**Preface to "Physiological and Molecular Characterization of Crop Resistance to Abiotic Stresses"**

Abiotic stress represents the main constraint for agriculture, affecting plant growth and productivity worldwide. Yield losses in agriculture will be potentiated in the future by global warming, increasing contamination, and reduced availability of fertile land. The challenge for agriculture of the present and future is that of increasing the food supply for a continuously growing human population under environmental conditions that are deteriorating in many areas of the world. Minimizing the effects of diverse types of abiotic stresses represents a matter of general concern. Abiotic stress in plants is a vast subject that can be addressed from different points of view and includes many different components, mainly environmental factors (e.g., soil, water, climate, irradiation, and even the influence of the moon). Plants have evolved a series of physiological and molecular mechanisms of response that may (or may not) allow them to adapt to and survive this broad range of stressful conditions. Understanding those mechanisms will help us to improve our interventions towards more sustainable and efficient agriculture. The papers included in this Special Issue cover a broad range of topics related to the effects of different abiotic stress types on crop plants, at the morphological, physiological, biochemical, and molecular levels, and the mechanisms of defense of the plants against these stresses. The methods employed were also diverse, from the analysis of agronomic traits based on morphological characteristics to omics approaches and the use of transgenics. Special attention was given to the screening for stress tolerance in local landraces, stress alleviation using different strategies, and the proposal of practical solutions for the agriculture of the (near) future, threatened by global warming and environmental pollution. The editors wish to thank the contributors, reviewers, and the editorial staff of MDPI for their professionalism.

> **Monica Boscaiu, Ana Fita** *Editors*

### *Editorial* **Physiological and Molecular Characterization of Crop Resistance to Abiotic Stresses**

**Monica Boscaiu 1,\* and Ana Fita <sup>2</sup>**


Received: 3 August 2020; Accepted: 26 August 2020; Published: 2 September 2020

**Abstract:** Abiotic stress represents a main constraint for agriculture, affecting plant growth and productivity. Drought and soil salinity, especially, are major causes of reduction of crop yields and food production worldwide. It is not unexpected, therefore, that the study of plant responses to abiotic stress and stress tolerance mechanisms is one of the most active research fields in plant biology. This Special Issue compiles 22 research papers and 4 reviews covering different aspects of these responses and mechanisms, addressing environmental stress factors such as drought, salinity, flooding, heat and cold stress, deficiency or toxicity of compounds in the soil (e.g., macro and micronutrients), and combination of different stresses. The approaches used are also diverse, including, among others, the analysis of agronomic traits based on morphological characteristics, physiological and biochemical studies, and transcriptomics or transgenics. Despite its complexity, we believe that this Special Issue provides a useful overview of the topic, including basic information on the mechanisms of abiotic stress tolerance as well as practical aspects such as the alleviation of the deleterious effects of stress by different means, or the use of local landraces as a source of genetic material adapted to combined stresses. This knowledge should help to develop the agriculture of the (near) future, sustainable and better adapted to the conditions ahead, in a scenario of global warming and environmental pollution.

**Keywords:** salinity; drought; heat stress; flooding; nutrient stress; ROS; cold stress

#### **1. Introduction**

Abiotic stress represents the main constraint for agriculture, affecting plant growth and productivity worldwide. Yield losses in agriculture will be potentiated in the future by global warming, increasing contamination, and reduced availability of fertile land [1]. The challenge of the present and future agriculture is to increase the food supply for a continuously growing human population under environmental conditions that are deteriorating in many areas of the world. Minimizing the effects of diverse types of abiotic stresses represents a matter of general concern [2].

The study of abiotic stress tolerance mechanisms is one of the most active lines of research in plant biology, given its undoubted academic interest and practical implications in agriculture. The different types of abiotic stresses imposed by the environment usually are interconnected and often have an osmotic component, affecting plant cell homeostasis [3].

To counteract abiotic stress, plants activate a series of stress responses, which are shared by both sensitive and tolerant plants as they use the same basic effectors [4]. The knowledge of the limits of tolerance to abiotic stress of different crops, and the understanding of their mechanisms of response to increasing environmental constraints are gaining importance in agronomic research [5]. Research on crop abiotic stress responses is diverse, as plants undergo specific changes in their gene expression, metabolism, and physiology in response to different environmental stress conditions [6].

In this Special Issue, 22 research papers and 4 reviews are presented covering different aspects of the responses of plants to abiotic stresses and their mechanisms of tolerance. However, what is considered abiotic stress? We can define it as any physical or chemical constraint to the potential development and growth of a plant not involving interactions with other living organisms. Abiotic stress in plants is a vast subject, which can be addressed from different points of view and includes many different components, mainly environmental factors, for instance: soil, water, climate, irradiation—even the moon influence! Plants have evolved a series of physiological and molecular mechanisms of response that may (or may not) allow them to adapt to and survive this broad range of stressful conditions. Understanding those mechanisms will help us to improve our interventions towards a more sustainable and efficient agriculture.

#### **2. Drought and Salinity**

Drought and salinity are major abiotic stresses that affect agricultural yields worldwide. The more frequent, longer, and more intense dry periods in many regions of the world, due to global warming, are associated with increasing salinization of land cultivated under irrigation. About 20% of irrigated land in the world, producing one-third of the global food, is affected by secondary salinization of the soil [7]. Drought and salinity have a common osmotic component and early responses to these two types of stress are practically identical [8]. Besides, salt stress causes ionic stress and Na<sup>+</sup> toxicity [3]. Like other types of stress, drought and salinity or their combination may trigger growth inhibition, including, for example, disturbances in mineral nutrition, alteration of membrane permeability and cellular osmotic balance, generation of oxidative stress by increasing reactive oxygen species (ROS) levels, or inhibition of different enzyme activities [9–11].

In the Special Issue is included a review on physiological changes under drought conditions that influence yields in several vegetable crops summarizing changes in the stomatal conductance and chlorophyll content of leaves for individual plants, but also the utility of water stress indices and spectral vegetation indices for predicting yields [12]. An overview by Ketehouli et al. [13] on the effects of salinity on plants and their tolerance mechanisms with particular emphasis on K<sup>+</sup> and Na<sup>+</sup> homeostasis and transport and their regulation is also here included.

Plants defense against abiotic stress starts within their roots [3], and a well-developed root system is essential to provide water uptake [12]. The ability of plants to change their root anatomy was found to improve water uptake and transport in peanut and, therefore, may be considered as a relevant drought tolerance mechanism in this species [14].

This Special Issue includes several papers on morphological, physiological, and biochemical responses to these two types of stress or their combination, and their use in screening for stress-tolerant cultivars. Increased activities of ROS-scavenging enzymes and a more balanced Na+: K<sup>+</sup> ratio was reported as the main mechanism of tolerance in wheat and barley [15]. Accumulation of proline and monovalent cations was related to salt tolerance mechanism in cultivated eggplant and its wild relative *Solanum insanum* [16]. Of special interest is the screening of neglected varieties and local landraces, as they can be a valuable source of allelic richness. Landraces evolved due to selection of traits specifically adapted to local conditions, often suboptimal or even highly stressful [17]. Therefore, such genotypes may enhance agronomic production under the foreseeable restrictive conditions imposed by climate change [2]. Proline was the marker used for screening of beans tolerant to water and salt stress [18], or antioxidant for salt-tolerant tomatoes with high nutraceutical value [19]. Proline and chlorophyll contents, in combination with several morphological and physiological traits, are optimal markers for screening drought tolerance in provitamin A maize, used in sub-Saharan Africa to combat vitamin A deficiency [20].

The irruption of transcriptomics, metabolomics, high-throughput DNA sequencing and high-density microarrays in the analysis of plants' responses to stress have brought new insights and allowed a better understanding on plants reactions to stressful conditions [21]. The stress-responding genes and their regulation pattern under drought were analyzed in common buckwheat cotyledons and roots [22] and female panicles in maize [23], and under salinity in roots and leaves of pomegranate [24].

Others papers published here deal with mitigation of the effects of drought in different crops, such as the synergistic effect of silicon and inoculation with an arbuscular mycorrhizal fungus on strawberries [25], transfer of a LEA gene of a Vietnamese maize landrace to transgenic maize and tobacco [26], and that of salinity by salicylic acid, yeast extract, and proline in sweet pepper [27].

