RETRACTED: Characterization of wheat (Triticum aestivum L.) genotypes unraveled by molecular markers considering heat stress

Abstract The current study focuses and emphasis on the potential of heat stress to negatively affect crop physiology. Here, we have screened 19 wheat (Triticum aestivum L.) genotypes for their tolerance of heat stress. Significant differences were observed among the genotypes for all the traits under consideration. Exploitable extent of genetic variability amongst the entries was present as revealed by considerably higher estimates of mean %. On the basis of Heat susceptibility Index, Halna, Mon’s Ald’s, genotypes Cuo/79/Prulla and K 307 were identified as heat-tolerant whereas SAWSN 3041, SAWSN 3101 and K 0583 were identified as heat-susceptible. The 17 wheat microsatellite markers were capable of detecting 89 alleles with an average of 4.6 alleles per locus. Polymorphism Information Content value ranged from 0.16 for the primer XGWM 516 to 0.83 for DUPW 117 with an average of 0.60. A perusal of similarity coefficients clearly reflected that a very high degree of similarity exists between wheat variety Mon’s Ald’s and SAWSN 3101 (0.70). On the other hand, the two most distantly related cultivars were found to be AKAW 4008 and PBW 343 (0.034). BARC 4, BARC 170, BARC 311, PSP 3058, WHE014.H04 and GWM 458 were strongly associated with the heat tolerance for traits TGW and BARC 311 was strongly associated with terminal heat tolerance for number of grains/plant respectively. Considering all the parameters it is adjudged that relatively stable genotypes may be evaluated at various agro climatic regions for grain yield and heat tolerance along with other contributing characters and ideal plant type.


Introduction
Wheat (Triticum aestivum L.) is one of the most widely grown cereal in the world. Its productivity in the coming decades, like that of most major crops, is threatened by impending climate change, the global mean temperature is predicted to rise by 0.3˚C per decade. Low temperature (<12°C) applied during germination and seedling emergence in wheat resulted in poor and uneven emergence , while spring wheat exposed to high temperature (30-35°C) resulted in a decrease in time to flowering, grain set, and physiological maturity, ultimately reducing grain yield . Currently crop models are evolving to accommodate the nonlinear response to temperature and the step reduction in grain number associated with a short periods (1-2 day) of heat or cold shock. The association mapping approach combines phenotypic with genotypic data to identify genomic locations associated with variation in a given trait. Its ability to resolve marker/ trait association depends upon the extent of linkage disequilibrium present in the association panel (Gupta et al. 2015). Linkage disequilibrium (LD) refers to the nonrandom association of alleles at two or more loci in a general population. More LD in A and B genome is due to recent evolution of D genome about 1-2 million years compared to A and B genome which evolved about 7 million years ago. The association mapping approach has been applied to a range of crop species, facilitating the discovery of many quantitative trait loci. In wheat, for example, loci associated with variation in kernel size and milling quality, grain yield, resistance to disease and various agronomic traits (Bordes et al. 2014) have been located in this way. Several association mapping studies have been conducted to dissect the genetic basis of grain yield in wheat (Kidane et al. 2017). Assessment of nature and magnitude of genetic variation, genetic resource characterization and identifying a marker for the tolerance would be a practicable solution for development of improved cultivars resistant to abiotic stresses and for efficient management and utilization of germplasms in regions where stress occurs commonly. Keeping all above consideration in view, the present investigation was carried out to identify polymorphic SSR markers to characterize the promising lines and to determine the extent of genetic divergence among the lines, followed by a survey of markers for their association to heat tolerance.

Plant material
A total of 19 diverse wheat genotypes from all over the world were selected for the evaluation (Table 1).

