Comparative plant biochemistry and soil biology of wild and cultivated cotton species

Purpose No attempts were made to analyze the diversity in soil and plant biology of wild cotton species (WCS) and cultivated cotton species (CCS), so far. Our study aimed to understand the differences in soil biological, plant biochemistry, and defense enzyme activities among the ten WCS and four CCS. Methods We studied the differences in soil biology, plant biochemistry, and defense enzyme activities among the ten WCS (Gossypium anomalum, G. aridum, G. australe, G. barbosanum, G. capitis-virides, G. davidsonii, G. raimondii, G. somalense, G. stocksii, G. thurberi) and four CCS (G. arboreum, G. herbaceum, G. hirsutum, and G. barbadense).


Introduction
Cotton (Gossypium spp.) is an important natural bre crop, which provides bre to the textile industries and supply oil and feed to society ). In Gossypium, there are 43 diploids (2n = 26 chromosomes), which have been classi ed into 7 genomes from A to G, and 7 tetraploids (2n = 52 chromosomes) with genome designation AD (Wendel and Grover 2015). Out of 50 Gossypium species, two diploids (G. arboreum and G. herbaceum) and two tetraploids (G. hirsutum and G. barbadense) are cultivated for their spinnable bre, worldwide, and the remaining 46 are wild species (Wendel and Cronn 2003;Campbell et al. 2010). Among the cotton-growing countries of the world, India is unique in terms of cultivating all four cultivated cotton species (CCS) (G. hirsutum, G. barbadense, G. arboreum and G. herbaceum) under diversi ed agro-ecological system (Gotmare et al. 2000;Gotmare 2021). Globally, cotton is grown in 75 countries, and breeding superior cotton to mitigate biotic and abiotic stresses for continued yield and quality enhancement is a primary goal of the cotton research community, worldwide (Shim et al. 2018). Among the Gossypium species, wild cotton species (WCS) serve as a genetic reservoir of several unique traits (wider adaptation, pest and disease tolerance, and yield contributing factors), which are useful to cotton breeders for prebreeding and genetic improvement in CCS (Narayanan et al. 2014;Mammadov et al. 2018). In cotton improvement, introgressive hybridization has played a signi cant role in transferring ber quality (length and strength), disease resistance (black arm, rust, wilt), insect resistance (jassids, boll weevil), drought resistance, and male sterility traits (cytoplasmic male sterility) from various wild and cultivated species (Wang et al. 2016). Further, the hybridization of cultivated plant species with wild species can also transfer bene cial traits, which can alter the composition of plant and microbial communities, potentially leading to plant and soil health (Bulgarelli et al. 2015).
Understanding the heritage of crop plants and their wild relatives has been a major focus for plant science, as it helps in crop domestication and improvement (Byrne et al. 2018). Several germplasm studies and breeding programs have shown the supremacy of wild species in enduring biotic and abiotic stresses compared to their cultivated counterparts (Byrne P et al. 2020; Hübner and Kantar 2021), although tolerance genes have conventionally been considered as negatively correlated with yield (Wise 2007). However, insu cient data on phenotypic and genotypic differences, and other genetic variations (ploidy, hybridization barrier, linkage, etc.) act as blockade in wild species utilization for crop improvement (Dempewolf et al. 2017), and a large portion of the natural variation in the wild species of cotton remain untapped (Shim et al. 2018). Though cotton-breeding program traditionally utilize phenotypic and genotypic information in selecting parents (from wild or cultivated species) for prebreeding and introgression studies, they overlook plant and soil innate attributes due to absence of data. Hence, generating information related to basic plant biochemical and soil biological traits further help breeders hasten the selection of traits of agronomic importance. However, no study has attempted to analyze the diversity in soil and plant biology of cultivated and wild cotton species, so far. Therefore, a comparative analysis was conducted to understand the differences in soil biological and plant antioxidant/defense enzyme activities among the cultivated and selected wild cotton species. Our study will provide a greater understanding of biochemical and soil biological differences between cultivated and wild cotton species, and eventually supplement the cotton database, and support cotton breeders in selecting unique traits for their crop improvement programs.

