*Article* **In Vitro Propagation Method for Production of Phenolic-Rich Planting Material of Culinary Rhubarb 'Malinowy'**

**Agnieszka Wojtania 1,\* and Monika Mieszczakowska-Fr ˛ac <sup>2</sup>**


**\*** Correspondence: agnieszka.wojtania@inhort.pl

**Abstract:** Culinary rhubarb is a popular vegetable crop, valued for its long, thickened stalks, very rich in different natural bioactive ingredients. Tissue cultures are a useful tool for vegetative propagation of virus-free rhubarb plants and rapid multiplication of valuable selected genotypes. The aim of this study was to develop an effective method for in vitro propagation of selected genotypes of Polish rhubarb 'Malinowy' characterized by high yield and straight, thick and intensive red stalks. Identification and quantification of anthocyanins and soluble sugars by the HPLC method in shoot cultures and ex vitro established plantlets were also performed. Shoot cultures were established from axillary buds isolated from dormant, eight-year-old rhizomes. Effective shoot multiplication of rhubarb 'Malinowy' was obtained in the presence of 6.6 μM benzylaminopurine or 12.4 μM *meta*-topolin. Both cytokinins stimulated shoot formation in a manner that depended on sucrose concentration. Increasing the sucrose concentration from 59 to 175 mM decreased the production of shoots and outgrowth of leaves by 3-fold but enhanced shoot length, single shoot mass and callus formation at the base of shoots. This coincided with increased accumulation of soluble sugars (fructose, glucose) and anthocyanins-cyanidin-3-O-rutinoside (max. 208.2 mg·100 g−<sup>1</sup> DM) and cyanidin-3-O-glucoside (max. 47.7 mg·100 g−<sup>1</sup> DM). The highest rooting frequency (94.9%) and further successful ex vitro establishment (100%) were observed for shoots that were earlier rooted in vitro in the presence of 4.9 μM indole-3-butyric acid. Our results indicated that anthocyanin contents in leaf petioles were influenced by developmental stage. Under in vitro conditions, it is possible to elicit those pigments by sucrose at high concentration and *meta*-topolin.

**Keywords:** anthocyanins; *meta-*topolin; micropropagation; *Rheum*; soluble sugars; sucrose concentration

#### **1. Introduction**

Rhubarb (*Rheum*) is a herbaceous perennial of the *Polygonaceae* family. The genus *Rheum* includes about 60 species, and most of them are native to the northern and central regions of Asia [1]. For thousands of years, *Rheum* has been cultivated in China for medicinal purposes. The dried rhizomes and roots of medicinal species (*R. palmatum* L., *R. officinale* Baill*, R. tanguticum* Maxim.) have been used to treat constipation, inflammation and ulcers [2,3].

Culinary rhubarb is popular as a vegetable crop, valued for its long, thickened stalks [4]. They are used in the production of desserts, cakes, jam, juices, fruit teas and wine. The use of rhubarb petioles for food was discovered at the beginning of the 18th century in Great Britain, but widespread consumption of rhubarb stalks began in the early 19th century. Culinary rhubarb cultivation spread to northern Europe, North America, Australia and New Zealand. Worldwide rhubarb production and consumption peaked just before the Second World War, then it came to be restricted [5]. Recently, there has been an increased interest in rhubarb production in Europe, including in Poland. This is due to the higher interest of consumers and the food industry in functional food [6]. It is known that

**Citation:** Wojtania, A.;

Mieszczakowska-Fr ˛ac, M. In Vitro Propagation Method for Production of Phenolic-Rich Planting Material of Culinary Rhubarb 'Malinowy'. *Plants* **2021**, *10*, 1768. https://doi.org/ 10.3390/plants10091768

Academic Editors: Laura Pistelli, Kalina Danova and Iyyakkannu Sivanesan

Received: 23 June 2021 Accepted: 20 August 2021 Published: 25 August 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/).

rhubarb stalks contain useful levels of organic acids (malic acid, citric acid, fumaric acid and ascorbic acid), dietary fiber, protein, potassium, calcium and magnesium [7]. They are also rich in polyphenolic compounds, such as stilbenes, anthocyanins and flavonols, which have a range of bioactivities relevant to human health [8–10]. Rhubarb juice may be used as a natural food preservative. It has been shown that the addition of rhubarb juice to strawberry jams or apple purees reduced unfavorable color changes and increased antioxidant properties [7]. Consumers and the food industry prefer red stalks that are sweeter and have a higher content of polyphenolics. It has been shown that the level and composition of phenolic compounds depend on the rhubarb cultivar [11].

The culinary rhubarb called *R. rhaponticum* L. is a hybrid between rhubarb species originally brought to Europe for medicinal purposes—*R. rhaponticum*, *R. undulatum* L. (syn. *R. rhabarbarum* L.) and *R. palmatum* L. [5,12]. At an early stage, growers in England started to develop culinary rhubarb cultivars, including the still-common 'Victoria' and 'Prince Albert'. Nowadays, the number of culinary rhubarb cultivars is difficult to estimate. Propagation of culinary cultivars by seed and careful selection of superior plants resulted in the development of many local cultivars with marked improvement in yield, quality and uniformity.

For successful commercial cultivation of rhubarb genotypes, a planting material of good quality is very important. In horticultural practice, seed propagation is not recommended because the obtained plants are usually not uniform and do not repeat parental traits. Plants differ in height, habit, shape and intensity of petiole color, as well as petiole size and number [13]. On the other hand, the production of rhubarb by division of crowns is limited by the low number of donor plants available and the risk of virus transfer. Rhubarb crops were found to be infected with several viruses. The most common are: *Turnip mosaic virus* (TuMV), *Arabis mosaic virus* (ArMV), *Cucumber mosaic virus* (CMV) and *Cherry leaf roll virus* (CLRV) [14–17]. A good alternative for the production of highquality planting material is in vitro propagation of virus indexed rhubarb plants. Moreover, tissue culture of rhubarb can be a source of bioactive compounds. In rhizomes and leaf petioles of *Rheum rhabarbarum*, the accumulation of catechin, gallic acid, p-cumaric acid, rosmarinic acid, isoquercitrin and resveratrol has been reported [18]. Several articles have been published on the vitro propagation of different culinary rhubarb selections [18,19] and cultivars, including 'Victoria' [20], 'Big Red' [16] and 'Karpov Lipskiego' [21]. For many genotypes, bacteria contaminations, slow culture establishment, low activity of axillary buds and hyperhydricity have been reported [16,18,20]. Mentioned difficulties caused that commercial micropropagation of some valuable rhubarb selections, including 'Malinowy', has not been attained.

The aim of this study was to develop an efficient in vitro propagation technology for the selection of rhubarb 'Malinowy'. Identification and quantification of anthocyanins and soluble sugars by the HPLC method in shoot cultures and ex vitro established plantlets were performed. The development of procedures for rapid in vitro clonal propagation of value rhubarb genotypes may be of great commercial value to the rhubarb industry.

#### **2. Results**

#### *2.1. Culture Initiation*

Shoot cultures were initiated from axillary buds isolated from dormant, eight-yearold rhizomes. We obtained 51.9–67.6% uncontaminated explants that developed shoots. Contamination of initial explants of *Rheum* 'Malinowy' partly depended on the sanitary status of the mother plant (Table 1, Figure 1A). In MS medium without growth regulators, no explant developed shoots (Table 2). Shoot tips and axillary buds cultured in MS media supplemented with BAP and GA3 or NAA started to develop into single shoots in 2–3 weeks (Figure 1B). After 3–4 subcultures in the same fresh medium, new shoot formation was observed (Figure 1C). The best growth and development of initial shoots from dormant buds was observed in the modified MS medium containing 75% nitrogen salts, 4.4 μM BAP and 0.3 μM GA3 (Table 2). After three subcultures in media containing

BAP + NAA, the developed shoots of 'Malinowy' were characterized by pale green leaves and a tendency to hyperhydricity, some of which died. The negative effect of auxin was enhanced when full-strength MS medium was used. It was observed that lowering the strength of nitrogen salts by one quarter in MS medium containing BAP + GA3 slightly enhanced new shoot formation from initial explants (Table 2).

**Table 1.** Effect of mother (donor) plants on contamination of initial explants of Polish rhubarb 'Malinowy' in vitro.


\* 'Malinowy'.

**Figure 1.** Initiation and stabilization stage of culinary rhubarb 'Malinowy' micropropagation: (**A**) 8-year-old mother plant collected from the plantation at the beginning of November; (**B**) initial shoot developed from axillary bud after 3 weeks of culturing in initiation medium; (**C**) the multiplied shoots at the stabilization stage.