#### **3. Other Significant But Less Studied Stresses**

Global warming alters the rainfall regime in many areas of the world [28], leading to increased floods and poorly drained, waterlogged soils; these conditions have a negative effect on crops by reducing oxygen availability for roots and soil microorganisms [29]. Escape and resilience strategies under flooding stress are presented in an extensive review, concluding that plants maintain their internal homeostasis by balancing hormonal cross-talk under excess water stress [30]. Besides, some treatments can help plants to cope with the stressful effects of waterlogging, for example, seed priming by sodium azide (NaN3) was found to enhance the performance of okra plants under waterlogged conditions [31].

Extreme temperatures pose another challenge for crops. Irregular weather patterns have increased their occurrence in the present climatic conditions; for example, more frequent heat waves are now reported worldwide [28]. One paper deals with the effect of heat stress in alfalfa and extensively discusses the effects of heat on plants [31]). In addition, cold is also a common stress which triggers sophisticated events that alter the biochemical composition of cells in order to protect them from damage [32,33]. Again, some treatments can reduce the negative effects of low temperatures. This is the case of studies on the physiological performance of plants, in which cold stress was alleviated by chitosan via enhancing the photosynthesis and carbon process in tea plant [34], or by 5-Aminolevulinic in cucumber [35].

#### **4. Combination of Di**ff**erent Stresses**

Usually, abiotic stresses come together. The association of drought and salinity is well known, but also that of drought with high temperatures. When different stresses combine, plants need to adjust their physiology to those specific conditions. Landraces, through their long process of farmers' selection in a pre-intensive agriculture period, offer a great opportunity to find appropriate combinations of genes and phenotypes tolerant to complex situations. The most stressful period in the Mediterranean region is summer, when drought is associated with increased temperatures, including heat waves, which are increasingly more frequent in recent years [36]. A comparative study on the responses of local landraces and a commercial cultivar of *Phaseolus lunatus* L. to different temperature and water stress regimes is presented here. The results indicated a better response and a marked competitiveness of one local cultivar [37]. Effects on agronomic traits of the same stresses and their combination was analyzed in African landraces of maize compared with drought and/or heat-tolerant lines [38], and some local landraces proved to be good candidates for improving stress tolerance in this crop.

#### **5. Soil Constrains**

Besides soil salinity, discussed above, there are several other soil constraints with an important impact on agriculture [39]. Of special interest are those related to nutrient conditions in the soil, such as soil P immobilization. Phosphorus is an essential element for plants, but is lacking in 40% of arable land. This nutrient is normally applied as P-enriched fertilizers, which contribute to increased eutrophication of water bodies [40]. Therefore, screening for cultivars with a good performance under low P-input conditions is of interest, as shown by an analysis of morphological traits in relation to P accumulation in pepper cultivars [41]. Zinc is a microelement necessary for plants, animals, and humans; when it is not present in the soil in sufficient amounts, it is necessary either to use varieties with a better uptake of this micronutrient, or its external application in the form of fertilizers and foliar sprays [42]. However, when in excess it has a toxic effect for plants [43]. Morphological and physiological traits,

in combination with the transcriptional regulation of aquaporin isoforms expression, were analyzed in pak choi subjected to two Zn concentrations [44].

Nitrogen is necessary for plant development; it is required in large quantities and, therefore, supplied to crops in fertilizers [45]. Nevertheless, an excessive N application was reported to decrease ROS scavenging ability, and to cause significant metabolic changes in wheat [46]. In the same species, the use of new ecofriendly polymeric-coated urea fertilizers insured a balanced proportion of N with beneficial effects [47].

Another paper deals with abiotic stress in crops imposed by treatments with herbicides and explores the possibility to control weeds with three natural compounds, analyzing the phytotoxic effects that they produce in weeds. The tree products demonstrated great possibilities as sustainable tools for integrated weed management [48].

Finally, this special issue also includes a review on some questions and beliefs that still impregnate a large part of agricultural traditions and agronomic practices, according to which the different lunar phases are beneficial or stressful to plant growth and development [49]. To address the possible link between the phases of the moon and agriculture from a scientific perspective, the authors analyzed physics and biology research papers and handbooks, focusing on those abiotic factors that have a proved influence on plant growth, searching specifically for any that could explain the influence of the moon on plant growth. They did not find any reliable, science-based evidence for such a relationship.

#### **6. Conclusions**

The papers included in this special issue cover a broad range of topics related to the effects on crop plants of different types of abiotic stress, at the morphological, physiological, biochemical, and molecular levels, and the mechanisms of defense of the plants against these stresses. The methods employed were also diverse, from the analysis of agronomic traits based on morphological characteristics to omics approaches and the use of transgenics. Special attention was given to the screening for stress tolerance in local landraces, stress alleviation using different strategies, and the proposal of practical solutions for the agriculture of the (near) future, threatened by global warming and environmental pollution.

**Author Contributions:** M.B. and A.F. equally contributed to organizing the special issue, editorial work, and writing this editorial. All authors have read and agreed to the published version of the manuscript.

**Funding:** No external funding was obtained.

**Acknowledgments:** Many thanks to the authors, reviewers, and to the editorial staff of MDPI for their professionalism. **Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


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

*Article*

### **Response to the Cold Stress Signaling of the Tea Plant (***Camellia sinensis***) Elicited by Chitosan Oligosaccharide**

#### **Yingying Li, Qiuqiu Zhang, Lina Ou, Dezhong Ji, Tao Liu, Rongmeng Lan, Xiangyang Li and Linhong Jin \***

State Key Laboratory Breeding Base of Green Pesticide and Agricultural Bioengineering, Key Laboratory of Green Pesticide and Agricultural Bioengineering, Ministry of Education, Guizhou University, Huaxi District, Guiyang 550025, China; gs.yingyingli17@gzu.edu.cn (Y.L.); gs.zhangqq18@gzu.edu.cn (Q.Z.); gs.lnou17@gzu.edu.cn (L.O.); gs.dzji19@gzu.edu.cn (D.J.); gs.taoliu18@gzu.edu.cn (T.L.); gs.rmlan19@gzu.edu.cn (R.L.); xyli1@gzu.edu.cn (X.L.)

**\*** Correspondence: lhjin@gzu.edu.cn; Tel.: +86-851-362-0521; Fax: +86-851-362-2211

Received: 26 April 2020; Accepted: 22 June 2020; Published: 26 June 2020

**Abstract:** Cold stress caused by a low temperature is a significant threat to tea production. The application of chitosan oligosaccharide (COS) can alleviate the effect of low temperature stress on tea plants. However, how COS affects the cold stress signaling in tea plants is still unclear. In this study, we investigated the level of physiological indicators in tea leaves treated with COS, and then the molecular response to the cold stress of tea leaves treated with COS was analyzed by transcriptomics with RNA-Sequencing (RNA-Seq). The results show that the activity of superoxide dismutase (SOD) activity, peroxidase (POD) activity, content of chlorophyll and soluble sugar in tea leaves in COS-treated tea plant were significantly increased and that photosynthesis and carbon metabolism were enriched. Besides, our results suggest that COS may impact to the cold stress signaling via enhancing the photosynthesis and carbon process. Our research provides valuable information for the mechanisms of COS application in tea plants under cold stress.

**Keywords:** tea plant; cold stress; chitosan oligosaccharide; physiological response; transcriptome

#### **1. Introduction**

The tea plant (*Camellia sinensis* (L.) O. Kuntze) is one of the most important commercial beverage crops in the world and an important revenue source in tea-producing countries [1]. The tea production in over 50 countries has reached over 5.95 million tons on 4.1 million hectares around the world [2]. Among them, the cultivar 'Anji Baicha' is a special green-revertible albino mutant widely cultivated in China, especially in Zhejiang, Hubei and Guizhou provinces, which exhibits periodic albinism during the development of young shoots [3,4]. It is rare and represent precious tea germplasm because of it special flavor, and also has high levels of total amino acids and low levels of polyphenols, which differs from conventional tea [3–8]. In addition, it has a higher commercial value than green tea [4].

The tea plant can grow in different agroclimates and adapted to optimal temperature of 18 to 30 ◦C and pH ranging from 4.5 to 5.5, but the thermophilic nature of tea plants confines their growth to temperate area [9–11]. Furthermore, tea plants that are exposed to a low temperature, such as a sudden frost in fall or early spring, may be at risk of cold stress [12]. Cold environment can adversely affect tea plants on their growth, development, and spatial distribution with decreasing yield and quality, which is one of the factors restricting the healthy development of the tea industry [13–15]. So, it is significant to explore the ways to improve the cold resistance of tea plants. Some studies have reported that the cold resistance of tea plant can be effectively improved by cultivating cold-resistant tea plant

varieties (e.g., Fudingdabai, Shuchazao), cold acclimation of tea plant and the application of exogenous substances [16–19].