Phenotypic Evaluation
The experimental material was sown in control and polytonal condition. The genotypes were grown in randomized block design with two replications. Cultivars were exposed to increased artificial temperature from anthesis onwards in plastic sheet tunnel with semi controlled temperature condition. The chamber was set at 35/30°C and 14/10 h day/night, 50/70% relative humidity and illumination of 335 µmol. For this, two replications were to be evaluated under natural conditions followed by exposure towards enhanced heat by covering it with plastic sheet tunnels, simulating the condition of greenhouse effect. All the recommended cultural and agronomical practices were followed uniformly to raise good crop. Data were recorded for each genotypes towards both the control and stressed environment excluding border effect for eight quantitative characters viz. grain filling duration (GFD), number of effective tillers, spike length (SL), number of spikelets, number of grains per main spike, thousand grain weight (TGW), grain yield (GY) and harvest index (HI) respectively.

Plant DNA Isolation
Genomic DNA from all the 19 wheat genotypes was isolated from 30 days old seedlings following modified CTAB extraction method described by (Doyle and Doyle 1990). A total of 30 markers were selected for the screening of genotypes.

Statistical analysis 2.4.1 Morphological Analysis
Mean performance on plot basis was determined by taking the average performance of the plants over the block of particular genotypes. ANOVA preparation was followed by the mean sum of squares where, genotypic mean square was tested against error mean squares by F-test for (v-1) and (v-1) (r-1) degree of freedom both at 1% or 5% level of significance. Standard error was calculated with the help of error mean square by the formula.
a a n a an Critical difference was calculated to know whether the

R E T R A C T E D
Ethical approval: The conducted research is not related to either human or animal use.

Analysis of variance
Analysis of variance was carried out for the performance of the individual under two environments for all the eight characters studied ( Table 2). The mean square due to genotypes were significant for all the given characters below.

Mean performance of genotypes
The mean performance of 19 genotypes with respect to eight quantitative characters studied in the present investigation under two environmental conditions has been furnished in (Table 3 and 4). Here we have assumed HD1733 as a check variety in compare with all the characters mentioned below.

Variability Range
The estimates of phenotypic coefficient of variation revealed a wide range of variation ranging from 8.17 for GFD to 74.59 for grain yield per plant. The range of genetic coefficient of variation (GCV) varied from 6.63 (GFD) to 71.56 (Grain yield / plant). The characters grain yield per, HI, number of grains per spike, TGW showed comparatively high genetic coefficient of variation, while other characters have shown moderate to low estimates of genetic coefficient of variation.

Heritability
High heritability was recorded for the characters viz., number of grains per spike, TGW and grain yield, whereas most of the characters under study had moderate to low heritability. The heritability percentage ranged from 67.06 (GFD) to 95.10 (TGW) in open and 60.13 (GFD) to 95.91 (TGW) in polytonal condition (Table 5). difference between two treatments was significant at 1% or 5% level. The genotypic variance (Vg or σ 1 g ) and phenotypic variance (Vp or σ 1 p ) were obtained from the analysis of variance. Coefficient of variability as suggested by (Burton and Devane 1951) were used. Heritability was expressed in terms of percentage as defined by (Lush 1949) and was calculated as per given by (Hanson 1956). Genetic advance i.e. expected genetic gain from the selection of 5% superior individuals was estimated and followed by the estimation of genetic advance a percentage of mean. Heat Susceptibility Index was computed by using the methods suggested by (Fisher and Maurer 1978).

Molecular Study
Marker polymorphism was used to measure the formativeness of the marker, the PIC for each marker was calculated according to the formula given by (Anderson et al. 1993). Genetic similarity estimation and diversity analysis and the genetic association among varieties were analysed by calculating the similarity coefficient (Jaccard 1908) for pair-wise comparisons based on the proportions of shared bands produced by primers. The dendogram based on similarity indices was obtained by UPGMA. The nature and extent of diversity between varieties were assessed by identifying the clusters at appropriate phenon levels. All calculations were performed using the NTSYS-pc version 1.10 m software.

