Introduction
Cotton, a prominent and leading source of fibre, is one of the most important economic crops grown in both dry land and irrigated areas of the world. Water is considered to be the key factor in any crop production system. However, sustainable cotton production is greatly threatened by the water-deficit conditions. Thus among the abiotic stresses, drought is recognized as the most devastating cause which markedly limits the fibre yield and lint quality in cotton production (Chattha et al., Reference Chattha, Shakeel, Malik, Saleem, Akram, Yaseen and Naeem2018). The flowering and boll development are the most critical stages requiring irrigation determining the yield in upland cotton (Gossypium hirsutum) (Rahman et al., Reference Rahman, Ullah, Ahsraf, Stewart and Zafar2008). The drought stress significantly reduces cotton production by affecting many agronomic traits such as reduction in size and number of bolls per plant, plant height, shoot fresh weight, seed cotton yield (SCY), etc. (Malik and Malik, Reference Malik and Malik2006). A common negative impact of drought stress on crop plants is the reduction in above ground fresh and dry biomass production (Chen et al., Reference Chen, Ma, Xia, Hou, Shi, Hao, Hafeez, Han and Luo2017). A decrease in the relative water content, chlorophyll contents (CC) and leaf area (LA) under drought condition has been reported in a variety of plant species (Nayyar and Gupta, Reference Nayyar and Gupta2006; Saleem et al., Reference Saleem, Malik, Shakeel, Amjad and Qayyum2015).
The importance of root systems in acquiring water has long been recognized as a tolerant mechanism against drought stress (Jamal et al., Reference Jamal, Shahid, Aftab, Rashid, Sarwar, Mohamed, Hassan and Husnain2014). The development of root system increased the water uptake and maintains requisite osmotic pressure through higher proline levels in Phoenix dactylifera (Basal et al., Reference Basal, Smith, Thaxton and Hemphill2005; Djibril, et al., Reference Djibril, Mohamed, Diaga, Diegane, Abaye, Maurice and Alain2005). Drought tolerance is a complex agronomic trait with multigenic components, which interact in a holistic manner in the plant system (Cushman and Bohnert, Reference Cushman and Bohnert2000). Therefore, the identification of drought stress-tolerant genotypes is an ongoing challenge for the breeders. However, plant stress tolerance can be developed by identifying and characterizing those morpho-physiological traits which contribute stress tolerance and determine their relative relationship with productivity under water-deficit condition. The prerequisite for success is based on the extent of genotypic variability within a species for these traits (Cooper et al., Reference Cooper, Podlich and Smith2005). Cotton breeders are trying to identify drought-tolerant cotton genotypes based on morphological, biochemical and physiological attributes to compensate for yield losses and improve productivity. Diploid cotton also known as desi cotton (Gossypium arboreum) is well known for possessing many favourable attributes for cotton yield that upland cotton lacks like for instance, improved drought tolerance, resistance to diseases and insect pests and well adaptability to dry land conditions amenable to low input cultivation practices (Mehetre et al., Reference Mehetre, Aher, Gawande, Patil and Mokate2003; Liu et al., Reference Liu, Guo, Lin, Nie and Zhang2006). However, a thorough elucidation of the root morphological, physiological and biochemical performance by which different cotton species respond to drought is still missing and poses a constant challenge for the breeders.
Thus the objective of the present study was to identify the root-related traits responsible for drought tolerance in G. arboreum so that these drought tolerance indicators could be used for screening and improving the G. hirsutum genotypes for drought tolerance.
Materials and methods
The experiment was conducted at research area of the Department of Plant Breeding and Genetics, The Islamia University of Bahawalpur, Pakistan from April to October in 2017 and 2018. Cotton was grown in polyvinyl chloride pipes (diameter: 20 cm and the total column height was 105 cm), each filled with 7 kg of a mixture of sand and organic-rich field soil (1:1). The pipes were spaced 90 cm in one direction and 60 cm in another.
Four cotton genotypes two from each species, i.e. G. hirsutum, (IUB-65 and IUB-13) and from G. arboreum, (PC-1 and COMILLA) were used for experimentation. The experiment was structured in a completely randomized design (CRD) factorial design with two factors (two moisture levels and four genotypes) with three replications. Twenty-four pipes were vertically buried in the field. Each genotype was accommodated in one pipe. Four seeds for each genotype were direct seeded in each pipe.
