).

**Table 1.** Dry weight (DW) and length of roots and shoots of barley cultivars (Sebastian and Scarlett). Treatments were 0 or 0.2 mM Al, combined with 0 or 2 mM Si. Values represent the mean of three replicates per treatment ± SD.


Different letters indicate statistically significant differences (*p* ≤ 0.05) among treatments. The lower case letters represent significant differences between treatments for one barley cultivar. The upper case letters indicate significant differences between barley cultivars for the same treatment. When uppercase letters do not appear, no significant differences between the same treatment in different cultivars were found. The significance of the interaction between cultivar and treatment was determined through the *p*-values: n.s, not significant; \* *p* ≤ 0.05; \*\* *p* ≤ 0.01; \*\*\* *p* ≤ 0.001.

### 3.1.2. Aluminum and Si Concentration

We observed that the interaction between cultivar and treatment had a significant effect on the Al and Si concentration in shoots and roots (Table 2). Roots of barley treated with 0.2 mM Al accumulated <sup>5232</sup> <sup>±</sup> 417 mg of Al kg−<sup>1</sup> DW for the Sebastian cultivar and 5285 <sup>±</sup> 167 mg Al kg−<sup>1</sup> DW for the Scarlett cultivar. Sebastian shoots showed 3153 <sup>±</sup> 417 mg Al kg−<sup>1</sup> DW and 4876 <sup>±</sup> 581 mg Al kg−<sup>1</sup> DW was recorded for Scarlett shoots. However, Si addition decreased the Al concentration in roots by 45% (Sebastian) and 68% (Scarlett), while the Al reduction in shoots was 49% (Sebastian) and 42% (Scarlett) as compared to the Al treatment without Si addition (Table 2). On the other hand, the Si concentration in plant tissues increased when plants were exposed to 2 mM of Si as compared to the control (Table 2). However, shoot Si concentration decreased with Al supply in both barley cultivars. Thus, Al addition reduced Si concentration in the shoots by 47% (Sebastian) and 37% (Scarlett), and by 55% (Sebastian) and 13% (Scarlett) in the roots.

**Table 2.** Aluminum and silicon concentrations in roots and shoots of the two barley cultivars (Sebastian and Scarlett). Treatments were 0 or 0.2 mM Al, combined with 0 or 2 mM Si. Values represent the mean of three replicates per treatment ± SD.


**Table 2.** *Cont*.


Different letters indicate statistically significant differences (*p* ≤ 0.05) among treatments. The lower case letters represent significant differences between treatments for one barley cultivar. The upper case letters indicate significant differences between barley cultivars for the same treatment. When uppercase letters do not appear, no significant differences between the same treatment in different cultivars were found. The significance of the interaction between cultivar and treatment was determined through the *p*-values: n.s, not significant; \* *p* ≤ 0.05; \*\* *p* ≤ 0.01; \*\*\* *p* ≤ 0.001.

### *3.2. The E*ff*ect of Al and Si on Phenol Production and Antioxidant Performance*

### 3.2.1. Total Soluble Phenols and Phenolic Profile

In general, the phenol concentration in plant tissues was affected by the interaction between the barley cultivar and Al/Si treatment (Figure 1A,B). Silicon addition did not alter the total phenol concentration in the Sebastian shoots (Figure 1A). However, when plants were exposed to 0.2 mM Al, lower concentrations of soluble phenols were recorded. Conversely, Sebastian roots showed an increase in phenols when Si was applied (Figure 1B). On the other hand, for Scarlett shoots, the highest phenol concentrations were observed in the +Al/−Si treatment (Figure 1A). Similarly, total phenols increased by 24% in Scarlet as a consequence of Si addition to the growth media, and it increased by 57% in the +Al/+Si treatment (Figure 1B).

**Figure 1.** Total phenols (**A**,**B**), free radical scavenging activity (**C**,**D**) and lipid peroxidation (**E**,**F**) in

shoots and roots of barley cultivars. Treatments were 0 or 0.2 mM Al, combined with 0 or 2 mM Si. Different letters indicate statistically significant differences (*p* ≤ 0.05) among treatments. The lower case letters represent significant differences between treatments for same barley cultivar. The upper case letters indicate significant differences between barley cultivars for the same treatment. When uppercase letters do not appear, no significant differences between the same treatment were found for different cultivars.

Seven flavonoids (flavone-glucosides) were identified in barley shoots (Table 3), but only four compounds reached quantifiable levels: (1) isoorientin-7-O-glucoside (lutonarin), (2) apigenin-pentoxide-hexoside isomer 1, (3) isovitexin-7-O-[6-sinapoyl]-glucoside, and (4) isovitexin-7-O-[6-feruloyl]-glucoside (Figure 2C). Caffeoylquinic acid isomer, a phenolic acid belonging to the chlorogenic acid family, was detected in barley roots (Table 3 and Figure 2A).


**Table 3.** Identification of phenolics from barley extracts by using HPLC–DAD–ESI-MS/MS.

The interaction between cultivar and treatment also had a significant effect on lutonarin, apigenin-pentoxide-hexoside and isovitexin-7-O-[6-sinapoyl]-glucoside concentrations (Figure 2D–F), but isovitexin-7-O-[6-feruloyl]-glucoside concentration was not affected by the interaction (Figure 2G). There were no differences in the lutonarin concentration in the shoots of Scarlett among treatments (Figure 2D). However, it decreased in all treatments for Sebastian shoots with respect to the control (Figure 2D). Similarly, apigenin-pentoxide-hexoside and isovitexin-7-O-[6-sinapoyl]-glucoside decreased in the shoots of both cultivars with the application of Si alone or in combination with Al (Figure 2E,F). Moreover, isovitexin-7-O-[6-feruloyl]-glucoside significantly increased in Scarlett shoots when the plants were exposed to the Al treatment (Figure 2G). In contrast, Sebastian plants treated with Si showed a decrease in the concentration of this phenol, irrespective of Al addition (Figure 2G).

In addition, we observed that the interaction between cultivar and treatment significantly affected the caffeoylquinic acid isomer (CQA) concentration. In the roots, a higher concentration of CQA was found for Sebastian compared to Scarlett in all treatments (Figure 2B). Sebastian showed a decrease in its concentration compared to control when plants were treated with Si and Al alone or in combination. Scarlett did not show any difference in the CQA concentration among the different treatments.

### 3.2.2. Radical Scavenging Activity

The interaction between cultivar and treatment significantly affected radical scavenging activity in shoots and roots (Figure 1C,D). In fact, radical scavenging activity in plant tissues of both cultivars increased following the addition of 0.2 mM Al compared to the control. A further increase was found in roots and shoots of barley cultivars simultaneously supplied with Al and Si. For Sebastian, the highest antioxidant capacity was observed in the roots (Figure 1D) of plants treated with Al and Si, showing a 22-fold increase as compared to the control. For Scarlett, the highest antioxidant capacity was also observed in shoots (Figure 1C) and roots (Figure 1D) as a consequence of the simultaneous addition of Si and Al, with a 13-fold and 1.7-fold increase, respectively, as compared to the control.

**Figure 2.** HPLC-DAD chromatograms (**A**,**C**) and individual phenolic concentration in shoots (**D**–**G**) and roots (**B**) of barley cultivars. Treatments were 0 or 0.2 mM Al, combined with 0 or 2 mM Si. Phenolic concentration values represent the mean of three replicates per treatment ± SD. Different letters indicate statistically significant differences (*p* ≤ 0.05) among treatments. The lower case letters represent significant differences between treatments for the same barley cultivar. The upper case letters indicate significant differences between barley cultivars for the same treatment. When uppercase letters do not appear, no significant differences between the same treatment in different cultivars were found.

### 3.2.3. Oxidative Damage

Lipid peroxidation was not significantly affected by the interaction between cultivar and treatment, but a significant effect of Al/Si treatments on oxidative damage was observed. Barley cultivars showed an increase in lipid peroxidation as a result of the application of 0.2 mM Al. For Sebastian this increase was about 44% in shoots and 29% in roots, whereas for Scarlett the oxidative damage increased by 28% in shoots and 57% in roots in Al-treated plants (Figure 1E,F). By contrast, lipid peroxidation decreased in Sebastian shoots (Figure 1E) when Si was applied alone (38%) or in combination with Al (35%). Likewise, in roots, this reduction was about 75% and 60%, respectively (Figure 1F). Similarly, a reduction in lipid peroxidation was observed in Scarlett following Si addition (29% in shoots and 60% in roots) compared to control. Moreover, plants supplied with 0.2 mM Al and 2 mM Si showed a reduction in lipid peroxidation of about 37% in shoots and 68% in roots, compared to plants exposed to 0.2 mM Al (Figure 1E,F).

### *3.3. Silicon Influence on Plant Structure*

### 3.3.1. Plant Stretching

We measured hypocotyl stretching by using an extensometer to evaluate cell wall creep in culms of plants cultivated under the different experimental treatments. Barley culm stretching was significantly influenced by the interaction between cultivar and treatment (Figure 3). For both cultivars, when Al was applied alone, greater stretching was evidenced compared to the control. However, Si addition decreased the stretching independent of added Al, thus improving the strength of the tissues.

**Figure 3.** Stretching culm of the two barley cultivars in all treatments (0 or 0.2 mM Al, combined with 0 or 2 mM Si). Values represent the mean of three replicates per treatment ± SD. Different letters indicate statistically significant differences (*p* ≤ 0.05) among treatments. The lower case letters represent significant differences between treatments for the same barley cultivar. The upper case letters indicate significant differences between barley cultivars for the same treatment. When uppercase letters do not appear, no significant differences between the same treatment in different cultivars were found.

### 3.3.2. Lignin Content and Composition

Sebastian and Scarlett shoots and roots showed greater lignin accumulation when Al or Si was applied (Figure 4). The highest accumulation was observed in both cultivars under the +Al/+Si treatment.

**Figure 4.** Visualization of lignin contents (green color) in barley roots and shoots of Sebastian and Scarlett cultivars, harvested after 21 days. Treatments were 0 or 0.2 mM Al, combined with 0 or 2 mM Si. The detection of safranine fluorescence was expressed as relative fluorescence unit (RFU). Values represent the mean of three replicates per treatment ± SD.

The effect of Al and Si on the total lignin calculated as the sum of monomers vanillyl (V), syringyl (S) and cinnamyl (C) and its composition were determined. The interaction between cultivar and treatment had a significant effect on lignin monomers and total lignin concentration of shoots and roots. Accordingly, Sebastian shoots showed an increase in cinnamyl and total lignin concentration (Figure 5A) as a consequence of Al addition. In the roots, the lignin concentration was reduced in the +Al/+Si treatment (Figure 6A). For Scarlett, increased concentrations of vanillyl phenols (shoots), cinnamyl phenols (roots) and total lignin (roots) were recorded in the +Al/+Si treatment compared to control (Figures 5B and 6B).

Sebastian shoots showed an increased cinnamyl:vanillyl (C/V) ratio when Al was supplied (Figure 5C), whereas the syringyl:vanillyl (S/V) ratio of roots exhibited higher values for the +Al/+Si treatment (Figure 6C). In contrast, the C/V (shoots) and S/V (roots) ratios of Scarlett were enhanced by 2 mM Si. Moreover, the C/V ratio of shoots was reduced when 0.2 mM Al and 2 mM Si were added, whereas it was increased in roots (Figures 5D and 6D). The acid to aldehyde ratio of vanillin in Sebastian shoots was enhanced by Si supply, irrespective of the Al dose (Figure 5E). Similarly, Scarlett roots in the +Al/−Si treatment increased the acid to aldehyde ratio of vanillin (Figure 6F). Conversely, plants treated with −Al/+Si reduced the acid to aldehyde ratio of syringyl (Figure 6F).

**Figure 5.** Lignin parameters of the shoots of Sebastian and Scarlett barley cultivars. The main lignin groups (**A**,**B**), comprising vanillyl [V], syringyl [S] and cinnamyl [C] phenols, and total lignin concentration [S + V + C]; their ratios (**C**,**D**); and acid to aldehyde ratios (**E**,**F**) of vanillyl [(Ac/Al)v] and syringyl [(Ac/Al)s]. Treatments were 0 or 0.2 mM Al, combined with 0 or 2 mM Si. Values represent the mean of three replicates per treatment ± SD. Different letters indicate statistically significant differences (*p* ≤ 0.05) among treatments. The lower case letters represent significant differences between treatments for the same barley cultivar. The upper case letters indicate significant differences between barley cultivars for the same treatment. When uppercase letters do not appear, no significant differences between the same treatment in different cultivars were found.

**Figure 6.** Lignin parameters of the roots of Sebastian and Scarlett barley cultivars. The main lignin groups (**A**,**B**), comprising vanillyl [V], syringyl [S] and cinnamyl [C] phenols, and total lignin concentration [S + V + C]; their ratios (**C**,**D**); and acid to aldehyde ratios (**E**,**F**) of vanillyl [(Ac/Al)v], and syringyl [(Ac/Al)s]. Treatments were 0 or 0.2 mM Al, combined with Si (0 or 2 mM). Values represent the mean of three replicates per treatment ± SD. Different letters indicate statistically significant differences (*p* ≤ 0.05) among treatments. The lower case letters represent significant differences between treatments for the same barley cultivar. The upper case letters indicate significant differences between barley cultivars for the same treatment. When uppercase letters do not appear, no significant differences between the same treatment in different cultivars were found.

### **4. Discussion**

It is well known that the effects of Al on plant growth vary markedly among plant species and cultivars [44]. In this respect, the Scarlett cultivar was more sensitive to Al than the Sebastian cultivar, and showed a higher Al concentration in the roots and shoots and a much lower plant dry weight (DW) than Sebastian (Tables 1 and 2). These findings agree with the greater Al-tolerance of Sebastian, and with earlier research showing different Al sensitivity for both barley cultivars in the short-term [17]. On the other hand, Si supply increased root DW in Sebastian and Scarlett cultivars. Nevertheless, both Si and Al concentrations were reduced in plant tissues when Al and Si were added simultaneously. Thus, our findings confirmed the improvement in plant growth and Al detoxification in plant tissues due to Si addition, which is in agreement with similar studies of other plant species [8,18,33,45,46].

In previous investigations, Si resulted in an improvement in the antioxidant system of plants subjected to abiotic stresses [1–3,11,12,14,47]. Under Si supply, different biochemical responses during Al exposure such as an increment in antioxidant compounds (e.g. phenols compounds, vitamins) and enzyme activities have been reported [30,48–51]. We observed an increase in both soluble phenol concentration and free radical scavenging activity in roots exposed to the +Al/+Si treatment (Figure 1B,D). This increase was accompanied by a reduction in lipid peroxidation (Figure 1F), which was more evident in the Al-sensitive (Scarlett) than in the Al-tolerant (Sebastian) cultivar. These responses could be at least partially associated with either the mitigation of Al stress through Al chelation by flavonoids at the cell wall level [16,52] or the incorporation of soluble phenols into the lignin biosynthetic pathway as demonstrated by the increase in lignin content under Al supply (Figure 4).

Additionally, we identified caffeoylquinic acid isomer (CQA), which belongs to the group of chlorogenic acids, in the roots of both barley cultivars (Figure 2A). In this context, chlorogenic acids function as intermediates in the lignin biosynthesis pathway, and they are regarded as powerful antioxidant compounds [53]. In all treatments, a higher CQA concentration was found in Sebastian roots as compared to those of Scarlett (Figure 2B). Nevertheless, Sebastian showed a reduction in CQA concentration compared to the control when plants were treated with either Si or Al alone or in combination, whereas in Scarlett the decrease in CQA was only found in the +Al/+Si treatment. Since chlorogenic acids are one of the main building blocks of lignin, the decrease of CQA due to the combined application of Si and Al might be associated with the greater accumulation of lignin in the roots (Figure 4).

The phenolic compounds identified in shoots belong to the group of flavonoids, specifically flavone-glucosides (i.e., compounds (1) to (7) mentioned above). Briefly, lutonarin, apigenin derivate (apigenin-pentoxide-hexoside) and saponarin derivatives (isovitexin-7-O-[6-sinapoyl]-glucoside, isovitexin-7-O-[6-feruloyl]-glucoside) were identified and quantified (Figure 2C). These compounds were similar to those already described for barley shoots [54,55]. In our study, the concentration of lutonarin and apigenin-pentoxide-hexoside was decreased in Sebastian shoots in all treatments compared to control, but no changes in lutonarin were observed in Scarlett (Figure 2D,E). However, the concentration of isovitexin-7-O-[6-feruloyl]-glucoside was increased 2.7-fold by Al addition (Figure 2G). Despite the reduction in flavones as a result of Al or Si addition, high antioxidant activity by flavonoids such as saponarin and lutonarin has been reported in barley shoots [56].

On the other hand, increased lignin accumulation in shoots following Al and Si supply was observed for both cultivars (Figure 4). In fact, a reduction in culm stretching after Si addition in barley plants confirmed there was an improvement in lignin accumulation in shoots irrespective of Al addition (Figure 3). In the Al-sensitive cultivar (Scarlett) total lignin concentration in roots, quantified as the sum of the monomers (V + C + S), was improved by Si addition under Al stress. This increase is in agreement with the higher intensity of safranine staining (Figure 4), and supports previous findings showing that Si has a mitigating effect due to increased production of lignin under stressful conditions [19,22,57]. Such an effect may be associated with either increased hydrogen peroxide production or peroxidase activity in cell walls [58,59]. Likewise, it has been demonstrated

that Si increases the activities of enzymes such as peroxidase, polyphenol oxidase, and phenylalanine ammonia lyase, which are involved in the lignin biosynthesis pathway [18,60].

Moreover, differences in lignin composition have been observed under stress conditions [61–65], but little is known about the role that lignin with different compositions might exert in vascular plants [66]. While the monomer composition allows us to distinguish the origin of vegetation (i.e., angiosperms or gymnosperms), C/V and S/V ratios could be used as indicators of the origin and degradability of lignin [67,68]. Thus, the range of individual monomers (V, C, S) obtained here (Figure 5A,B and Figure 6A,B) agree with those reported in grasses [69]. In Scarlett roots, increased C/V and S/V ratios were observed in plants treated with Si, irrespective of the Al addition (Figure 6D). A similar trend in the S/V ratio in Sebastian roots was found due to the combined application of 0.2 mM Al and 2 mM Si (Figure 6C). In addition, Si decreased the acid to aldehyde ratio of V in Scarlett roots (Figure 6F). Higher proportions of C and S units indicate lower lignin stability and may influence biogeochemical cycling differently after it is returned to soil. Further studies under field conditions are needed to confirm this hypothesis.

Hence, Si fertilization in barley can be envisaged as a key strategy for counteracting Al toxicity. Changes in soluble phenols and lignin production/composition mediated by Si appear to be involved in improving the performance of barley cultivars, since an enhancement in root growth and plant antioxidant ability was observed when Si was supplied to Al stressed plants. Some hypotheses have been proposed to explain the possible linkage between Si and phenol metabolism. For example, Williams [70] proposed that OH groups of phenols are condensed with Si(OH) in biological systems, and Inanaga et al. [24] suggested that Si may be associated with lignin-carbohydrate complexes in the wall of epidermal cells. Silicon might also be involved in signal transduction pathways, thus inducing lignin production [21]. Despite the evidence regarding the impact of Si supply on the production of antioxidant or structural phenolic compounds, the mechanisms implicated in the modulation of phenolic metabolism by Si need to be investigated further.

**Author Contributions:** Data curation, I.V.; Formal analysis, I.V., C.R., A.R. and D.F.C.; Funding acquisition, P.C.; Methodology, I.V.; Supervision, P.C.; Visualization, I.V.; Writing—original draft, I.V.; Writing—review & editing, I.V., C.R., A.R., D.F.C., M.d.l.L.M. and P.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by FONDECYT projects 1161326 and 1201257.