Association study
A set of four genotypes, both from heat-tolerant and susceptible groups, categorised on the basis of HSI score for the considered traits above were selected for the association analysis with markers. The marker showing the values of 2 χ probability more than 0.90 were considered as strongly associated, values from 0.80 to 0.90 were considered as moderately associated, from 0.70 to 0.80 weakly associated and below 0.70 were considered as not associated. The association between phenotypic data and molecular data were analysed on the basis of following statistical model given below.
a a n a an For two-fold table (1× r table), a a n a an

Identification of Heat stress tolerant genotype
HSI was conducted by keeping the five characters under consideration (Table 6). HIS for GFD ranged from 0.31 (VL 914) to 1.69 (Pusa Gold). Genotypes C 306, Halna, Iepaca Rabe, Mon's Ald's, VL 914, HD 1888, Cuo/79/prulla, K307, and check HD 1733 showed value of HSI below unity i.e. these genotypes were heat-tolerant and VL 914 which showed minimum HSI was most heat-tolerant genotype for GFD. The remaining 10 genotypes were heat-susceptible and Pusa Gold had maximum value of HSI and thus was most heat-susceptible genotype.
Number of grains per spike showed range of heat susceptibility index from-3.14 (Iepaca Rabe) to 1.18 (C306 and    to be clustered together. Therefore 5 di-genotypic and 2 multi-genotypic clusters were obtained when phenon line was drawn at 50 similarity units. The di-genotyic cluster A consists of variety Pusa Gold and SAWSN 3041, B accommodated variety Halna and Iepaca Rabe while E, F and G consisted variety AKAW 4008 and HD 1733, Cuo/79/ Prulla and K 0583, and K 307 and HD 1733 respectively. When phenon line was drawn keeping 85 similarity units as the cut-off point to discriminate the entries, each of the clusters C and D was further dissociated into 2 subclusters. The cluster C was dissociated into 2 sub-clusters with di-genotypic cluster C-I and mono-genotypic cluster C-II. The mono-genotyic sub-cluster C-II had the variety AKAW 4189-3, while the di-genotypic sub-cluster C-I accommodated Sonalika and Kauz/AA/Kauz. Similarly, cluster D was further dissociated into 2 di-genotyic sub-

Analysis of genetic divergence
Allelic diversity data was used to produce a dendogram in order to elucidate the relationship among the 19 wheat varieties (Figure 1). A perusal of similarity coefficients clearly reflected that a very high degree of similarity exists between wheat varieties Mon's Ald's and SAWSN 3101 (0.70). On the other hand, the two most distantly related cultivars were AKAW 4008 and PBW 343 (0.034). The consensus tree showed that it divided the wheat genotypes into 2 main clusters, the first di-genotypic cluster consists of variety PBW 343 and C 306 and second multi-genotypic cluster included the rest of 17 wheat varieties. The multigenotypic groups were further divided into 7 clusters by drawing phenon line at 50 similarity units and allowing the entries with increasingly similar pattern for markers Two markers, PSR 6511 and GWM 516 were moderately associated for heat tolerance for the character HI at 1 and 1 degree of freedom respectively (probability 0.10-0.10). The alleles 177 and 140 of the markers PSR 6511 and GWM 516 were linked to heat-tolerant group while, alleles 195 and 146 were linked to susceptible group.