Chitosan oligosaccharide (COS) prepared from chitosan, is an environmentally friendly plant growth regulator and stress tolerance inducer [20–24]. Chitosan is a linear polysaccharide composed of β-1,4-glucosamines. The hydrolysis of the glycosidic chitosan chains yields oligosaccharides, including the water-soluble oligochitosan [21,22]. Chitosan and COS have a rich history of being researched for applications in agriculture, primarily for plant defense and yield increase [23,24]. As a natural biocontroller and elicitor of defense responses, COS can boost the innate ability of plants to defend themselves by stimulating secondary metabolite synthesis, and increasing the chlorophyll content and photosynthetic ability [20,21], enrich the soluble sugar in plant [25], and enhancing the activities of antioxidant enzymes [25–27]. COS stimulated the signaling pathways involved in disease resistance in rice [28], and its role in tobacco mosaic virus (TMV) resistance in *Arabidopsis* has been investigated [29]. And studies have shown that COS enhances carbon metabolism, nitrogen metabolism, photosynthesis, and defense against abiotic stress in plants [30]. As reported, COS was able to mitigate the effects of abiotic stresses in plant, including salt, cold and drought [25–27,31,32]. The mechanism of COS in increasing abiotic stress tolerances was summarized as: enhancing the activities of antioxidant enzymes [25], photosynthesis, and stimulate secondary metabolite synthesis [31]. For example, COS has been applied to wheat seedlings for improved chilling tolerance by enhancing antioxidant activities of superoxide dismutase (SOD) and peroxidase (POD) and increasing content of chlorophyll.

These physiological responses of plants elicited by COS are closely related to the regulation of plant gene expression. Transcriptome sequencing has been widely applied to tea plant, which is has the advantage of highly accurate, highly efficient and sensitive profiling in recent years [33]. RNA sequencing (RNA-Seq) technology for measuring transcriptomes of organisms can analyze genes related to abiotic and biotic stress responses, growth, development and metabolites [34–37], to improve our understanding of the molecular mechanism of the tea plant [13–16,38], and RNA-Seq will also be a valuable tool to reveal the role of exogenous substances in tea plant cold resistance.

Though many investigators provided valuable information to cold stress in tea plant, the action mode of COS eliciting responses to cold stress of tea plant is unclear. Therefore, in this report, we studied the effect of exogenous COS on the molecular mechanism of tea plant under low temperature stress. Herein, the physiological parameters of tea plants with and without COS-treatment were compared. The molecular response to cold resistance within tea plant was analyzed by RNA-Seq technology. This research improves the understanding of the cold resistance mechanism of COS-treated tea plant and provides important guidance for COS application under low temperature stress.

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

#### *2.1. Plant Materials and Cold Treatments*

Two-year-old albino tea cultivar (*Camellia sinensis* (L.) O. Kuntze cv. 'Anji Baicha') were used in the experiment from AnShun County, Guizhou Province, China. Additionally, the tea plants were transplanted into the plastic pot. Plants were grown in a growth chamber at the experimental of Guizhou University, Guizhou Province, China (16 h day/8 h night at 25 ◦C/20 ◦C and relative humidity of 70%). After a month, tea plants were treated with 10 mL of following elicitors by surface spraying with sterile distilled water (control, CK), or with 1.25 mL/L COS solution (COS comes from Hainan Zhengye Zhongnong High-tech Co., Ltd., Haikou, Hainan Province, China). After 24 h, the two groups of tea plants were separately maintained in a chamber at −4 and −8 ◦C at cold treatment for 24 h, with one group maintained under normal room temperature conditions. Three independent biological repeats were collected for each treatment. Fresh leaves from the stable stage (re-greening stage) of chlorophyll development of Anji Baicha were harvested at 24 h and frozen immediately in liquid nitrogen and stored at −80 ◦C for further study.

#### *2.2. Physiological Response Assay*

Physiological indexes of tea leaves (containing 1st, 2nd, 3rd leaf and old leaves), involving the activities of SOD and POD, and content of chlorophyll and soluble sugar, were determined. Additionally, the assay kits used included the SOD assay kit, the POD assay kit, the chlorophyll assay kit, the soluble sugar assay kit (Solarbio, Cat. No. BC0175, BC0095, BC0995, BC0035, respectively, Beijing, China). All assays were performed according to the manufacturer's instructions.

#### *2.3. cDNA Library Construction and Sequencing*

We selected tea leaves from control and COS treatment on −4 ◦C for RNA-Seq analysis. Total RNA was extracted from tea leaves using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instruction. Poly (A) + mRNA was purified with oligo (dT) beads. The mRNA was randomly cut into short fragments using Fragmentation Buffer, which were used as a template for the short fragment mRNA, first-strand cDNA was synthesized with 6 bp random primers, and then the Buffer, dNTPs and DNA polymerase I were added to synthesize the second-strand cDNA. RNA Integrity was confirmed using 1.5% agarose gel. RNA quality was checked by a NanoDropTM OneC spectrophotometer (Thermo Fisher Scientific, New York, NY, USA). RNA qualified was measured by QubitTM RNA BR Assay Kit in Qubit® 2.0 (Life Technologies, Carlsbad, CA, USA). The cDNA library construction and Illumina sequencing of the samples were performed using a 150 bp paired-end Illumina Nova-seq 6000 (Illumina, San Diego, CA, USA) by Seqhealth Technology Co., Ltd. (Wuhan, China).

#### *2.4. RNA-Seq Data Analysis*

The raw reads were first filtered to obtain the clean reads by removing the adaptor sequences, unknown sequences "N" and low-quality reads using Trimmomatic (version 0.36). After filtering, the clean reads were mapped to the reference genome of *Camellia sinensis* using STATR software (version 2.5.3a).

#### *2.5. Identification of Di*ff*erentially Expressed Genes*

The expression levels of each gene were calculated and normalized by the corresponding Reads Per Kilobase of transcript per Million mapped reads (RPKM). The RPKM method can eliminate the influence of gene length and sequencing amount differences on gene expression. FeatureCounts (version 1.5.1) was used to count the read numbers mapped to each gene [39]. Additionally, differentially expressed genes (DEGs) were identified with the edge R package (version 3.12.1) [40]. The resulting *p*-values were adjusted using Benjamini and Hochberg's method for controlling the false discovery rate (FDR). Genes with *p*-value < 0.05 and a logarithm two-fold change |log2FC| > 1 were defined as DEGs.

#### *2.6. Gene Ontology and KEGG Pathway Analysis*

Gene ontology (GO) analysis and Kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis of DEGs were both implemented by KEGG orthology based annotation system (KOBAS) software (version 2.1.1) with *p*-value < 0.05 to judge statistically significant enrichment [41].

#### *2.7. Quantitative RT-PCR (qRT-PCR) Analysis*

To verify the RNA-Seq analysis, we randomly selected five unigenes and used qRT-PCR to confirm their participation in the high-temperature reaction. RT-qPCR was conducted on ABI ViiATM 7 Real-Time PCR System (Applied Biosystems, Foster, CA, USA) using GoTaq® qPCR Master Mix (Promega, Madison, WI, USA). The PCR amplifications were consisted of 95 ◦C for 3 min, followed by 40 cycles of 95 ◦C for 15 s, 60 ◦C for 30 s, and then 72 ◦C for 30 s. Gene expression was normalized using the glyceraldehyde-3-phosphate dehydrogenase (GADPH) as an internal reference gene, and the relative changes of gene expression were calculated using the 2−ΔΔCt method. The list of primers is presented in Table S1.

#### *2.8. Statistical Analysis*

Data were expressed as the mean ± standard error, and the data were subjected to one-way analysis of variance (ANOVA) (*p* < 0.05) followed by a significant difference test (LSD) using SPSS statistics v17.0 (SPSS Inc., Chicago, IL, USA).

#### **3. Results**

#### *3.1. Physiological Parameter Response to a Low Temperature*

To analyze the effects of COS on tea plant growth, we measured the change in activity of SOD, and POD enzymes and content of chlorophyll, soluble sugar in COS-treated tea plant and their respond to low temperature stress, with sterile distilled water served as control. As shown in Figure 1, under a low temperature, the tea plant responds to cold stress with all the physiological parameters changed and COS-enhanced freeze protection. As in the control group, a low temperature caused increases in those physiological parameters. As shown in Figure 1A, the enzyme activity of SOD was significantly increased by 24.04% at −4 ◦C and 32.68% at −8 ◦C. Similarly, the enzyme activity of POD was significantly increased by 38.05% at −4 ◦C and 8.81% at −8 ◦C. Cold stress significantly reduced the chlorophyll content by 20.18% and 21.96% at −4 and −8 ◦C, respectively (Figure 1C). Moreover, soluble sugar content was significantly increased by 29.87% at −4 ◦C and 28.16% at −8 ◦C, respectively (Figure 1D). The results show that cold stress consistently increased SOD and POD activity, and soluble sugar content, when the temperature was switched from 25 ◦C to −4 ◦C or −8 ◦C, but POD activity was highest at −4 ◦C.