Materials And Methods
Site description and sampling Soil and plant sampling were conducted during the pre-monsoon (February) 2021 at wild species garden, ICAR-Central Institute for Cotton Research, Panjari Farm, Nagpur (21° 02′ 8′′ N and 79° 03′ 32′′ E) in Central India. This farm is situated at 309 m above the mean sea level and has a mean annual rainfall of 1200 mm. The study site has a deep black Vertisol (Typic Haplusterts) classi ed as sub humid moist bioclimate (10.2) under Agro-ecological sub-regions of India. For our analysis, we selected ten wild (Gossypium anomalum (Wawra & Peyritsch), G. aridum (Skovsted), G. australe (F.Muell.), G. barbosanum (Phillips & D.Clement), G. capitisvirides (Mauer), G. davidsonii (Kelogg), G. raimondii (Ulbrich), G. somalense (Hutchinson), G. stocksii (Mast. & Hook.), G. thurberi (Todaro)) cotton species (based on their economic traits) and four cultivated (Gossypium arboreum L., G. herbaceum L., G. hirsutum L., and G. barbadense L.) cotton species. The detailed description about the genome, distribution, and economic attributes of wild and cultivated cotton species used in this study is presented in Table 1. The rhizosphere soil samples were collected at 0-60 cm soil depth, while, the shoot and root samples were collected from the representative cotton species, and were labelled and transported back to the laboratory in polyethylene bags/ice boxes, and stored at 4°C before analysis.

Soil biological analysis
Soil basal respiration rate (SBR) was measured following the method outlined by Anderson (1982). Soil microbial biomass carbon (MBC) was determined using the CHCl 3 fumigation-extraction method (Vance et al. 1987), and the MBC was calculated using the equation Biomass C = 2.64 EC (extractable carbon), where EC = (organic C in K 2 SO 4 from fumigated soil) -(organic C in K 2 SO 4 from unfumigated soil). The microbial metabolic quotient (qCO 2 ) an indicator of heterotrophic respiration was determined according to Anderson and Domsch (1986) by calculating the respiration-to-biomass ratio for the samples analyzed. The easilyextractable glomalin related soil proteins (EE-GRSP) was estimated according to Wright and Upadhyaya (1998). The total soil polysaccharides (SPS) and soil proteins (SoP) from wild and cultivated cotton species grown rhizospheric soils were quanti ed following the procedure outlined by Lowe (1994) and Bradford (1976), respectively.

Soil enzyme assay
The soil dehydrogenase activity (DHA) was determined by the colorimetric measurement of the reduction of 2,3,5-triphenyltetrazolium chloride to triphenylformazan according to the method of Casida et al. (1964). The activities of acid (AcP) and alkaline phosphatases (AlkP), and β-glucosidase (BG) were assayed according to Tabatabai (1982).
The POD and PPO activity was determined spectrophotometrically by reading the absorbance at 470 and 546 nm, respectively, and expressed as units mg −1 protein min −1 . The catalase (CAT; EC 1.11.1.6) activity in the shoot and root samples was estimated according to the procedure outlined by Bergmeyer (1970). CAT activity was determined by reading the absorbance at 240 nm with a time scan of 0-60 s and expressed as units mg −1 protein min −1 . The L-phenylalanine ammonia lyase (PAL; EC 4.3.1.5) activity in the shoot and root samples was determined by the method described by Whetten and Sederoff (1992). The absorbance (290 nm) was measured using the upper phase and PAL activity was expressed in moles t-cinnamic acid mg −1 protein min −1 . The phenolic concentrations in the shoot/root samples were estimated using Folin-Ciocalteu reagent (Singleton et al. 1999). The intensity of the blue colour developed was measured spectrophotometrically at 660 nm. The amount of phenol in the shoot/root samples was calculated from a standard curve prepared using caffeic acid. The total phenol was expressed as milligrams of caffeic acid equivalent (CAE) gm −1 fresh weight. The total carbohydrate and protein content in shoot and root samples of wild and cultivated cotton species were determined by the method outlined by DuBois et al. (1956) and Lowry et al. (1951), respectively.