**Table 2.** Influence of nitrogen salts (75% and 100% MS medium) and growth regulators on the survival rate and initial shoot formation of rhubarb 'Malinowy' after 3 subcultures in the same fresh medium.


<sup>1</sup> Means marked with the same letter do not differ significantly (*p* = 0.05) according to Duncan's test; the lowest value is marked with "a".

#### *2.2. Shoot Multiplication*

The results of our study showed that adenine-type cytokinin (BAP and mT) was effective in the stimulation of shoot formation of rhubarb 'Malinowy' in vitro. It was observed that BAP was the most effective in the stimulation of shoot formation when applied at a concentration of 6.6 μM and *meta*-topolin at 12 μM (Figure 2). The next experiment showed the cytokinin effect on rhubarb shoot formation and quality significantly depended on sucrose supply. The highest multiplication rate (4.8–4.9 shoots/explant) was observed in medium with the lowest sucrose content (59 mM) and in the presence of 6.6 μM BAP or 12.4 μM mT, respectively. Increasing the sucrose supply from 59 to 175 mM in the cytokinin medium decreased the production of shoots and outgrowth of leaves by 3-fold but enhanced shoot length, single shoot mass and callus formation at the base of shoots (Table 3). Progression of shoots from a juvenile to adult phase (shoots with large, dark-green leaf petioles, single root) was also observed (Figure 3A). At 117 mM sucrose in the medium, leaf petioles started to turn red, and increased sucrose supply intensified the color of rhubarb leaf petioles under in vitro conditions. Despite the highest shoot formation rate of rhubarb

'Malinowy' in medium with low sucrose content (59 mM), cyclic multiplication, especially in the presence of mT, resulted in the formation of pale green shoots, with small leaf blades and a tendency to hyperhydricity and deformation. On the other hand, *meta*-topolin stimulated more juvenile shoot formation at higher sucrose levels compared to BAP. This resulted in higher shoot production at 88 mM sucrose and higher shoot quality at high exogenous sucrose supply. Shoots produced in BAP medium with high sucrose content (175 mM) showed a tendency to leaf yellowing and swelling shoot bases (Figure 3B).

**Figure 2.** Effect of different cytokinin types (BAP, mT) and concentrations (4.4, 6.6, 12.4 μM) on shoot formation of rhubarb 'Malinowy' after a 4-week subculture period. Sucrose concentration in the medium was 88 mM. Medium without cytokinin was the Control. Means marked with the same letter do not differ significantly (*p* = 0.05) according to Duncan's test.

**Table 3.** Effect of cytokinin type (BAP, mT) and sucrose levels on shoot formation, fresh mass of shoots, leaf number and shoot length of rhubarb 'Malinowy' after a 4-week subculture period.


<sup>1</sup> Means indicated with the same letter within cytokinin type and sucrose levels do not differ significantly (*p* = 0.05) according to Duncan's test; the lowest value is marked with "a".

#### *2.3. Soluble Sugar and Anthocyanin Contents in Leaf petioles*

Among tested soluble sugars, fructose and glucose were dominant in leaf petioles of rhubarb 'Malinowy' after a 4-week subculture period in multiplication medium. Sorbitol was not detected in rhubarb 'Malinowy'. The content of soluble sugars in leaf petioles varied significantly depending on sucrose concentration in the medium and cytokinin type. Increasing sucrose concentration (from 59 to 175 mM) resulted in an increase in endogenous soluble sugar content by 60% and 75% in the presence of BAP and mT, respectively. Generally, BAP influenced higher fructose and glucose accumulation compared to *meta*-topolin, especially at the lowest sucrose level in the medium. In medium with BAP and high sucrose content (175 mM), the accumulation of a small amount of sucrose was also observed (Figure 4).

**Figure 3.** In vitro shoot formation, rooting and acclimatization of culinary rhubarb 'Malinowy'. (**A**) Shoots multiplicated in mT medium containing different levels of sucrose (from the left: 59, 73, 88 and 175 mM); (**B**) shoots from BAP medium containing 175 mM sucrose; (**C**) leaf petioles from mT medium containing 175 mM sucrose; (**D**) shoot cultures; (**E**) roots three weeks after transfer to medium with 0.49 μM IBA; (**F**) roots three weeks after transfer to medium with 4.9 μM IBA; (**G**) shoots rooted in the presence of 4.9 μM IBA; (**H**) roots after 4 weeks in medium with 4.9 μM IBA; (**I**–**J**) plantlets four weeks after transfer to ex vitro conditions (growth room); (**K**–**L**) plantlets six weeks after transfer to the greenhouse; (**M**–**O**) plantlets nine weeks after transfer to the greenhouse; (**N**–**P**) leaf petioles after different periods of growing ex vitro.

**Figure 4.** Contents of soluble sugars in culinary rhubarb 'Malinowy' after a 4-week subculture period in medium containing different cytokinin types: BAP (6.6 μM) and mT (12.4 μM) and sucrose levels (59, 73, 88, 117 and 175 mM); \* not detectable. Duncan's test was used independently for each type of cytokinin and sucrose concentration. Different letters indicate significant differences among sucrose treatments (*p* = 0.05).

We found that exogenous sucrose levels and cytokinin type had a significant effect on anthocyanin accumulation in rhubarb 'Malinowy' under in vitro conditions (Figures 3A and 5). The most abundant anthocyanin compound was cyanidin-3-O-rutinoside (max. 208.2 mg·100 g−<sup>1</sup> DM), followed by cyanidin-3-O-glucoside (max. 47.7 mg·100 g−<sup>1</sup> DM). Both cyanidin levels were enhanced in rhubarb leaf petioles by increased sucrose concentration in the medium. At a high sucrose level (175 mM), the production of anthocyanins was significantly stimulated by cytokinin. It was found that *meta*-topolin resulted in 50% higher content of anthocyanins in rhubarb petioles compared to BAP (Figure 5).

**Figure 5.** Contents of anthocyanin in culinary rhubarb 'Malinowy' after a 4-week subculture period in medium containing different cytokinin types: BAP (6.6 μM) and mT (12.4 μM) and sucrose levels (59, 73, 88, 117 and 175 mM). Means indicated with the same letter within cytokinin treatment do not differ significantly (*p* = 0.05) according to Duncan's test.

#### *2.4. Rooting and Acclimatization*

Under in vitro conditions, roots emerged 1–2 weeks after transfer to the rooting media. As shown in Table 4, rooting of culinary rhubarb 'Malinowy' was significantly affected by IBA levels. In auxin-free medium, roots formed with the effectiveness of 40%. Application of IBA resulted in an increase in root formation in a concentration-dependent manner (Table 4). The best rooting response (94.9% rooting frequency; 10.7 roots/shoot) was observed in the presence of IBA at the concentration of 4.9 μM (Figure 3F). In all treatments, root length progressively increased with time (Figure 3F) and resulted in their damage during transfer to ex vitro conditions (Figure 3H).


**Table 4.** Rooting response of rhubarb 'Malinowy' shoots after three-week growth in MS medium without auxin (control) and supplemented with IBA at different concentrations (0.49, 2.5, 4.9 μM).

<sup>1</sup> Means of each parameter marked with the same letter do not differ significantly (*p* = 0.05) according to Duncan's test; the lowest value is marked with "a".

We then compared the efficiency of acclimatization of rhubarb shoots rooted in vitro and directly ex vitro. The development of the root system was evaluated on a three-point scale (Figure 6). It was observed that in vitro rooted shoots rapidly developed a root system (Table 5, Figure 3J). Three weeks after transfer to ex vitro conditions, roots were visible through the walls of the peat plugs. This resulted in more vigorous growth of shoots and higher ability to tolerate low humidity compared to shoots directly rooted and acclimatized. After four weeks of acclimatization, in vitro rooted shoots were transferred to the greenhouse. Ex vitro rooted shoots needed at least 7–8 weeks to develop a root system and acclimatize. However, after this period of time, the rooting rate was poorer than after 4-week acclimatization of shoots rooted in vitro. The ex vitro establishment rate of in vitro rooted shoots was 100%, while those rooted ex vitro survived with 83%.

**Figure 6.** Development of the root system of culinary rhubarb 'Malinowy' on a 1–3 scale.

**Table 5.** Effect of microcutting status (rooted, unrooted) on survival, shoot growth and root system development of *Rheum* 'Malinowy' after 4 weeks of growth ex vitro (growing room).


<sup>1</sup> Means of each parameters indicated with the same letter do not differ significantly (*p* = 0.05) according to Duncan's test; the lowest value is marked with "a".