When exogenous COS was used, it consistently enhanced SOD and POD activities, and the soluble sugar content and chlorophyll content in the tea plant. For example, COS improved SOD activity by 11.75% at 25 ◦C, 25.93% at −4 ◦C and 9.21% at −8 ◦C, respectively, as compared with the control. Similarly, POD activity was enhanced by 19.91%, 19.23% and 30.09% on 25 ◦C, −4 ◦C and −8 ◦C, respectively.

**Figure 1.** Effect of chitosan oligosaccharide (COS) on physiological parameters of tea leaves. (**A**) Superoxide dismutase (SOD) activity; (**B**) peroxidase (POD); (**C**) chlorophyll content; (**D**) soluble sugar content. The data represent the means ± SD of three replicates samples. Different letters indicate significant differences at *p* < 0.05.

For all the tested parameters, the effects of COS were more pronounced under cold stress. When tea plants were treated with COS combined with cold stress, SOD enhanced by 56.21% and 44.91% at −4 and −8 ◦C, respectively. Similarly, POD increased 37.26% and 18.04%. The content of soluble sugar also increased by 45.22% and 40.25% at −4 and −8 ◦C, respectively. Chlorophyll content was decreased by 13.47% and 14.99%, respectively. The results show that COS treatment consistently increased chlorophyll content, but three parameters of SOD, POD and soluble sugar were highest at −4 ◦C of cold stress combined with COS.

#### *3.2. Transcriptome Sequencing and Assembly*

To understand the response of the tea plant to cold stress and the effect of COS on the molecular level, we compared the transcriptomes between COS treatment and the control group at −4 ◦C by RNA-Seq. Replicate samples of the control group (ConT3\_1/2/3) and COS-treatment group (TreT3\_1/2/3) were included in this study. We obtained 5.59–6.60 million raw reads in control and 5.79–6.77 million raw reads in the COS-treatment group. After filtering and removing low-quality reads, the clean reads were limited 5.26–6.21 million and 5.45–6.34 million, respectively. Of these clean reads, the GC content was 46.46–47.21% and the Q30 values were over 98.45%. The ratio of total mapped reads between the control and COS-treatment groups was 94.69–94.90% and 94.85–95.20% for *Camellia sinensis* according to the Genome Database. Unique mapped reads were 91.48–92.10% in the control group and 88.02–90.66% in the COS-treatment group (Table 1).

**Table 1.** Statistical analyses and mapping results of RNA sequencing reads.


#### *3.3. Di*ff*erentially Expressed Genes Analysis*

In order to verify the correlation of gene expression level between samples, we demonstrated that the biological repeatability between samples was great through spearman correlation analysis based on the RPKM of different samples. Genes with *p*-value < 0.05 and |log2(Foldchange)| > 1 were defined as differentially expressed genes between control and COS. There were identified 4503 differentially expressed genes (DEGs) between the control and COS, including 1605 up-regulated and 2898 down-regulated genes in the leaves of tea plant (Figure 2 and Table S2).

**Figure 2.** Volcano plot of differentially expressed genes (DEGs) showed up-regulated and down-regulated between control and COS under −4 ◦C treatment. The red dots represent up-regulated genes, the blue dots represent down-regulated genes, and the gray dots represent no significant difference. The horizontal coordinates indicate the change in multiple expression, the longitudinal coordinates indicate the magnitude of differences.

#### *3.4. Gene Ontology (GO) Annotation*

The differentially expressed mRNAs were analyzed by GO enrichment, as shown in Figure 3 and Table S3. The differentially expressed genes were mostly enriched in biological process (Figure 3). In the biological process categorization, functional enrichment mainly focuses on metabolic processes and nutrient synthesis processes, such as "single-organism biosynthetic process" (GO: 0044711), "metabolic process" (GO: 0008152), "carbohydrate metabolic process" (GO: 0005975) and "carbohydrate derivative biosynthetic process" (GO: 1901137). The molecular function category includes the expression of transmembrane transporters and catalytic enzyme-related genes, such as "catalytic activity" (GO: 0003824), "transporter activity" (GO: 0005215), "transmembrane transporter activity" (GO: 0022857), and "ion transmembrane transporter activity" (GO: 0015075). Besides, "serine-type endopeptidase activity" (GO: 0004252) was mostly enriched in the molecular function category.

**Figure 3.** Gene ontology (GO) classification analysis based on DEGs induced by COS under −4 ◦C treatment. The horizontal coordinates indicate GO terms, the longitudinal coordinates indicate rich factor, rich factor represents the ratio between the number of different genes enriched in the term and the background genes in GO term.

#### *3.5. Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathway Annotation*

The KEGG enrichment scatter plot is a graphical representation of the statistical analyses that visualizes the pathway enrichment (Figure 4). The degree of KEGG enrichment was measured in terms of richness factor, *p*-value, and the number of genes in the pathway. The important enriched pathways with high generation, low *p*-value and large numbers of genes are shown in the Figure 4 and Table S4. As shown in Figure 4, these enriched pathways, including "photosynthesis" (ko00195), "carbon fixation in photosynthetic organisms" (ko00710), "photosynthesis–antenna proteins" (ko00196), "ribosome" (ko03010), "carbon metabolism" (ko01200).

Compared with the control group, 71 genes were significantly induced to up-regulated by COS treatment, including PSII, PSI, cytochrome b6/f complex, photosynthethic electron transport and F-type ATPase (Table S5). In the carbon metabolism pathway, a total of 77 genes were differentially expressed, including 52 up-regulated and 25 down-regulated (Table S6). A total of 43 genes were assigned to

the plant hormone signal transduction pathway, including 16 genes that were up-regulated in auxin, abscisic acid, ethylene, salicylic acid (Table S7). These results suggest that the addition of COS at a low temperature have a complex effect on biological process and metabolism of the tea plant.

**Figure 4.** Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis based on DEGs induced by COS under −4 ◦C treatment. The significance of enrichment is shown on the horizontal coordinates (represented by −log10 (*p*-value), the greater the value, the more significant the enrichment), and the KEGG pathway is shown on the longitudinal coordinates. The size of the dots indicates the number of different genes contained in the KEGG pathway, and the color of the dots indicates the degree of rich factor enrichment.

#### *3.6. qRT-PCR Validation of Di*ff*erentially Expressed Transcripts from RNA-Seq*

Five transcripts were randomly selected for qRT-PCR analysis, which used to confirm validity and accuracy the RNA-Seq data. The results show that the trend of qRT-PCR is consistent with the results of RNA-Seq in Figure S1.

#### **4. Discussion**

Cold stress affects photosynthetic activities and metabolic functions in plants, which further affected growth, development, and metabolism. It has a negative effect on the yield and quality of tea. Anji Baicha is a temperature-sensitive albino tea cultivar. When the environment temperature is below 20 ◦C in early spring, the white shoots phenomenon will appear. After about two weeks, the plant gradually turns as green, as does those of common tea cultivars [4–6]. The change of leaf color was mainly due to chloroplast development in the albescent stage, the etioplast–chloroplast transition was blocked, and the accumulation of chlorophyll was inhibited under low temperature [4–8,37]. In this study, we chose Anji Baicha in the stable stage of chlorophyll development as a research object,

the results revealed that COS could enhance antioxidant activity, increase accumulation of sugar content and chlorophyll content in tea plant. It is confirmed that COS could play an important role in improving stress tolerance of Anji Baicha.