**Acknowledgments:** The authors gratefully acknowledge the Institute of Ecology and Environmental Sciences of Paris, Institute of Plant Production and Protection of Universidad Austral de Chile, University of Castilla-La Mancha, Spain, Centro de Investigación en Micorrizas y Sustentabilidad Agroambiental and Scientific and Technological Bioresource Nucleus of Universidad de La Frontera, for providing access to specialized equipment used in phenolics, lignin and plant stretching analyses. The authors also thank the Maltexco company for contributing the barley seeds used in this research and Dirección de Investigación of Universidad de La Frontera. CR acknowledges the MEC-CONICYT project 80180025.

**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*

## **Improved Growth and Yield Response of Jew's Mallow (***Corchorus olitorius* **L.) Plants through Biofertilization under Semi-Arid Climate Conditions in Egypt**

**Ahmed Fathy Yousef 1,2,**†**, Mohamed Ahmed Youssef 3,**†**, Muhammad Moaaz Ali 1, Muhammed Mustapha Ibrahim 1,4, Yong Xu 5,6,\* and Rosario Paolo Mauro 7,\***


Received: 17 October 2020; Accepted: 13 November 2020; Published: 16 November 2020

**Abstract:** This study was conducted to comparatively assess the effects of fertilization typology (organic, inorganic, and biofertilization) on the growth, yield, and compositional profile of Jew's mallow. The experiment was carried out over two growing seasons, under semi-arid climate conditions on silty loam soil. We adopted three fertilization strategies: (1) inorganic NPK fertilizer (146, 74, and 57 kg ha−<sup>1</sup> for N, P2O5, and K2O, respectively), (2) farmyard manure (36 m<sup>3</sup> ha−1), and (3) a biofertilizer (a set of mixed cultures of *Bacillus* spp., *Candida* spp., and *Trichoderma* spp. at 36 L ha−1). Treatment combinations were control (without fertilization, T1), NPK fertilizer (T2), farmyard manure (FYM, T3), biofertilizer (T4), NPK+biofertilizer (T5), and FYM+biofertilizer (T6). The T5 treatment maximized both plant and leaf biomass (up to 31.6 and 8.0 t ha−1, respectively), plant height (68.5 cm), leaf area (370 cm m−2), leaf protein content (18.7%), as well as N, P, and K concentration in leaves (2.99, 0.88, and 2.01 mg 100 g<sup>−</sup>1, respectively). The leaves' weight incidence was lower in T5 treatment (36.7%) as compared to the unfertilized plants (T1). The results revealed that the combined application of inorganic NPK plus biofertilizer is most beneficial to increase growth, yield, and nutrient accumulation in Jew's mallow plants.

**Keywords:** leafy vegetable; mineral nutrients; soil structure; chlorophyll content; cation exchange capacity

### **1. Introduction**

Over the years, mineral fertilizers have helped agriculture enhance crop productivity to meet the ever-increasing demand for food. However, the overutilization of inorganic fertilizers poses a negative impact on the environment and soil functioning and fertility [1]. Moreover, it leads to the high cost of crop production. Therefore, many researchers have tried to restore soil fertility through the use of organic materials of plant or/and animal origin, in the forms of organic fertilizers. Organic fertilization involves the use of naturally occurring material that includes animal manures and agricultural residues [2]. These materials have been proposed to boost the supply of inorganic nutrients, which can bridge fertilizer demand due to economic and environmental reasons [3]. Organic manure increases the status of soil nutrients via the gradual release of minerals to the soil as well as enhancing its physical, biological, and chemical properties [4,5]. Also, organic manure has been shown to improve the agronomic performances of many crops [6].

Biofertilizers are substances containing living organisms and organic materials that can be utilized to increase soil nutrients availability and promote plant growth and productivity. They are also considered an eco-friendly way toward sustainable agriculture because they do not cause pollution [7,8]. Biofertilizers have become a preferable alternative or supplement to organic and inorganic fertilizers. Therefore, to increase soil productivity, the utilization of biofertilizers has become increasingly important, because they help in stimulating plant growth hormones, thereby enhancing nutrients uptake and increasing tolerance towards several abiotic stressors too [9]. Biofertilizers can be applied to seeds, soils, rhizosphere, or plant surfaces. Moreover, they are less costly and sometimes more effective as compared to inorganic fertilizers [10–12] Jew's mallow (*Corchorus olitorious* L.) belongs to the Malvaceae (Tiliacea) family and classed in the genera of about 40–100 species of the flowering plants [13]. It is also known as jute mallow in English and called Mulukhiyah in Egypt. The leaves are edible either fresh, dried, or frozen by many Egyptians because it is a quite cheap vegetable and forms part of the national Egyptian dishes [8]. It is one of the popular tropical green leafy vegetables of great importance in most countries in the Middle East and Latin America [14], Africa, and Asia [15].

Jew's mallow is a source of income for smallholders and poor families in Egypt, farmers cultivate Jew's mallow in many marginal areas. They use their seeds, which consequently result in genetic diversity in Jew's mallow distribution in Egypt [16]. Recently, Jew's mallow, which is a neglected and underutilized crop species (NUS), has received great international recognition because of its role in providing food and nutrition security and income opportunities among smallholder farmers. Moreover, NUS can be utilized to adapt agriculture and food systems to climate change [17]. Jew's mallow plays an important role in humans nutrition because its leaves contain an average 13–15% dry matter, 4.7 mg vitamin A, 259–266 mg Ca, 250–261 Mg, 4.5–8 mg Fe, 4.8–6 g protein, 92 μg foliates, 105 mg ascorbic acid, 1.5 mg nicotinamide, 0.9 g folic acid, 0.7 g oil, 5 g carbohydrate, and 1–5 g fiber per 100 g of edible leaves [13,18]. Additionally, the seeds of *C. olitorius* can be integrated into livestock feeds and human diets [19].

Jew's mallow performs well in marginal areas, even without the addition of organic and/or inorganic fertilizers, as well as under fertilized conditions, especially with application of N [20]. In this regard, Olaniyi and Ajibola [21] found that the use of N, P, and K fertilization significantly increased plant height, fresh shoots biomass, number of leaves, and dry matter content of Jew's mallow above the control (no fertilization). Thus, it is concluded that the yields and growth of the crop could significantly be improved by soil application of N, P, and K fertilizers at the optimum rate of 45, 30, and 20 kg ha−1, respectively. Also, Aisha, et al. [22] found that application of 70% (100, 100, and 80 kg ha−<sup>1</sup> NPK, respectively) of inorganic fertilizer recommended rate on spinach plants gave rise to the longest harvest period, the highest total weight of leaves and its various organs and improve leaves nutritional values, including N, P, K, and protein contents. However, using biofertilizers in Jew's mallow cultivation has not received adequate attention, whether singly or in integrated use with organic and inorganic fertilizers. Similarly, the effects of these combinations on the nutrient uptake require proper understanding and documentation, which is still lacking in the reported literature. Therefore, this study aimed to assess the bio-agronomical response of Jew's mallow to the combined soil incorporation of organic, inorganic and biofertilization, so checking the possibility to obtain a more sustainable fertilization technique for the crop.

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

### *2.1. Experimental Site*

A two-year field experiment was carried out under semi-arid climate conditions on silty loam soil at the Research Farm of the College of Agriculture, Al-Azhar University, Assiut branch. The location is (27◦12 16.67 N; 31◦09 36.86 E) in Assiut governorate, Egypt. Table 1 shows some physical and chemical properties of the soil at the experimental site, collected at a depth of 0–30 cm and analyzed as described by [23].


**Table 1.** Some physical and chemical properties of the experimental soil in 2017 and 2018.

Each value represents a mean of three replicates. E.C: electrical conductivity; O.M: organic matter.

### *2.2. Experimental Design and Treatments*

Treatments were laid out using a randomized blocks design with three replications. Each plot unit included a totally flat area of 10.5 m2. The seeds of Jew's mallow were sprinkled on 25 March 2017 and 2 April 2018 for each season, respectively. The irrigation of experimental units was immersed-way once per 10 days, as per local custom. Weeds were removed manually at 20 and 40 days after sowing (DAS) in both growing seasons, before irrigation was affected. The treatments application comprised three fertilization types (alone or in combinations), namely an organic fertilization (farmyard manure, FYM), an inorganic NPK fertilization, and a biofertilizer. The organic fertilizer was obtained from the animal Production Farm, College of Agriculture, Al-Azhar University, Assiut, and was incorporated into the soil during plowing at the recommended dose of 36 m<sup>3</sup> ha<sup>−</sup>1. Its chemical composition was reported by Silva [24] and presented here in Table 2. For the inorganic fertilization, the recommended P2O5 dose of 74 kg ha−<sup>1</sup> (as Ca super phosphate) was incorporated into the soil during plowing, while 146 kg ha−<sup>1</sup> (as urea) and 57 kg ha−<sup>1</sup> K2O (as potassium sulfate) were divided in two equal applications at 10 and 20 DAS, as commonly used for growing Jew's mallow plants, recommended by the Ministry of Agriculture [25]. The liquid biofertilizer (T.S) contains of molasse as organic material carrier of microorganisms, and a set of mixed cultures of *Bacillus circulans*, *B. poylmyxa, B. megatherium,* *Candida* spp., and *Trichoderma* spp., whose amount in terms of living cells was <sup>&</sup>gt;0.5 <sup>×</sup> 109 cfu ml<sup>−</sup>1, <sup>&</sup>gt;<sup>2</sup> <sup>×</sup> <sup>10</sup><sup>7</sup> cfu ml<sup>−</sup>1, >1.5 <sup>×</sup> 10<sup>9</sup> cfu ml−1, >1.5 <sup>×</sup> 107 cfu ml−<sup>1</sup> and >0.5 <sup>×</sup> 106 cfu ml−1, respectively. The biofertilizer was added at 36 L ha−<sup>1</sup> with irrigation in three equal doses at 20, 30 and 40 DAS. The biofertilizer was obtained from the directorate of Agriculture in Assiut. Overall, the trials comprised an unfertilized control (T1), inorganic NPK fertilization (T2), farmyard manure (FYM) (T3), biofertilizer (T4), inorganic NPK+biofertilizer (T5), and FYM+biofertilizer (T6).

**Table 2.** Chemical composition of farmyard manure used in the experiments on dry weight basis.


EC: Electrical Conductivity.

### *2.3. Data Collection*

Data were collected using plant samples from 0.5 m2 in the middle of each experimental unit. Plant height was taken from the base of the rhizome to the top of the plant using a ruler. The fresh biomass of total plants, fresh weight of leaves, and dry weight of leaves was weighed using an electronic balance (0.01 g). Fresh biomass of total plants and fresh weight of leaves were put in paper bags and transferred to a drying oven at 70 ◦C until constant weight to obtain the dry weight. Leaf area was estimated as described by Pandey and Singh [26], whereas leaf weight incidence, expressed on a percentage basis, was calculated by using the following Equation (1).

$$\text{Leaves drag} \left( \% \right) = \frac{\text{leaves dry weight (g)}}{\text{plant dry weight (g)}} \times 100\tag{1}$$

Harvesting was done in the two seasons at 28-May and 5-Jun, respectively. The following soil properties were determined after harvest: cation exchange capacity (CEC) and organic-C, determined according to Clark, et al. [27]. The soil bulk density was calculated by using Equation (2).

$$\text{Soil bulk density} = \frac{\text{Dry weight of bulk sample (g)}}{\text{the volume of soil core (cm3)}} \tag{2}$$

Leaf samples from each experimental unit during two seasons were collected, the fifth leaf from the top of 20 plants after 65 DAS (the first season) and 62 DAS (second season), and washed three times with distilled water, before chemical analysis. N-content in leaves was determined using the Kjeldahl procedure according to Motsara and Roy [28]. P-content in leaves was determined by the colorimeter method (ammonium molybdate) using a JENWAY 6305 UV/Visible Spectrophotometer at 643 nm (OD643) [28]. K-content in leaves was determined photometrically using a Flame Photometer (BWB Model BWB-XP, 5 Channel) as described by Motsara and Roy [28]. Protein content in leaves (expressed on a percentage basis) was calculated as N content (%) X 6.25. Leaf chlorophyll content was determined using a mobile chlorophyll meter (SPAD-502-m Konica Minolta, Inc., Tokyo, Japan). Before taking the readings, the performance of the chlorophyll meter was calibrated according to the manufacturer' instructions. At the measurement date, 6 readings from each replicate were taken at 65 DAS (the first season) and 62 DAS (second season), using the youngest fully expanded leaves.

### *2.4. Statistical Analysis*

All data collected were subjected to one-way analysis of variance (ANOVA) using SPSS statistical software package version 16.0 (SPSS Inc., Chicago, IL, USA). Significantly different means were separated using Duncan's multiple range test at the *p* ≤ 0.05 level of probability [29]. Mean values were presented as mean ± SD.

### *2.5. Weather Condition during the Experiment*

During the first year of experiment (2017), the average mean temperature was 26.3 ◦C, with a gradual increase from 3 April (17.3 ◦C) to 15 May (29.7 ◦C), whereas average minimum and maximum temperatures fluctuated between 11–28.4 ◦C and 24.6–40.8 ◦C, respectively. The average relative humidity varied between 20% and 60%, with the lowest value recorded at 20 April and the highest one at 13 April (Figure 1). During the second year of experiment (2018), the average mean temperature was 28.6 ◦C, with a gradual increase from 20.8 ◦C to 31.5 ◦C, whereas average minimum and maximum temperatures fluctuated between 12–28 ◦C and 20.8–46 ◦C, respectively. The average relative humidity varied between 16% and 58%, with the lowest value recorded at 6 May and the highest one at 6 April (Figure 1).

**Figure 1.** Weather conditions during the two growing periods of Jew's mellow cultivation.

### **3. Results**

### *3.1. Growth Variables*

Tables 3–5 show the effect of organic, inorganic, and biofertilizers supplementations on the growth and yield of Jew's mallow plants. The results showed no significant difference in the plants height under T2, T3, T5, and T6, but they were higher than those under the other treatments in the mean of both growing seasons (Table 3). There were statistically significant differences between the treatments, where the maximum fresh plants weight, leaves fresh weight, plant dry weight, leaves dry weight, and leaves area (3.16 kg.m−2, 797.88 g.m−2, 646.79 g.m−2, 223.35 g.m−2, and 369.5 cm2.m−2, respectively) were showed by the plants treated with NPK with biofertilizers (T5), and that of without fertilization, T1 treatment gave the lowest values (Tables 3 and 4) in the mean of both growing seasons.

The highest leaves weight incidence was observed in plants treated with biofertilizer (T4) having non-significant difference among FYM+biofertilizer (T6), NPK fertilizer (T2), and T1 during first growing season, while in second growing season the plants treated with T1 showed significant highest values (Table 5). There was no significant difference in dry matter content of plants under all the treatments except T1, but T4 was higher than other treatments in the first season, while in the second season, there was no significant difference in dry matter contents of plants in T1, T4, T5, and T6 treatments, but T4 was highest compared to other treatments in the mean of both growing seasons.

### *3.2. Compositional Variables*

As shown in Tables 3–5, the accumulation of protein in Jew's mallow plants in both seasons was the significantly highest under NPK with biofertilizers (T5). Data presented in Table 6 shows that the average accumulation of N, P, and K in leaves were under T5 higher than other treatments in both growing seasons, while T1 gave the least N, P, and K accumulation.


**Table 3.** Effect of organic, inorganic, and biofertilizers on plant growth characteristics (plant height, plant fresh weight, and leaves fresh weight) of Jew's mallow

Values are means of three replicates; different letters in the same column indicate significant differences according to Duncan's multiple range test at *p* ≤ 0.05. Where without(T1), NPK fertilizer (T2), farmyard manure (T3), biofertilizer (T4), NPK+biofertilizer (T5), and Farmyard manure +biofertilizer (T6).

**Table 4.** Effect of organic, inorganic, and biofertilizers on plant growth characteristics (plant dry yield, leaves dry yield, and leaf area) of Jew's mallow (*Corchorus olitorius* L.) plants.


Values are means of three replicates; different letters in the same column indicate significant differences according to Duncan's multiple range test at *p* ≤ 0.05. Where(T1),NPKfertilizer(T2),farmyardmanure(T3),biofertilizer(T4),NPK+biofertilizer (T5),andFarmyardmanure+biofertilizer(T6).

 without fertilization


**Table 5.** Effect of organic, inorganic, and biofertilizers on plant growth characteristics and chemical contents (leaves weight incidence, dry matter content,

Values are means of three replicates; different letters in the same column indicate significant differences according to Duncan's multiple range test at *p* ≤ 0.05. Where without (T1), NPK fertilizer (T2), farmyard manure (T3), biofertilizer (T4), NPK+biofertilizer (T5), and Farmyard manure +biofertilizer (T6).

**Table 6.** Effect of organic, inorganic, and biofertilizers on compositional variables (N-Content, P-Content, and K-Content) of Jew's mallow (*Corchorus olitorius* L.) plants.


Values are means of three replicates; different letters in the same column indicate significant differences according to Duncan's multiple range test at *p* ≤ 0.05. Where without fertilization(T1), NPK fertilizer (T2), farmyard manure (T3), biofertilizer (T4), NPK+biofertilizer (T5), and Farmyard manure+biofertilizer (T6).

and

The average leaf chlorophyll content the first season was higher than in the second season in all variants, where the highest value of leaf chlorophyll content 41.21 mg g−<sup>1</sup> was obtained in T5, while the lowest one (29.47 mg g<sup>−</sup>1) was recorded in T1 (Figure 2).

**Figure 2.** Effect of organic, inorganic, and biofertilizers on leaf chlorophyll content of Jew's mallow (*Corchorus olitorius* L.). Each column represents the mean of three replicates; different letters on similar columns indicate significant differences using Duncan's multiple range test at *p* ≤ 0.05.

### *3.3. Soil Properties at the End of the Experimental Period*

The result of soil properties after harvesting Jew's mallow plant showed that the soil was variably influenced by the different treatments. The average values for the soil organic-C (%) contents were influenced by the individual treatments. The highest value for soil organic-C, as shown in Figure 3a, was observed in T3 in both of seasons and in T6 in the 2018 season, while the lowest values were recorded in T1 and T2. As shown in Figure 3b, the application of organic manure with biofertilizers significantly enhanced the CEC value. The highest average values CEC were noticed in T6 (17.98 cmol kg−1), and T3 (17.93 cmol kg−1) which were statistically undifferentiated. The lowest value (15.72 and 16.01 cmol kg<sup>−</sup>1) were obtained in control (T1) and NPK fertilizer (T2), respectively. The treatment effects on the average soil bulk density for two seasons are presented in Figure 3c. These treatments (T3–T6) had positive and significant effects on soil bulk density. The bulk density was reduced in the T3–T6 treatments (1.39 g cm<sup>−</sup>3, on average) and showed statistically lower values than obtained in the control (1.46 g cm<sup>−</sup>3).

**Figure 3.** Effect of organic, inorganic, and biofertilizers on the soil properties [O-carbon (**a**), cation exchange capacity (**b**) and bulk density (**c**)] on which Jew's mallow (*Corchorus olitorius* L.) was grown. Each column represents the mean of three replicates; different letters on similar columns indicate significant differences using Duncan's multiple range test at *p* ≤ 0.05.