Statistical analysis
Experimental data in triplicate related to soil biological parameters, plant storage products, and plant antioxidant/defense enzymes were statistically analysed by one-way analysis of variance ANOVA with WASP version 2.0 (Web Agri Stat Package, Indian Council of Agricultural Research, India). Signi cant differences at P ≤ 0.05, after ANOVA, were followed by single means differentiated by Tukey's honestly signi cant difference (HSD) test at P ≤ 0.05. Principal component analysis was performed through XLSTAT program.

Plant carbohydrate and protein
Signi cant differences (P < 0.05) in the carbohydrate concentration were recorded in root and shoot tissues of WCS and CCS, however, shoot samples exhibited higher carbohydrate concentrations (Fig. 3a). Between wild and cultivated species, WCS recorded signi cantly (P < 0.05) higher carbohydrate concentration in shoot (SC) and roots (RC). Among the WCS, higher SC was observed in G. anomalum (97.2 mg g −1 ) followed by G. australe (86.3 mg), and the lowest SC was recorded in G. somalense (33.7 mg). In cultivated species, higher SC was recorded in G. hirsutum (45.9 mg) followed by G. arboreum (38.7 mg), and the lowest in G. barbadense (32 mg). The percent increase in SC compared with G. hirsutum was 112% and 88%, respectively, for G. anomalum and G. australe. Among the WCS, higher root carbohydrate (RC) was recorded in G. australe (47.4 mg), followed by G. aridum (46.1 mg), and the lowest RC was observed in G. thurberi (20.2 mg). In cultivated species, G. herbaceum (35.4 mg) recorded higher RC, and the lowest was recorded in G. barbadense (13.3 mg). The percent increase in RC compared to G. hirsutum was 152%, 146%, and 89%, respectively, for G. australe, G. aridum, and G. herbaceum.
Signi cant differences (P < 0.05) in the protein concentration were recorded in root and shoot tissues of WCS and CCS, however, shoot samples exhibited higher protein concentration in the wild and cultivated species (Fig. 3b). Cultivated species recorded signi cantly (P < 0.05) higher protein content concentration in shoot (SP) and roots (RP) compared to WCS. Among the CCS, higher SP was observed in G. herbaceum (74.2 mg g −1 ) followed by G. arboreum (66.8 mg), and the lowest SP was recorded in G. hirsutum (58 mg). In wild species, higher SP was recorded in G. australe (73.3 mg) followed by G. anomalum (66 mg), and the lowest in G. capitis-virides (35 mg). The percent increase in SP compared to G. hirsutum was 28%, 26%, and 15%, respectively, for G. herbaceum, G. australe, and G. arboreum. Among the CCS, higher RP was observed in G. barbadense (25.9 mg g −1 ) followed by G. herbaceum (22.9 mg), and the lowest RP was recorded in G. hirsutum (20. 8 mg). In wild species, higher RP was recorded in G. anomalum followed by G. stocksii (23.8 mg) and G. raimondii (23.8 mg). The lowest RP was recorded in G. capitis-virides (13.9 mg). The percent increase in RP compared with G. hirsutum was 25%, 10%, 17.5, and 14.7%, respectively, for G. barbadense, G. herbaceum, G. anomalum, and G. stocksii/ G. raimondii.
Plant antioxidant/defense enzymes WCS recorded signi cantly (P < 0.05) higher peroxidase activity (POD) in shoots and root tissues compared to CCS (Fig. 4a). Among the WCS, higher shoot POD was observed in G. aridum (7.54 units (U) g −1 protein min −1 ) followed by G. somalense (4.37 U), and the lowest in G. australe (3.26 U). Among the CCS, higher shoot POD was observed in G. arboreum (3.80 U) followed by G. hirsutum (3.0 U), and the lowest in G. barbadense (2.34 U). In wild species, G. aridum (10.7 U) followed by G. somalense (8.2 U) recorded higher root POD, while, lowest activity was recorded in G. barbosanum (4.64 U). Among the CCS, higher root POD was observed in G. arboreum (6.93 U) followed by G. hirsutum (4.5 U), and the lowest activity in G. barbadense (3.53 U). The percent increase in POD in G. aridum and G. arboreum compared with G. hirsutum was 144% & 139% and 23% & 54%, respectively, for shoot and root tissues.