Cold stress can cause excessive production of reactive oxygen species (ROS), disrupt the normal physiological and metabolic balance of plants, lead to the increase of membrane lipid peroxidation and damage to vital biomolecules [42,43]. Plants have evolved complex mechanisms to combat against the damage induced by ROS, including improve the antioxidant enzymes [44,45]. In this study, under cold stress, the tea plant natively reacted to protect themselves by increasing the activity of SOD and POD enzyme, and the application of COS provided external assistance plant. Chlorophyll content in COS-treated tea plant was higher than in control, which indicated that COS application mitigated the cold-induced decline in chlorophyll content. Soluble sugar can maintain the osmotic balance, and the soluble sugar in COS treated tea plant was higher than that without COS treatment, suggesting that COS can stabilize cell membrane and enhance cold resistance of plant. Those results indicated that the utilization of COS can positively affect these physiological parameters in tea plants, and beneficially regulate the natural defense system and improve growth and developmental processes of tea plants under cold stress. Moreover, this was also demonstrated in wheat seedlings where the application of COS could enhance the activities of antioxidant enzymes and the content of chlorophyll and alleviate the damage of abiotic stress in wheat [25–27,46]. In wheat, COS could enhance the activities of antioxidant enzymes and the content of chlorophyll, alleviate plant the damage of abiotic stress [25–27,46]. These differentially expressed genes indicate that the application of COS has complex effects on metabolism and signaling pathways of tea plants at low temperature. From RNA sequencing, we found that COS significantly altered the level of gene expression involved in photosynthesis and carbon metabolism under cold stress.

The up-regulated differentially expressed genes could be important for the pathology and biological processes of response to cold stress. Chlorophyll content is an important parameter frequently used to indicate chloroplast development, and which is sensitive to abiotic stresses [47]. COS can increase chlorophyll content under cold stress, which is consistent with the observations from RNA-Seq. Compared with the control group, COS treatment may increase the photosynthesis of plants by significantly up-regulating photosystem I (PSI), photosystem II (PSII)-related genes (Table S5). In the PSII core complex, PsbR is an important link, which can stabilize the assembly of the oxygen-evolving complex protein PsbP [48]. In the present study, PsbR was up-regulated, which was consistent with the action of chitosan heptamer response in wheat seedling [49]. Besides, Chlorophyll a/b-binding protein can participate in light uptake, transfer energy to the reaction centers of the photosystem I and photosystem II, and regulate the excitation energy distribution to maintain the structure of the thylakoid membrane [50], and all of 23 chlorophyll a/b-binding protein genes were also up-regulated, which can imply the recovery of photosynthesis activities by COS treatment under cold stress [51]. These results indicate that COS may enhance photosynthesis via the upregulation of related proteins to improve the cold resistance of tea plant.

In the carbon metabolism pathway, genes encoding ribulose bisphosphate carboxylase small subunit (rbcS), phosphoglycerate kinase, glyceraldehyde-3-phosphate dehydrogenase, triosephosphate isomerase were up-regulated significantly (Table S6). RbcS is one of the subunits of Ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCo), and the activity of rbcS decreased to inhibit photosynthesis under cold stress [52]. This result was consistent with previous research demonstrating the application of COS to regulate the photosynthetic mechanism and carbon metabolism and thereby the plant growth [53].

During plant development, the response of plants to endogenous and environmental signals is mediated by several hormones, which are involved in almost every aspect of plant growth. For example, plants respond very quickly to auxin, including cell growth and the activation of multiple auxin-responsive genes [53]. Indole-3-acetic acid (GH3) and the ethylene receptor (ETR) were up-regulated genes in the plant hormone signal transduction pathway (Table S7). GH3 is an

important response gene of auxin-responsive protein (IAA), which can encode a class of IAA-amido synthetases responsible for balancing endogenous free IAA content and plays an important role in IAA-regulated plant growth and development [54,55]. The ETR responds to ethylene and abscisic acid (ABA) signaling. ETR is the most important ethylene receptor protein in plants, and the lack of ETR will hinder the transduction of ethylene signal cascade reaction, resulting in the insensitivity to ethylene in plant [56–58].

The application COS can improve antioxidant enzyme activities, and the content of chlorophyll and soluble sugar. Besides, compared with the control group, the addition of COS significantly changed the photosynthesis pathway and carbon metabolism of tea plants under low temperature stress, which may contribute to COS' ability to improve the cold tolerance of tea plants. These results may represent that COS participates in the specific regulatory mechanism related to cold adaptation in the cold resistance of Anji Baicha. As for the comparison of cold resistance between Anji Baicha and other tea plants (e.g., Xiaoxueya, Fudingdabai), we are further carrying out relevant experimental verification.

#### **5. Conclusions**

In summary, low temperature will impact the key physiological and developmental processes that determine the yield of tea. This study indicates that the utilization of COS can positively affect these physiological parameters in tea plants by improving antioxidant enzyme activities, and the content of chlorophyll and soluble sugar. Hence, COS can beneficially regulate the natural defense system and improve the growth and developmental processes of tea plants under cold stress. With transcriptome sequencing and differentially expressed genes analysis, we identified 1605 up-regulated and 2898 down-regulated genes in COS compared to the control, and photosynthesis and the carbon metabolism pathway of enrichment may play a role in the COS-improved cold resistance of a tea plant. The results may provide the foundation for further research on the regulation mechanism of COS on plant cold tolerance.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/6/915/s1, Table S1: Primer sequences used for qRT-PCR. Table S2: The list of different expression genes. Table S3: GO enrichment list of different expression genes. Table S4: KEGG pathway enrichment list of different expression genes. Table S5: Differentially expressed genes in photosynthesis related pathway. Table S6: Differentially expressed genes in carbon metabolism pathway. Table S7: Differentially expressed genes in plant hormone signal transduction pathway. Figure S1. Verification of relative expression levels of DEGs in transcriptome date by qRT-PCR between control and COS.

**Author Contributions:** Y.L. conducted the experiments; Y.L., L.O. and D.J. designed and performed the experiments; Y.L., Q.Z., T.L. and R.L. analyzed the data; X.L. and L.J. conceived and supervised the project. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by science and technology project of Guizhou province ([2015]5020) and scientific research projects of major agricultural industries of Guizhou province ([2019]006).

**Acknowledgments:** We are grateful to Xia Zhou, Guizhou University, for the fruitful discussions and helpful comments on earlier draft.

**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* **The Use of Proline in Screening for Tolerance to Drought and Salinity in Common Bean (***Phaseolus vulgaris* **L.) Genotypes**

### **Sugenith Arteaga 1, Lourdes Yabor 1,2, María José Díez 3, Jaime Prohens 3, Monica Boscaiu 4,\* and Oscar Vicente <sup>3</sup>**


Received: 30 April 2020; Accepted: 5 June 2020; Published: 9 June 2020

**Abstract:** The selection of stress-resistant cultivars, to be used in breeding programmes aimed at enhancing the drought and salt tolerance of our major crops, is an urgent need for agriculture in a climate change scenario. In the present study, the responses to water deficit and salt stress treatments, regarding growth inhibition and leaf proline (Pro) contents, were analysed in 47 *Phaseolus vulgaris* genotypes of different origins. A two-way analysis of variance (ANOVA), Pearson moment correlations and principal component analyses (PCAs) were performed on all measured traits, to assess the general responses to stress of the investigated genotypes. For most analysed growth variables and Pro, the effects of cultivar, treatment and their interactions were highly significant (*p* < 0.001); the root morphological traits, stem diameter and the number of leaves were mostly due to uncontrolled variation, whereas the variation of fresh weight and water content of stems and leaves was clearly induced by stress. Under our experimental conditions, the average effects of salt stress on plant growth were relatively weaker than those of water deficit. In both cases, however, growth inhibition was mostly reflected in the stress-induced reduction of fresh weight and water contents of stems and leaves. Pro, on the other hand, was the only variable showing a negative correlation with all growth parameters, but particularly with those of stems and leaves mentioned above, as indicated by the Pearson correlation coefficients and the loading plots of the PCAs. Therefore, in common beans, higher stress-induced accumulation of Pro is unequivocally associated with a stronger inhibition of growth; that is, with a higher sensitivity to stress of the corresponding cultivar. We propose the use of Pro as a suitable biochemical marker for simple, rapid, large-scale screenings of bean genotypes, to exclude the most sensitive, those accumulating higher Pro concentrations in response to water or salt stress treatments.