### **4. Discussion**

### *4.1. Growth Variables*

The positive effects of NPK with biofertilizers (T5) on growth variables may have been due to the efficiency of the microorganism in the biofertilizer in immobilizing N for a longer time in the form of NH4 <sup>+</sup>, which helped in the nutrient uptake by the plant [30]. According to Alori and Babalola [31], a biofertilizer is a living organisms that is added into the soil as inoculant that helps to provide certain nutrients for crop growth. Furthermore, these positive effects may be related to the increased availability of nutrients provided by mineral fertilization, which also served as an energy source for

the microbial community [32]. Similar to our findings, Al-Zabee and Al-Maliki [32] reported that the combination of mycorrhizal fungi, algae, and yeast with a higher rate of chemical fertilization (120 kg N, 60 kg P, and 200 kg K per hectare) was beneficial to soil microbial metabolism and potato yield. Besides, Asmamaw, et al. [33] reported that the application of dry cyanobacterial biofertilizer could serve as an auxiliary N source to inorganic fertilizer for pepper, maize, and kale production. It was also noted that the use of biofertilizers in combination with chemical N fertilizers increased growth, productivity, and chemical compositions of the dill plant (*Anethum graveolens L.*) compared to the untreated control, where the highest values of plant growth were recorded when biofertilizer was used in combination to 97.6 kg.ha−<sup>1</sup> N [34]. Observations have also shown that the most effective treatment for growth characteristics of barley cultivars (Giza-128 and Giza-129) under newly reclaimed sandy soil was 178.57 kg N ha−<sup>1</sup> + Yeast [35]. Moreover, Sen, et al. [36] reported that the combined use of 100% of the recommended dose [714.3 kg.ha−<sup>1</sup> ammonium sulfate (20.5% N), 476.2 kg ha−<sup>1</sup> calcium superphosphate (15.5% P2O5), and 119.05 kg ha<sup>−</sup><sup>1</sup> potassium sulfate (48% k2O)] of inorganic fertilizers with biofertilizer was optimal for increasing oil yield (33.22 mg g<sup>−</sup>1) of cumin black (*Nigella sativa* L.). The co-application of biofertilizers like *Azospirillum* and *Phosphobacteria* spp. and inorganic fertilizers had a significant effect on the growth variables of cucumber (*Cucumis sativus L*.) [37]. Application of biofertilizer at 300–400 kg ha−<sup>1</sup> dose combined with inorganic fertilizer at 75% of crop requirement dose was the best combination for increasing NPK nutrient uptake for rice crop and weight of milled dry rice. Marlina, et al. [38] recommended the use of dry cyanobacterial biofertilizer which serve as a supplementary N source in place of inorganic fertilizer for rice production in inception soil of lowland swamp area.

### *4.2. Compositional Variables*

The data in Tables 5 and 6 indicated differences in the average proportion of protein, N, P, and K content in leaves among treatments. The NPK with biofertilizers (T5) treatment significantly increased these variables in both seasons. The results presented by Hellal, Mahfouz and Hassan [34] showed that the highest NPK-accumulation were recorded after the combination of biofertilizer with 476.2 kg ha−<sup>1</sup> ammonium sulphate (20.5% N) in the Dill plant. The pronounced positive effect on protein, N, P, and K in leaves resulting from T5 addition may be attributed to the increased uptake of N by plants, and thus, the biosynthesis of protein was increased. Moreover, Tisdale et al. [39] reported that the addition of N in combination with adequate P tended to increase K-uptake by plants. They also showed that K concentration may be high in the NH4 <sup>+</sup>-nourished plants as it is adsorbed by soil colloids, so it does not get leached from the soil. This gave the plant a greater chance of taking up N, and thus some nutrients, to build the dry matter. Also, data in Figure 2 for leaf chlorophyll content supported the results of Hellal, Mahfouz and Hassan [34], where it was observed that the highest values of chlorophyll content were recorded where biofertilizer was used in combination to 97.6 kg ha−<sup>1</sup> N in the dill plant. Moreover, Sen, Choudhuri, Chatterjee and Jana [36] reported that the combination of 100% of the recommended dose of inorganic fertilizers with biofertilizer increased the leaf chlorophyll content (13.18 mg.g−1) in cumin black (*Nigella sativa* L.) in the eastern Himalayan region of West Bengal. Moreover, Youssef, et al. [40] reported that the combined application of organic manures and biofertilizer (EM) had a synergistic effect on the total chlorophyll content of plants.

### *4.3. Soil Properties at the End of the Experimental Period*

The combined application of the biofertilizer with the organic or the inorganic fertilizer was beneficial for the physical and chemical properties of soil and were important for the quality and productivity of the soil. The application of organic fertilizer in T3 and its combination with biofertilizer in T6 increased the soil organic-C content of the soil at the end of experimental period by 91.25 % and 68.75%, respectively, over the control treatment (without fertilization T1). This organic fertilizer in the soil can increase the soil organic-C due to higher soil organic matter added from organic fertilizer. This serves as nutrient sources for plants and improves physical, chemical, and biological properties

of the soil through improved structure and stable aggregates. This is because organic matrices are a natural chelating material with high moisture retention capacity [41,42]. These results are in agreement with Nesgea, et al. [43], who reported that the application of organic fertilizer increased the soil organic-C content after harvest by up to 65%.

As for the cation exchange capacity of the soil, our results showed that the application of organic fertilizer with biofertilizer increased in CEC of the soil after crop harvest, which were statistically undifferentiated with T3 and T5. The increased CEC might be attributed to the addition of organic fertilizer with the biofertilizer, which might have helped in releasing more nutrients into the soil. This could be an indication of increased exchange sites on the surface of the soil colloids. In line with this result, Tana and Woldesenbet [44] reported that CEC significantly increased with increasing organic fertilizer (15 ton ha<sup>−</sup>1) with inorganic fertilizer.

The data in Figure 3c indicated that there were no differences between treatments (T3, T4, T5, and T6), but the highest average reduction in soil bulk density was recorded in farmyard manure +biofertilizer (T6). Soil bulk density was reduced after the combined use of organic manure with biofertilizer (T6) compared to soil amended with only inorganic fertilizer (T3). This could be due to improved soil aggregation as a result of decreased soil bulk density. Several studies have shown that the appropriate addition of combined biofertilizers, inorganic and organic, improved soil porosity and decreased its bulk density. Our results are in harmony with Khan, et al. [45] who reported that organic fertilizer improved soil organic matter content and decreased soil bulk density.

### **5. Conclusions**

Biofertilizers play a significant role in improving soil structure, and inorganic fertilizers are important due to their ability to provide essential nutrients, resulting in the better growth and productivity of crops. The results of the present study, conducted on a silty loam soil and in semi-arid climate conditions, revealed that the Jew's mallow plants treated with the combined application of biofertilizer and NPK fertilizer showed maximum growth and productivity among all other treatments. Although the current study unfolded the performance ability of a neglected crop under the application of different kinds of fertilizers, there is a further need to understand the molecular mechanism behind it and to improve the fertilization techniques and material according to the need for crops.

**Author Contributions:** M.A.Y. and A.F.Y. equally contributed in experimentation and draft preparation; M.M.A. and M.M.I. helped in data analysis and editing; Y.X. reviewed and editing; R.P.M. helped in draft preparation and revised the manuscript. All authors have read and agreed to the published version of the manuscript.

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

**Acknowledgments:** Thanks to all field technicians at College of Agriculture, Al-Azhar University branch Assiut, Egypt and College of Horticulture, Fujian Agriculture and Forestry University, Fuzhou, China.

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

### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 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*

## **Assessing the Potential of Jellyfish as an Organic Soil Amendment to Enhance Seed Germination and Seedling Establishment in Sand Dune Restoration**

### **Iraj Emadodin \*, Thorsten Reinsch, Ra**ff**aele-Romeo Ockens and Friedhelm Taube**

Group Grass and Forage Science/Organic Agriculture, Institute for Crop Science and Plant Breeding, Christian-Albrechts-University, 24118 Kiel, Germany; treinsch@gfo.uni-kiel.de (T.R.); r.ockens@gmx.de (R.-R.O.); ftaube@gfo.uni-kiel.de (F.T.)

**\*** Correspondence: iemadodin@gfo.uni-kiel.de

Received: 6 May 2020; Accepted: 16 June 2020; Published: 18 June 2020

**Abstract:** Worldwide, sandy coastlines are affected by extensive wind and water erosion. Both soil quality and periodic drought present major problems for sand dune restoration projects. Hence, soil amendments are needed to improve soil quality and enhance soil restoration efficiency. The jellyfish population has increased in some aquatic ecosystems and is often considered as a nuisance because of their negative impacts on marine ecosystem productivity as well as coastal attractiveness. Thus, development of new products derived from jellyfish biomass has received attention from researchers although utilization is still at a preliminary stage. Herein, our main objective was to test seed germination, seedling establishment, and seedling vitality of annual ryegrass (*Lolium multiflorum* L.) when supplied with organic soil amendment from two different jellyfish species (*Aurelia aurita* and *Cyanea capillata*) in comparison with an unfertilized control and mineral fertilizer treatment. We hypothesized that jellyfish dry matter as an organic soil amendment would improve seed germination and seedling establishment in sand dune environments. Germination and seedling growth experiments were conducted in the laboratory and greenhouse. The results indicate that jellyfish enhanced seedling growth and establishment in sand dune soil significantly (*p* < 0.05 and *p* < 0.01) under water scarcity conditions. Therefore, jellyfish may have potential for an auxiliary role in sand dune restoration projects in coastal areas in the future.

**Keywords:** seed germination; jellyfish; blue fertilizer; soil restoration; soil amendments; water use efficiency

### **1. Introduction**

Human activities have damaged sandy beaches and coastal dunes. With further population growth, it is expected that human impacts on coastal ecosystems will only increase [1]. Furthermore, periodic environmental stress, including drought, is predicted to increase water scarcity problems worldwide. This creates additional and serious functional limitations for plant growth and also has significant adverse influences on crop yields [2,3]. Sand dune plants play an important role in dune stabilization and protect the adjacent coastal beaches from wind and water erosion but they are subjected to severe environmental stresses such as drought [4,5]. Therefore, organic soil amendments are important for ecological maintenance of the plants in sand dune restoration plans, especially in arid and semi-arid land where soils are under fragile conditions because of water scarcity and poor organic matter.

According to Donohue et al. [6], seed germination is a highly important stage in plant life cycles and factors affecting germination significantly impact subsequent seedling establishment, as well as plant environmental adaptation [7]. Seed germination is the initial process through which seeds develop

into plants and it commences by the seed absorbing water under adequate temperature, which leads to the creation of a new plant [8]. According to Kildisheva et al. [9], taking seed germination into account in restoration planning is important to certify seed use efficiency and management. Seedling emergence is the most important phenological event that influences the success of plants [10] and it starts with seed germination. Seedling survival is also one of the most critical stages in plant growth and it is often adversely affected by drought and soil dryness [11]. According to Manz et al. [12] and Bewley et al. [13], water uptake by a seed includes three phases: first, the rapid initial uptake of water; second, a plateau phase (metabolic preparation for germination); and third, a further increase of water uptake. Therefore, soil moisture (available and adequate water) plays an important role in seed germination and seedling establishment.

Jellyfish, when supplied as an organic fertilizer, have a potential to promote seed germination, as reported by Emadodin et al. [14]. The term jellyfish as a group of marine invertebrates commonly refers to the medusae form of planktonic marine members of the class Scyphozoa or Cubozoa. Jellyfish populations have been a recurrent topic of debate over the past few decades, as mass aggregation (called blooms) of jellyfish are reported more frequently and have often been linked to human-induced environmental changes [15,16]. For example, blooms of some jellyfish species have been reported in the East Asian marginal seas [17], Red Sea [18], Mediterranean Sea, and Black Sea [19].

Positive effects from using jellyfish as inputs for agricultural production have been recorded by several researchers. Fukushi et al. [20] indicated the potential usefulness of two jellyfish species (*Aurelia aurita* and *Chrysaora melanaster*) as a source of fertilizer for vegetable production. Hossain et al. [21] introduced desalinated-dried jellyfishes (*Nemopilema nomurai* and *Aurelia aurita* from the Sea of Japan) as an alternative material to replace herbicides and chemical fertilizers in rice production. The possible use of jellyfish as pesticides was also investigated by Hussein et al. [22]. However, some jellyfish species may have different effects on plant growth processes due to their chemical components. Therefore, in this study, we evaluated the influences of two different jellyfish species (*Aurelia aurita* and *Cyanea capillata)* from the Baltic Sea Coast of Germany, on the early growth stages of annual ryegrass (*Lolium multiflorum* L.). It was hypothesized that the addition of jellyfish dry matter to the soil will enhance seed germination as well as seedling growth and establishment in sand dunes. This was considered to be particularly important in terms of soil restoration projects in dry lands where there is a need to establish new plant communities on degraded soil with very low productivity, as well as drought conditions. However, it should be emphasized that due to various ecological conditions and different species, several experiments (under greenhouse and field conditions) are required to investigate this hypothesis. Here, we report results from a petri plate germination experiment and a greenhouse pot experiment. The experiments are carried out, which is running under a European Union (EU) project entitled: Development of products from jellyfish biomass (Gojelly) with funding by the EU Horizon 2020 Research and Innovation Program.

### **2. Material and Methods**

Two jellyfish species (*Aurelia aurita* and *Cyanea capillata*) were collected from two sites on the Baltic Sea Coast of Germany (54◦25 N, 10◦10 E and 54◦47 N, 9◦84 E) during summer of 2018 and 2019. The samples were put into plastic bags separately and stored at −20 ◦C before further processing. In this investigation, two drying methods including oven-dried and alcohol-dried methods have been conducted. Jellyfish were oven-dried at 50 ◦C until constant weight was reached. Alternatively, the alcohol-dried method of Pedersen et al. [23] was applied in which the fresh or frozen jellyfish were exposed to ethanol (70%) then after around one-hour the jellyfish were removed and put into distilled water. After about 30 min, jellyfish were taken out of the water and dried under room temperature (around 21 ± 1 ◦C).

The dry matter was homogenized in a ball mill (Retsch MM2000, Haan, Germany). The carbon and nitrogen content of the material was measured via dry combustion (Vario Max CN, Elementar Analysensysteme GmbH, Hanau, Germany).

For chemical analysis, 200 mg of dried and finely ground jellyfish material was digested with 10 mL 15.6 M HNO3 (ROTIPURAN® Supra) at 190 ◦C for 45 min in 1800 W microwave oven (MARS 6, Xpress, CEM, Matthews, MC, USA). After digestion, the concentrations of Ca2<sup>+</sup>, Cu2+, Fe2+, K+, Mg2+, Na+, and Zn2<sup>+</sup> were quantified with an atomic absorption spectrometer (AAS 5EA Thermo Electron S, Carl Zeiss, Jena, Germany). The citric acid extractable P was determined by using citric acid (2% *v*/*v*) and a jellyfish/solution ratio of 1:10. After shaking with an end-over-end shaker (Type RA 20, C. Gerhardt GmbH and Co. KG, Bonn, Germany) for 30 min, the citric acid extracts were filtered (MN 619 G<sup>1</sup> <sup>4</sup> , Machery-Nagel GmbH and Co. KG, Düren, Germany). P concentrations in all extracts were determined photometrically with a continuous flow analyzer (Skalar Analytical B.V., Breda, The Netherlands) by using the modified molybdenum–ascorbic acid blue method. Results of the chemical analysis are given in Table 1.


**Table 1.** Some macro- and microelement compounds of jellyfishes (*Aurelia aurita* and *Cyanea capillata*).

Seed of annual ryegrass *(Lolium multiflorum* L.) was used in this investigation. The soil used in this experiment was beach sand and this was collected from the Baltic Sea Coast of Germany (54◦25 N, 10◦10 E). Nitrogen and carbon contents of the soil were around 0.006% and 0.17%, respectively, with a C:N ratio around 28. Calcium ammonium nitrate (CAN) was used as a fertilizer treatment for comparison with jellyfish with regard to impacts on seed germination, seedling plant establishment and vitality. CAN is widely used as an inorganic fertilizer on grassland and other crops.

The pH and EC values of the jellyfish liquids that were provided for each treatment in pot experiment were the same as with a petri plate experiment. In order to test seed germination rate as well as seedling growth and establishment, a petri plate experiment and pot experiment were conducted as follows.

### *2.1. Petri Plate Experiment*

The petri plate method was used to test germination rate. This method helps to monitor the processes of germination under a controlled environment. In pre-treatment the dry matter of *A. aurita*, *C. capillata* and calcium ammonium nitrate (CAN) was dissolved in distilled-water (0.5 g DM in 40 mL distilled water). Pure distilled water was also used as a control treatment. Filter paper was put in each petri plate (8.5 cm diameter) wetted by different aqueous solutions. In total, 20 annual ryegrass (*Lolium multiflorum* L.) seeds were placed in each petri plate. All plates were covered by plastic foil to mitigate evaporation and put in darkness at 21 ± 1 ◦C. The plates were controlled every day and observations (seed germination number and assessment of vitality) were recorded. The petri plate experiment was carried out with eight treatments and with four replications (Table 2). The pH and electronic conductivity (EC) of the different treatments were measured with a PC60 Premium Multi-Parameter tester (Apera instruments, Europa, GmbH; Table 3). Although the petri plate test is considered as a standard work for assessment of seed germination, it may not give an accurate prediction of seedling emergence in the field [24]. Therefore, a pot cultural experiment was also conducted in the greenhouse.


**Table 2.** Petri plate experiment treatments.

**Table 3.** pH and electrical conductivity (EC) values of the treatments.


### *2.2. Pot Experiment*

This experiment was conducted in the greenhouse in summer 2019 at the University of Kiel. Plastic pots (around 11 cm top diameter, 6.5 cm bottom diameter, 10 cm height) were filled with 700 g of sand. In total, 20 uniform seeds of annual ryegrass (*Lolium multiflorum* L.) were placed in each pot and covered by 0.5 cm of sand material and irrigated with six different solutions (oven and alcohol dried materials from *Aurelia aurita* and *Cyanea capillata* and CAN (0.5 g DM in 40 mL distilled water) and 40 mL distilled water for control (Table 4). In order to reduce evaporation, the sand was covered by seagrass. The pot experiment was conducted in six treatments with four replications (Table 4). All treatments were irrigated three times (days 10, 15, and 17) during the experimental period, with 40 mL of tap water.

**Table 4.** Pot experiment treatments.


### *2.3. Statistical Analysis*

The petri plate test and the pot experiment were both conducted in a completely randomized design. The data were subjected to analysis of variance (ANOVA) and Welch's *t*-test as an adaptation of student's *t*-test [25]. The level of significance was declared at *p* < 0.05, *p* < 0.01 and *p* < 0.001.

### **3. Results**

### *3.1. Petri Plate Experiment*

Petri plate experimentation for testing germination rate and estimating seed viability is a standard practice [24]. In the petri plate experiment, the initial evidence of radicle protrusion (germination) appeared after two days. The germination rate was significantly less (*p* < 0.05; *p* < 0.001) than the control in Aur (Alc) and Aur + Cya (Alc) (Figure 1). Other treatments showed no indication of significant differences (*p* > 0.05) in germination rates relative to the control. Germination under the control (water) treatment started earlier than other treatments. However, after four days, germination rates under Cya (Ov) showed slightly more (2.5%) than control. All shoots under jellyfish treatments showed more vitality than water and chemical fertilizer treatments (according to the visual estimation).

**Figure 1.** Total germination in percent depending on the different treatment applications in the petri plate experiment. The error bars represent the ± standard deviation. Welch test in two pairs only with control. Same letters indicate no significant differences with control.