After transfer to the greenhouse, growth of the plantlets tended to decrease, and yellowing of the oldest leaves was observed. As shown, soluble sugar and anthocyanin contents in leaf petioles were very low at this stage of growth in the greenhouse (Table 6). On sunny days, the plantlets needed shading. After two weeks, new leaves started to develop. Six weeks after transfer to the greenhouse, the average length of leaf petioles was enhanced by 15.6% and leaf area by 47.8%. A rapid increase in leaf (petioles and blades) thickness was also observed (Figure 3K). Leaf petioles of all rhubarb plantlets turned red (Figure 3M–N). During six-week growth in the greenhouse, contents of anthocyanins and soluble sugars in the leaf petioles were enhanced by 4 and 7 times, respectively. They contained sucrose, glucose and fructose in a ratio 1:2:2. Total anthocyanin content in leaf petioles of 'Malinowy' was 188 mg·100 g−1, including 94% cyanidin-3-O-rutinoside and 6% cyanidin-3-O-glucoside (Table 6). No morphologically aberrant plantlets were found. The ten-week-old rhubarb plantlets were transferred (at the end of April) to the plantation for further observation.

**Table 6.** Morphological and biochemical characteristics of *Rheum* 'Malinowy' during a 6-week growth period in the greenhouse.


t.a.—trace amounts.

#### **3. Discussion**

The quality and quantity of rhubarb crops are significantly dependent on planting material. Tissue cultures are a useful tool for vegetative propagation of virus-free rhubarb plants and rapid multiplication of valuable selections with the highest possible content of bioactive ingredients. For commercial micropropagation of planting material, a successful in vitro propagation protocol including all stages (initiation of aseptic culture, shoot multiplication, rooting of microshoots, ex vitro acclimatization) is very important.

Axillary bud development has proven to be the most often applied system for producing true-to-type plantlets. In the present study, dormant axillary buds of culinary rhubarb 'Malinowy' were used successfully for the initiation of shoot cultures. By using a two-step disinfection procedure, we obtained an average of 59.5% uncontaminated explants from eight-year-old rhizomes that developed shoots of high quality. Given the age of the rhizomes and the associated large amount of dead tissue and secreted mucus, as well as the presence of endogenous bacteria, this is a high efficiency of culture establishment. Buds of 'Big Red' treated with 2% sodium hypochlorite for 15 min showed 51% establishment success. Additional methods for disinfection in sodium dichloroisocyanurate (300 mg·L−<sup>1</sup> for 20 min and 48 h) and 4% chlorine dioxide did not reduce the amount of contamination [16]. Clapa et al. [18], by using bleach solution of 20% (ACE-Protector) for 20 min, obtained 20% contaminated explants, but 36% were necrotic and only 44% were viable.

Data in the literature indicate that initial growth of culinary rhubarb shoots from axillary buds depends on different factors, including growth regulators, term of explant collecting, bud size and genotype [16–21]. The most important factor stimulating the activity of rhubarb buds was BAP added singly [16] or in combination with auxin [19,20]. We obtained the best growth response of dormant buds of rhubarb 'Malinowy' in modified MS medium containing 75% nitrogen salts, 4.4 μM BAP and 0.3 μM GA3. The combination of cytokinin and GA3 was found to increase the activity of axillary buds of many plant species, especially perennials and woody plants [22–24]. On the other hand, we found that NAA presence in the initial medium significantly decreased shoot quality. This might be related to the inappropriate type and concentration of auxin or lack of need of auxin for rhubarb 'Malinowy' shoot induction. We also observed that the negative effect of auxin was enhanced when full-strength MS medium was used. Murashige and Skoog medium with KNO3 (at 1900 mg·L−1) and NH4NO3 (at 1650 mg·L−1) is the most common medium for micropropagation of many horticultural plants, but these high concentrations of nitrogen salts are supraoptimal for some plant species and can induce different physiological disorders, including hyperhydricity [25]. Ogura-Tsujita and Okuba [26] reported that rhizome explants of *Cymbidium kanran* growing in low-nitrogen medium produced 55% less ethylene than those coincident with enhanced shoot production.

It has been reported that among the cytokinins tested, BAP was more effective than isopentenyladenine (2iP), kinetin and thidiazuron (TDZ) for axillary multiplication of various culinary rhubarb genotypes [19,21]. The optimal BAP concentration ranged from

2.2 to 22.2 μM according to rhubarb cultivar [16,20]. Kozak and Sałata [21] showed that 'Karpow Lipskiego' cultured in BAP medium was characterized by the largest number of leaves, basal tissue and fresh mass of shoots. However, the use of kinetin and 2iP enhanced the length of shoots. The genotype-dependent multiplication rate and shoot morphology of some culinary rhubarb cultivars and clones have already been observed [16]. For example, rhubarb 'Malinowy' cultured in BAP medium formed much lower shoots but with an enhanced number of leaves per explant compared to 'Karpow Lipskiego' [21]. Similar to a study on *Rheum rhabarbarum* [18], we observed that *meta*-topolin was of similar efficiency in the stimulation of axillary multiplication of rhubarb 'Malinowy' compared to BAP. Additionally, our study showed that both cytokinins stimulated shoot formation of rhubarb in a manner dependent on sucrose concentration. Increasing sucrose supply (from 59 to 175 mM) in the growing medium resulted in a reduction in the multiplication rate by 70% and the stimulation of mature shoot production. Sucrose inhibition of shoot formation was previously observed in different plant species, including *Helleborus niger* L., *Paeonia lactiflora* Pall*,* [24,27], *Pelargonium hortorum* L.H. Bailey [28] and *Rosa* 'Konstancin' [29].

It is known that sugars play important roles in in vitro cultures as energy and carbon sources and osmotic agents. They can also act as signaling molecules and/or as regulators of gene expression [30]. Sugar-mediated signals indicate carbohydrate availability and regulate metabolism with sugar usage and storage [31]. As shown in our study, the exogenous sucrose supply in cytokinin medium affected the accumulation of fructose and glucose in rhubarb leaf petioles and coincided with a reduced multiplication rate and plantlet transition from juvenile to mature phase. At high exogenous sucrose content in the medium, BAP influenced the formation of more mature shoots compared to mT. Moreover, our study showed that high exogenous sucrose content promotes the accumulation of anthocyanins in leaf petioles of rhubarb in vitro.

Anthocyanins are an important class of flavonoids that represent a large group of plant secondary metabolites. They are recognized for their diverse function in plant development and beneficial effect on human health. Generally, anthocyanin accumulation in fruits and vegetables are accompanied by their maturation and is regulated by both developmental and environmental cues [32]. Sugar-induced anthocyanin accumulation has been observed in many plant species. In *Petunia* Juss, sugars were shown to be required for the pigmentation of developing corollas [33], while in grape berry skin, sugars were found to induce most genes involved in anthocyanin synthesis [34]. Enhanced anthocyanin accumulation caused by high sucrose content in the medium has been previously reported in *Melastoma malabathricum* L. [35], *Clematis pitcheri* [36] and *Petunia* [37]. It is well known that phytohormones can also modulate anthocyanin biosynthesis by regulating the expression of genes involved in the flavonoid biosynthetic pathway [38,39]. As shown in our study, the sucrose-induced accumulation of anthocyanin in rhubarb petioles in vitro was stimulated by cytokinin, mainly *meta*-topolin. Potential of mT in stimulating the accumulation of proanthocyanidins has been previously observed for the shoot cultures of *Musa* 'William' [40]. The authors demonstrated that *meta*-topolin was the most effective in this process compared to other topolins and BAP.

During the last decade, *meta*-topolin has been widely used in micropropagation of many plant species, improving multiplication and rooting efficiency [41–43]. Our previous observation showed that the shoot of rhubarb 'Malinowy' treated with *meta*-topolin had higher rooting capacity than those cultured in BAP medium before rooting (data not shown). It was shown that the rooting ability of rhubarb in vitro is also dependent on genotype. For example, 'Karpow Lipskiego' produced 100% rooted shoots in hormone-free medium and spontaneously formed multiple roots in multiplication medium [21]. We obtained only 40% rooted shoots in auxin-free medium, but the application of IBA at 4.9 μM resulted in 100% rooting frequency and multiple root formation. Thomas [16] reported 86% rooting frequency for 'Big Red' in the presence of 2.9 μM 3-indoleacetic acid (IAA). Our study results are in agreement with those by Thomas [16], showing that rhubarb shoots

rooted in vitro showed better shoot growth and higher survival rate after transfer to ex vitro conditions compared to direct rooting and acclimatization.