**Keywords:** abiotic stress biomarkers; bean landraces; osmolytes; plant breeding; salt stress; salt stress tolerance; water deficit; water stress tolerance

#### **1. Introduction**

Drought and soil salinity are amongst the most restrictive environmental factors affecting agriculture worldwide. Even moderate degrees of water deficit or salt stress can lead to a reduction of 50–70% in average yields in most crops when compared with registered record yields [1–3]. Drought, brought about by the scarcity of rain, affects more than half of the agricultural land of our planet and is often linked to secondary salinisation of farmland due to intensive irrigation [4,5]. Cropland salinisation is becoming one of the major constrains for agriculture in many parts of the world, especially in arid and semi-arid regions. At the beginning of this century, it was estimated that around 20% of the irrigated lands were salinised [6], but this figure is increasing yearly, mainly due to anthropogenic alterations, such as irrigation with brackish water or the abusive and indiscriminate use of chemical fertilisers [4]. On the other hand, the scarcity of good-quality water for irrigation, mainly as a consequence of the effects of global warming, will mean more-significant crop losses in the near future, which will especially affect subsistence agriculture in developing countries [7]. Legumes are some of the most important crops, representing a significant component of the human diet. Globally, legumes complement cereal crops as the main sources of plant minerals and proteins [8]. Among the leguminous crops, *Phaseolus* L. is a large and diverse genus comprising about 70 American species [9], five of which have been domesticated (*Phaseolus vulgaris* L., *Phaseolus dumosus* Macfady, *Phaseolus coccineus* L., *Phaseolus acutifolius* A. Gray and *Phaseolus lunatus* L); moreover, a few additional species show signs of incipient domestication [10].

The common bean (*P. vulgaris*) is the most-consumed legume in human nutrition; it is an essential component of the diet, especially in developing countries, as a source of proteins, vitamins, minerals and fibre [8,11]. The species has a natural distribution area from northern Mexico to northwestern Argentina. It was domesticated independently in Central America and the Andes [12,13], but now it is cultivated practically all over the world. Beans from both origins were introduced to Spain in the 16th century [14–16], where they had to adapt to the new environmental conditions, which were very different from those in their native areas. The cropping system in small farms, spread in proximal areas, allowed the genetic flow between genotypes of Mesoamerican and Andean origin [17]. Due to centuries of bean cultivation, the Iberian Peninsula has become a secondary centre of diversification of this species [18].

*Phaseolus vulgaris* is not considered as very tolerant to water stress [19]; nevertheless, it is cultivated under diverse environmental conditions, including relatively dry areas [20,21]. In fact, globally, only a small percentage, around 7%, of the cropland planted with common bean receives adequate rainfall [11], and in some areas, drought causes yield losses of up to 80% [22] Like practically all cultivated plants, the bean is a glycophyte, sensitive to soil salinity even at electric conductivity values below 2 dS·m−<sup>1</sup> [23]. However, just as there are cultivars that are more resistant to water stress, some respond better to high soil salinity [24,25].

As for other common crops, many bean genotypes no longer grown in the fields or cultivated only locally at a small scale (landraces, local varieties, heirlooms or minor commercial cultivars) are available from small farmers or germplasm banks and represent a rich source of genetic variability. Landraces appeared over time due to selection of traits specifically adapted to local conditions, often suboptimal or even highly stressful. Therefore, such genotypes are probably more competitive in low-input agriculture and represent a source of allelic richness that may enhance agronomic production under the foreseeable restrictive conditions imposed by climate change [26]. There is an increasing interest for the recovery of local landraces by consumers and markets, not only concerning global warming but also because of the commercial demand for local products, considered as tastier and healthier [27]. Unfortunately, many autochthonous varieties have been lost, and many others are at risk of extinction, due to genetic erosion. Screening this type of varieties for tolerance to stresses represents an interesting strategic path for the agriculture of the future.

The screening of a large number of genotypes would be greatly facilitated by identifying a suitable stress biomarker, easily quantified by simple, rapid and non-destructive assays, and unequivocally associated to the relative resistance of the cultivars to water deficit or salt stress. Proline (Pro), one of the commonest plant osmolytes [28,29], could be an appropriate candidate because a significant increase in Pro contents in response to water deficit, high salinity or other stressful conditions has been

detected in beans—as in many other species. However, it is not yet clear whether Pro accumulation in *P. vulgaris* is associated with enhanced or reduced tolerance to stress since contradictory results are available in the literature. Some reports correlated higher Pro contents with a relatively higher stress tolerance when comparing different bean cultivars [30–37], whereas in other cases higher Pro concentrations were measured in the relatively more stress-sensitive cultivars [38–40]. All these studies were based on the comparison of a few genotypes. Only a wider analysis, based on a considerably higher number of cultivars, grown under the same experimental conditions and subjected to the same stress treatments, could establish whether responses to stress based on Pro accumulation are relevant, or not, for stress tolerance in *P. vulgaris*, and how Pro could be used as a reliable abiotic stress biomarker in this species.

Based on the ideas mentioned above, we have applied specific water deficit and salt stress treatments, under controlled greenhouse conditions, to a relatively large number of common bean cultivars, obtained from germplasm banks. The aims of this study were (i) to determine the overall response of the analysed genotypes to controlled water and salt stress treatments, (ii) to establish the role of Pro in bean stress responses, either as a mere stress biomarker or as an osmolyte directly involved on stress tolerance mechanisms and (iii) based on the results obtained, to propose Pro as a suitable biochemical marker for the rapid selection of bean cultivars with a (relatively) higher tolerance (or sensitivity) to drought or salinity.

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

#### *2.1. Plant Material*

The study included 47 accessions of common bean (*P. vulgaris*), from Spain (23), Colombia (19) and Cuba (5), provided by the Germplasm Bank of Universitat Politècnica de València (UPV), the International Center for Tropical Agriculture (CIAT) and the Bioplants Center, University of Ciego de Ávila, respectively.

Spanish genotypes are represented by local landraces, with geographic origins indicated in Table 1. Materials from Cuba are commercial varieties or experimental lines from INIFAT (Alexander Humboldt Institute for Basic Research in Tropical Agriculture) or IIHDL (Liliana Dimitrova Horticultural Research Institute, La Habana, Cuba), and those from Colombia are lines reported to be relatively resistant to drought and high temperatures.


**Table 1.** Origin of the analysed *Phaseolus vulgaris* accessions and duration of the applied stress treatments.


**Table 1.** *Cont.*

a: commercial varieties; b: experimental lines.

#### *2.2. Plant Growth and Stress Treatments*

The plants were obtained by seed germination. Several seeds of each genotype were germinated in trays with peat, perlite and vermiculite (2:1:1). When the first trifoliate true leaves were formed, the seedlings were transplanted to individual 1.6 L-pots with the same substrate in the greenhouse; Hoagland's nutrient solution [41] was used for irrigation. When the plants reached a height of at least 20 cm and had two to five true leaves, plants were selected for the treatments and placed in 55 × 40 cm plastic trays (10 pots per tray). Irrigation was performed twice a week by adding to each tray 1.5 L deionised water or a 150 mM NaCl solution, for the control and salt stress treatments, respectively. The water stress treatment was applied by completely withholding irrigation of the plants. Five individual plants (biological replicas) of each genotype were used per treatment. Treatments were stopped after two weeks for 18 genotypes when plants showed clear wilting and general decline symptoms in the water deficit treatment, but before plant mortality was observed; salt treatments of these cultivars were stopped at the same time. The remaining, relatively more resistant 29 genotypes were treated for an additional week. The two groups of plants were analysed independently. All treatments were carried out under controlled conditions in the greenhouse: long-day photoperiod (16 h of light), temperature set at 23 ◦C during the day and 17 ◦C at night. Once the treatments were finished, whole plants were harvested, collecting separately their roots, stems and leaves. Several growth parameters were measured in all plants: the diameter of the stem (SD), the length of the roots (RL) and stems (SL), the number of trifoliate leaves (Lno) and the fresh weight of roots (RFW), stems (SFW) and leaves (LFW).

Part of the fresh material of roots, stems and leaves was weighed (FW), placed at 65 ◦C in an oven for three days, and weighed again to determine the dry weight (DW). The water content percentage (WC%) of the three organs was calculated according to the formula:

$$\text{WC } (\%) = \text{[(FW - DW)/Fw]} \times 100\tag{1}$$

#### *2.3. Quantification of Proline Contents*

Leaf Pro concentrations were quantified using dry plant material, according to the ninhydrin-acetic acid method [42]. Pro was extracted in a 3% (*w*/*v*) aqueous sulfosalicylic acid solution; the sample was mixed with the acid ninhydrin solution, incubated for 1 h at 95 ◦C, cooled on ice and extracted with toluene. Samples with known Pro amounts were assayed in parallel to obtain a standard curve. The absorbance of the supernatants was read at 520 nm using toluene as a blank. Pro concentration was finally expressed as μmol g−<sup>1</sup> DW.

#### *2.4. Statistical Analysis*

Plants from the two- and three-week treatments were analysed separately. A two-way analysis of variance (ANOVA) was performed for all determined traits, to check the effects of the 'cultivar' and 'treatment' factors, and the interaction between treatment and genotype. Pearson moment correlations were also performed for all measured parameters, and a principal component analysis (PCA) was used to check the similarity between the responses to the different types of stress within each cultivar, and the similarity between accessions. Data were analysed using Statgraphics Centurion v.16 software (Statpoint Technologies, Warrenton, VA, USA).