### *3.2. Pot Experiment*

In the pot experiment, the effect of jellyfish on seed germination and seedling growth varied with jellyfish species (*p* < 0.05). Seedling emergence occurred earliest in the control treatment, yet mortality increased sharply in the following days (Figure 2). Seedlings in the jellyfish treatment showed greater vitality with better establishment rate. The sprouts and seedlings had a stronger and greener appearance than those of the water (control) and chemical fertilizer treatments.

**Figure 2.** Mortality of seedlings (%) in the pot experiment during the first period of the control (two days).

Results of total germination and final seedling emergence, including surviving and non-surviving seedlings, for the different treatments at the end of the experiment confirmed there was less mortality and higher rates of seedling establishment in the Aur (Alc) and Aur (Ov) treatments (Figures 3 and 4). The results also indicated that the length of grass seedlings changed under different treatments significantly (*p* < 0.05 and *p* < 0.01, Figure 5). The maximum length of the grass seedlings recorded under the Aur (Alc) treatment was around 14 cm. Plant from Aur (Alc), Cya (Alc) and Cya (Ov) treatments were also significantly taller than control (*p* < 0.05; Figure 5).

**Figure 3.** Total germination for final seedling emergence (including surviving and non-surviving seedlings) for each of the different treatments.

**Figure 4.** The proportion of surviving seedlings (as percent) among the different treatments in the pot experiment. Error bars represent the ± standard deviation. Welch test in two pairs only with control. Same letters indicate differences are non-significant with control.

**Figure 5.** Average length of seedling (cm) under different treatments. Treatments followed by different letters according to Welch test significantly different at *p* < 0.05 and ANOVA test shows significantly different at \* *p* < 0.05 and \*\* *p* < 0.01.

### **4. Discussion**

In this investigation both of the jellyfish species *Aurelia aurita* and *Cyanea capillata* were considered as potential material for promoting seed germination. In the petri plate experiment, there were negative effects of Aur (Alc) and Aur + Cya (Alc) on germination, which may be related to the impacts of alcohol or the jelly form of Aur (Alc) liquid that may delay the time of germination. This effect did not occur in the pot experiment.

According to Bewley [8], the time for germination and post-germinative growth varies from several hours to many weeks, and it depends on the plant species and the germination conditions (Figure 6). Results from our investigation indicated that jellyfish amendments provided conditions that are likely to have caused a delay in germination in comparison with the control. However, evidence showed that in phase 3 (post-germinative growth), seedlings appeared to show greater vitality in treatments with the jellyfish amendments compared with the control. The delay in germination could be related to

the salinity of jellyfish, which is indicated by high sodium concentrations in the jellyfish dry matter (Table 1). Delayed germination has been shown previously in many crops as a consequence of high salinity [26].

**Figure 6.** Germination and post-germinative growth divided into three phases regarding water uptake (adapted after Bewley [8]).

The electrical conductivity of alcohol dried jellyfish was shown in Table 3 to be less than other treatments. This could be related to the washing out of the ions by alcohol. The morphology of sprouts and seedlings under the jellyfish treatment shows greater vitality than in the control. This may be related to the presence of additional essential elements provided by jellyfish as well as enhancing soil water holding capacity. There is a good correlation between soil biological activities and soil water content [27]. Thus, there could be a key role for the use of jellyfish as an organic amendment in this context. The germination under control (water) treatment commenced earlier or proceeded faster to seedling emergence than other treatments. Therefore, it is assumed that jellyfish may affect germination by absorbing water and, if so, this may reduce the amount of water available for the seed in the initial stage of germination.

According to Smith and Doran [28], pH values from 5 to 8 represent the optimum range for most soil microorganisms. Hence the pH rates measured for different jellyfish liquids show no harmful effects in this case. The electrical conductivity was lower in treatments dried with alcohol. This may be attributed to different salinity levels and cation exchange capacities. According to Rawls et al. [29], soil organic matter content has impacts on soil structure as well as water adsorption properties. Thus, the application of jellyfish may also enhance soil water retention through enhancing soil organic carbon and collagen content, as well as provide some essential bio-, macro-, and microelements. According to Carter [30], using chemical fertilizer under conditions of low soil moisture content has harmful effects on seedling establishment. Our investigation also showed the same result. According to Killham [27], the most commonly used index to show resource quality is C:N ratio, and low C:N ratios indicate rapid rates of decomposition. The jellyfish used in this study also showed low C:N ratio = 3.5 (Table 1) in comparison with the green manures such as seagrass (C:N = 14) [16,31] that were also used as soil amendment materials.

### **5. Conclusions**

Jellyfish generally did not reduce the germination rate and provided favorable conditions for seedling survival in sand dunes. However, the positive effects might depend on the species of jellyfish, drying process methods, natural environment (e.g., temperature), edaphic conditions, and plant types. In this study, a positive effect of *Aurelia aurita* was observed on seedling establishment of *Lolium multiflorum*, seedling length, and the vitality of seedlings under conditions of water scarcity. In the context of this investigation, where there is a local surplus of jellyfish, it can be regarded as a local sustainable resource and its use can be considered an innovative organic soil amendment for sand dune restoration projects.

**Author Contributions:** Conceptualization, visualization, investigation, collected data, analyzed data, and writing original draft, I.E.; investigation, collected data, and analyzed data, R.-R.O.; project administration, investigation, reviewed and edited the manuscript, T.R.; supervision, reviewed and edited the manuscript, F.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by European Union's Horizon 2020 research and innovation program (Grant agreement no. 774499).

**Acknowledgments:** The authors would like to thank anonymous reviewers for their constructive comments and suggestions. We also acknowledge financial support by DFG within the funding programme Open Access Publishing.

**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/).

## *Communication* **Suitability of Black Soldier Fly Frass as Soil Amendment and Implication for Organic Waste Hygienization**

### **Thomas Klammsteiner 1,\*, Veysel Turan 2, Marina Fernández-Delgado Juárez 1, Simon Oberegger <sup>1</sup> and Heribert Insam <sup>1</sup>**


Received: 20 September 2020; Accepted: 14 October 2020; Published: 15 October 2020

**Abstract:** Because of its nutritious properties, the black soldier fly has emerged as one of the most popular species in advancing circular economy through the re-valorization of anthropogenic organic wastes to insect biomass. Black soldier fly frass accumulates as a major by-product in artificial rearing set-ups and harbors great potential to complement or replace commercial fertilizers. We applied frass from larvae raised on different diets in nitrogen-equivalent amounts as soil amendment, comparing it to NH4NO3 fertilizer as a control. While the soil properties did not reveal any difference between mineral fertilizer and frass, principal component analysis showed significant differences that are mainly attributed to nitrate and dissolved nitrogen contents. We did not find significant differences in the growth of perennial ryegrass between the treatments, indicating that frass serves as a rapidly acting fertilizer comparable to NH4NO3. While the abundance of coliform bacteria increased during frass maturation, after application to the soil, they were outcompeted by gram-negatives. We thus conclude that frass may serve as a valuable fertilizer and does not impair the hygienic properties of soils.

**Keywords:** animal feedstuff; circular economy; fertilizer; greenhouse; insect larva; organic waste

### **1. Introduction**

In recent years, the use of saprobic insect larvae from the mealworm beetle (*Tenebrio molitor*), the black soldier fly (*Hermetia illucens*; BSF), or the house fly (*Musca domestica*) has attracted interest in the face of rising prices of animal feedstuff and accumulating amounts of waste [1,2]. In the European Union, green waste and food waste largely contribute to an annual amount of 118 to 138 million tons of organic wastes [3]. Especially BSF larvae (BSFL) have been shown to efficiently convert organic wastes into high quality fat and protein [4]. The economic potential and meaningful reintroduction of otherwise wasted nutrients into the biosphere via a circular economy enticed researchers, investors, and the public to contribute to a more efficient recycling of organic wastes by exploiting the potential of insect larvae on a large scale [5,6]. BSFL could also play a valuable role for smaller decentralized waste management systems operated by e.g., hobbyists or farmers in areas where the fly occurs naturally [7–9]. Additionally, the exploitation of BSF and its by-products could create an affordable opportunity for revenue generation by entrepreneurs and smallholder farmers in low-income countries [9–11]. The main by-product in the bioconversion of wastes into high quality protein for animal feedstuff is summarized as 'frass'. Frass in general describes insect excretions, but in a commercial context it often refers to a mixture of mainly insect feces, substrate residues, and shed exoskeletons. It is an inevitable

side-stream during the mass-rearing of insects that can add up to 75% of the fed substrate [12] and is often merchandised as a fertilizing product. In recent years, an increasing number of studies started focusing on meaningful applications of insect frass [9,13–15], and the first large-scale field studies provided promising perspectives for its application in agriculture, especially in terms of plant nutrient availability [10,11,16].

The substrate used to grow insects affects the properties of the frass, since undegradable residues remain unused, while the digested fraction is modified by the gut microbiota when passing through the gastrointestinal tract [17,18]. Wang et al. [19] used frass from *T. molitor* for subsequent rearing of BSFL to exploit leftover nutrients that *T. molitor* could not take up or digest. In substrates carrying a high bioburden like human feces and manure, BSFL have shown to reduce pathogenic bacteria such as *Salmonella enterica* [20,21] and *Escherichia coli* [20,22], which is attributed to their production of antimicrobial peptides [23]. In the wild, frass from various insects can help to increase the chances of survival and reproduction by either deterring [24,25] or attracting [26,27] conspecifics. Frass can act as a vector for phytopathogenic microorganisms [28,29] and as a source of probiotic yeasts [30]. Its effect on the insects' environment can be observed in forests, where frass deposition goes hand in hand with insect canopy herbivory. It has been shown that frass has an impact on C and N dynamics, and has beneficial effects for tree growth by increasing soil total C, N, and NH4 <sup>+</sup>, as well as microbial soil respiration [31,32]. In industrial environments, frass pyrolyzed to biochar has been successfully tested as a bioadsorbent for wastewater detoxification [33]. According to recent studies, frass' agriculturally and economically most meaningful potential could lie in its application as fertilizer [9,34].

In this study, we assessed the fertilizing potential of process residues (frass) from three generic diets degraded by BSFL. Two of the diets represent major streams of organic waste, namely grass-cuttings (GC) and fruit/vegetable (FV) mix, while the chicken feed (CF) control diet is a commonly used insect breeding substrate. We hypothesized that (I) microbial colonization increases with frass maturation and (II) frass may serve as a valuable alternative to mineral fertilizer by inducing beneficial effects on plant growth.

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

### *2.1. Black Soldier Fly Frass Collection*

The frass was collected from a preparatory feeding experiment conducted at 27 ◦C, 60% relative humidity (Figures S1 and S2, Table 1). The chicken feed (CF; Grünes Legekorn Premium, Unser Lagerhaus, Klagenfurt, Austria) was processed with a Fidibus flour mill (Komo Mills, Hopfgarten, Austria) and mixed with water in a 40:60 ratio.

**Table 1.** Feeding experiment termination summary. The feeding experiment was terminated after a total of 23 days when more than 90% larvae from one treatment group transitioned to prepupal stage. Different lower-case letters indicate differences between treatments (*p* ≤ 0.05) according to the Tukey's HSD test. (*n* = 4; average ± standard deviation; CF = Chicken feed diet, GC = Grass-cuttings diet, FW = Fruit/Vegetables diet).


The fruit/vegetable mix (FV; cucumber, tomato, apple, orange, in ratio 0.5:1:1:1) and fresh grass-cutting diet (GC) were shredded and homogenized using a Total Nutrition Center blender (Vitamix, Olmsted Township, United States). Feeding was done in organic content-equivalents (100, 250, and 370 mg larvae−<sup>1</sup> day−<sup>1</sup> for CF, GC, and FV). After termination of the feeding experiment, the black soldier fly frass (BSFF) from each treatment was collected in plastic bags and stored at room temperature until further use.2.2. Soil Preparation and Greenhouse Set Up

A greenhouse trial using soil collected from an agricultural site (47◦15 54" N, 11◦20 20" E; Table 2) was set up to evaluate the fertilizing effect of the BSFF on the soil. The neutral-to-slightly basic soil (pH 7.3 <sup>±</sup> 0.4) had an electrical conductivity of 78.0 <sup>±</sup> 2.7 <sup>μ</sup>s cm−<sup>1</sup> and a volatile solids content of 78.0 <sup>±</sup> 26.6 g kg<sup>−</sup>1. In addition to a Ptotal content of 823 <sup>±</sup> 190 mg kg−<sup>1</sup> (Pbioavailable proportion 6.88 <sup>±</sup> 1.28 mg kg<sup>−</sup>1), elemental analysis determined a C/N ratio of 24 (40 g Ctotal kg<sup>−</sup>1, 1.7 g Ntotal kg<sup>−</sup>1). The soil classified as a calcaric Fluvisol (IUSS Working Group WRB, 2015) was sieved (Ø < 4 mm) and homogenously mixed with a vermiculite/sand blend (1:1; *v*:*v*) at a ratio 2:1 *w*:*w* (soil:blend).

**Table 2.** Characterization of the soil used for the greenhouse trial. Values expressed on a dry mass basis for *n* = 3 (average ± standard deviation). pH (pH CaCl2), EC (Electrical conductivity), VS (Volatile solids), Ctot (Total carbon), Ntot (Total nitrogen), Ptot (Total phosphorous), Pav (Plant available P).


The four experimental treatments were performed in 500 mL pots: soil was mixed with (1) mineral fertilizer (which served as control); (2) GC BSFF; (3) FV BSFF; (4) and CF BSFF. The mineral fertilizer (NH4NO3) and the different types of BSFF (Table 3) were added in an amount of 40 mg N kg−<sup>1</sup> soil, which is equivalent to 80 kg N ha<sup>−</sup>1, considering the soil bulk density of 1 g cm−<sup>3</sup> and a plough depth of 20 cm as described by Goberna et al. [35]. Thereby, all treatments received the same dose of total N.

**Table 3.** Main properties of the three different black soldier fly frass fractions (CF-F: Chicken feed frass; GC-F: Grass-cuttings frass; FV-F: Fruit/Vegetables frass). Values expressed on a dry mass basis for *n* = 3 (average ± standard deviation). Different lower-case letters indicate differences between treatments (*p* ≤ 0.05) according to the Tukey´s HSD test. Different capital letters indicate significant differences between treatments (*p* ≤ 0.05) according to the Mann–Whitney test. EC (Electrical conductivity), VS (Volatile solids), Ctot (Total carbon content), Ntot (Total nitrogen content).


After an equilibration period of 16 h at 4 ◦C, pots were randomly placed in a greenhouse. Ryegrass (*Lolium perenne*; seed amount based on 30 kg seeds ha−1) was sown and left to develop. During the incubation period of 28 days, at an average temperature of 20 ◦C with a light/darkness cycle of 10/14 h, the soil moisture was kept at field capacity (moisture of the soil after drainage by gravity). All treatments were applied in four replicates, resulting in a total of 16 pots in this study. After the

incubation period, plants were removed from the pots, and soil samples were sieved (Ø < 2 mm) and immediately stored at +4 ◦C until analyses (Table 4).

**Table 4.** Physicochemical and biological properties of the control (C-S: NH4NO3) and the frass amended soils (CF-S: Chicken feed frass + soil; GC-S: Grass-cuttings frass + soil and FV-S: Fruit/Vegetables frass + soil). Values expressed on a dry mass basis for *n* = 4 (average ± standard deviation). Different lower-case letters indicate differences between treatments (*p* ≤ 0.05) according to the Tukey´s HSD test. Different capital letters indicate significant differences between treatments (*p* ≤ 0.05) according to the Mann–Whitney test. EC (Electrical conductivity), VS (volatile solids), Ctot (Total carbon content), Ntot (Total nitrogen content), NH4 <sup>+</sup> (Ammonium content), NO3 − (Nitrate content), DOC (Dissolved organic carbon), DC (Dissolved carbon), DN (Dissolved nitrogen), Pav (Plant available phosphorous content), Ptot (Total phosphorous content), BR (Basal respiration), qCO2 (Metabolic quotient).


### *2.2. Frass and Soil Analyses*

Frass and soil samples (10 g fresh weight) were placed into a glass Petri dish and oven-dried (105 ◦C) for 24 h to determine the content of total solids. Volatile organic solid (VS) content was determined from the weight loss following ignition in a muffle furnace (CWF 1000, Carbolite, Neuhausen, Germany) at 550 ◦C for 5 h. Total C and N contents were analyzed in dried samples using a CN analyzer (TruSpec CHN, LECO, St. Joseph, MI, USA). EC and pH were determined in distilled water and 0.01 M CaCl2 extracts (1:2.5, *w*/*v*), respectively.

Soil inorganic nitrogen (NH4 <sup>+</sup> and NO3 <sup>−</sup>) was determined in 0.0125 M CaCl2 extracts as described by Kandeler [36,37]. Soil total P (Ptot) and plant available P (Pav) were determined as described by Illmer et al. [38]. To estimate dissolved organic carbon (DOC), dissolved carbon (DC), and dissolved nitrogen (DN), 10 g of field-moist soil were shaken in 40 mL distilled water, filtered, and immediately measured using a TOC-L analyzer (Shimadzu, Kyoto, Japan). Soil basal respiration (BR) and microbial ¯ biomass (Cmic) were measured according to Heinemeyer et al. [39]. The metabolic quotient (qCO2) was calculated from BR and Cmic according to Anderson and Domsch [40]. At the end of the trial, aboveground plant biomass was determined by cutting plant shoots at the soil surface and drying them at 60 ◦C for 48 h. Samples were then re-weighted to determine the dry biomass.

### *2.3. Preparation of Media*

For the assessment of the total cultivable bacterial colony forming units (CFUs), we used standard methods agar (0.5% peptone, 0.25% yeast extract, 0.1% glucose, 1.5% agar, pH adjusted to neutral). To determine the abundance of *Salmonella* sp., *E. coli*, coliforms, and other gram-negative bacteria,

XLT-4 and ChromoCult® coliform agar (Merck, Darmstadt, Germany) were prepared according to the enclosed recipe.

### *2.4. Pathogen Quantification*/*Assessment of Microbial Colonization in Frass and Soil*

An amount of 2 g frass or soil sample was added to 18 mL sterile saline solution (0.95% NaCl) and placed on a rotation shaker at 200 rpm for 15 min. Samples were diluted to 10−<sup>2</sup> and 10−<sup>3</sup> for soil, and 10−<sup>5</sup> and 10−<sup>6</sup> for frass using sterile 0.95% NaCl. From each dilution, 50 μL was plated using the spread plate technique. Plates were then incubated at 37 ◦C for 24 h, and the CFUs were counted.

### *2.5. Statistical Analyses*

The effect of the BSFF application on soil parameters was tested with a one-way analysis of variance (ANOVA). In case of significant F-values, a Tukey's HSD (honestly significant difference) post hoc test (*p* < 0.05) was performed. Prior to analysis, the homogeneity of the variances was tested (Levene's test), and data were also tested for normality. Non-normal data were subjected to non-parametric tests for several independent samples (Kruskal–Wallis test), and pairwise comparisons between treatments were performed using the Mann–Whitney U test (*p* < 0.05). Statistical analyses were performed using the SPSS v. 23.0 Software (IBM, Armonk, NY, USA). Principal component analysis was performed in R [41] using the vegan package [42]. Analysis of similarity (ANOSIM) on the physicochemical data (999 permutations) was also conducted with vegan. All graphical representations of data were created with ggplot2 [43].