We found plantlets of 'Malinowy' showed a rapid increase in leaf size, petiole length and diameter when they were transferred to the greenhouse. Anthocyanin analyses revealed that ten-week old 'Malinowy' plantlets contained a very high amount of cyanidyn-3- O-rutinoside (176.7 mg·100 g<sup>−</sup>1) and cyanidin-3-O-glucoside (11.3 mg·100 g<sup>−</sup>1). Cyanidin derivatives were found previously to be the main anthocyanin in rhubarb stalk grown in the field. It has been demonstrated that rhubarb cultivars significantly differ in their anthocyanin content and percentage of the two main cyanidins [7,9]. The study presented by Takeoka et al. [9] showed that of twenty-one cultivars, total anthocyanin content ranged from 19.8 ('Crimson Red') to 341.1 mg·100 g−<sup>1</sup> ('Valentine'). Similarly, as in our study, cyanidyn-3-O-rutinoside was the main anthocyanin present in culinary rhubarb 'Red Malinowy' [11]. Moreover, in this genotype, twenty other phenolic compounds, including flavan-3-ols, flavonols and gallotannins, were identified. Among them, the most abundant was catechin (112 ng·mg−<sup>1</sup> dry mass). The study of Clapa et al. [18] showed that phenolic composition can differ between in vitro and field-grown *Rheum rhabarbarum* plant extract. Further observations and analyses of phytochemical and nutritional compounds for rhubarb 'Malinowy' plants growing in vitro and in the field are needed.

#### **4. Materials and Methods**

#### *4.1. Plant Material*

Two donor plants of Polish garden rhubarb 'Malinowy' were carefully selected from a plantation in Ka ´nczuga, located in the Subcarpathia Province, in southeastern Poland (WGS-84: 49◦59 0" N, 22◦24 42" E). Among other plants, they were distinguished by high yield and straight, thick and intensive red stalks. The selected plants were harvested at the beginning of November. Before culture initiation, the donor plants were tested by enzyme-linked immunosorbent assay (DAS-ELISA) with commercially available antibodies against ArMV, TuMV, *Cucumber mosaic virus* (CMV), *Cherry leaf roll virus* (CLRV), *Tobacco ring spot virus* (TRSV), *Tomato ring spot virus* (ToRSV), *Tomato spotted wilt virus* (TSWV), *Tomato black ring virus* (TBRV), *Strawberry latent ringspot virus* (SLRV) (LOEWE Biochemica, Sauerlach, Germany) and *Tobacco mosaic virus* (TMV) (Agdia, Elkhart, IN) by the method of Clark and Adams [44]. The plants were found to be virus free.

#### *4.2. Culture Initiation*

Shoot cultures were established from axillary buds isolated from dormant, eightyear-old rhizomes. First, they were divided into small parts then washed thoroughly with running tap water and soaked in fungicide. Buds were isolated and sterilized by soaking in commercial bleach (ace 4 mL/water 100 mL) for 20 min and then in 0.1% HgCl2 for 5 min. After rinsing in sterile water, the explants were placed in 50 mL Erlenmeyer flasks in Murashige and Skoog [45] (MS) medium containing different levels of nitrogen salts (100% and 75%), 100 mg·L−<sup>1</sup> myo-inositol, vitamins (nicotinic acid, pyridoxine, thiamine (1.0 mg·L−<sup>1</sup> each)), 2 mg·L−<sup>1</sup> glycine, 88 mM sucrose and 6 g·L−<sup>1</sup> agar (Biocorp, Poland). The effect of 4.4 μM benzylaminopurine (BAP) added together with 0.3 μM gibberellic acid (GA3) or 0.1 μM 1-naphthaleacetic acid (NAA) was studied. Full-strength MS medium without growth regulators was the control. The pH of the medium was adjusted to 5.8 before autoclaving. After 3 subcultures (each 3 weeks), the survival rate and the number of developed shoots were determined.

The shoots in all in vitro experiments were maintained at the temperature of 20 ± 2 ◦C under a standard 16/8 h photoperiod of 40 μmol m−<sup>2</sup> s−<sup>1</sup> (warm-white fluorescent lamps).

#### *4.3. Shoot Multiplication*

Murashige and Skoog medium with nitrogen salts reduced by a quarter containing 100 mg·L−<sup>1</sup> myo-inositol, vitamins (nicotinic acid, pyridoxine and thiamine (1.0 mg·L−<sup>1</sup> each) and 2 mg·L−<sup>1</sup> glycine was used throughout the experiments. To obtain effective, cyclic

shoot multiplication of rhubarb 'Malinowy', first, the effect of two adenine-type cytokinins (BAP and hydroxybenzylaminopurine (*meta*-topolin (mT)), added in concentrations of 4.4 μM, 6.6 μM and 12.4 μM, was examined (Experiment 1). After two subcultures (each lasting 4 weeks), the multiplication rate was determined. Then, the interaction between cytokinins (BAP or mT) and sucrose added at different concentrations (59, 73, 88, 117 and 175 mM) was studied. In Experiment 2, BAP was used at 6.6 μM and mT at 12.4 μM. After two subcultures, fresh mass, number and length of shoots, and number of leaves were determined. Leaf petioles were collected, lyophilized, and crushed into a homogenous powder using a laboratory mill (A 11, IKA, Staufen, Germany). The samples were then subjected to qualitative and quantitative analyses of sugars and anthocyanins by the HPLC method.

#### *4.4. Rooting and Acclimatization*

The aim of the experiments was to compare the efficacy of in vitro and ex vitro rooting and acclimatization. In the last subculture before rooting, shoots were grown in medium containing 4.1 μM *meta*-topolin.

For in vitro rooting, the selected shoots approximately 4 cm long were placed in modified MS medium containing 88 mM of sucrose and indole-3-butyric acid (IBA) at different concentrations (0, 0.49, 2.5 and 4.9 μM). After 3 weeks, rooting frequency, number of roots per explant and length of roots were determined. For the acclimatization experiment, shoots containing at least 5–6 roots were selected.

For ex vitro rooting and acclimatization, the shoots were taken directly from mT medium. The bases of shoots (approx. 4 cm long) were dipped in commercial rooting powder (Rhizopon AA 0.5%, Poland). Both types of microcutting (in vitro rooted and unrooted) were planted in multicellular trays 30 mm in diameter with a mixture of peat (Alonet Substrat, SIA Florabalt, LV-5106 Valle, Latvia) and perlite (2:1) in plastic plug boxes covered with transparent plastic caps to prevent dehydration. Ex vitro rooting and acclimatization took place in a growth room (25 ± <sup>2</sup> ◦C; PPFD—50 <sup>μ</sup>mol m−<sup>2</sup> <sup>s</sup>−1). The microcuttings were hardened by gradually decreasing air humidity. After 10 days, they were fed with 0.1% Kristalon (Yara Vlaardingen B.V., Netherlands) containing 18:18:18 (*w/w/w*) NPK. Four weeks after transfer to ex vitro conditions, the following data were collected: length of plantlets and the visual estimation of the root system on a threepoint scale (1—no roots or single, short roots; 2—poorly developed root system, 3—welldeveloped root system, roots out of trays) (Figure 6).

After four weeks of acclimatization, in vitro rooted shoots were transferred to a greenhouse (at the end of February). They were grown at a temperature of 20/18 ◦C (day/night) with a 16 h photoperiod provided by additional lighting. Plants were manually watered and fed with 0.1% Kristalon. Plantlet growth (length of leaf petioles and leaf area) and biochemical status (soluble sugars and anthocyanin contents) were assessed after 1 and 6 weeks of growth in the greenhouse.

#### *4.5. Soluble Sugar Analysis*

The sugars were quantified by calibration curve for sucrose, glucose, fructose and sorbitol standards in the concentration range of 20–250 mg/100 mL, and the results are expressed as mg·100 g−<sup>1</sup> dry mass (DM). An example chromatogram of sugar separation is shown in the Supplementary Materials (Figure S1). HPLC analysis of sugars was carried out with the HP 1200 system (Agilent Technologies, Waldbronn, Germany) equipped with an RI Detector with a BioRad Aminex-87C column (300 × 7.5 mm) according to European Standard EN 12630. The mobile phase was water purified by the MiliQ System (Milipore, Molsheim, France), isocratic flow was 0.6 mL·min−<sup>1</sup> and column temperature was 80 ◦C. The lyophilized powder (100 mg) was extracted in 4 mL of redistilled water for 20 min in an ultrasonic bath. The suspension was then centrifuged (7000× *g*, 10 min). The resulting extract for sugar determination was filtered through a Sep-Pak® PLUS C18 filter (Waters, Ireland).