On germination, only one seedling was retained in each pipe. The seedlings were raised under optimum moisture conditions for 30 d. After 30 d of growth, half of the pipes were supplied with optimum irrigation (normal), while in the other half, irrigation was withheld to develop water stress. A total of eight tensiometers were installed (at 30 cm depth); one for each genotype under both conditions (normal and drought stress). The normal was irrigated when moisture level declined to 60% while drought stressed pipes were irrigated when moisture level reduced to 40% throughout the season. The optimum 60 and 40% levels were determined from a preliminary experiment conducted (data not shown) providing enough differences in growth.
A dose of basal fertilizer (2.76 g of N, 9.36 g of P2O5, 6.38 g of K2O per pipe) was mixed in the 10–20 cm layer before sowing. An additional dose of 18 g/pipe of N was applied in three split doses till maturity.
Data collection
On the 180th day after sowing, the pipes were removed carefully and soaked in water overnight to loosen the soil. Next day plants were carefully extracted from the pipes without damaging the roots. To measure the root traits, each pipe was carefully dug and cut down into 20 cm segments at the top to bottom. The segments were immersed in water for 1/2 h and the roots from each segment were rinsed, removed carefully and rinsed with tap water. Debris, weeds as well as dead roots were separated from live roots by hand according to the method given by Gwenzi et al., Reference Gwenzi, Veneklaas, Holmes, Bleby, Phillips and Hinz2011. So, live roots from each pipe column were evenly spread in a plastic tray holding deionized water. Data were collected on following parameters;
Primary root length (PRL)
PRL (cm) was measured for individual plants using a meter rod.
Lateral root numbers (LRN)
LRN were counted manually for individual plant root.
Root fresh weight (RFW)
Meanwhile, fresh weight was also taken for the root of individual plants by weighing the balance in grams after exposure to sunlight for 1/2 h.
Root dry weight (RDW)
Fresh roots were also split into small pieces and oven dried at 80°C for 3 d in a fan-forced oven to measure dry root weight (g).
Leaf area (LA)
To measure LA (cm2), the selected leaflets were separated from petiole and numbered. Each leaflet area was measured by a portable Leaf Area Meter.
Chlorophyll contents (CC)
CC (μmol/m2) of fully expanded leaves of selected plants were measured in the early morning from 7.00 to 9.00 a.m. through a portable fluorometer plant analyser (Hansatech Instruments, King's Lynn, Norfolk, UK).
Plant height (PH)
At harvest, the distance between the highest point of the plant and the ground was recorded with the help of measuring tape in centimetres for each selected plant.
Monopodial branches/plant (MBP)
The monopodial branches are the vegetative types of branches in cotton. At maturity, the monopodial branches per plant were counted for all the plants.
Above ground fresh biomass (AGFB)
At maturity, each individual plant was removed from each pipe and fresh plant weight (g) was calculated after the separation of the root.
Above ground dry shoot biomass (AGDB)
Fresh shoot were chopped into small pieces and oven dried at 80°C for 3 d in a fan-forced oven to measure dry shoot biomass (g).
Sympodial branches/plant (SBP)
The sympodial branches are the fruit branches, i.e. bearing the bolls. At maturity, the sympodial branches on each plant were counted for all selected plants.
Seed cotton yield (SCY)
SCY was manually picked from each plant and SCY per plant was weighted in grams.
Lint percentage/ginning out-turn (GOT)
Lint percentage or GOT is the weight of lint that can be obtained from a given weight of seed cotton and is expressed as a percentage. Samples of seed cotton obtained from individual plants were weighed and ginned separately with a single roller electrical gin in the laboratory. Lint was weighed and GOT was calculated as the percentage of lint.
Fibre traits
For fibre trait analysis, the ginned samples were re-conditioned by placing samples in blow room (65% humidity and 18–20°C temperature) using a humidifier. High Volume Instrument (HVI-900-SA; Zellweger Ltd., Switzerland), available at Cotton Research Institute (CRI), Multan, Pakistan, was used to analyse staple length (SL) (mm), fibre strength (FS) (g/tex) and fibre fineness (FF) (micronaire).
Statistical analyses
Data from the response of four cotton genotypes to two moisture levels were analysed as a 4 × 2 factorial experiment with CRD layout, with three replications. Data were subjected to analysis of variance (ANOVA) using the Statistix 8.1 computer software package. Statistical significance is reported at 5 and 1% levels of probability. Since there were non-significant differences for the Genotype × Year and Genotype × Year × Treatment, data were averaged for 2 years and used for further statistical analysis.
Mean comparisons were made through Tukey's Test (Statistix 8.1 computer software package). Standard error of mean was calculated through Microsoft Excel Program.