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

### *3.1. Assessment of Microbial Load in Frass and Frass-Amended Soils*

The high moisture content of substrates and air, as well as the pleasantly warm temperature common in insect breeding, favor microbial growth. While the type of diet is known to directly influence the BSFL gut microbiome [17,44], the excrements in turn may influence the microbiome in the frass. It is likely that by agitating and mixing their surrounding substrate with feces and their inherent microorganisms, the larvae have an impact on their habitat. Similar effects are known from the widely used earthworms (*Eisenia fetida*), which can stabilize organic wastes and introduce ammonia-oxidizing microorganisms, thereby boosting nitrification and increasing nitrate concentrations in the resulting vermicompost [45]. Other insect species inoculate the soil with excreted microorganisms and provide beneficial effects for its quality both in wild and artificial settings [46–48].

Before and after applying frass as soil amendment, the number of cultivable *E. coli*, coliform, and other gram-negative bacteria were assessed (Figure 1). While frass counted up to 109 CFU g−1, the count in soil was down to 103–105 g<sup>−</sup>1. With the nutrient media used in this study, untreated soil contained no cultivable *E. coli* or coliforms, and only low abundances of cultivable gram-negatives with 102 CFUs g−1. In particular, frass from the CF treatment acted as a reservoir for coliforms with a CFU count of 1.9 <sup>×</sup> 109, thereby exceeding CFU counts recorded on larval surfaces (Figure S2). Gram-negative bacteria predominated the cultivable microbiota in frass-amended soil with highest CFU counts of up to 10<sup>5</sup> in soil treated with frass from a FV diet. High microbial load and dominance of coliforms in frass shifted to lower CFU numbers and predominantly gram-negative bacteria in the frass-soil mix, indicating that the autochthonous soil microbiota outcompeted allochthonous microorganisms introduced with frass [49–51].

**Figure 1.** Colony forming units counted for gram-negative, coliform, and *Escherichia coli* from frass samples after collection from the feeding experiment and soil samples after having mixed the soil with frass (*n* = 4). CF = Chicken feed, FV = Fruit/vegetable mix, GC = Grass-cuttings.

### *3.2. Black Soldier Fly Frass Properties, Soil Quality and Plant Performance*

The physicochemical properties of frass were influenced by the larval diet (Table 3). Especially CF frass was more alkaline, had a higher EC, and a higher content of VS. While total C contents were similar in all types of frass, FV frass showed a C:N-ratio of 26.6, compared to 18.5 and 18.2 in CF frass and GC frass, respectively. Similar C:N ratios as found in CF and GC frass have been reported by other studies that used brewery spent grains as larval substrate [11,16]. A C:N-ratio > 20 bears the risk of soil N immobilization, which may favor plants with a more efficient N exploitation attributed to their rhizobiome [46,52]. The addition of biochar to the larval waste conversion process might further improve the frass' N retention, while at the same time increasing larval biomass yield [10]. Moreover, larvae pass through six instars continuously shedding their exoskeleton. Chitin, an N-acetylglucosamine-based polymer (C8H13O5N)n, may influence not only the C:N ratio, but its degradation product chitosan may also provide underrated benefits for plant health and pathogen resistance [53,54]. The C:N ratio is one of the major parameters to consider when it comes to deciding whether frass should be used as soil amendment or as co-substrate in anaerobic digestion or composting [55,56]. Chitin utilization by insects is often associated with chitinolytic gut symbionts [57], which still needs to be further investigated in the context of BSF larvae. Chitin-containing fertilizers have previously been found to serve as splendid nitrogen sources [58].

Frass addition to the soil before planting *Lolium perenne* was adjusted on a basis of N-equivalence (80 kg N ha−1; Tables 3 and 4). Soil amended with CF frass exhibited a higher Ptot content than the other frass-amended soils; however, Pav was not significantly different. Principal component analysis (Figure 2) highlighted the parameters that influenced the properties of the soil-frass mix the most, which was further confirmed by ANOSIM (R = 0.5061, *p* < 0.001). The three frass-amended soils clustered closely together, with Ptot and Pav, pH, DC, Ntot, C:N ratio, BR, and the qCO2 being the most influential parameters for their similarity.

**Figure 2.** Principal component analysis of samples from control soil and soil mixed with the three different frass types. Data points represent replicates, and arrows show the most influential parameters for the spread of the data. NO3 = Nitrate, DN = Dissolved nitrogen, CN = Carbon/Nitrogen ratio, Pbio = Phosphorus bioavailability, BR = Soil basal respiration, qCO2 = Metabolic quotient, Cmic = Microbial biomass, Ntot = Total nitrogen, Pav = Plant available phosphorus, DC = Dissolved carbon, EC = Electric conductivity.

The qCO2 describing the microbial soil respiration per unit Cmic is known to be tightly connected to the C:N ratio and increases when less N is available [59]. Higher qCO2 can indicate stress or disturbances within the soil because, although C sources are readily available, microbial metabolism and substrate decomposition are limited by N [60]. NO3 and DN, on the other hand, were the major drivers for the deviation of the control group from the frass treatment groups, since they were both significantly higher in control soil.

In our study, the frass treatments were compared with a control that received an equivalent of 80 kg ha−<sup>1</sup> nitrogen in the form of NH4NO3. In a similar experiment, Ros et al. [61] found that such an amount of mineral N increased the maize yield by 33% compared with an unfertilized control, while N-equivalent additions of compost yielded only 15% increase. Recent observations at field-scale by Beesigamukama et al. showed that even at lower application rates of 30 kg N ha<sup>−</sup>1, BSFF exceeded the performance of mineral N fertilizer in terms of grain yield and nitrogen fertilizer replacement values when applied at the same rates [16]. Compared with commercial fertilizers, nitrogen recovery rates and nitrogen use efficiency of plants have been shown to be improved when amended with BSFF [11]. Additionally, the higher P concentrations in the frass could facilitate N accumulation in plants by improving N uptake, as P plays an important role in energy transfer [62,63].

Using BSFL instead of aerobic windrow composting has additionally been shown to reduce the global warming potential of treating organic wastes by 50% [64]. The addition of frass did not lead to significant differences in plant growth compared to the mineral fertilizer (Figure 3). In fact, the similar growth progress indicates that the nutrients from frass are readily available for uptake and have no detrimental impact on plant growth. These results, however, do not support the findings of Alattar et al. [13], who reported that the development of plant height and leaves in corn (*Zea mays*) was inhibited by the addition of BSFL frass. In their study, they attributed the negative effects to the low porosity of larval residues that may have created anaerobic conditions. The moisture content of the frass harvested from our preliminary feeding experiment was only 10% (Table 3), thereby facilitating aeration and miscibility in soil. Insufficient oxygen supply can occur when frass has a high moisture content and is not subjected to adequate post-processing. In an environment specialized on insect rearing, a multi-step treatment of frass could increase the efficiency of degradation. With additional downstream composting or anaerobic digestion [65,66], the recovery as soil amendment represents the economically most promising option.

**Figure 3.** Plant biomass yield of *Lolium perenne* after application of black soldier fly frass (BSFF) obtained from the degradation of various organic substrates. CF BSFF = Chicken feed frass, FV BSFF = Fruit/vegetables frass, GC BSFF = Grass-cuttings frass (*n* = 4).

### **4. Conclusions**

The valorization of organic wastes by insect larvae generates frass as a side-product. From our study we conclude that frass may serve as a soil nutrient source and does not impair soil hygiene. In some cases, however, frass post-processing through anaerobic digestion or composting may be advised to avoid soil nitrogen deficiencies or impairing soil gas permeability. In the light of the increasing importance of insect rearing, the agricultural utilization of frass is demanding further research, in particular, long-term studies.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/10/1578/s1, Figure S1: Influence of three different diets on larval biomass increase, Figure S2: Microbial colonization of larval surfaces.

**Author Contributions:** Conceptualization, T.K. and M.F.-D.J.; formal analysis, V.T., T.K., and S.O.; funding acquisition, H.I., V.T., and T.K.; investigation, S.O. and V.T.; methodology, V.T. and M.F.-D.J.; resources, H.I.; supervision, T.K. and M.F.-D.J.; visualization, T.K.; writing—original draft, T.K. and M.F.-D.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Austrian Science Fund (FWF; project number: P26444). Thomas Klammsteiner was supported by a PhD grant from the Vizerektorat für Forschung of the Universität Innsbruck (Doktoratsstipendium aus der Nachwuchsförderung). Veysel Turan was supported by a post-doctoral fellowship from the Scientific and Technological Research Council of Turkey (TUBITAK, grant number, 1059B191601133).

**Acknowledgments:** The authors show their gratitude to Carina D. Heussler for providing the black soldier fly larvae for this study. Open Access Funding by the Austrian Science Fund (FWF).

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

### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 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/).

## *Communication* **Earthworms (***Lumbricus terrestris* **L.) Mediate the Fertilizing E**ff**ect of Frass**

### **Anne-Maïmiti Dulaurent 1,\*, Guillaume Daoulas 2, Michel-Pierre Faucon <sup>1</sup> and David Houben 1,\***


Received: 29 April 2020; Accepted: 29 May 2020; Published: 31 May 2020

**Abstract:** With the forecasted dramatic growth of insect rearing in the near future, frass (insect excreta) has been increasingly considered a sustainable resource for managing plant nutrition in cropping systems and a promising alternative to conventional fertilizer. However, the impact of soil fauna on its fertilizing effect has not been investigated so far. In this study, we investigated the effect of earthworms (*Lumbricus terrestris* L.) on nitrogen (N), phosphorus (P), potassium (K) and calcium (Ca) uptake and crop growth in the presence of frass from mealworm (*Tenebrio molitor* L.). Using a pot experiment, we found that earthworms increased N, P, K and Ca concentration in barley (*Hordeum vulgare* L.) in the presence of frass, suggesting that earthworm activity enhances the short-term recycling of nutrients from frass. Compared to treatments with and without frass and earthworms, the specific leaf area of barley was the highest in the presence of both earthworms and frass. This confirms that earthworms and frass have a synergistic effect on soil fertility. Overall, our study shows that earthworms may improve the efficiency of organic fertilizers and argues therefore for the importance of developing sustainable agricultural practices that promote earthworm populations.

**Keywords:** earthworms; frass; insect excreta; insect farming; nitrogen; phosphorus; soil fauna; soil fertility; waste management

### **1. Introduction**

In the context of the massive increase in the human population at an unprecedented level, insect rearing represents an opportunity to answer the growing demand for proteins with a low ecological footprint [1]. Although insect production is highly efficient in converting by-products into biomass, it also yields a waste stream consisting especially of insect feces ("frass"). Given the "zero waste" context and the need to contribute to the circular economy, the possibility of recovering frass as a fertilizer has recently been considered by researchers [2–4]. For instance, Houben et al. [3] have found that frass from mealworm (*Tenebrio molitor* L.) might be as efficient as conventional mineral fertilizer to sustain crop growth due to its rapid mineralization after its incorporation into the soil and the presence of nutrients in a readily-available form. A couple of studies have suggested that microbial activity might partially control the effect of frass on soil fertility, either in natural conditions [5–7] or in cropping systems [2,3]. However, the impact of soil fauna has not been considered so far. It is known that soil fauna, especially earthworms, may positively affect plant growth [8,9] due to, among others, changes in soil structure and water regime [10], improvement of soil organic matter and nutrient cycling [11,12], and stimulation and dispersal of beneficial microorganisms [13]. Moreover, adding exogenous organic amendments generally stimulates earthworm activity which reciprocally improves the fertilizing effect of these amendments [13–15], even though some contradictory results have also been found [12,16].

Since frass is an organic amendment, it is therefore likely that its effect on soil fertility might also be mediated by earthworm activity. Therefore, the aim of this study was to investigate the impact of the earthworm presence on the fertilizer potential of frass. For this purpose, we carried out a pot experiment to determine the effect of earthworms on the frass fertilizing effect. Barley (*Hordeum vulgare* L.) was grown in greenhouse conditions with or without frass from mealworm (*Tenebrio molitor* L.) in the presence or absence of earthworms (*Lumbricus terrestris* L.).

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

A pot experiment was conducted to determine the effect of earthworms on the fertilizer potential of frass. Frass (ŸnFrass) from mealworm (*Tenebrio molitor* L.) was provided in the form of powder by Ÿnsect (Paris, France), an industrial company farming this insect at a large-scale. Chemical characteristics of frass are presented in Table 1.


**Table 1.** Chemical characteristics of frass (data from Houben et al. [3]).

The studied soil was sampled in Beauvais (Northern France) and was classified as a Haplic Luvisol (IUSS Working Group WRB, 2015), a soil with properties suitable for soil fauna activity [17]. Soil characterization was carried out by Houben et al. [3] following the procedures described elsewhere [18] and revealed that organic C was 1.54%, total N was 0.18%, the cation exchange capacity (CEC) was 12.5 cmolc kg<sup>−</sup>1, and pH was 7.8.

The experimental device was based on our previous study which aimed at estimating the fertilizer potential of frass [3]. Briefly, plastic plant pots were filled with 3500 g of either soil or a mixture of soil and frass at a rate of 10 Mg dry matter ha-1 (hereafter called "Frass" treatment), or untreated soil (hereafter called "Control"). Three earthworms (*Lumbricus terrestris* L.) were added in half of the pots (hereafter called "Frass + earthworms" or "Control + Earthworms" treatments) representing biomass of 12.05 ± 0.24 g and 12.25 ± 0.17 g in Frass + Earthworms and Control + Earthworms treatments respectively, according to the recommendations by Vos et al. [19]. Each of the four treatments was replicated four times.

Eight seeds of barley (*Hordeum vulgare* L.) were sown in each pot. After 10 days, excess germinated seedlings were removed (first harvest) so that only four uniform plants per pot were allowed to grow for the following eight weeks (ca. 120 plants m<sup>−</sup>2). The trials were conducted under controlled greenhouse conditions (temperature 18–25 ◦C, 16 h photoperiod) with daily sprinkler watering to maintain the soil moisture at field capacity. After 9 weeks, the shoots were harvested with ceramic scissors. Three fully-grown young leaves per replicate were scanned at 600 dpi and then dried at 60 ◦C for 72 h to determine specific leaf area (SLA). All aboveground biomass was dried at 60 ◦C for 48 h in a similar manner and weighed. The concentrations of P, K, and Ca in aerial parts were analyzed by inductively coupled plasma-atomic emission spectroscopy (ICP-AES; Jarrell Ash) after *aqua regia* digestion. The concentration of N in aerial parts was analyzed using the Dumas combustion method. Earthworms were extracted from pots, counted, and weighed. As suggested by Coulis et al. [20], available P concentration in soil was assessed using water extraction (soil:water 1:60; w-v) following the procedure described by Sissingh [21]. Available K and Ca concentrations were determined using the acetate ammonium-ethylenediamine tetraacetic acid (AAEDTA) [18,22]. Soil pH was measured in water (soil:water 1:5; w-v).

All recorded data were analyzed using descriptive statistics (mean ± standard error) and normality was determined using the Shapiro-Wilk test. One-way ANOVAs and Tukey's multiple comparison tests or Kruskal-Wallis and Mann-Whitney tests were used to compare biomass, SLA, and nutrient concentrations in the shoot and soil according to whether the distribution was normal or not, respectively. Pearson's correlation coefficient was used to analyze the relationship between SLA and N concentration. All statistical analyses were performed using R software version 3.5.0 [23] and the package Rcmdr [24].

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

### *3.1. Earthworm Survival*

At harvest, earthworm survival was 100% for all the treatments and their burrowing activity was clearly visible (Figure 1). In addition, their number and biomass per pot at the end of the experiment were not significantly (*p* > 0.05) different from that before their incorporation into the soil. This indicates that frass had no toxic effect on earthworms and allows us to ascribe the following results to the actual presence of earthworms. The similar earthworm biomass between the beginning and the end of the experiment contrasts with Sizmur et al. [25] who found in a 12-week microcosm experiment a continuous decrease of earthworm (*L. terrestris*) biomass in soil with no amendment or with organic amendments including farmyard manure, anaerobic digestate, and compost. However, Sizmur et al. [25] carried out their experiment without plants, which could possibly explain the discrepancy with our study. Since *L. terrestris* may feed on plant roots [26,27], it is likely that, by providing an additional source of food, the presence of plants contributed to maintaining earthworm biomass all over the experiment.

**Figure 1.** Representative pictures of soil collected in pots at the plant harvest illustrating the intense burrowing activity of earthworms.

### *3.2. Impact of Earthworms on Nutrient Uptake and Crop Growth*

Many studies have reported that earthworm activities significantly increase N concentration in plant tissues [13,28], predominantly due to an earthworm-induced stimulation of N mineralization, which in turn, enhances N availability for plants [29,30]. For instance, Amador et al. [31,32] showed higher N mineralization in the drilosphere of *L. terrestris*, leading to an accumulation of nitrate in earthworm burrow soil. This increase of nitrate can result in higher N uptake, as observed for oilseed rape grown in an earthworm-inoculated (*Metaphire guillemi*) soil [33]. In agreement with these researchers, our results showed that irrespective of the treatment, N concentration in barley shoot was higher with than without earthworms (Figure 2).

More importantly, our results suggest a synergistic effect between frass and earthworms since the Frass + Earthworms treatment displayed the highest N concentration in barley shoot (Figure 2). The positive effect of frass on N uptake by plants has been previously discussed and was attributed to its very rapid mineralization after its incorporation into the soil [3]. Here, our results indicate that earthworms induced a higher uptake of N in the presence of frass. Although the present study did not allow us to identify the pools from which N was taken up by plants, it is likely that the presence of earthworms stimulated the release of N from frass since earthworms generally promote N

mineralization from organic fertilizer [8]. For instance, Postma-Blaauw [12] showed that *L. terrestris* enhanced the release of N from crop residue by increasing its mineralization while it had no effect on the mineralization of soil organic matter-derived N. Using 15N, Amador and Görres [34] found that *L. terrestris* could double the amount of litter-derived N taken up by maize grown in mesocosms. In another study, N uptake by maize was 26 and 74% higher from manure and compost treatments, respectively in the presence of earthworms (*Pheretima hawayana*) compared to control without earthworms [14]. This was attributed to an increase of the decomposition of organic N by earthworms which enhanced the N mineralization from the manure and compost treatments, as also observed by Rashid et al. [35]. Besides increasing microbial metabolic activity [3], frass, like other organic amendments, might also have promoted earthworm activity [36], which could further increase N mineralization.

**Figure 2.** Concentrations of N, P, K, and Ca, specific leaf area (SLA), and biomass of barley. Values are average (n = 4) ± standard error. Columns with the same letter do not differ significantly at the 5% level.

Unlike N, the presence of earthworms decreased P concentration in the shoot of the control (Figure 2), which can be related to the decrease of available P concentration in soil (Figure 3). Earthworms have been reported to enhance P availability in the short run due to changes in complexes induced by competition for sorbing sites between orthophosphates and carboxyl groups of the mucus produced in the gut [37]. However, after three weeks of incubation, Le Bayon and Binet [38] found a dramatic decrease of P availability in the presence of *L. terrestris*, which was ascribed to the immobilization of P by microorganisms. Phosphorus availability may also be reduced due to soil pH increase brought about by earthworm activities [39]. Our results indicate, however, that soil pH was unaffected by the presence of earthworms (Figure 3), which therefore suggests that the lower P availability in the Control + Earthworm treatment would predominantly result from P immobilization by microorganisms.