#### *4.6. Estimation of Anthocyanins by HPLC*

HPLC analysis of anthocyanins was performed according to the method described by Nielsen et al. [46] with some modifications. In short, 5 μL of the eluate was analyzed using an Agilent HPLC Model HP 1200 equipped with a diode array detector (DAD). Separation was performed on a Phenomenex®Fusion RP column (250 mm × 4.6 mm; particle size = 4 μm) using a mobile phase consisting of water/ formic acid (95:5 *v/v*) (A) and acetonitrile (B). Elution profile: 0–16 min, 3%–9% B; 16–30 min, 12% B; 30–54 min, 33% B; 54–58 min, 90% B; 58–62 min, isocratic 90% B. The anthocyanins in the eluate were detected at 520 nm and a temperature of 25 ◦C. Their amounts were quantified by calibration with the standards of cyanidin-3-O-glucoside and cyanidin-3-O-rutinoside and expressed in mg·100 g−<sup>1</sup> dry mass (DM). An example chromatogram of anthocyanins separation is shown in Supplementary Material (Figure S2).

The lyophilized powder (50 mg) was extracted in 2 mL of 60% methanol acidified with 1% formic acid for 20 min in an ultrasonic bath. The suspension was then centrifuged (7000× *g*, 10 min). The resulting extract for phenolic compound determination was filtered through a PTFE filter (0.45 μm, 15 mm).

#### *4.7. Statistical Analysis*

For the multiplication and in vitro rooting experiments, 30 shoots (6 shoots × 5 glass jars) were used in each treatment. For ex vitro rooting and acclimatization, 50 rooted and 50 unrooted shoots were used. The experiments were carried out twice. The final data were the means of the two replicated experiments. The data were subjected to one- (rooting and acclimatization experiment) or two-factor analysis of variance (ANOVA). The significance of the differences between means was evaluated by Duncan's test at *p* = 0.05.

#### **5. Conclusions**

We developed a practical protocol for mass propagation of the selected genotype of rhubarb with a high content of anthocyanins. It enhanced the availability of phenolic-rich planting material for the establishment of commercial rhubarb plantations. The study indicates that growth and development of rhubarb shoots in vitro and secondary metabolite production are modulated by cytokinin and sucrose concentration. It was found that *meta*-topolin is a very useful cytokinin for rhubarb 'Malinowy' in vitro, which can be a good alternative for cultivars revealing multiplication and rooting difficulty in the presence of BAP. Although direct ex vitro rooting is possible, rhubarb shoots rooted in vitro showed better ex vitro establishment and growth in the greenhouse. Hence, in vitro rooting of rhubarb 'Malinowy' is preferred for mass production. Our results indicated that anthocyanins in leaf petioles of rhubarb plantlets were influenced by plant developmental stage and in vitro conditions. It is possible to elicit anthocyanins by high sucrose concentration combined with *meta*-topolin. Finally, the results obtained give important suggestions for introduction into the market and cultivation for rhubarb 'Malinowy', following the latest research conducted globally also on other interesting wild and cultivated plants [47–49].

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/plants10091768/s1, Figure S1: HPLC chromatograms of: (**A**) sugars in standard solution; (**B**) sugars in rhubarb 'Malinowy'. Peak identification: 1—sucrose (Rt −8.1 min.), 2—glucose (Rt −9.8 min), 3—fructose (Rt −12.7 min), 4—sorbitol (Rt −21.4 min.). Figure S2: HPLC chromatograms of: (**A**) antocyanins in standard solution; (**B**) antocyanin in rhubarb 'Malinowy'. Peak identification: 1—cyanidin-3-O-glucoside (Rt −22.3 min.), 2—cyanidin-3-O-rutinoside (Rt −25.7 min); Rt—retention time.

**Author Contributions:** A.W.—designed and performed all experiments, collected and analyzed data, prepared figures and tables, writing—review and editing of the paper M.M.-F.—performed analyses of soluble sugars and anthocyanins, collected data and revised the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the European Union under the "European Agricultural Fund for Rural Development: Europe investing in rural areas", Grant Number 00009.DDD.6509.00020.2019.09.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Acknowledgments:** The authors thank Prof. Mirosława Cie´sli ´nska for virus analyses and Prof. Małgorzata Podwyszy ´nska for kindly revising this manuscript.

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

#### **References**


#### *Article* **Repression of Carotenoid Accumulation by Nitrogen and NH4 + Supply in Carrot Callus Cells In Vitro**

**Tomasz Oleszkiewicz \*, Michał Kruczek and Rafal Baranski**

Department of Plant Biology and Biotechnology, Faculty of Biotechnology and Horticulture, University of Agriculture in Krakow, 31-425 Krakow, Poland; kruczek.michael@gmail.com (M.K.); rafal.baranski@urk.edu.pl (R.B.)

**\*** Correspondence: tomasz.oleszkiewicz@urk.edu.pl

**Abstract:** The effect of mineral nutrition on the accumulation of the main health beneficial compounds in carrots, the carotenoid pigments, remains ambiguous; here, a model-based approach was applied to reveal which compounds are responsible for the variation in carotenoid content in carrot cells in vitro. For this purpose, carotenoid-rich callus was cultured on either BI (modified Gamborg B5) or R (modified Murashige and Skoog MS) mineral media or on modified media obtained by exchanging compounds between BI and R. Callus growing on the BI medium had abundant carotene crystals in the cells and a dark orange color in contrast to pale orange callus with sparse crystals on the R medium. The carotenoid content, determined by HPLC and spectrophotometrically after two months of culture, was 5.3 higher on the BI medium. The replacement of media components revealed that only the N concentration and the NO3:NH4 ratio affected carotenoid accumulation. Either the increase of N amount above 27 mM or decrease of NO3:NH4 ratio below 12 resulted in the repression of carotenoid accumulation. An adverse effect of the increased NH4 <sup>+</sup> level on callus growth was additionally found. Somatic embryos were formed regardless of the level of N supplied. Changes to other media components, i.e., macroelements other than N, microelements, vitamins, growth regulators, and sucrose had no effect on callus growth and carotenoid accumulation. The results obtained from this model system expand the range of factors, such as N availability, composition of N salts, and ratio of nitrate to ammonium N form, that may affect the regulation of carotenoid metabolism.

**Keywords:** *Daucus carota*; carotene; nitrate; ammonium; somatic embryogenesis

#### **1. Introduction**

The carrot is a well-known vegetable grown around the world for its nutritious storage root. The roots of the most common carrot varieties accumulate carotenoids, mainly β-carotene and α-carotene, which give them their orange color. Both carotenes have provitamin A activity and, together with other carotenoids, play beneficial roles in human health [1]. The high carotene content makes carrots one of the most important sources of carotenoids in the human diet [2], and knowledge on carotenoid biosynthesis, accumulation, and regulation of these processes is essential for the development of highquality carrot varieties.

Carotenoids exist widely in nature. They are 40-carbon molecules built from eight base isoprenoid units. They are classified to two main groups: carotenes, being hydrocarbons such as β-carotene and α-carotene, and xanthophylls, which are oxidized carotenes [3]. The processes of carotenoid biosynthesis in plants, including carrot, have been well described [4]. In recent years, research has focused on understanding the regulation of biosynthetic pathway and carotenoid sequestration. Currently, it is known that developmental and environmental factors, such as light, influence carotenoid accumulation in carrot cells [5,6]. Field conditions and genotype have a pronounced effect on carotenoid

**Citation:** Oleszkiewicz, T.; Kruczek, M.; Baranski, R. Repression of Carotenoid Accumulation by Nitrogen and NH4 <sup>+</sup> Supply in Carrot Callus Cells In Vitro. *Plants* **2021**, *10*, 1813. https://doi.org/10.3390/ plants10091813

Academic Editors: Laura Pistelli and Kalina Danova

Received: 10 August 2021 Accepted: 27 August 2021 Published: 31 August 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/).

accumulation in carrot storage roots in contrast to plant fertilization [7]. However, fertilization with calcium ammonium nitrate increased carotenoid content [8]. The effect of other N salts was equivocal unless an additional foliar nutrition with a complex fertilizer was applied [9]. Variation in carotenoid content among carrot cultivars was also reported depending on the applied urea dose [10]. Thus, the conclusions regarding the effect of nutrition on carotenoid metabolism in carrot remain ambiguous. Recently, it was demonstrated that ammonium ions negatively affect carotenoid accumulation in *Calendula officinalis* callus cultured in vitro [11]. Another genetic research based on callus response to changes in the composition of mineral medium indicated that N supply affected pigment accumulation [12].