Signi cantly (P < 0.05) higher polyphenol oxidase activity (PPO) was recorded in WCS compared to CCS in the shoot and root tissues (Fig. 4b). G. thurberi (0.489 U g −1 protein min −1 ) followed by G. australe (0.419 U) recorded higher PPO in shoot and root samples among the WCS, while, the lowest PPO in shoot and root sample was recorded in G. capitis-virides (0.237 U) and G. barbosanum (0.145 U), respectively. Among the CCS, higher shoot and root PPO were observed in G. arboreum (0.487 & 0.229 U) followed by G. hirsutum (0.319 & 0.210 U), respectively, and the lowest PPO was observed in G. barbadense (0.252 U) and G. herbaceum (0.185 U) respectively for the shoot and root samples. The percent increase in PPO in G. thurberi, G. australe, and G. arboreum compared with G. hirsutum was 54% & 31%, 55% & 38%, 53% & 9%, respectively, for shoot and root tissues.
Signi cantly (P < 0.05) higher catalase activity (CAT) was recorded in WCS compared to CCS in the shoot as well as root tissues (Fig. 4d). G. aridum ( arboreum compared with G. hirsutum was 138% & 112% and 16% & 18%, respectively, for shoot and root tissues. Both shoot and roots showed signi cant differences (P < 0.05) in the phenolic content, however, the shoot samples exhibited much higher values in the WCS and CCS (Fig. 4e). Among the WCS, the shoot phenolic concentrations were signi cantly higher in G. barbosanum (4.57 mg caffeic acid equivalent (CAE) g −1 fresh wt.) followed by G. capitis-virides (4.50 mg) and G. anomalum (4.41 mg), and the lowest phenol content was recorded in G. davidsonii (2.30 mg). In CCS, G. herbaceum (3.91 mg) showed the highest shoot phenolic concentration followed G. arboreum (3.74 mg), and lesser phenols were recorded in G. hirsutum (2.97 mg). There were no much differences in mean root phenolic concentrations between WCS and CCS.
However, the root phenolic concentrations varied signi cantly (P < 0.05) among the species. Among the WCS, higher phenols were recorded in G. anomalum (2.40 mg) followed by G. aridum (2.08 mg), and lowest in G. stocksii (1.20 mg). G. herbaceum (2.19 mg) recorded higher root phenols in CCS, and the lowest was recorded in G. arboreum (1.41 mg). The percent increase in shoot phenol content in G. barbosanum and G. herbaceum compared with G. hirsutum was 54% & 32%. Similarly, the percent increase in root phenol content in G. anomalum and G. herbaceum compared with G. hirsutum was 65% & 51%.

Discussion
Presently, several Gossypium germplasm, including wild species and land races are conserved through national gene banks worldwide to serve as a source for introgression studies for ongoing and future crop improvement programs. However, no study has attempted to analyze the diversity in soil and plant biology of cultivated and wild cotton species, so far. Our study aimed to understand the differences in soil biological attributes and plant antioxidant and defense activities between WCS and CCS.
From our study, we found signi cant differences in soil, plant and microbial functional attributes between WCS and CCS (Fig. S1). CCS recorded signi cantly higher soil basal respiration rate and microbial biomass carbon than WCS. Since, both the CCS and WCS where grown in the similar conditions in the wild species garden, the difference in respiration rate and biomass carbon can only be attributed to plant biomass quantity and litter quality (biochemical constituents), which have favoured enhanced heterotrophic respiration in CCS compared to WCS. Higher phenolic contents in leaf and root biomass of WCS slow down the microbial growth and subsequent decomposition of biomass reducing soil CO 2 emissions.