As reported by Houben et al. [3], application of frass to soil improved P nutrition (Figure 2), which is due to the presence of P in a readily available form as well as to the slightly acidifying effect of frass which can, in turn, increase P solubility (Figure 3). By contrast to the control, the presence of earthworms in the frass treatment increased P concentrations in shoots, suggesting that earthworms promoted the recycling of P from frass. Interestingly, available P concentration was not increased in the Frass + Earthworm treatment and pH was unaffected by earthworms (Figure 3), which indicates that the higher P concentration in barley shoot in this treatment would not result only from a change in the biogeochemical status of P. Improvement of P concentration in barley shoot might be explained by a better distribution of P within the soil due to earthworm activities. Earthworms facilitate P transfer of organic fertilizer within the soil [15,38,40], which can in turn increase the root accessibility to P, especially for plants such as barley, whose spatial soil exploration by roots plays an important role in the acquisition of P from organic fertilizer [41]. Similar to P, available Ca and K concentrations in soil in the presence of frass were not increased by earthworms (Figure 3) while, as for P, the Frass + Earthworm treatment showed the highest Ca and K concentrations in the shoot (Figure 3). This, therefore, suggests that mechanisms responsible for the earthworms-induced recycling of Ca and K from frass are similar to that for P.

**Figure 3.** Available P concentration (water extraction), available K and Ca concentrations (AA-EDTA extraction), and pH in the soil. Values are average (n = 4) ± standard error. Columns with the same letter do not differ significantly at the 5% level.

The synergistic effect between earthworms and frass on plant nutrition was reflected by an increase of SLA of barley shoot (Figure 2). Being related to the relative growth rate of plant species [42], SLA is widely used as a target trait to unravel plant responses to soil properties, especially those linked to soil fertility [43], and can explain plant productivity [44]. As a leaf functional trait, its characterization allows us to elucidate the plant response to changes in soil properties at an individual scale. SLA is usually well correlated to N availability and N concentration in plants [45,46], which was also found in our study (r = 0.82; *p* <0.001). Therefore, its improvement in the Frass + Earthworms treatment corroborates our findings that earthworms improve the fertilizer potential of frass. It is noteworthy that, unexpectedly, shoot biomass was not improved by the presence of earthworms in the frass treatment. Shoot biomass is known to be less sensitive than SLA to a change of soil fertility as many factors can drive it [47]. In the present study, the lack of biomass improvement in the Frass + Earthwoms treatment, in spite of a higher soil fertility status, could be explained by competition for light induced

by the higher SLA [48,49]. The perspective is, therefore, to elucidate the plant density which optimizes canopy light interception, crop yield, and nutrient use efficiency from frass. The lack of biomass improvement could also have been due to herbivory by earthworms. Some studies reported *L. terrestris* to commonly consume roots [27], especially in situations of low litter availability [26], which might, in turn, reduce the aboveground biomass production, as shown for other organisms [50]. Therefore, another perspective will be to investigate how the root consumption by earthworms may be affected by organic amendments such as frass.

### **4. Conclusions**

With the forecasted growth of insect farming in the near future, frass is increasingly considered a promising resource for the sustainable management of plant nutrition in cropping systems and an enticing alternative to conventional fertilizer. In this study, we evidenced that earthworms enhance the fertilizer potential of frass. Indeed, their activity increases soil fertility and nutrient (N, P, K and Ca) concentrations in barley in the presence of frass, likely by improving the short-term recycling of nutrients from frass. More generally, our study highlights that, as key biological agents in the transformation of organic matter and waste, earthworms may improve the efficiency of organic fertilizers. Coupled with the other well-documented ecosystem services delivered by earthworms, our findings further argue for the importance of developing sustainable agricultural practices that promote earthworm populations.

**Author Contributions:** Conceptualization, A.-M.D., G.D., M.-P.F. and D.H.; methodology, A.-M.D. and D.H.; investigation, A.-M.D. and D.H.; writing—original draft preparation, A.-M.D. and D.H..; writing—review and editing, G.D. and M.-P.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by a private corporation, Ÿnsect.

**Acknowledgments:** We thank the "ASET 158" students, Céline Roisin, Aurore Coutelier and Vincent Hervé for technical assistance. L. Dulaurent is heartily acknowledged for his contribution to the weightless cloud.

**Conflicts of Interest:** Although the research was funded by a private corporation, Ÿnsect, we ensure the research is free of bias.

### **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* **Biochar-Compost Interactions as Affected by Weathering: Effects on Biological Stability and Plant Growth**

**Marie-Liesse Aubertin 1,2,\*, Cyril Girardin 2, Sabine Houot 2, Cécile Nobile 3, David Houben 3, Sarah Bena 4, Yann Le Brech <sup>4</sup> and Cornelia Rumpel 1,\***


**Abstract:** Biochar addition to compost is of growing interest as soil amendment. However, little is known about the evolution of material properties of biochar-compost mixtures and their effect on plants after exposure to physical weathering. This study aimed to investigate the physicochemical characteristics of fresh and weathered biochar-compost mixtures, their biological stability and their effect on ryegrass growth. To this end, we used the contrasting stable isotope signatures of biochar and compost to follow their behavior in biochar-compost mixtures subjected to artificial weathering during 1-year of incubation. We assessed their impact on ryegrass growth during a 4-week greenhouse pot experiment. Weathering treatment resulted in strong leaching of labile compounds. However, biochar-compost interactions led to reduced mass loss and fixed carbon retention during weathering of mixtures. Moreover, weathering increased carbon mineralization of biochar-compost mixtures, probably due to the protection of labile compounds from compost within biochar structure, as well as leaching of labile biochar compounds inhibiting microbial activity. After soil application, weathered mixtures could have positive effects on biomass production. We conclude that biochar-compost interactions on soil microbial activity and plant growth are evolving after physical weathering depending on biochar production conditions.

**Keywords:** biochar; compost; isotopic signature; carbon mineralization; plant growth

### **1. Introduction**

According to the last report of the Intergovernmental Panel of Climate Change (IPCC), global temperatures have increased by 1 ◦C above pre-industrial levels due to human activity [1]. Further increase should be limited to 1.5 ◦C in order to prevent dangerous climate change. To achieve this goal, active carbon dioxide removal from the atmosphere and its storage is needed [1]. Soil carbon sequestration and biochar application to soils may be used for this purpose. As negative emission technologies (NETs), their implementation may be able to achieve long-term carbon sequestration and may have advantages over the other NETs related to their effect on land use, water use and energy requirement [2].

Soil carbon (C) sequestration may be enhanced by the addition of organic amendments. While organic residues such as plant material or manure are usually transformed into amendments through composting, they may also be the feedstock for biochar production [3]. Biochar is a solid pyrolysis product intended to be used as soil amendment [4]. It is mainly composed of aromatic C and has favourable properties such as large porosity and surface area in addition to high cation exchange capacity, depending on feedstock, pyrolysis

129

**Citation:** Aubertin, M.-L.; Girardin, C.; Houot, S.; Nobile, C.; Houben, D.; Bena, S.; Brech, Y.L.; Rumpel, C. Biochar-Compost Interactions as Affected by Weathering: Effects on Biological Stability and Plant Growth. *Agronomy* **2021**, *11*, 336. https:// doi.org/10.3390/agronomy11020336

Academic Editor: Domenico Ronga

Received: 20 January 2021 Accepted: 9 February 2021 Published: 13 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 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 (https:// creativecommons.org/licenses/by/ 4.0/).

conditions and particle size [5–7]. Biochar is known to improve soil properties such as water retention under drought conditions [8], and soil aggregate stability and porosity [9,10]. Due to its low nutrient content, biochar should be combined with nutrient additions through mineral fertilizers, compost and/or growth promoting micro-organisms to further increase its beneficial effects on plant growth when applied to soil [3]. On the other hand, compost is rapidly mineralized after soil application and its carbon sequestration potential may be enhanced by combination with organic and inorganic additives [11]. Mixtures of both materials may therefore be an innovative practice, leading to more efficient soil amendment as compared to their single use.

Biochar combination with other organic amendments may have synergistic effects on organic C retention, which were attributed to physical protection of compost by its occlusion into aggregates or adsorption on biochar surface [12–14]. Other studies found that biochar and mature compost mixtures induced a negative priming effect [15] or a neutral effect [16] on C mineralization when compared to application of compost. Soil addition of biochar-compost mixtures was shown to promote plant growth, biomass accumulation, yield and to improve soil properties such as water holding capacity [17–22]. Yet the synergistic effects of freshly applied biochar-compost mixtures on plant growth and performance are still under debate [23]. Indeed, application of fresh biochar-compost mixture has been found to have neutral [18] or even antagonisms effects [23]. This may be due to release of toxic compounds contained in the biochars' labile fraction [24–27] or to low availability of nutrients due to the biochars' high sorption capacity [23].

When applied to the field and exposed to weathering, the mixture effects may prevent carbon and nitrogen losses as compared to the single use of compost and biochar [28]. Physical weathering may increase the biological stability of biochars and reduce their priming effect on native SOM mineralization [29]. Moreover, weathering may change the biochar structure [30] and its effects on soil properties [8]. These effects may also change the compost-biochar interactions in mixtures and their amendment effects. Indeed, several studies observed an alleviation of beneficial effects of biochar-compost addition on biomass production over time [31–33]. However, to the best of our knowledge, no studies have focused on the effect of weathering on biochar-compost mixture properties and their biological stability.

Therefore, the aim of the present study was to investigate the effect of artificial weathering on chemical characteristics and biological stability of biochar-compost mixtures and the consequences for plant biomass production after soil amendment. We used two industrially produced biochars from maize and *Miscanthus*, a green-waste compost and the corresponding biochar-compost mixtures. The mixtures and pure media were subjected to a physical weathering to mimic natural aging mechanisms. Thanks to contrasting stable carbon isotope ratios of biochars derived from C4 plants and compost derived from C3 plants, we were able to monitor the mineralization of the two components of the mixtures during a 1-year of laboratory incubation with a soil inoculum. In addition, we investigated in a 4-weeks pot experiment the effect of fresh and weathered biochar-compost mixtures on ryegrass growth growing on two different soils. We hypothesized that (i) biochar addition to compost would induce synergetic effects on biological stability and plant growth and that (ii) physical weathering would weaken these interactions.

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

### *2.1. Biochar and Compost*

Biochars were produced from maize cobs (*Zea mays* L.) and elephant grass (*Miscanthus* × *giganteus*, Greef and Deuter), through pyrolysis without oxygen during 10 min at respectively 450 and 550 ◦C. Pyrolysis was performed by VTGreen (Allier, France), using an industrial pyrolysis reactor (Biogreen®Pyrolysis Technology, ETIA, Oise, France). The compost was made from green wastes at the platform of Fertivert (Normandy, France). The composting process consisted of 4 months fermentation and 2 months maturation. Three compost turnings were applied. The biochar from maize cobs and the compost are the

same than the ones used in Nobile et al. [34]. General parameters of the biochars and the compost are listed in Tables 2 and 3. Biochar-compost mixtures were prepared by mixing 20 % (*w*/*w*) of each biochar with 80% (*w*/*w*) of the compost. The biochars and mixtures were air-dried at ambient temperature and the compost was stored at 4 ◦C.

### *2.2. Physical Weathering*

The mixtures and pure media were subjected to a physical weathering through wetdrying and freeze-thawing cycles to mimic natural aging mechanisms. The weathering procedure was inspired by Naisse et al. [29]. Briefly, we placed 100 g (d.w.) of compost or biochar-compost mixtures in PVC cylinders (ø 9.5 cm). Two PVC cylinder (ø 5 cm) were used for the weathering of 30 g of maize and *Miscanthus* biochars. We covered the bottom of all tubes with a polyamide canvas with 20 μm mesh size (SEFAR-Nitex, Sefar AG, Haiden, Switzerland) and placed them on smaller tubes of 10 cm height to elevate the device. All was then put in a 10 cm ø beaker, in order to recover the lixiviates (Supplementary Material, Figure S1). We mimicked weathering processes through three successive cycles including three cycles of wetting/drying and three cycles of freezing/thawing. Wetting/drying steps consisted of saturating the samples with distilled water, leaving them at room temperature during 3 h followed by drying of the sample at 60 ◦C overnight. Freezing/thawing steps consisted of saturating samples with distilled water with the same amount as for the previous cycles, freezing at −20 ◦C overnight and thawing during 6–7 h at 28 ◦C. We replicated these experiments 2 times. At the end of the weathering procedure, we dried the solid samples at 60 ◦C during 2 days and lixiviates until complete evaporation. Mass and carbon loss after artificial weathering were assessed by mass balance.

### *2.3. Material Properties: Physico-Chemical, Elemental and Thermogravimetric Analysis*

To measure pH and electrical conductivity (EC), 2 g of sample were mixed with 40 mL of distilled water and centrifugated for 1 h. The pH (780 pH meter, Metrohm, Herisau, Switzerland) was measured in the supernatant and the mixtures were filtered (glass microfibres paper, Fisherbrand) before EC (InLab® 738-ISM, Mettler Toledo, Columbus, Ohio, USA) measurement. We evaluated the effect of weathering on dissolved organic carbon content (DOC) and elemental content. For DOC determination, 2 g of dried samples were sieved at 2 mm and mixed with 40 mL of distilled water, (1:20 w/v) ratio. The samples were shaken during 1 h, centrifugated at 4750 t/min during 20 min and the supernatant recovered by filtration (glass microfibres paper, Fisherbrand). DOC was analysed using a Total organic carbon analyzer (TOC-5050A, Shimadzu, Marne-la-Vallée, France). The determination of C, H, N and O of solid samples was performed using a CHN-O analyzer (FlashEA 1112 Series, Thermo-Fisher Scientific, Illkirch, France).

Ash content, volatile matter and fixed carbon of dry matter were determined by thermogravimetric analyses (TGA/DSC1 STAR System, Mettler-Toledo, Viroflay, France). The samples (in 70 μL crucibles, approx. 6–7 mg) were first heated at 105 ◦C during 30 min to determine the moisture content. Thereafter, the temperature was increased by 15 ◦C min−<sup>1</sup> to 900 ◦C during 40 min under N2 atmosphere to determine volatile content. Temperature was then kept at 900 ◦C under air flux (50 mL min<sup>−</sup>1) for 6 min to determine ash content.

### *2.4. Biological Stability: Incubation*

Laboratory incubation was carried out under optimum conditions after the addition of a microbial inoculum (4 mL soil inoculum per 100 g of sample). The inoculum was prepared with 50 g of soil from a cropland field (Haplic Luvisol [35], Beauvais, Northern France), by preparing a water extract with 200 mL of distilled water. The soil was not carbonated, contained 154 mg g−<sup>1</sup> organic C, 18 mg g−<sup>1</sup> total N and had a pH (water) of 7.7 (Table 1). After inoculum addition, 20 g of sample were placed in 100 mL glass vials and covered with rubber septa. We carried out the incubation in triplicate for 8 treatments (2 biochar/compost mixtures, a compost and one biochar (all fresh and weathered) at 20 ◦C during 12 months. As we hypothesized that pure biochars will behave similarly, we used only *Miscanthus* biochar as control sample. We adjusted the water content to 60 % at the beginning of the incubation, when the flask's atmosphere was free of CO2. We monitored the decomposition of the materials by measuring release of CO2-C using a micro-GC (490 Micro-GC, Agilent Technologies, Les Ulis, France) and the stable carbon isotope ratio of CO2-C with an isotopic ratio mass spectrometer (Vario isotope select, Elementar, UK-Ltd, Cheadle, UK) at day 1, 3, 7, 16, 24, and then once a month until the end of the incubation. At each CO2-C measurement date, we also determined the isotopic signature of the CO2 emitted by compost, biochar and compost-biochar mixtures. Thanks to the isotopic 13C signature of the C4-biochar, which is distinctly different from C3 compost, we were able to determine the contribution of carbon mineralized from biochar or compost in CO2 emitted from the biochar-compost mixtures. After each measurement, we flushed the bottles with synthetic CO2 free-air. The results are expressed as cumulated CO2-C emitted form fresh and aged samples in terms of initial total C content of the compost or biochar within the fresh samples.



### *2.5. Effect on Biomass Production: Pot Experiment*

A pot experiment was carried out with fresh and weathered compost and mixtures added to two different agricultural soils sampled in Beauvais (Northern France) and classified as a silt loam Haplic Luvisol and a clay loam Calcaric Cambisol [35]. Soil characteristics are shown in Table 1.

After sieving the soil (4 mm), the composts and mixtures were applied at respectively 16t ha−<sup>1</sup> and 20 t ha−<sup>1</sup> to 0.4 kg of soil. Both fresh and weathered amendments were applied to soil at a similar rate, considering the mass loss during the weathering treatment. The pots were sown with 0.15 g pot−<sup>1</sup> of Italian ryegrass (*Festuca perennis* Lam. ex *Lolium multiflorum*) seeds. Thereafter, they were kept in a growth chamber under controlled conditions: 16 h day−<sup>1</sup> of light, a temperature of 24 ◦C (day) and 20 ◦C (night) and addition of distilled water every two days (Supplementary Material, Figure S2). We harvested the plants 4 weeks after sowing by cutting at 2 cm from soil surface. Biomass production was determined gravimetrically after 72 h drying at 60 ◦C.

### *2.6. Calculations and Statistics*

The stable C isotope signatures were used to estimate the contribution of biochar and compost to the mixtures and the CO2 emissions from the mixtures. The partitioning was done with Equation (1):

$$\mathbf{C\_{biochar,mix}} = (\delta^{13}\mathbf{C\_{mix}} - \delta^{13}\mathbf{C\_{compost}}) / (\delta^{13}\mathbf{C\_{biochar}} - \delta^{13}\mathbf{C\_{compost}}) \tag{1}$$

where Cbiochar,mix is biochar carbon in the mixture or in CO2-C emitted from the mixture (%); δ13Cmixture is the stable C isotope signature of the mixture, δ13Cbiochar is the stable C isotope signature of biochar and δ13Ccompost is the stable isotope signature of compost.

To evaluate interactions between biochar and compost in mixtures, we calculated expected values for the mixtures according to Equation (2). The comparison between the expected and the measured values of the mixtures were used to assess interactions between biochar and compost.

$$\mathfrak{m}\_{\text{biochar},\text{mix}}/\mathfrak{m}\_{\text{mix}\text{true}} = \mathbb{C}\_{\text{mix}\text{true}} \times \mathbb{C}\_{\text{biochar},\text{mix}}/\mathbb{C}\_{\text{biochar}}\tag{2}$$

where mbiochar,mix is the mass of biochar within the mixture (g); mmixture is the mass of the mixture (g); Cmixture is the C content of the mixture; and Cbiochar is the C content of biochar.

To calculate differences between fresh and weathered materials, we tested for normality using the Shapiro-Wilk test. For the normally distributed data, we performed analysis of variances (ANOVA) and Tukey multiple comparison. When data did not follow a normal distribution, we used Kruskal-Wallis tests with Bonferroni corrections. The level of significance was set at *p* = 0.05. We performed all statistical analyses using the R software (version 3.5.2).

### **3. Results**

### *3.1. Leaching Due to Physical Weathering*

Material losses ranged from about 20 mg g−<sup>1</sup> for maize biochar to about 150 mg g−<sup>1</sup> for compost (Figure 1). Artificial physical weathering thus resulted in twice as much material loss from compost as compared to biochars. Mass losses for both mixtures were around 75 mg g−1. They were about two times lower than expected from the losses of individual materials (Figure 1).

**Figure 1.** Total mass loss during physical weathering of compost, biochars and their mixtures. Data are presented as mean ± sd (*n* = 2 for the compost and the mixture and *n* = 1 for the biochars). Expected values for mixtures were calculated based on mass losses measured for individual components.