Reduction/change in morphological, yield, root and fibre quality traits under drought as compared to normal was calculated using the following formula:
$$\eqalign{&{\rm Reduction}\;{\rm in}\;{\rm trait}\lpar \% \rpar {\rm under}\;{\rm drought} \cr & \quad = \left( {{\rm 1}-\displaystyle{{{\rm Performance}\;{\rm under}\;{\rm drought}\;{\rm condition}} \over {{\rm Performance}\;{\rm under}\;{\rm normal}\;{\rm condition}}}} \right) \times 100.}$$Results
Assessment of the genetic variation under normal and water-deficit conditions
ANOVA results described in Table 1 showed that there existed highly significant differences under both levels of significance (P < 0.01 and P < 0.05) among genotypes and moisture treatments indicating that there could be potential for drought tolerance among the genotypes for all the traits. The Genotype × Year and Genotype × Year × Treatment were non-significant for all traits, which imply that the genotypes performed quite consistently between the 2 years and therefore average values for 2 years were taken and discussed.
Table 1. Mean square values for various morphological, physiological and fibre quality traits of two Gossypium arboreum and two Gossypium hirsutum genotypes evaluated for 2 years under two treatments (normal and drought stress conditions)
MBP, monopodial branches/plant; SBP, sympodial branches/plant; PRL, primary root length (cm); LA, leaf area (cm2); PH, plant height (cm); LRN, lateral root numbers; AGFB, above ground fresh biomass; AGDB, above ground dry biomass; RFW, root fresh weight (g); RDW, root dry weight (g); SCY, seed cotton yield (g); CC, chlorophyll contents (μmolm−2); GOT, ginning out-turn (%); SL, staple length (mm); FF, fibre fineness (micronaire); FS, fibre strength (g/tex); Geno, genotypes; Treat, treatments; SOV, source of variation; df, degrees of freedom.
*P < 0.01, **P < 0.05.
Root traits related to drought tolerance such as LRN, PRL, RDW and RFW were considerably enriched in G. arboreum, as compared to G. hirsutum genotypes under drought (Table 2). For PRL trait, the reduction (%) under drought was considerably less in G. arboreum genotypes (−49.9 and −50.5%) in PC-1 and COMMILA, respectively. While in G. hirsutum genotypes, IUB-13 and IUB-65, it was −30.0 and −22.3%, respectively. Regarding LRN trait, the average reduction (%) under drought in G. arboreum genotypes PC-1 showed −56.8% and COMMILA showed −75.8% reductions. While it was −24.9 (IUB-13) and −37.5% (IUB-65) in G. hirsutum genotypes. Regarding RFW and RDW, the average reduction (%) in G. hirsutum genotypes was −18.0% (IUB-13), −29.9% (IUB-65) for RFW and −25.9% (IUB-13) and −27.0% (IUB-65) for RDW. Likewise, for G. arboreum genotypes, it was −41.1% (PC-1), −44.4% (COMMILA) for RFW and for RDW it was −57.5 and −51.1 in PC-1 and COMMILA, respectively.
Table 2. Mean values ± standard errors for root related traits under normal, drought stress and reduction (%) under drought
LRN, lateral root numbers; PRL, primary root length (cm); RDW, root dry weight (g); RFW, root fresh weight (g).
Note: Mean comparisons lettering was applied across rows among genotypes for each treatment separately, i.e. normal and drought stress.
Morphological traits such as PH, MBP, SBP, AGFB and AGDB were higher in G. hirsutum genotypes (IUB-13 and IUB-65) as compared to G. arboreum genotypes (PC-1 and COMMILA) (Table 3). However, under drought stress condition, less reduction in these traits was observed in G. arboreum genotypes than those of G. hirsutum. For instance, for PH trait, the average reduction (%) was 25.0% (PC-1) and 28.1% (COMMILA) in G. arboreum genotypes, while it was 36.5 and 33.7% in IUB-13 and IUB-65, respectively. Regarding MBP trait, the average reduction (%) in PC-1 and COMMILA was 12.6 and 9.0%, respectively. IUB-65 and IUB-13 showed 36.6 and 41.5% reduction under drought, respectively. Similarly, for SBP, the average reduction (%) was also higher in G. hirsutum genotypes (51.6 and 54.2%). While it was 19.2 (PC-1) and 28.1% (COMMILA) in G. arboreum genotypes. AGFB trait was maximum (168.6 and 129.3 g) in PC-1 under normal and drought conditions, respectively, followed by COMMILA (156.8 and 121.3 g) under normal and drought stress conditions, respectively. While IUB-13 showed 290.9 g (normal) and 142.1 g (drought) and IUB-65 showed 237.4 g (normal) and 149.3 g (drought). The reduction was 22.6 and 23.3% in COMMILA and PC-1, respectively. Similarly, IUB-65 and IUB-13 showed 37.1 and 51.2% reduction, respectively. AGDB was 63.6 and 48.5 g in PC-1 under normal and drought conditions, respectively, and in COMMILA, it was 54.8 and 45.3 g (normal and drought, respectively). Likewise, IUB-13 and IUB-65 produced average AGDB as 93.0 and 61.6 g and 81.5 and 54.3 g under normal and drought stress conditions, respectively. Reduction was also higher in G. hirsutum genotypes being 33.4 and 33.8% in IUB-65 and IUB-13, respectively. Between G. arboreum genotypes, it was 17.4 and 23.8% in COMMILA and PC-1, respectively.