#### **3. Results**

#### *3.1. Analysis of Variance of Registered Traits*

Some cultivars (18) were apparently more sensitive to both salt and water stress and therefore treatments were stopped after two weeks. For the remaining genotypes (29), treatments were extended to three weeks (Table 1). All growth parameters and the leaf Pro concentration of control and stressed plants, for each cultivar, are summarised in Supplementary Table S1. Notwithstanding quantitative differences between genotypes, the overall picture is that plants of most cultivars were affected by both types of stress, water deficit and salinity, which inhibited growth as indicated by the general relative reduction observed in the measured morphological variables. Under the specific stress conditions applied in the experiments, in most cases, growth inhibition was more accentuated in the water-stressed plants than in the salt-stressed ones. Again for most cultivars, leaf Pro contents increased significantly in response to both types of stress. To assess the general responses to stress of the selected cultivars, a two-way ANOVA was performed considering the effect on each parameter of cultivar and treatment, and their interaction (Table 2).

**Table 2.** Two-way analysis of variance (ANOVA) of cultivar, treatment and their interactions for the parameters considered. Numbers represent percentages of the sum of squares at the 5% confidence level. Abbreviations: RL, root length; RFW, root fresh weight; RWC, root water content; SD, stem diameter; SL, stem length; SFW, stem fresh weight; SWC, stem water content; Lno, leaf number; LFW, leaf fresh weight; LWC, leaf water content; Pro, proline content. Asterisks indicate the degree of significance: \*\* *p* < 0.01; \*\*\* *p* < 0.001, ns = not significant.


For most analysed variables, the effects of cultivar, treatment and their interactions were highly significant (*p* < 0.001). The only non-significant value was found in the two-week treatment and the trait 'root length', for the interaction cultivar × treatment. In plants subjected to the two-week treatment, relatively stronger contributions to the sum of squares were those of 'cultivar' for the variables root fresh weight (RFW) and stem length (SL), and 'treatment' for root water content (RWC), stem fresh weight

(SFW), leaf fresh weight (LFW) and leaf water content (LWC). For stem water content (SWC) and Pro, both factors, cultivar and treatment, contributed similarly to the sum of squares (SS). On the other hand, most of the variation observed for root length (RL), and stem diameter (SD) was due to uncontrolled variation, as shown by the higher SS percentage of the residual (Table 2).

The ANOVA of data obtained from the three-weeks-treated plants showed somewhat different results. The effect of 'cultivar' was the most substantial contributor to SS for the variables SL, LWC and Pro, and that of 'treatment' for RWC and SFW. The relative contributions of cultivar and treatment were similar for LFW, and those of cultivar and the interaction of both factors, for SWC. The most-significant contribution to variation of RL, RFW, SD and the number of leaves (Lno) is accounted for by the residual source of variation.

Disregarding the individual responses to water and salt stress of the selected bean genotypes, which vary quantitatively (Table S1), a general analysis was performed, including all cultivars and using the mean values calculated for all measured growth variables and Pro contents (Table 3). After the water stress treatments, either for two or three weeks, all morphological parameters determined in the stressed plants showed a significant decrease with respect to the corresponding values of the well-watered controls. The strongest reductions, down to less than 30% of the controls, were observed for root and leaf fresh weight. The effect of water deficit was relatively weaker regarding the reduction of root length and stem parameters (SD, SL and SWC), especially in the three-week treatments. Leaf Pro concentration, on the contrary, significantly increased in response to water stress, about 2.7-fold and 2.1-fold, as average, for the plants treated for two and three weeks, respectively (Table 3).



Under the specific conditions of our experiments, salt stress had a smaller effect than water deficit on the average growth inhibition of the bean cultivars, reflected mostly in a sharp reduction (>70%) of the leaf fresh weight with respect to the control, followed by that of stem fresh weight (about 50%). Other parameters, such as root and stem water content, stem diameter or root fresh weight (in the two-week treatment) did not change significantly or decreased only slightly in response to increased salinity. The mean values calculated for most growth variables were similar for both treatment times. Pro contents also rose significantly, about 2.6-fold over control values, in the salt-treated plants (Table 3).

#### *3.2. Correlation Analysis*

Pearson moment correlation between the analysed traits for salt and water stress are presented separately for plants from the two-week (Figure 1a) and three-week (Figure 1b) treatments. Correlations between all morphological variables were in most cases positive, for both stresses and the two treatment times, although the correlation coefficients varied widely, from *r* < 0.1 to *r* > 0.9. Considering specifically the two-week treatment, the strongest correlations (*r* ≥ 0.8) for the salt stress treatment were found between root water content (RWC) and stem fresh weight (SFW) or water content (SWC); or between SFW, leaf fresh weight (LFW) and the number of leaves (Lno) (Figure 1a). Under conditions of water stress, the strongest positive correlations were also found between SFW, LFW and Lno; between water contents of roots, stems and leaves (RWC/SWC/LWC) or between SFW and stem diameter (SD) (Figure 1a). On the other hand, Pro contents showed negative correlations with all growth parameters (except for RWC in the salt stress treatment), most significantly with leaf water content, but also with LFW and stem growth parameters (SFW and SWC). Correlations followed a similar pattern for both types of stress but were weaker (lower '*r*' values) in the case of salt stress (Figure 1a).

**Figure 1.** Heatmap of Pearson moment correlation coefficients (r) between the analysed traits in *Phaseolus vulgaris* cultivars submitted to two weeks (**a**) and three weeks (**b**) of water and salt stresses. Dark blue denotes high correlation (*r* → 1), dark red high negative correlation (*r* → −1). Abbreviations: RL, root length; RFW, root fresh weight; RWC, root water content; SD, stem diameter; SL, stem length; SFW, stem fresh weight; SWC, stem water content; Lno, leaf number; LFW, leaf fresh weight; LWC, leaf water content; Pro, proline content.

Correlations between the different measured variables, generally positive for growth parameters and negative between Pro contents and the rest of variables, were maintained, qualitatively, when comparing the two- and three-week treatments, and for both stresses, but with lower relative significance for the longer treatment time (Figure 1b).

#### *3.3. Principal Component Analysis (PCA)*

A PCA was performed, separately for the cultivars subjected to the two-week and three-week treatments, and including the mean values of all measured parameters and the three applied conditions (control, water stress and salt stress) (Table 4, Figure 2).

**Table 4.** Component weights in the PCA performed on cultivars subjected to two and three weeks of treatment. Abbreviations: RL, root length; RFW, root fresh weight; RWC, root water content; SD, stem diameter; SL, stem length; SFW, stem fresh weight; SWC, stem water content; Lno, leaf number; LFW, leaf fresh weight; LWC, leaf water content; Pro, proline.


The PCA corresponding to the two-week treatments detected two components with Eigenvalues higher than 1, which explained 70.1% of the total variability of data (56.5% and 13.6% for the first and second components, respectively). All growth parameters—most significantly the fresh weights of stems (SFW) and leaves (LFW), followed by the water contents of both organs (LWC and SWC)—were positively correlated with the first component, whereas the only one negatively correlated was Pro concentration in leaves. Regarding the second component, some morphological variables (especially SWC and RFW) were positively correlated, whereas for others (e.g., SL or Lno) the correlation was negative (Table 4, Figure 2a).

Two components with an Eigenvalue higher than one were also detected in the PCA corresponding to the three-week treatments, the first explaining 44.3% and the second 14.0% of the total variability; that is, together explaining 58.3% of the total variation. Correlations of the different variables followed similar patterns to those observed for the cultivars treated for two weeks, for example regarding the negative correlation of Pro with the first component, and the positive correlations of all growth variables, with SFW and LFW showing the highest significance (Table 4, Figure 2b).

The 18 cultivars from the shorter treatment period (Figure 3a) were dispersed onto the two axes of the scatterplot, indicating high variability in the selected genotypes. There was, however, good separation between the different treatments, not only when looking individually at each cultivar, but also considering the overall behaviour of all genotypes. Plants from the control (green symbols) and water stress (pink) treatments were clearly separated, with almost no overlapping between the two conditions. Those symbols (blue) corresponding to the salt stress treatments appear located in the scatterplot in-between the control and water stress samples, which was in agreement with the weaker effect (on average) of the salt treatments as compared to water deficit, under the specific conditions used in our experiments. The scatterplot corresponding to the 29 cultivars that were subjected to the more prolonged (three-week) treatment (Figure 3b) showed the same general picture, maybe with more overlapping of the water- and salt-stressed plants. More-significant dispersion of the scores was found, for both treatment times, in the controls indicating a high variability of morphological traits of the different cultivars. Under salt stress, the separation between scores was not so pronounced as under water stress, suggesting a more homogeneous general response of the bean genotypes to salinity than to drought, at least under the conditions of our experiments (Figure 3).