### *3.2. Properties of the Fresh and Weathered Materials*

### 3.2.1. Elemental Composition

Fresh compost was composed of 226 mg g−<sup>1</sup> C, 20 mg g−<sup>1</sup> H, 112 mg g−<sup>1</sup> O and 23 mg g−<sup>1</sup> N (Table 2). Fresh biochars contained at least twice more C than the fresh compost, with biochar from maize and *Miscanthus* containing respectively 591 and 778 mg g−<sup>1</sup> C (Table 2). Hydrogen content of biochars were similar to compost, whereas O and N content of biochars were at least twice lower than for compost. Following the mixing ratio, carbon content of the mixtures ranged between 298 mg g−<sup>1</sup> and 332 mg g−<sup>1</sup> and all other elemental components had similar values for both mixtures. The mixtures showed similar C/N ratios independently from biochar feedstocks.


**Table 2.** Elemental composition of fresh (F) and weathered (W) compost, biochars and biocharcompost mixtures. Expected (exp) values were calculated for the weathered mixtures. Data are presented as means ± sd (*n* = 3). The letters represent differences among treatments.

Compost weathering induced decreasing contents of all elements, while mostly C and O were affected for biochars. As a result of weathering, C content respectively increased and decreased for the maize and *Miscanthus* biochars, while O content more than doubled for both biochars. The expected C content of the weathered mixtures were slightly lower than the measured ones ranging between 321 and 355. As for biochars, weathering affected mainly the C and O contents of the mixtures; O contents of the weathered mixtures were slightly lower than the expected values. For both mixtures, weathering increased the C/N ratio (Table 2).

### 3.2.2. Physico-Chemical Properties, Dissolved Organic Carbon and Stable δ13C Ratio

Table 3 shows physico-chemical properties and the dissolved organic carbon content (DOC) of the materials. pH and electrical conductivity (EC) ranged from 8.1 to 10.5 and from 109 to 1598 μS cm<sup>−</sup>1, respectively. Compost had lower pH (8.4), and EC (944 μS cm<sup>−</sup>1) than both biochars. Both biochars showed similar pH (around 10.5), but maize biochar had higher EC than *Miscanthus* biochar. The pH and EC of fresh mixtures were in between the values from compost and biochars.

Fixed C content ranged between 0.6 and 67.8 %, DOC varied between 2.2 and 277.2 mg g−<sup>1</sup> C, whereas ash content ranged between 13.6 and 59.3 % and volatile matter content between 17.8 and 38.8 %. Compost showed lower fixed C and higher DOC, ash content and volatile matter than biochars. Both biochars had similar volatile C but varied in ash content and fixed C; maize biochar presented a twice-higher ash content and a lower fixed C content (45.6 vs. 63.6%) than *Miscanthus* biochar. We assumed that differences between the two biochars were mainly driven by production temperature rather than initial feedstock, as it has been found to be the main driver of biochar chemical composition [36–38]. Maize mixtures showed higher pH (9.1 vs. 8.9) and ash contents (54.0 vs. 51.2%) and lower volatile matter contents (35.1 vs. 38.2%) compared to *Miscanthus* mixture.

Weathering induced an increase of fixed C from around 10% to 17.1% and 16.6% for maize and *Miscanthus* mixtures. In contrast, EC and DOC showed 4 times lower values after weathering. When compared to the expected values, slightly higher EC values than expected were recorded for both mixtures after weathering. In addition, the weathered mixture with maize biochar showed lower DOC (50.1 vs. 57.6 mg g C<sup>−</sup>1) and higher fixed C (17.1 vs. 11.6%) than expected. The weathered *Miscanthus* mixture showed higher volatile matter than expected (37.1 vs. 31.3%) (Table 3). During weathering, the isotopic signatures remained unchanged for compost, biochars and the mixture containing maize biochar, but decreased for the mixture containing *Miscanthus* biochar. The δ13C ratios of the weathered mixtures (21.9‰) were lower than expected (25.4 and 25.2‰).

**Table 3.** Chemical characteristics of fresh (F) and weathered (W) compost, biochars and biochar-compost mixtures. Expected (exp.) values were calculated for the weathered mixtures. EC: electric conductivity; DOC: dissolved organic carbon. Data are presented as means <sup>±</sup> sd (*<sup>n</sup>* = 3) for pH, EC, DOC and <sup>δ</sup>13C. Proximate analysis was carried out for 1 sample. The letters represent differences among treatments.


\* standard deviations of pH were <0.05.

### *3.3. Biological Stability*

Cumulative CO2-C released during 1-year of incubation from fresh and weathered compost, *Miscanthus* biochar and both mixtures are presented in Figure 2. After 1 year of incubation, the fresh compost showed the highest cumulative C mineralization with values up to 30 mg g−<sup>1</sup> of initial carbon. In contrast, very few C was mineralized from *Miscanthus* biochar. The isotopic signatures of carbon were used to assess the origin of C mineralized from biochar-compost mixtures. The data indicated that compost released between 15 and 20 mg g−<sup>1</sup> C when incubated in mixtures, while biochar released between 10 and 15 mg g−<sup>1</sup> C when incubated in mixtures. Compost showed lower C-mineralization in mixture compared to individual incubation. Conversely, biochar showed higher Cmineralization when combined with compost compared to individual incubation.

After weathering, cumulative compost C mineralization amounted to 10 mg g−<sup>1</sup> C, which was significantly lower than C mineralization of fresh compost (Figure 2). Biochar C-mineralization was not significantly affected by weathering when individually incubated. When combined with compost it mineralized significantly less than in fresh mixtures. In contrast, compost mineralized significantly more in weathered mixtures as compared to fresh mixtures and reached values between 20 and 25 mg g−<sup>1</sup> C after 1-year incubation.

### *3.4. Ryegrass Growth*

Biomass of Italian ryegrass was higher when grown on Haplic Luvisol as compared to Calcaric Cambisol, as shown for the unamended controls (Figure 3). All organic amendments stimulated ryegrass growth, when applied to Calcaric Cambisol. However, when applied to Haplic Luvisol, organic amendments induced neutral or negative effects on biomass. For both soils, application of fresh biochar-compost mixtures did not lead to significant differences in ryegrass biomass as compared to fresh compost alone. Physical weathering decreased the effect of compost addition to Calcaric Cambisol on biomass, but

the effect was still positive as compared to the control. Concerning the Haplic Luvisol, compost addition tended to decrease biomass. For both soils and after weathering, the mixture containing *Miscanthus* biochar induced significantly higher biomass than the compost alone, while the mixture containing maize biochar showed similar effects as compost alone.

**Figure 2.** Cumulative CO2-C mineralized from biochar and compost when incubated alone or in mixture. Turquoise and red colors represent C mineralized from compost and biochar respectively. Data represent means from 3 replicated samples. The colored ribbon represents the standard deviations. The letters represent the significant differences from a two-ways ANOVA analysis (*n* = 3).

**Figure 3.** Biomass of ryegrass after addition of compost or its mixture with maize and *Miscanthus* biochars, grown on two soil types. Data are presented as means ± sd (*n* = 3). The letters represent the significant differences from a one-way ANOVA analysis (*n* = 4) within each treatment and soil type.

### **4. Discussion**

### *4.1. Weathering Effects on Material Properties*

Physical weathering induced much higher mass loss from compost as compared to biochar and mixtures. This may probably be explained by the high leaching losses. Biochar mass loss amounted to 75 mg g−1, which is much lower than observed for gasification biochar [29]. This may be due to the lower friability of biochar produced by pyrolysis making it less prone to particle losses [30]. Lower mass loss for the mixtures than ex-

pected (Figure 1), may be explained by protection of compost from leaching losses by its association with the biochar structure [12,39]. Both weathering cycles may affect release of dissolved organic matter and cause cracking on biochar-surfaces, thus leading to changes in pore structure [40]. While DOC was lower than expected in weathered mixtures, EC values were higher than expected (see below). We therefore suggest that there may be interactions between biochar and compost leading to solid particles retention during weathering treatment.

Compost weathering induced a decrease of the content of all main elements, following strong leaching due to weathering treatment (Table 2). However, weathering of biochars affected only C and O contents and led to decreasing C content and increasing O content. Our results are consistent with data of Naisse et al. [29], who suggested that these observations may indicate oxidation processes induced by weathering [41]. In contrast, weathering of the mixtures increased their C contents, while it decreased their O contents. This might be related to a preferential elimination of O relative to C in the labile fraction of the mixtures. This hypothesis may be supported by the visual observation of high loss of soluble compounds during weathering. Indeed, strong decreases of DOC and EC of the remaining substrates indicated that soluble compounds were removed by leaching during artificial weathering (Table 3). In contrast to the mixture containing *Miscanthus* biochar, the DOC content of the mixture containing maize biochar decreased slightly stronger than expected. The strong decrease of EC as a result of weathering is consistent with the results of Yao et al. [42], who evidenced a rapid decline of EC from 0.7 to 0.2 mS cm−<sup>1</sup> following leaching losses from biochar. EC reduction after weathering may be due to the leaching of mineral biochar compounds. This is supported by the lower ash content of the material remaining after weathering. Ashes and volatile compounds were both partly removed during weathering, except for volatile compounds of maize biochar. Both ashes and volatile compounds compose the labile fraction of all materials and are more likely to be leached than the more stable compounds. In particular, ash represents the mineral material contribution, which may be an indicator of nutrient content [43].

Fixed C slightly decreased for compost and biochars following weathering treatment, while it increased for the mixtures (Table 3). Fixed C is mainly composed by fused aromatic C structures and may be used as an indicator of the C sequestration potential of biochars [44]. Higher fixed C of the mixtures than the expected values after weathering might result from the increasing chemical recalcitrance of the materials due to labile compounds leaching. These observations are in agreement with the lower than expected δ13C ratios of the mixtures, might indicate preferential leaching of 12C enriched compounds, e.g., C3-compost or labile polysaccharides, which are 13C enriched compared to recalcitrant compounds [45].

### *4.2. Biological Stability*

### 4.2.1. Biological Stability of the Fresh Materials

During the incubation, compost showed the highest cumulative C-mineralization, while biochar C hardly mineralized. C-mineralization of the mixtures ranged between those of its individual components. These results are in agreement with other studies [13,14,16] and may be explained by a higher content of labile C in compost than in biochar [5]. It was interesting to note that compost showed a lower C-mineralization when combined with biochar than when incubated individually. Two mechanisms could explain observation: the adsorption of labile fraction on the biochar surface [13], and the presence of phenolic compounds or salts originating from biochar [24,25,27], which might inhibit microbial activity in compost-biochar mixtures. The opposite effect was observed for biochar, since biochar C mineralized more when combined with compost than when individually applied. Indeed, several studies showed positive priming effect when labile substrates were added to biochar [46–48].

### 4.2.2. Effect of Weathering on the Biological Stability

The cumulative C-mineralization from compost after 1 year of incubation was significantly lower for weathered compost compared to fresh compost when individually incubated (10 vs. 30 mg g<sup>−</sup>1). This negative effect of weathering on C-mineralization from compost was attributed to the strong leaching of easily mineralizable labile components. On the other hand, the absence of weathering effects on biochar C mineralisation may be explained by the high stability of biochar with only few labile compounds [48].

C-mineralization from compost in the mixture increased significantly after weathering, when compared to the fresh mixtures (Figure 2). This may be due to the protection of labile compounds by biochar and/or the removal of biochar compounds, which inhibited microbial activity and thus C-mineralization from compost (see above). Indeed, fresh biochar may contain large amounts of salts, which may inhibit microbial activity when applied to soil [49–51]. This could lead to the negative priming effect of biochar on native C often observed immediately after soil addition [52].

Weathering also reduced biochar C-mineralization, within the mixtures (Figure 2), most probably due to the leaching of easily mineralizable C and nutrients from compost, which stimulated biochar C-mineralization before weathering (see above). Our results thus indicate that weathering affects biochar-compost interaction in mixtures, which might also impact their effects on plant growth.

### *4.3. Ryegrass Growth*

### 4.3.1. Effect of the Fresh Media on Ryegrass Growth

Higher ryegrass biomass was recorded when grown on Haplic Luvisol as compared to Calcaric Cambisol, regardless the organic amendment (Figure 3). Moreover, the addition of organic amendments containing compost had positive effects on biomass when applied on Calcaric Cambisol, but the effects were neutral or negative when applied to Haplic Luvisol (Figure 3). Our results were consistent with the results of Von Glisczynski et al. [53], who also did not find any plant growth promoting effect of biochar-compost mixtures application on Haplic Luvisol. As reviewed by Faucon et al. [54], organic amendments such as compost may promote plant growth by providing readily available nutrients or releasing them through mineralization. The available P concentration of the Calcaric Cambisol was much lower than that of the Haplic Luvisol (19.66 vs. 71.18 mg kg<sup>−</sup>1) (Table 1), suggesting a possible P-limitation for plant growth in this soil, which might have been alleviated by compost application.

Addition of biochar compost mixtures led to similar ryegrass biomass than compost along (Figure 3). As reported in the literature, the combination of biochar with compost can have synergic [32,55], antagonistic [23,56] or neutral effects [16,18,23,57,58] on plant growth. Several factors may impact plant growth after biochar-compost mixtures addition and the mechanisms are still poorly understood [17]. It was suggested that pre-treatment of biochar may be beneficial for plant growth before its soil application [59]. Moreover, it was shown that weathering may alter biochar properties [29]. Therefore, we tested in the following, if weathering of biochar/compost mixtures influenced plant growth.

### 4.3.2. Effect of Weathered Amendments on Ryegrass Growth

Irrespective of the soil type, weathered compost had negative or neutral effects on biomass when individually applied (Figure 3). This is most likely due to the weatheringinduced loss of readily-available nutrients and easily-mineralizable C compounds (Table 3 and Figure 2).

The addition of weathered biochar-compost mixtures to both soils had neutral or positive effects on biomass compared to the effect of compost applied individually depending on the biochar feedstock (Figure 3). The positive effect of the weathered *Miscantus* mixture on biomass may result from better compost mineralisation through the removal of compounds, which inhibit microbial activity as discussed above (Section 4.3.1). However, the weathered maize mixture showed neutral effect on biomass when compared to the effects

of weathered compost alone. Our results showed that weathering of biochar-compost mixtures could lead to positive growth effect. These results are in agreement with a recent field study, showing positive growth effects on the second crop after soil application [60]. In addition, our results also showed that neutral effects of weathering depending on biochar feedstocks and/or soil type may occur [60,61]. Further studies would be needed to investigate the mechanisms controlling the variation of biochar-compost interactions on plant growth over time.

### **5. Conclusions**

We investigated the effect of two biochar-compost mixtures and weathering on their material properties, biological stability and on plant growth after addition to two contrasting soils. Our results showed that the physical weathering led to the alteration of material properties of the mixtures, in particular through leaching of labile compounds. These effects could impact the mineralisation of the mixture and also plant growth after soil addition. We suggest that the mixtures contained inhibitive compounds for microbial activity in their labile fraction, as shown by the negative effect on compost mineralisation when combined with biochar. The increase of compost mineralisation within the mixtures after weathering may have provided more plant available nutrients, which could promote plant biomass production when compared to individual compost application. On the other hand, biochar mineralisation was also affected by weathering, indicating that weathering may influence its C sequestration potential.

We conclude that biochar-compost interactions are evolving after physical weathering most probably due to its effect on leaching of soluble compounds. The effect of fresh and weathered biochar-compost mixtures on plant growth depend on biochar production conditions. Further studies should focus on mechanisms influencing the nutrient supply of biochar-compost mixtures.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2073-439 5/11/2/336/s1, Figure S1: Experimental set up used for physical weathering of organic amendments, Figure S2: Pot experiment with ryegrass.

**Author Contributions:** M.-L.A., C.G., S.H., C.N., D.H. and C.R. designed the study. M.-L.A. and C.N. carried out the laboratory work, and exploited the results. Y.L.B. and S.B. contributed data. M.-L.A. wrote the first draft of the manuscript. All authors discussed the results and commented on the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** The study was funded by FUI (project BIOCHAR21). ADEME provided a PhD grant.

**Data Availability Statement:** Data will be made available upon request.

**Acknowledgments:** We thank FUI for funding under the framework of the project BIOCHAR21. MLA acknowledges ADEME for providing her PhD grant. We thank Valérie Pouteau for help with the laboratory work.

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

### **References**


*Article*

## **Cover Crop Selection by Jointly Optimizing Biomass Productivity, Biological Nitrogen Fixation, and Transpiration E**ffi**ciency: Application to Two** *Crotalaria* **Species**

### **Verónica Berriel 1,\*, Jorge Monza <sup>2</sup> and Carlos H. Perdomo <sup>3</sup>**


Received: 24 May 2020; Accepted: 24 July 2020; Published: 1 August 2020

**Abstract:** *Crotalaria spectabilis* and *Crotalaria juncea* are cover crops (CC) that are used in many different regions. Among the main attributes of these species are their high potential for biomass production and biological fixation of nitrogen (BNF). Attempting to maximize these attributes, while minimizing water consumption through high transpiration efficiency (TE), is a challenge in the design of sustainable agricultural rotations. In this study, the relationship between biomass productivity, BNF, and TE in *C. spectabilis* and *C. juncea* was evaluated. For this purpose, an experiment was carried out under controlled conditions without water limitations and using non-inoculated soil. BNF was determined by the natural abundance of 15N, while TE was estimated by several different methods, such as gravimetric or isotopic method (13C). *C. juncea* produced 42% less dry matter, fixed 28% less nitrogen from the air, and had 20% less TE than *C. spectabilis*. TE results in both species were consistent across methodologies. Under simulated environmental conditions of high temperature and non-limiting soil water content, *C. spectabilis* was a relatively more promising species than *C. juncea* to be used as CC.

**Keywords:** *Crotalaria spectabilis*; *C. juncea*; 15N natural abundance; 13C isotopic composition; transpiration efficiency

### **1. Introduction**

The use of legumes as cover crops (CC) in agricultural rotations makes it possible to reduce the production costs associated with a lower use of nitrogenous fertilizers, which also results in environmental benefits [1,2]. CCs are also used to reduce soil erosion caused by high precipitation, minimize surface runoff, and provide channels to the subsurface layers of the soil, allowing an increased infiltration rate [3,4].

The use of the genus *Crotalaria*, in particular *C. juncea* and *C. spectabilis*, as CCs has been recommended for warm and temperate regions [5]. Some of the main attributes of these species are their rapid and high productivity of biomass (8 Mg ha<sup>−</sup>1) [6–8] and their high content of foliar nitrogen, obtained by biological nitrogen fixation (BNF) at an average of 150 kg N ha−<sup>1</sup> [9–11]. In addition, a characteristic of these species is that they have the ability to establish a promiscuous and functional symbiosis with the native rhizobia of the soil [12]. The biomass production of CCs, including *C. juncea*

and *C. spectabilis*, is positively correlated with the recycling of nutrients, the entry of carbon (C) into the soil [13–15], and a decrease in the rate of erosion [3]. Furthermore, high concentrations of foliar N derived from BNF determine a low C/N ratio, which favors the rapid decomposition of plant remains [16,17]. The ease of degradation of this material also facilitates net N mineralization, which can be used by subsequent crops [18].

For these reasons, in a sustainable production system, it is necessary that plant species used as CCs, if they are legumes, have a high BNF and also high biophysical gain rates (biomass productivity) in relation to the consumed or transpired water [19,20]; in other words, a high water use efficiency (WUE) or transpiration efficiency (TE). A low TE and excessive water consumption can not only waste soil water reserves, but can also induce a water deficit in the subsequent cash crop and reduce its yield [21]. For the genus of *Crotolaria*, there is little information about TE, so it was interesting to evaluate this attribute and its relationship with others that have been more studied, such as biomass production and BNF [6–8,10].