Table 3. Mean values ± standard errors for morphological and physiological traits under normal and drought stress conditions and reduction (%) under drought
AGDB, above ground dry biomass; AGFB, above ground fresh biomass; MBP, monopodial branches/plant; SBP, sympodial branches/plant; LA, leaf area (cm2); PH, plant height (cm); SCY, seed cotton yield (g); CC, chlorophyll contents (μmolm−2); GOT, ginning out-turn (%).
Note: Mean comparison lettering was applied across rows for each treatment separately, i.e. normal and drought stress.
For LA, the reduction in G. arboreum was 13.1% (PC-1) and 16.4% in COMMILA. Likewise, IUB-65 and IUB-13 had much higher reduction (27.0 and 30.4%, respectively) under drought stress condition. CC was higher in G. hirsutum genotypes. IUB-13 had 35.2 and 24.6 µmol/m2 under normal and drought conditions, respectively. Similarly IUB-65 had 34.6 µmol/m2 (normal) and 26.0 µmol/m2 (drought). Between G. arboreum genotypes, PC-1 had CC 22.0 and 20.7 µmol/m2 under normal and drought conditions, respectively. However, reduction was lower in G. arboreum genotypes, i.e. 5.9 (PC-1) and 10.0% (COMMILA) than in G. hirsutum genotypes, i.e. both IUB-13 (30.1%) and IUB-65 (24.9%) showed much higher reductions under drought.
Under normal irrigation, the total SCY of G. hirsutum was 69.7 g (IUB-13) and 70.6 g (IUB-65); while in G. arboretum, the SCY was 42.3 g (PC-1) and 38.7 g (COMMILLA). In contrast under drought condition, the SCYs were greatly reduced in G. hirsutum, i.e. 30.5 and 29.0 g for IUB-13 and IUB-65, respectively. However, drought stress less affected the yield of G. arboreum as it was 30.5 g (COMMILLA) and 28.5 g (PC-1). Therefore, SCY showed less reduction (%) in G. arboreum genotypes as was 21.2% in COMMILA and 32.5% in PC-1. While between G. hirsutum genotypes, it was 56.2% (IUB-13) and 59.0% (IUB-65) much higher than that of G. arboreum genotypes. For GOT, reduction in G. arboreum genotypes was 3.8% (PC-1) and 4.0% (COMMILA), likewise, IUB-65 (6.1%) and IUB-13 (6.5%).
Fibre traits were also of no exception. For FS trait, less reduction was recorded in G. arboreum genotypes (7.5 and 8.3%) in COMMILA and PC-1, respectively, while IUB-65 and IUB-13 (G. hirsutum) showed higher reduction as 10.4 and 11.6%, respectively. Similar reduction trend was observed for SL with less reduction for G. arboreum genotypes; 4.5% (PC-1) and 5.3% (COMMILA). Between G. hirsutum genotypes, IUB-13 showed 8.7% and IUB-65 had a 9.2% reduction. In contrast, for FF, less reduction for both species under drought stress was observed as G. arboreum genotypes had −11.6 and −13.7% (PC-1 and COMMILA, respectively). In the same way, G. hirsutum genotypes, i.e. IUB-65 and IUB-13 had −27.2 and −28.4% reduction under drought, respectively (Table 4).
Table 4. Mean values ± standard errors for fibre quality traits under normal and drought stress condition and reduction (%) under drought
FF, fibre fineness (micronaire); FS, fibre strength (g/tex); SL, staple length (mm).
Note: Mean comparisons lettering was applied across rows among genotypes for each treatment separately, i.e. normal and drought stress.
Discussion
Cotton plant growth mostly depends on environmental factors, changes in weather condition, management, genotype characteristics and planting pattern. Weather is the most critical feature of crop growth and yield in any particular area diverging from small scale to large agro-ecological zone (Manjunatha et al., Reference Manjunatha, Shalephyati, Koppalkar and Pujari2010).