For more than 60 years, the carrot has been considered a model species in research on totipotency, somatic embryogenesis, and horizontal gene transfer (for review see [13] and [14]), while broad genetic research, including the recent genome sequencing project [15], led to the development of high-quality varieties [16]. Carrot is amenable to cell and tissue culture on mineral media in vitro. Callus can be induced in in vitro culture from various explants by the supplementation of mineral medium with growth regulators and then it can be easily propagated [17], hence, it has become a convenient material for research on stress factors, genetic transformation, and genome editing, including genes of the carotenoid pathway [18,19]. Carrot callus usually accumulates low amounts of carotenoids, as do the storage root meristematic cells used to induce callus [20,21]. However, the development of carotenoid-rich callus was also reported [22–24], and recently, it has been successfully used for structural studies of carotene crystals [25,26], regulation of carotenoid biosynthesis, sequestration, and interaction of carotenoid and cell wall composition when it was subjected to targeted mutagenesis using novel tools of genome editing, i.e., clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR associated (Cas9) proteins [19,24].

Inorganic components in the medium are necessary for the growth of plant tissues and organs in vitro. Their contents and composition influence tissue and plantlet development, hence, the proper balance of medium components is required [27]. For the induction of carrot callus development, and for its further propagation, a medium based on Gamborg B5 [28] mineral salts and vitamin composition is more effective and more often used than the Murashige and Skoog (MS; [29]) medium [30]. The MS medium is recommended for somatic embryogenesis and carrot plant regeneration [31,32]. Our preliminary experiments showed that attempts of plant regeneration using a modified MS medium, the R medium, led to a visually paler color of callus due to a reduced carotenoid accumulation in comparison to callus grown on the B5-based BI medium. Both media, BI and R, differ in the composition of salts, vitamins, plant growth regulators, and in the sucrose concentration. Thus, in this study we sought for an answer to which component of the R medium is responsible for the repression of carotenoid accumulation. For this purpose, we used a model carotenoid-rich callus [23] and exposed it to media with modified compositions. A substantial reduction of carotenoid content was observed on the B5 medium with altered composition of nitrogen salts, thus, we show here that the amount of N and NO3:NH4 ratio are key factors affecting carotenoid accumulation in carrot cells.

#### **2. Results**

#### *2.1. Callus Growth and Morphology*

Carrot callus cultured on both BI (modified Gamborg B5) and R (modified MS) media (Supplementary Table S1) grew with a similar rate. Callus cultured on the BI medium retained its characteristic morphology throughout all experiments. It had dense and lumpy structure, with small parts being more friable. It retained orange color, although rarely, friable callus was paler (Figure 1a). Callus growing on the R medium differed from callus on the BI medium in color, which became paler and eventually light orange over the course of time (Figure 1b). Microscopic observations revealed that regardless of the medium, callus cells were densely packed in aggregates (Figure 2a,b). Carotene crystals, clearly

distinguishable due to their intense orange color, were sequestered in the cells of dark orange callus on the BI medium (Figure 2c). Cells of callus maintained on the R medium contained only small crystals, and they were not abundant (Figure 2d). Additionally, proembryogenic tissue was identified in callus on the R medium. Embryoid structures were visible (Figure 1b inset); however, their development was arrested at early stages. They did not convert into plants, turned brown, and often died or dedifferentiated to new callus cell layers.

**Figure 1.** Carrot callus growing on mineral media. (**a**) BI medium; (**b**) R medium; (**b-inset**) embryo-like structure.

**Figure 2.** Densely packed carrot callus cells (not macerated tissue) growing on either the BI (**a**) or R medium (**b**). Easily noticeable carotene crystals in cells growing on the BI medium (**c**) and sparse crystals in cells growing on the R medium (**d**). Bar = 100 μm.

#### *2.2. Carotenoid Content in Callus*

Two main carotenes were identified in callus using HPLC (Table 1 and Supplementary Figure S1). The sum of α- and β-carotene contents in callus growing on the BI medium was very high (2264 μg/g DW) and exceeded that in callus on the R medium (425 μg/g DW) by 5.3 times, which corresponded to differences in color observed between calli on both media. The β/α carotene ratio (2.6) was also higher for the BI medium than for the R medium (1.5) (Table 1). The sum of α- and β-carotenes determined by HPLC highly correlated (*r* = 0.98, *p* < 0.001) with the total carotenoid content, determined spectrophotometrically, although it was usually higher. A linear relationship was described by a well fitted regression line (*p* < 0.001) with the coefficient of determination R2 = 0.96 (Figure 3). Hence, quantitative determination of carotenoid content in further experiments was done using spectrophotometry.

**Medium** α**-carotene <sup>1</sup>** β**-carotene <sup>1</sup>** β**:**α **Ratio Total Carotenoids <sup>2</sup>** BI (modified Gamborg B5 medium) <sup>627</sup> <sup>±</sup> 73 a 1637 <sup>±</sup> 305 a 2.6 1169 <sup>±</sup> 89 a R (modified MS medium) 172 ± 55 b 253 ± 113 b 1.5 404 ± 62 b BI/MS-macro (BI with macroelements as in MS (R)) 304 ± 35 b 529 ± 101 b 1.7 531 ± 44 b R/B5-macro (R with macroelements as in B5 (BI)) 725 ± 55 a 2191 ± 182 a 3.0 1322 ± 99 a BI:R ratio 3.6 6.5 nd <sup>3</sup> 2.9

**Table 1.** Carotenoid content [μg/g DW] in callus growing on the BI and R media.

<sup>1</sup> determined by HPLC, <sup>2</sup> determined spectrophotometrically; means ± std. error (*<sup>n</sup>* = 4); means followed by different letters within column are significantly different at *p* = 0.05; <sup>3</sup> not determined.

**Figure 3.** Fitted linear regression model for carotenoid content (μg/g DW) in carrot callus, determined based on HPLC (sum of α- and β-carotene) and spectrophotometric measurements (total carotenoids).

#### *2.3. Effect of Medium Composition*

To identify compounds that affected the carotenoid content in callus, 12 media varying in the composition of main compound groups were compared. For this purpose, the composition and amounts of macroelements, microelements, vitamins, growth regulators, or sucrose in the BI medium were replaced by the corresponding compound groups, and in the same amounts, as present in the R medium (Table 2). Analogous modifications were applied to the R medium by replacing compound groups to be the same as in the BI medium. Callus growing on the modified media showed changes in color and carotene contents (*p* < 0.001). The replacement of macroelements in the BI medium caused callus discoloration and the decrease of carotenoid content by 54.6% to a level similar to the R medium (Table 2). The effect of B5 macroelements added to the R medium was also significant. Callus growing on the R/B5-macro medium had an intense orange color and over 3-fold increased carotenoid content in comparison to the unmodified R medium, reaching the carotenoid level present in callus on the BI medium (Table 2). Thus, the Gamborg B5 composition of macroelements stimulated carotenoid accumulation in contrast to the MS formulation of macroelements. Any other changes done to either the BI or R media composition (microelements, vitamins, growth regulators, sucrose) did not significantly affect carotenoid content (Table 2). These results indicated that macroelement composition in the medium was critical for carotenoid accumulation in callus. The only other effect of modified media was the formation of embryoid structures on the BI medium free of 2,4-D and kinetin that resembled structures observed in callus on the R medium.




**Table 2.** *Cont.*

<sup>1</sup> %BI—carotenoid content expressed as the percentage of the content in callus growing on the BI medium; <sup>2</sup> *P*—significant at *p* < 0.05 (\*) or not significant (ns) difference from either BI or R considered as the reference (ref) according to the Dunnett test; <sup>3</sup> Modified Gamborg B5 [28] medium; <sup>4</sup> Modified MS [29] medium.

#### *2.4. Effect of Macroelements*

Various salts of macroelements were present in the BI and R media, thus, any element or their combination could affect carotene content. The contents of individual macroelements in the BI medium were modified to get the same molar concentrations as in the R medium. Modifications to either P, K, Ca, or Mg contents did not result in changes either of callus morphology, color, or carotenoid content. A noticeable callus discoloration was observed only on the medium with a modified composition of N salts. Callus exposed to the BI/MS-N medium developed more white or pale orange cell aggregates. It accumulated almost 40% less carotenoids than callus on the BI medium and had similar amounts of carotenoids as callus on the R medium (Table 2). The N content depends on the amounts of NH4NO3 and KNO3 salts, and KNO3 supplementation results in the increased concentration of both N and K. To verify the effect of K on carotenoid content, the R and BI/MS-N media were supplemented with K2SO4 (R+K and BI/MS-N+K, respectively). No effect of additional K amounts on carotenoid level was found.