However, as there are different methodological approaches to assess TE, we needed to find a simple but robust indicator for these species. The reference technique consists of computing the ratio between total biomass productivity and transpired water during the whole crop cycle [20,22], providing an integrated value of TE for the entire plant growing period. Two other methods provide only a one-time "snapshot" of TE. The instantaneous foliar WUE is the ratio of the photosynthetic rate (A) to the transpiration rate (E), while the intrinsic foliar WUE is the proportion of A to stomatal conductance (g) [23,24]. In contrast, the 13C isotopic composition (δ13C) of plants with C3 photosynthetic metabolism has also been used to estimate the TE of plants in a time-integrated manner [25,26]. Through models, it is possible estimate from δ13C the intrinsic WUE (iWUE) [25,27,28].

In a previous work, we compared the biomass productivity and the WUE of these two *Crotalaria* species, but under conditions of a moderate deficit of water in the soil. We found *C. spectabilis* showed superior behavior [29]. In this work, under controlled conditions and non-limited water, our objective was to relate the productivity of the biomass, BNF, and TE in these species. In addition, another secondary objective was to study the consistency between the methodologies that estimate TE, to understand its robustness and precision.

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

### *2.1. Plant Materials and Growing Conditions*

*Crotalaria juncea*, *Crotalaria spectabilis* (obtained from Brseeds Sementes Co., Araçatuba, Brazil), and corns the seeds were planted in plastic pots containing 4 kg Argiudol soil from the south of Uruguay (latitude—34.6 S and longitude—55.6 W). Soil characteristics: soil organic carbon = 11.6 g kg−<sup>1</sup> soil; organic matter = 20.0 g kg−<sup>1</sup> soil; sand = 245 g kg−<sup>1</sup> soil; silt = 487 g kg−<sup>1</sup> soil; clay = 268 g kg−<sup>1</sup> soil). The plants were not inoculated and noduled with the rhizobia in the soil. Ten days after the initial emergence of seedlings, the plants were thinned to one per pot, and perlite was placed on the soil surface to minimize water evaporation. The pots were kept in a growth chamber at 30 ± 3 ◦C, with variable relative humidity between 30% and 50%, and a light intensity of 1200 μmol m−<sup>2</sup> s−<sup>1</sup> with a 16/8 h cycle (light/dark). The growth chamber was continuously monitored by a computer system.

Soil moisture was kept constant at 100% (*w*/*w*) at container capacity for 75 days. The amount of water needed to achieve soil water capacity was estimated daily as the difference between the target gravimetric content and the actual water content in the soil. The sum of these daily differences was the evapotranspiration (ET) accumulated during the plant growing cycle. Transpiration (T) was determined as the accumulated loss of water from pots with plants, minus the average value determined in pots without plants and with perlite on the surface.

### *2.2. Biomass Productivity and Characteristics of Nodules*

Seventy five days after starting the experiment (before flowering), the aerial parts of the plants (leaves, stems, and leaves + stems = shoots) were harvested and dried at 60 ◦C until they reached a constant weight, and then the dry mass of each plant was weighed. The roots were washed and the nodules were considered, according to their size, as larger or smaller nodules, the latter being about half the size of the large ones.

### *2.3. Determination of Transpiration E*ffi*ciency*

### Gravimetric method

The *TE* was calculated based on Equation (1) as the quotient between the biomass produced by the aerial part (shoot) and the accumulated plant transpiration throughout the experiment:

$$TE = \frac{\text{shot dry mass}}{T} \,\text{s}\tag{1}$$

### *2.4. Gas Exchange Measurements*

Intercellular CO2 concentration, A, g, and E were determined using the youngest fully expanded leaf of all plants 70 days after sowing. These determinations were made using a portable photosynthesis system (LI-6400, LI-COR Inc., Lincoln, NE, USA); the photosynthetically active radiation was set to 1200 μmol m−<sup>2</sup> s−2, and the leaf temperature at 25 ◦C. The CO2 concentration of the chamber was adjusted to 400 μL L<sup>−</sup>1.

### *2.5. Determination of Nitrogen Concentration and Stable Isotopic Composition of Plant Parts*

Samples from different plant parts (leaves, stems, and leaves + stems = shoots) were first ground with a fixed and mobile knife mill (Marconi MA-580) until a particle size of less than 2 mm was achieved, and then with a rotary mill (SampleTek 200 vial Rotator). Determination of N-total concentration and natural abundance of 13C and 15N was determined on a Flash EA 1112 elemental analyzer coupled to a Thermo Finnigan DELTAplus mass spectrometer (Bremen, Germany). Isotopic relationships were expressed in delta notation (δ) in parts per thousand (‰), using the following equation [30]:

$$\delta^{13}\mathcal{C}\text{ or }\delta^{15}N = \left(\frac{R\_{sample}}{R\_{standard}} - 1\right) \times 1000. \tag{2}$$

Carbon 13C isotope discrimination (Δ13C) was calculated according to Farquhar et al. [25], where <sup>δ</sup>13Catmosphere is the <sup>δ</sup>13C value of air (−8‰) and <sup>δ</sup>13Cplant is the <sup>δ</sup>13C value of the plant sample:

$$
\Delta^{13}\text{C} = \left(\frac{\delta^{13}\text{C}\_{atmospler} - \delta^{13}\text{C}\_{plant}}{1 + \frac{\delta^{13}\text{C}\_{atmosplac}}{1000}} - 1\right) \times 1000. \tag{3}
$$

The ratios between the intercellular (in the plant) and air CO2 concentration and the intrinsic WUE (iWUE) were determined from the following equations [25]:

$$
\delta \mathcal{W} \text{UE} = \frac{\text{Ci}}{\text{Ca}} = \frac{\Delta^{13} \mathcal{C}\_{plant} - 4.4}{22.6} [4]. \tag{4}
$$

Biological nitrogen fixation was estimated with Equation [6], according to Unkovich et al. [25]:

$$BNF = \left(\frac{\delta^{15}N\_{ref} - \delta^{15}N\_{fix}}{\delta^{15}N\_{ref} - B}\right) \times 100,\tag{5}$$

### where:

*BNF* is the percentage of N in the plant, derived from BNF. δ15Nref is the δ15N value of the non-fixing reference plant. δ15Nfix is the δ15N value of the fixing plant. *B* is the δ15N value of a fixing plant growing in N-free growth medium.

Corn was the non-fixing reference plant used, with an <sup>δ</sup>15N isotopic composition of <sup>−</sup>8‰ (average value of 12 plants), while in *C. juncea* and *C. spectabilis*, the reported B values of −2.25‰ [31] and −1.0‰ [32] were respectively assumed.

### *2.6. Experimental Setup*

A completely randomized design was used; the pot was the experimental unit and the species was considered the treatment. The experiment was repeated in the same plant growth chamber in two time periods (with the same set of environmental parameters and the same duration in time), that were named batch 1 and batch 2. Nine pots of each *Crotalaria* species were used in each batch. Close to the *C. spectabilis* and *C. juncea* pots, six pots with corn plants and eight with soil but without plants were randomly placed. Between the two batches, 17 plants of *C. spectabilis* and 14 of *C. juncea* plants culminated the experiment. The scheme of the experiment is shown in Figure 1.

**Figure 1.** Design of the experiment. Circles represent the pots in the plant growth chamber.

### *2.7. Statistical Analysis*

In order to test if there was a difference in five variables (foliar concentration of N, T, TE, A, and E) in each *Crotalaria* species between the two batches, we carried out a Shapiro–Wilks test to evaluate normality, an F-test and a Student's *t*-test. According to the results obtained, the F-test showed that the variances could be considered as equal because the p-value was superior to 0.05. In the Student's *t*-test, the null hypothesis (the differences between means is equal to 0) could not be rejected in any of the species at a significance level of 0.05. Within a specie, no statistically significant difference at α = 0.05 was found between batches for any of the evaluated parameters. For this reason, the data for the two batches were pooled for each species.

In the pooled data, also the normality was evaluated with the Shapiro–Wilks test, while the assumption of equality of variances was evaluated with Levene's test. After, the species effect was analyzed by ANOVA in those variables with a normal distribution (N, T, TE, A, E, and A/E), and by the Kruskal Wallis test for variables without a normal distribution (shoot dry mass, g, A/g, δ15N, BNF, δ13C, and iWUE). Pearson correlation analyses were also performed. The statistical packages InfoStat [33] and XLSTAT [34] were used in the statistical analyses.

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

### *3.1. Biomass and Nitrogen Productivity from Fixation*

In simulated environmental conditions, with a high temperature and non-limited soil water availability, the two species differed both in terms of biomass productivity (Tables 1 and 2) and foliar N concentration (Tables 1 and 3). *C. spectabilis* was the species that produced the highest biomass and had the higher leaf N concentration (Table 1). All *C. spectabilis* plants and 57% of *C. juncea* presented large pink nodules. The remaining 43% of the *C. juncea* plants also had pink nodules, but these were small. The same trend with respect to nodulation was observed between the two analyzed batches of *C. juncea* plants, most of them presented larger and a minority smaller nodules.

Due the species of the genus *Crotalaria* sp. showing promiscuous behavior and establishing more or less efficient symbiosis with rhizobia from the soil, the plants were not inoculated. Therefore, in this experiment, the symbiotic efficiency of the rhizobia strains present in the soil was evaluated. The difference in the size of *C. juncea* nodules may be a consequence of its nodulation by less efficient and competitive strains, as has been observed in white clover [35].

When were compared the biomass productivity and leaf N concentration in the two *C. juncea* groups (with larger and smaller nodules), a statistically significant difference in favor of the group with larger nodules was found (Tables 1 and 3). Furthermore, shoot dry matter and foliar N concentration were correlated positively with each other (shoot dry mass = 2.4415 <sup>×</sup> [N] + 0.0286, R2 = 0.3783, *p* = 0.0004). This finding is in agreement with the findings of Adams et al. [36], which stated that an increase in foliar N concentration favors photosynthetic capacity [37].

The 15N isotopic composition of the leaves (δ15N) significantly varied between the two species; while the δ15N mean in *C. spectabilis* was negative, in *C. juncea* it was positive (Table 1). Contrarily, when only the *C. juncea* group with large nodules was included in this comparison, no significant difference was found (Table 2). In turn, the mean values of δ15N in the *C. juncea* groups with larger and smaller nodules were different, being negative in the first group and positive in the second (Table 1), although they were always less than the δ15N values of the reference plant. Negative values of δ15N would indicate that the main N source was atmospheric N2 acquired by BNF, while positive values seem to point to the soil as the main N source.

The BNF proportion, estimated form the average δ15N values of whole plants, was higher in *C. spectabilis* than in *C. juncea* (Table 1). On the contrary, there was no difference in BNF between these two species when only the *C. juncea* plants with large nodules were compared with *C. spectabilis* plants (Table 2). Within the *C. juncea* plants, the BNF values were close to 85% in the group with larger nodules, but decreased to 45% in the group with smaller nodules (Table 1). In *C. spectabilis*, on the other hand, all individuals had BNF values equal to or greater than 90% (Table 1). In any case, the BNF proportion was high for both species, which is in agreement with reports from Brazilian authors [11,38]. Overall, this result suggests that *C. spectabilis* maintained high BNF values in the simulated environment, while *C. juncea* showed high variability among plants. This result contrasts, however, with that of another Uruguayan field study, in which these species, despite having been inoculated, failed to nodulate [17].


*Agronomy* **2020** , *10*, 1116

**Table 1.** Mean values of total dry matter (Total DM), transpired water mass (T), foliar N

concentration

 (Nleaf), net

photosynthesis

 (A), leaf stomatal conductance

 (g),


**Table 2.** Statistical results of the Kruskal–Wallis analysis for total dry matter (Total DM), isotopic composition of 15N (δ15N), proportion BNF (BNF), leaf stomatal conductance (g), intrinsic leaf water-use efficiency (A/g), isotopic composition of 13C (δ13C), and intrinsic plant water-use efficiency (iWUE) in *Crotalaria spectabilis* and *C. juncea,* evaluated in plants with large (+) and small nodules (−).

Means with a common letter are not significantly different (*p* > 0.05), NS: not significant.

**Table 3.** Statistical results of an ANOVA for foliar N concentration (Nleaf), transpired water (T), transpiration efficiency (TE), net photosynthesis rate (A), instantaneous transpiration rate (E), and instantaneous water-use efficiency (A/E) in *Crotalaria spectabilis* and *C. juncea,* evaluated in plants with large (+) and small nodules (−).


Means with a common letter are not significantly different (*p* > 0.05), NS: not significant.

On the other hand, the two *Crotalaria* species did not differ in terms of photosynthetic rate (Table 3), stomatal conductance (Table 2), and transpiration rate (Table 3). However, the transpiration and photosynthetic rate were significantly higher in *C. juncea* plants with large nodules and a higher BNF (Table 3). Moreover, the transpiration rate (E) in the *C. juncea* group with higher nodulation was significantly higher than in *C. spectabilis* (Table 3).

The mass of transpired water (T) during the plant growing cycle was higher in *C. spectabilis* than in *C. juncea* (Tables 1 and 3), and besides, T was positively correlated with the aerial biomass (Figure 2). This result was consistent with what was reported for these two same species when they grew under controlled conditions but went through a period of moderate water deficit [29]. Contrarily, no significant T difference was found when *C. spectabilis* plants were compared with *C. juncea* with larger nodules (Table 3). The T mean, however, was significantly higher in the *C. juncea* group with larger nodules and a higher BNF.

**Figure 2.** Relationship between shoot dry mass and water transpiration expressed for *Crotalaria spectabilis* (rhombuses) and *Crotalaria juncea* (circles). *C. juncea* was evaluated at two nodulation levels. Plants with large nodules are identified with gray circles, and those with small nodules with white circles. Regression lines: *<sup>y</sup>* <sup>=</sup> 2.893*<sup>x</sup>* <sup>−</sup> 0.2. R2 <sup>=</sup> 0.7896 (*<sup>p</sup>* <sup>&</sup>lt; 0.0001).

The water footprint, which corresponds to the amount of water used to generate 1 kg of dry matter, was on average 515 and 342 L water/Kg dry matter for *C. juncea* and *C. spectabilis,* respectively. Therefore, *C. juncea* was less efficient in the use of water resources than *C. spectabilis*. If the water supply of these crops in the field were only rainwater, the water footprint of both species could be classified as green [39].

The isotopic composition of 13C, evaluated as δ13C, was different between species and lower in *C. juncea* (Tables 1 and 2), which was due to the greater isotopic fractionation of 13CO2 in this species [40]. As comparisons between species were made in the same environment and developmental circumstances, the δ13C values are related to genetic differences [41]. In addition, the 13C isotopic composition within *C. juncea* plants was not related to BNF, because there were no differences between the groups with the largest and smallest nodules; that is, plants that fixed more and less N (Table 2).

### *3.2. Transpiration E*ffi*ciency and Water Use E*ffi*ciency*

In both species, the mean values of the different WUE indicators evaluated in this work (TE, A/E, A/g, iWUE) were consistent, and showed that *C. spectabilis* was more efficient than *C. juncea* in the use of water resources (Table 1). Interestingly, the mean TE of *C. spectabilis* was higher than that of *C. juncea*, (Table 1), regardless of the size of the nodules and the BNF values of the latter species (Table 3). Regarding A/E, A/g, and iWUE, significant differences were observed between the species, but not between *C. juncea* plants with different nodule sizes (Table 2).

When both species were grouped, positive correlations between iWUE and the other instantaneous WUE indicators, such as A/g, were found (Figure 3; Table 4). This outcome agrees which the findings of Johnson et al. [42] and Read et al. [43]; they found negative correlations between A/g and Δ13C in different *Agropyron desertorum* clones, observed both under conditions without hydric limitation and under drought conditions. Overall, these results highlight the robustness of the isotopic methodology for the study of these parameters.

**Figure 3.** Relationship between the integrated intrinsic water use efficiency (iWUE) and foliar water use efficiency [quotient: photosynthesis (A) and stomatal conductance (g)] for *Crotalaria spectabilis* (rhombuses) and *Crotalaria juncea* (circles). *C. juncea* was evaluated at two nodulation levels. Plants with large nodules are identified with gray circles, and those with small nodules with white circles. Regression lines in a): *y* = 0.43*x* + 42.2. R2 = 0.66 (*p* < 0.0001).

**Table 4.** Pearson's correlation matrix of transpiration efficiency (TE) in *C. spectabilis* and *C. juncea*, efficiency in the use of leaf intrinsic water (A/g), isotope composition of 15N (δ15N), proportion of biological fixation of N (BNF), foliar N concentration (N), and efficiency in the use of intrinsic water from the entire plant (iWUE).


\*\*\* Significant at the 0.001 level (2-tailed), \*\* Significant at the 0.01 level (2-tailed), \* Significant at the 0.05 level (2-tailed), NS: non-significant.

A positive correlation was also established between BNF and iWUE (Table 4), as also reported by Kumarasinghe et al. [44]. These authors found a negative correlation between BNF and 13C isotopic discrimination in different *Glycine max* cultivars subjected to saline stress conditions. However, Knight et al. [45], working in greenhouse conditions, reported a positive correlation between both variables. They attributed this result to the 13C depletion that occurred at the leaf level, which was caused by isotopic fractionation mechanisms within N-fixing plants.

The foliar N concentration was also positively correlated with TE and iWUE (Table 4). Results obtained by Evans et al. [36] through metadata analysis of multiple plant species suggested that low Δ13C values (or high δ13C values) in fixing plants with high N contents were a consequence of relatively high A/g ratios.

The results indicate that *C. spectabilis* is more promising than *C. juncea* for use as a CC in this evaluation under controlled conditions. Although the results in these conditions may not be fully extrapolated to field conditions, it is important to highlight that the plants were able to nodulate with rhizobia present in soil with no history of these CCs. This is auspicious for regions where there is no commercial availability of specific rhizobia for *Crotalaria*. Similarly, the plants were harvested in the same phenological state as that used in the field to finish the CC, so it is expected that the same trends will be maintained regarding the evaluated attributes. In any case, although this first approach is necessary, field evaluation must also be carried out with the use of the same isotopic technique used

in this work to determine TE, given its consistency with other forms of evaluation of this attribute and being that its main advantages are the simplicity of sampling and the precision of the results.

### **4. Conclusions**

This study shows that under simulated conditions of high temperature and non-limiting soil water content *C. spectabilis* has advantages for use as a CC over *C. juncea* in terms of biomass production, BNF, and transpiration efficiency. Furthermore, these results suggest that the 13C isotopic technique is a robust indicator to differentiate TE between these species. In *C. juncea,* the 13C isotope indicator was not useful to distinguish between plants with low and high TE. In contrast, the 15N isotope was useful to detect differences in TE between plants. Finally, although these results are valid only for these two species, this methodology of selecting legumes based on multiple objectives could also be applied to other species or cultivars—not only those destined to be used as CCs, but also cash crops.

**Author Contributions:** Conceptualization, V.B. and C.H.P.; methodology, V.B.; formal analysis, V.B and C.H.P.; investigation, V.B and C.H.P.; resources, V.B.; data curation, V.B and C.H.P.; writing—original draft preparation, V.B, J.M. and C.H.P.; writing—review and editing, V.B, J.M. and C.H.P.; visualization, V.B. and C.H.P.; supervision, J.M. and C.H.P.; project administration, V.B.; funding acquisition, V.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by National Research and Innovation Agency of Uruguay, Funds: María Viñas, grant number FMV\_125492; University of Republic of Uruguay Funds: Fellowship CAP grants; and Faculty of Science, Funds: Fellowship Biotechnology Postgraduate. The APC was funded by the National Research and Innovation Agency of Uruguay.

**Acknowledgments:** The authors give thanks to J. Berriel and G. Galindo for the experimental work in growth plant chamber, G. Quero for photosynthesis measurements, and S. Álvarez for the graphical design.

**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/).