**Figure 2.** Loading plot of the principal component analysis (PCA) conducted with the analysed traits, in *P. vulgaris* cultivars subjected to control, water deficit and salt stress treatments. Two-week treatments (**a**); 56.5% and 13.6% of the total variability are explained by the first (*x*-axis) and the second (*y*-axis) components, respectively. Three-week treatments (**b**); 44.3% and 14.0% of the total variability are explained by the first (*x*-axis) and the second (*y*-axis) components, respectively. Abbreviations: RL, root length; RFW, root fresh weight; RWC, root water content; SD, stem diameter; SL, stem length; SFW, stem fresh weight; SWC, stem water content; Lno, leaf number; LFW, leaf fresh weight; LWC, leaf water content; Pro, proline.

**Figure 3.** Scatter plot of the PCA scores. Plants treated for two (**a**) or three weeks (**b**); control (green), water deficit (pink) and salt stress (blue) treatments. (**a**) 1–9, cultivars from Spain; 10–15, from Colombia and 16–18, from Cuba and (**b**) 19–32, cultivars from Spain; 33–45, from Colombia and 46 and 47 from Cuba.

Based on the PCA scatter plot in the two-week trial (Figure 3a), we identified four accessions (7, 5, 13 and 17) with highly negative values for the first component (i.e., with high concentrations of Pro and low values for growth and water content parameters), both for the water deficit and salinity treatments; these cultivars can be considered as highly susceptible to both stresses. On the other hand, three accessions (25, 31 and 32) were detected in the three-week scatter plot (Figure 3b), showing positive values for the first component (low Pro contents and limited growth inhibition), both for the drought and salt stress treatments, indicating that these accessions can be considered as the most tolerant to both stresses. Similarly, the relative position of other accessions along the *x*-axis should allow a ranking of their tolerance to water deficit and to salinity, within each group of cultivars (treated for two or three weeks).

#### **4. Discussion**

In the present study, responses to drought and salinity have been analysed in 47 *Phaseolus vulgaris* genotypes of different origins. Large variability was observed in the size and morphology of the plants of the different bean cultivars—as seen when comparing their growth parameters (Supplementary Table S1) individually and also by their dispersion in the PCA scatterplots (Figure 3)—making it difficult to determine, at first sight, the variables that are more relevant for assessing the relative degree of stress-induced growth inhibition and, therefore, for ranking the different cultivars according to their relative sensitivity or resistance to water deficit and salt stress. However, the statistical analyses performed with all experimental data provided a clear overall picture of the responses to stress of the *P. vulgaris* cultivars. Both 'cultivar' and 'treatment', as well as their interaction, had a highly-significant effect on (practically) all growth traits analysed, and on Pro contents, for the twoand three-week treatments of both water deficit and salt stress. In all cases, growth inhibition was mostly reflected in the stress-induced reduction of fresh weight and water contents of stems (SFW and SWC) and leaves (LFW and LWC), as reported in the same species [39,40] or other species of this genus [43]. These parameters are the growth variables most significantly correlated, positively, with the first principal component in the PCA. Pro, on the other hand, was the only variable showing a negative correlation with all growth variables, but particularly with those of stems and leaves mentioned above—as indicated by the Pearson correlation coefficients and the loading plots of the PCAs.

When comparing the stress tolerance of related taxa, for example, different cultivars of a particular crop, measurements of growth parameters are often complemented with the determination of several biochemical stress markers, associated with increased (or lower) tolerance; they include compatible solutes or osmolytes [44–47]. Proline (Pro) is a common osmolyte in plants, which accumulates in response to different types of abiotic stress, including drought and salinity, in a variety of plant species [28,48–50]. Besides its role in cellular osmotic adjustment, Pro has additional functions as 'osmoprotectant'; it directly stabilises sub-cellular structures, such as membranes and proteins, scavenges free radicals buffering redox potential, alleviates cellular acidosis and acts as a signalling molecule in the responses to stress [51,52]. Proline also plays essential roles in the absence of stress, being involved in many developmental processes; for example, Pro concentration increases during pollen and seed maturation. However, Pro can be toxic for certain tissues if it is partially catabolised to pyrroline-5-carboxylate (P5C), leading to apoptosis [53]. Considering the multiple functions of Pro, it is logical to assume that Pro accumulation would be associated with higher stress tolerance, and this has indeed been demonstrated for many plants, both wild species [54,55] and crops [43,56]. However, other comparative studies on related taxa, such as species of the same genus or cultivars or varieties of the same species, revealed higher Pro accumulation under stress in the less-tolerant genotypes [57,58]. There is some confusion, often found in the literature, between the concepts of 'stress responses' and 'stress tolerance'. Even though stress tolerance mechanisms are based on specific stress responses, not all responses are relevant for tolerance. On this line, Pro accumulation can be considered as a general 'response' to abiotic stress in many plant species, but Pro may or may not be involved in stress tolerance mechanisms, depending on the species.

Common bean is clearly a Pro accumulator species, as numerous reports have shown significant increases in Pro contents in *Phaseolus* plants in response to either salt stress [38,39,59] or water stress [40,60,61] treatments. Also, Pro appears to be a good bioindicator in other types of stress in beans, such as that induced by excess nitrogen dosage [62], herbicides [63] or heavy metals [64]. Moreover, exogenous application of Pro was shown to alleviate the salt stress deleterious effects in beans [65]. However, there are some contradictory data in the literature regarding the function of Pro in the mechanisms of stress tolerance in *Phaseolus.* Some published reports indicated higher Pro contents in more drought-tolerant [30,34–37] or salt-tolerant [31–33] cultivars than in less tolerant ones; that is, Pro accumulation correlates positively with the degree of stress resistance, suggesting a direct contribution to stress tolerance mechanisms. Other reports, on the contrary, showed that, under stress conditions, the less tolerant genotypes had a higher concentration of this osmolyte than the more resistant cultivars [38–40,66]; therefore, in this case, Pro is simply a marker of the level of stress affecting the plants, accumulating at higher concentrations in the more stressed—the more sensitive—cultivars, but is not directly involved in the mechanisms of tolerance. This was also the conclusion of previous work from our laboratory, comparing three commercial cultivars (two of *P. vulgaris* and one of *P. coccineus*) and one Spanish common bean landrace [39,40]. All these latter studies, based on the comparison of a few bean genotypes, generally some commercial cultivars, have been confirmed in the present work, using a much larger number of cultivars of different origins and an extensive statistical analysis of the experimental data.

Our results showed a strong negative correlation of Pro levels and growth variables, especially the fresh weight and water content of the aboveground organs of the plants; these are the most relevant parameters to evaluate the inhibition of growth induced under water deficit and high salinity conditions. Therefore, there is an unequivocal association of higher Pro contents with stronger growth inhibition; that is, with a higher sensitivity to stress of the bean cultivars.

#### **5. Conclusions**

*Phaseolus vulgaris* cannot be considered as drought- or salt-tolerant. It is even more sensitive to stress than many other crops such as barley or cowpea [67,68]. However, amongst the extremely high number of available genotypes of *P. vulgaris*, some will show a relatively higher resistance and could be used as parental lines in bean breeding programmes aimed at enhancing stress tolerance in this major crop. The identification of common bean accessions in the extremes of variation for susceptibility and tolerance to water deficit and salinity is of great interest for further studies on the physiological mechanisms of tolerance to both stresses. Also, the development of segregating generations after hybridisation between both types of materials can lead to the identification of genomic regions involved in tolerance to these stresses.

Proline concentrations in stressed plants can be determined by a simple and rapid spectrophotometric assay, requiring only small amounts of leaf material. From a practical point of view, our results support the use of Pro as a biochemical marker for the initial, large-scale screening of bean cultivars, to exclude the most sensitive, those accumulating higher Pro concentrations in response to water or salt stress.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/6/817/s1, Table S1: Variation of morphological parameters and proline concentrations in 47 accessions of common bean (*Phaseolus vulgaris*) under salt stress and water stress.

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

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

**Acknowledgments:** We are indebted to Steve Beebe, from Alliance Biodiversity CIAT, for providing the Colombian bean cultivars and for his helpful comments on the manuscript.

**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/).