Soil respiration is the key mechanism through which soil carbon is released into the atmosphere at a global scale, and hence it acts as an important indicator of carbon cycling in the ecosystem (Song et al. 2021), and help in the prediction of warming-related increases in soil CO 2 emissions and climate change (Davidson and Janssens 2006). However, it is still poorly understood how genotypic changes in cotton species can affect this soil respiration process. It has been proposed that soil respiration has in uenced either by plant functional groups (genotypes/diversity) or through differences in nutrient concentration in the plant biomass (Dias et al. 2010). Path analysis revealed that species richness predominantly regulates soil respiration through variations in plant productivity and community structure (Xu et al. 2015), rather than differences in species composition (Dias et al. 2010). However, soil respiration is controlled by soil temperature, soil moisture, soil aeration (Cook and Orchard 2008), soil nitrogen content (Song et al. 2021), differences in canopy structure, plant biomass, and associated litter quality (Xu et al. 2015), which in uences soil physico-chemical properties (Aponte et al. 2012) and soil microbial community structure, and eventually result in soil autotrophic and heterotrophic respiration (Xu et al. 2015). In cotton, soil respiration was greatly in uenced by soil water content and external irrigation, which directly affects the root respiration activity and soil CO 2 production and exchange (Yu et al. 2015). Previously, Bhattarai et al. (2006) reported low oxygen and lesser respiration can lead to lesser cotton yields on a heavy clay soil.
Oxygation (irrigation of crops with aerated water, through air injection in the root zone) of cotton increased the soil oxygen content and soil respiration eventually water use e ciency, enhancing cotton plant Peroxidase (POD) is a key enzyme involved in the biosynthesis of lignin and suberin (Thakker et al. 2013), and its activity in plants is induced by pathogens (Sasaki et al. 2004), and higher POD activity in plants has been correlated with the plant defense response to external stresses, which leads to ligni cation of plant tissues, suberization, cross-linking of glycoproteins and phenols, and phytoalexin production (Agrios 2005).
Plant POD was also implicated in the inhibition of pathogens through the production of reactive oxygen species (Passardi et al. 2005). Conservation of adequate antioxidant enzymes preceding heat stress is reported to ease the leaf temperature increase in cotton (Snider et al. 2010). Polyphenol oxidase (PPO) is a key enzyme involved in catalyzes of phenols to quinones, which exhibits antimicrobial activity by several mechanisms including alkylation, reduced bioavailability of cellular proteins to the pathogen, production of reactive oxygen species (Thakker et al. 2007). Pathogen-induced PPO activity has been reported in several plants, including monocots and dicots (Raj et al. 2006). Furthest, PPO production in plants was reported to decrease the nutritional quality of food and reducing protein digestibility by conversion of phenols to quinones (Felton et al. 1994). Phenylalanine ammonia lyase (PAL) enzyme is a key enzyme involved metabolism of aromatic amino acids (MacDonald and D'Cunha 2007), and was shown to act as defense molecules against pests and disease infestation (War et al. 2012). PAL activity is induced by signalling molecules viz., ethylene, salicylic acid, and jasmonic acid (Kim et al. 2007) and biotic and abiotic stresses (Lafuente et al. 2003). Catalase (CAT) enzyme and their isomers play a major role in the scavenging of Among the Gossypium, WCS serve as a genetic reservoir of diverse unique traits, which are useful for genetic improvement in CCS. Though cotton-breeding strategy largely uses phenotypic and genotypic data for parent selection and prebreeding, they overlooked plant biochemistry and soil biology attribute due to the absence of data. Further, no attempts were made to analyze the diversity in soil and plant biology of WCS and CCS, so far. Our study aimed to understand the differences in soil biological, plant biochemistry, and defense enzyme activities among the ten WCS and four CCS. Our study suggests that the difference in soil biological, plant biochemistry, and defense enzyme activities among the WCS and CCS are attributed to the inherent genetic makeup of WCS and CCS, which in uences consequent plant and soil attributes. Further studies on identi cation of genetic mechanisms and genes involved in modulating the plant defense enzymes and nutrient cycling in WCS and CCS will help conventional and molecular breeding in cotton.  GraphicalAbst.eps