#### *2.5. Effect of N Concentration and NO3:NH4 Ratio*

The N content in the R medium (60.02 mM) was more than doubled in comparison to the BI medium (26.76 mM) (Table 3). To verify the effect of N concentration on carotenoid accumulation, media differing in composition of N salts were compared (Supplementary Table S2). The increase of N amount in the range from 27 mM to 80 mM did not affect callus growth but significantly reduced carotene content from 1252 μg/g DW to 411 μg/g DW. Such changes were highly significant independent from whether the NO3:NH4 ratio in all comparing media was the same, i.e., 12.19 (the same as in the BI medium) or it was increasing in the range from 12.19 up to 38.5 (both *p* < 0.001). In both sets of media, the observed reduction of carotene content followed similar trends described by logarithmic functions with R<sup>2</sup> of 0.8249 and 0.9490, respectively (Figure 4).

The R medium contained NH4NO3 not present in the BI medium, which had NH4 + ions supplied in a low amount of (NH4)2SO4 not present in the R medium (Table 3). In consequence, the R medium had more N, mainly due to the use of an amount 10.2 times higher of the ammonium form while the nitrate form was only 1.6 times higher. The effect of N form in the medium on the carotene content was verified by using media containing 26.76 mM N, the same as in the BI medium, adjusted by using both nitrate and ammonium salts in various ratios. While decreasing the NO3:NH4 ratio down to 1.91 (the same as in the

R medium), callus grew similar to the callus exposed to the BI medium with the NO3:NH4 ratio of 12.19. Further elevation of the NH4 <sup>+</sup> amount in the media restricted callus growth, which was eventually inhibited on the medium with the 1:1 NO3:NH4 ratio. The increasing NH4 <sup>+</sup> level also highly reduced carotenoid content in callus (*p* < 0.001). In comparison to the BI medium, the carotenoid content was reduced 3.0-fold at the NO3:NH4 ratio of 1.91 (the same as in the R medium), and a further increase of NH4 <sup>+</sup> to the 1:1 NO3:NH4 ratio reduced the carotenoid content by 11.5-fold to the level of 115 μg/g DW. Such changes in the carotenoid content followed a trend described by a well fitted logarithmic function with R<sup>2</sup> = 0.9685 (Figure 5).


**Table 3.** The BI and R media composition with regard to nitrogen content.

<sup>1</sup> not determined.

**Figure 4.** Carotenoid content in callus grown on the BI media with modified N content and with the constant (12.2) NO3:NH4 ratio (squares, solid line) or with increasing NO3:NH4 ratio from 12.2 to 38.5 (dots, dashed line). Lines represent logarithmic functions: y = 3551.5 − 749.5ln(x) for the constant NO3:NH4 ratio and y = 3852 − 820.7ln(x) for the increasing NO3:NH4 ratio; whiskers—std. error.

**Figure 5.** Carotenoid content in callus grown on BI media with a modified NO3:NH4 ratio and with the constant N content (26.8 mM). The line represents a logarithmic function y = 107.65 + 478.27ln(x); whiskers—std. error.

#### **3. Discussion**

Nutrient supply and their uptake by plants determine yield and quality of agricultural products, including carrot, and available data indicate that fertilization, in particular with N, may also affect carotenoid accumulation in carrot storage roots. Previous evaluations of NPK fertilization showed that genotype and environmental conditions affected carotenoid accumulation rather than N supply [7]. These conclusions were supported by results of a multiyear field trial using various N fertilizers, although significant increase of carotenoid content was achieved after foliar nutrition [9]. Fertilization with urea suggested variation in carotenoid content in used cultivars depending on the urea dose, although differences between overall means were insignificant [10], while fertilization with calcium ammonium nitrate increased carotenoid accumulation in two cultivars in a two-year trial [8]. The conclusions of field studies on plant nutrition remain ambiguous as the results are highly affected by complex environmental factors, additionally interacting with variety.

Experiments utilizing cell and tissue culture in vitro allow to apply controlled conditions that, in particular, are essential in plant nutrition research, and which are not possible to obtain in field conditions. Therefore, we have applied a research model to elucidate at cellular level the role of nutrition on the accumulation of the main carrot health beneficial compounds, the carotenoid pigments. The MS-based mineral media had already been used to culture cell suspension or to induce callus for carotenoid research. The cell suspension or callus from a red storage root carrot variety accumulated mainly β-carotene and lycopene; the level of these pigments highly varied and was clone-dependent [33,34]. For *Arabidopsis thaliana*, the carotenoid content in wild type callus cultured on the medium containing MS salts was low, 200–550 μg/g DW [35]. For *Tagetes erecta* [36], individual carotenoids were identified and not quantified, but the assessment of color and HPLC profiles also indicated low amounts of pigments. The Gamborg B5-based mineral media were used to induce development of light-orange [24] and dark-orange, carotenoid-rich, carrot callus accumulating up to the same amounts of carotenoids (2150 μg/g DW) as the storage root from which

such callus was derived [23]. Thus, carotenogenesis was ongoing in materials cultured on the MS-based media, but the B5-based media were much more efficient for pigment accumulation. The observed color variation of carrot callus cultured on different mineral media in vitro have indicated that accumulation of carotenoid pigments is stimulated or repressed by media components. The mineral compositions of BI and R media, used in this work, differed significantly as they were essentially based on the Gamborg B5 and MS formulations, respectively. Both media differed mainly in their N salts composition. The amount of N was 2.24 times higher in the MS medium, and N was supplied in NH4NO3 and KNO3 salts, of which the former was present in a higher concentration, thus, the NO3:NH4 ratio in MS was 1.91 (Table 3). The B5 medium was richer in the nitrate salt by 32% but contained a low amount of ammonium (NH4)2SO4 salt, thus, the NO3:NH4 ratio in B5 was 12.19. Hence, the B5 medium contained 10.2 times less ammonium N form than MS. In this study, we found that callus grown on the BI medium and on any modified BI medium containing N salts according to the Gamborg B5 formulation accumulated many more carotenoids than when using MS-based N salts.

Both BI and R media differed also in the composition of other compounds. The MS medium contained three times more CaCl2, 50% more MgSO4, and had KH2PO4 instead of NaH2PO4. Although these differences are less pronounced than differences in N salts composition, they were also taken into account in this study. It was previously reported that a reduction of Ca supply promoted carotenoid accumulation in the roots of carrot plants, but this effect was variety dependent, with the most significant effect on lycopene content in a lycopene accumulating variety [37]. When using a lycopene accumulating carrot cell suspension, it was shown that increasing the initial P content in the medium or resupplying P during the culture increased carotenoid accumulation [38]. It was also shown that the increase of 2,4-D up to 10 ppm promoted carotenoid accumulation in carrot cells [33]. A higher sucrose concentration increased carotenoid content, with 3%−5% sucrose being optimal, while 8% sucrose had adverse effects on carrot cells growth and their size [39]. Additionally, the carotenoid accumulation increased in *Calendula officinalis* callus when sucrose concentration was raised from 4% to 7% [11]. In our work, no significant changes in carotenoid accumulation in carrot callus was found when modifying the compositions of Ca, P, K, and Mg salts. Further BI medium modifications by replacing microelements, vitamins, elimination of growth regulators, and reduction of sucrose from 3% to 2%, as present in the R medium, had no significant impact. No response of carrot callus to all these modifications supports the conclusion that N availability is the prime factor affecting carotene accumulation. This finding is in contrary to the results presented by Hanchinal et al. [40], who modified N, P, and sucrose concentrations and used a response surface methodology to optimize β-carotene production by carrot cells in suspension. A doubled N concentration to 50 mM with increased sucrose content from 2% to 3% increased β-carotene production up to 13.61 μg/g DW. However, it must be underlined that they used cells accumulating very low amounts of carotenoids, two magnitude lower than callus in our work, which may highly bias the conclusions.

Our results showed also that a gradual increase of N from 26.76 mM to 80.04 mM restricts carotenoid accumulation, which eventually decreased 3-fold. Nitrogen was supplied mainly in the form of KNO3, thus, the amounts of N and K in the medium were interrelated. Further media adjustments with K salts to keep this element at the same level while increasing N concentration showed that K did not affect carotenoid accumulation, confirming that the N amount in the medium is a critical factor and, moreover, its effect is independent on the NO3:NH4 ratio in the range from 12.2 to 38.5. Additionally, the amount of N did not alter callus growth. Recent study on grape callus showed that the reduced N amount in the MS medium from 60 mM to 40–50 mM enhanced accumulation of other pigments, anthocyanins, in red-pod okra callus; however, further reduction to 30 mM had an adverse effect [41]. Additionally, N starvation promoted anthocyanin accumulation in grape callus [42]. In contrary, other reports showed that the increase of total N content by doubling KNO3 in the MS medium increased anthocyanin content by 135% [12], which

was congruent with results showing the highest accumulation of anthocyanins using an elevated N amount (70 mM) in the medium [43].