The present study was aimed to compare four genotypes belonging to two different species, i.e. G. arboreum and G. hirsutum for their ability of drought tolerance. These genotypes were evaluated for different morphological, physiological, fibre and root-related trait. Traits such as AGDB, AGFB, CC, LA, SCY, SBP, FS, GOT were higher in G. hirsutum genotypes (IUB-13 and IUB-65) as compared to G. arboreum genotypes (PC-1 and COMMILA). Polyploids have often been found to be associated with plant growth vigour, higher yields and enhanced fitness with an increase in the organ size and improvement in biomass production in several crops (Comai, Reference Comai2005; Corneillie et al., Reference Corneillie, De Stotme, Van Acker and Fangel2019). This increase in IUB-13 and IUB-65 (G. hirsutum genotypes) than in COMMILA and PC-1 (G. arboreum genotypes) might be due to an increase in ploidy level from diploid (AA) (G. arboreum) to tetraploid (AADD) (G. hirsutum). Higher AGDB and AGFB were recorded in G. hirsutum than G. arboreum, but in contrast, the PH trend was opposite. It could be due to the reason that canopy of G. hirsutum is dense with less PH than G. arboreum and consequently higher biomass was attained in G. hirsutum genotypes. The higher MBP but less SBP were recorded in G. arboreum. So, the higher SCY in G. hirsutum might also be due to larger fresh and dry biomass and higher sympodial branches as well.
Despite relatively poor yield and fibre quality, the diploid species G. arboreum has some attractive agronomic traits too (Liu et al., Reference Liu, Guo, Lin, Nie and Zhang2006). For example, it is tolerant of abiotic environmental stresses, including drought and resistance to biotic stresses (Mehetre et al., Reference Mehetre, Aher, Gawande, Patil and Mokate2003; Liu et al., Reference Liu, Guo, Lin, Nie and Zhang2006; Maqbool et al., Reference Maqbool, Abbas, Rao, Irfan, Zahur, Bakhsh, Riazuddin and Husnain2010; Ullah et al., Reference Ullah, Akhtar, Moffett, Mansoor and Briddon2014). Drought tolerance is the ability of a plant to maintain high water status for sustaining better function and the capacity of the plant to sustain metabolic functions at low water status (Ullah et al., Reference Ullah, Rahman, Ashraf and Zafar2008). Plants maintain favourable water balance by morphological or physiological characters that either reduce water loss by transpiration and/or increase water absorption through an extensive and deep root system. The latter has been suggested as an adaptive feature of plants surviving under water stress because it allows plants to divert assimilates and energy, otherwise used for growth, into protective molecules to battle stress (Zhu, Reference Zhu2002). This feature is most likely to be relevant for crops intended for drought-susceptible areas (Chaves et al., Reference Chaves, Maroco and Pereira2003). Less reduction in aboveground morphological features, i.e. ABGD, AGFB, PH, MBP and SBP, was recorded for G. arboreum, than G. hirsutum L under drought condition. Less reduction in above ground canopy traits is an indicator of drought tolerance (Pettigrew, Reference Pettigrew2004). In the same way, less reduction in SCY was also observed in G. arboreum genotypes. CC and LA also showed less reduction in G. arboreum, than G. hirsutum genotypes, under drought condition. Yield and fibre traits such as SCY, GOT FS, SL had also shown less reduction in G. arboreum than G. hirsutum genotypes. In contrast, we identified that root traits PRL, LRN, RFW and RDW were higher in G. arboreum than G. hirsutum genotypes under drought stress. This means that these traits related to drought tolerance are not significantly improved in G. hirsutum genotypes and need to be improved. Under limited water conditions, roots tend to grow deeper into the soil. Consequently, more assimilates will be partitioned from shoot growth to root growth into moist zones (Taiz and Zeiger, Reference Taiz and Zeiger2002). At very low water potentials, when shoot growth is completely inhibited, some roots may still elongate. This is thought to be an adaptive mechanism to ensure survival during times of drought, thus enabling plants to cope with unexplored water resources which are typically found in deeper soil layers (Sawkins et al., Reference Sawkins, Julien and Ribaut2006). G. arboreum genotypes performed better for drought tolerance-related root traits. Efforts to transfer the desirable characters from G. arboreum to G. hirsutum through conventional hybridization have not been encouraging due to abortion of the hybrid embryo/endosperm in the early stages of seed development (Pundir, Reference Pundir1972; Stewart and Hsu, Reference Stewart and Hsu1978). So, the interspecific hybridization between the two species might be attained through ovule culture (Gill and Bajaj, Reference Gill and Bajaj1987). Furthermore, hybridization between these two species can also be done through bridge crossing.
Conflict of interest
The authors have declared that no competing or conflicts of interest exist.