The N content in a medium depends on the combination of supplied ammonium and nitrate salts. The ammonium N form is preferred by plants as it can be directly used, and its incorporation by a cell requires less energy. However, it can become toxic to plant cells at higher concentrations, and plant sensitivity to ammonia varies greatly depending on species, plant age, and environment pH [44]. Hairy roots of carrot, red beet, and madder in the presence of NH4 <sup>+</sup> available in the amounts in the MS medium had a reduced growth [45]. A similar effect of restricted growth was observed for anthocyanin accumulating carrot callus cultured on MS [12]. No growth changes were reported only when carrot cell suspension was exposed to a doubled amount of ammonium N form than present in MS [46]. The comparison of a wide range of NO3:NH4 ratios in our work demonstrates that when keeping the optimum N level (26.76 mM) for carotenoid accumulation, as in the B5 medium, the callus growth is restricted with increasing amounts of NH4 +, as is the carotenoid content. This adverse effect of the ammonium form led to an over 10-fold reduction of carotenoids and such response intensified logarithmically with the NH4 <sup>+</sup> concentration. A similar adverse effect of high NH4 <sup>+</sup> concentration was found in *Calendula officinalis* callus. The 50% decrease of NH4 <sup>+</sup> amount, in comparison to MS, induced carotenoid biosynthesis, and the complete removal of NH4 <sup>+</sup> from the medium further promoted carotenoid accumulation [11]. An analogous effect of increasing NH4 + concentration was reported in relation to anthocyanin accumulation. Lower anthocyanin contents were recorded in carrot cells in suspension exposed to media with a low NO3:NH4 ratio, and the increase of the ratio to 4:1 led to the highest pigment content [43]. In callus induced from rose leaves, the reduction of NH4 <sup>+</sup> and increase of NO3 − concentrations in the MS-based medium enhanced anthocyanin accumulation [47]. Conversely, doubling the concentration of NH4 <sup>+</sup> in the MS medium restricted the anthocyanin content by one third in carrot callus [12].

Nitrogen is required by plants for their growth, but the composition and concentrations of N salts in a culture medium affect also morphogenesis and embryogenesis [27]. It was previously shown that a reduced form of N, present in high amounts in MS, is required for the development of somatic embryos from carrot hypocotyl explants. Three N salts with reduced N were compared and two of them, NH4NO3 and NH4Cl, favored somatic embryogenesis, while (NH4)2SO4 did not [48]. However, contrary to these results, nearly 20 times more plants developed on the B5 medium, which contained much less of the reduced N form and was supplied with (NH4)2SO4 salt only, than on the MS-based medium [49]. The comparison of several media free of growth regulators in our work has shown that the formation of proembryogenic tissue and globular embryos is ongoing regardless of the medium mineral composition. Somatic embryos were observed on both, carotenoid-rich callus and low carotenoid accumulating callus, thus, there was no clear relationship between somatic embryogenesis and N salt composition, hence, carotenoid accumulation, although quantitative comparison was not the subject of this work.

#### **4. Materials and Methods**

#### *4.1. Plant Material, Media Preparation, and Experiment Design*

Dark-orange callus derived from the root of DH1 (doubled haploid) carrot (*Daucus carota* L.) line accumulating high amounts of carotenoids and described previously [23] was used. Callus was maintained on filter paper disks laid down on the surface of solidified mineral medium in 9 cm Petri dishes and cultured at 26 ◦C in the dark. In each experiment, callus was grown for eight weeks with one transfer to a fresh medium after four weeks. Two main media were used: (1) the BI medium consisting of Gamborg B5 macro- and microelements with vitamins, 30 g/L sucrose, 1 mg/L 2,4-D, and 0.0215 mg/L kinetin, and (2) the R medium consisting of MS macro- and microelements with vitamins (with an increased glycine content to 3 mg/L), 20 g/L sucrose, and free of growth regulators (Supplementary Table S1). Both media had pH adjusted to 5.8 and were solidified with

2.7 g/L phytagel. Macro- and microelements, including vitamin mixtures, separate macroand microelement mixtures, vitamins, and plant growth regulators were purchased from Duchefa Biochemie (Haarlem, The Netherlands). Media modifications were done by exchanging group of components between BI and R media (12 media variants; Table 2), or by exchanging individual compounds (eight variants; Table 2), or by changing nitrogen salts composition and their concentration (14 variants) (Supplementary Table S2).

#### *4.2. Microscopic Observations*

To observe callus structure, small callus pieces were placed on a microscopic slide in a water drop under the cover slide. Carotenoid crystals were observed in single cells after tissue maceration in 1N HCl at 50 ◦C for 5 min. Observations were done in a bright-field using the Zeiss Axiovert S100 microscope with ×10 objective. Images were collected by using the attached digital camera.

#### *4.3. Determination of Carotenoid Content*

Eight-week-old callus was lyophilized and ground into a fine powder in a beading mill for 5 min. Carotenoids were extracted from 5–10 mg samples with 500 μL of acetone in 1.5 mL tubes. Samples were vortexed for 30 s and centrifuged at 18,000 g for 5 min. Acquired extracts were transferred to fresh tubes. The procedure was repeated to ensure complete extraction, and then, obtained extracts were combined. The absorbance of extracts was measured in a 1 cm QS quartz cuvette (Hellma Analytics, Müllheim, Germany) at 450 nm using the NanoDrop 2000c (ThermoScientific, Waltham, MA, USA) spectrophotometer. Extractions and measurements were done for each sample in triplicate and the readouts were averaged before statistical analysis. The total carotenoid content was calculated based on the β-carotene extinction coefficient (*A*1% <sup>1</sup>*cm* = 2500) using the formula:

$$\frac{\frac{\text{absolute (450 nm)}}{2500} \times \text{extract volume (ml)} \times 10000}{\text{sample mass (g)}}$$

The results are presented in μg of carotenes per gram of callus dry weight.

High performance liquid chromatography (HPLC) measurements were done using the same callus samples as for spectrophotometry. HPLC was performed as previously described [19]. Briefly, the extraction was performed using ethanol:*n*-hexane (1:1, *v*:*v*) and HPLC was performed using Shimadzu LC–20AD chromatograph equipped with a C18 RP (5 μm) column and the Shimadzu SPDM–20A–DAD photodiode-array detector. The identification of β-carotene was based on the retention time of the standard and confirmed by analysis of absorption spectra. The identification of α-carotene was based on the analysis of the absorption spectra. Quantification of β-carotene was done using a standard curve, while α-carotene was quantified in relation to β-carotene.

#### *4.4. Statistical Analysis*

Each experiment was set up in four replicates, each consisting of five calli, and having a completely randomized design. A one-way ANOVA was performed to test effects of media composition on carotene contents in callus using the Statistica v.13.1 software (TIBCO; Palo Alto, CA, USA). Differences between means were verified at the significance level *p* = 0.05 using the Dunnett test. Means are presented with their standard errors.

#### **5. Conclusions**

In this study, we sought for the answer to which component of the culture medium affects carotenoid accumulation in carrot callus. A compound by compound replacement in the MS and Gamborg B5 media have revealed that the only critical element is nitrogen, and either the increase of the total N concentration or the decrease of NO3:NH4 ratio restricts carotenoid accumulation. Thus, the highest carotenoid content was achieved in the medium with 26.76 mM N and 12.19:1 NO3:NH4 ratio, regardless of whether the other

media components were supplied according to the MS or Gamborg B5 formulation. These model-based obtained results pave the way for further elucidation of biological processes related to regulation of carotenoid metabolism. The observed effects might be limited to a simplified in vitro model; hence, further confirmation in planta may be required. However, they can be useful for research or application purposes using cell or tissue culture where stimulation of valuable secondary metabolites, such as carotenoids, is required.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/plants10091813/s1, Table S1: Composition of Gamborg B5 and Murashige and Skoog (MS) media and their modified variants, BI and R, respectively; Table S2: Nitrogen salts composition of BI media with a modified N content or NO3:NH4 ratio; Figure S1: HPLC chromatogram at 452 nm of the sample from callus grown on a BI (control) medium.

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

**Funding:** This Research was financed by the Ministry of Science and Higher Education of the Republic of Poland. This research was supported by the National Science Centre, Poland (grant No. 2018/31/N/NZ9/02368).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** Tomasz Oleszkiewicz received financial funding for a doctoral scholarship from the National Science Centre, Poland (Etiuda No. 2019/32/T/NZ9/00463).

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

#### **References**