Discussion
Analysis of variance indicated significant differences among genotypes for all characters under study in both the environments. Five genotypes (PBW 343, Pusa Gold, Halna, AKAW 4008, K 307) in polytonel condition showed significantly higher harvest index than HD 2733(C). The investigation involved the study of genetic variability through coefficient of phenotypic and genotypic variation as an index for the amount of potential variability present in the individual economic traits. The range of total variation observed in the eight traits exhibited a wide variation for the traits TGW. (Bergale et al. 2001) and (Sahu et al. 2005) also reported wide variation in TGW. Higher value of phenotypic coefficient of variation was obtained in all the cases than the genotypic coefficient of variation for all the characters. Phenotypic coefficient of variation which measures the total variation in different characters were higher for grain filling duration, total number of effective tillers, spike length and grain yield per plant, while the characters number of spikelets per ear head, number of grains per spike, TGW and harvest index showed low value of phenotypic co-efficient of variation. The result obtained in this study is in agreement with the result of (Kumar et al. 2003) who reported high GCV than PCV for number of effective tillers per plant.
High heritability was obtained for number of grains per spike, TGW and grain yield per plant. (Sahu et al. 2005) found high heritability for total number of effective tillers per plant, spike length, number of grain per spike and TGW. The maximum genetic advance was observed for harvest index followed by TGW while the minimum was recorded for total number of effective tillers. GA as per cent of mean was found maximum for grain yield per plant followed by harvest index and TGW while rest of the characters were found to show comparatively low estimates. The present result is in agreement with the result obtained by (Subhashandra et al. 2009) who found high genetic advance for yield per plant and productive tillers per plant. If high heritability is followed by high genetic advance it indicates predominance of additive gene action (Panse 1957). Therefore the selection for clusters. The sub-cluster D-I had variety Mon's Ald's and SAWSN 3101, while sub-cluster D-II consisted variety VL 914 and HD 1888. Clustering pattern was found to be exactly same, when phenon line was drawn at 95 similarity units as the cut-off point.

Association analysis
The test of independence was carried out for determining the relationships between molecular markers and different quantitative traits (Table S8). It showed that BARC 311 was weakly associated at 1 degree of freedom (Probability 0.10-0.30) for the heat tolerance of the character GFD. The allele 116 of this marker was linked to heat-tolerant and allele 194 was linked to heat-susceptible group. The marker BARC 311 was strongly associated for the heat tolerance of the character number of grains/plant at 1 degree of freedom (probability 0.01-0.05). The allele 194 of this marker was linked to tolerant group and 116 were linked to susceptible group. However, SSR marker GWM 516 was weakly associated at 1 degree of freedom (Probability 0.10-0.30). The allele size 146 of this marker was linked to both heat tolerance and susceptible group at 75 and 100% probability respectively. Four markers, BARC 311, PSR 6469, PSP 3058 and WHE014.H04 were strongly associated with the heat tolerance of the character TGW (probability 0.01-0.05) at 1 degree of freedom. The alleles 194,177,196 and 193 of BARC 311, PSR 6469, PSP 3058 and WHE014.H04 were linked to heat-tolerant group while alleles 116, 190, 195, and 181 were linked to susceptible group of TGW. Markers BARC 170, BARC 4 and GWM 458 were also strongly associated at 3, 1 and 1 degree of freedom (probability 0.01-0.05, 0.01-0.01, 0.05-0.10) respectively. The alleles 187, 104 and 113 were linked to tolerant group and allele 190, 111 and 131 were linked to susceptible groups of the character TGW. The marker TAE hwm004.H07 was weakly associated at 1 degree of freedom (probability 0.10-0.30). The allele 167 of this marker was 50% linked to tolerant group and 100% linked to susceptible group of TGW. The SSR marker PSP 3103 was moderately associated at 3 degree of freedom (probability 0.10-0.10) for the heat-tolerance of the character grain yield /plant. The allele 167 of this marker was linked to tolerant group and allele 183 was linked to susceptible group. The microsatellite markers BARC 110 and DUPW 138 were weakly associated for the heat tolerance of the character grain yield /plant at 3 and 1 degree of freedom(Probability 0.10-0.30), respectively. The alleles size 106 and 100 of BARC 110 and alleles 110 and 116 of DUPW 138 were linked to heat-tolerant and susceptible group of the character grain yield /plant. H04, BARC 170, BARC 4 and GWM 458 were strongly associated for heat tolerance of the character 1000 grain weight, while the marker TAE hwm004.H07 was weakly associated with the character 1000 grain weight. The marker PSP 3103 was moderately associated while BARC 110 and DUPW 238 were weakly associated with the terminal heat tolerance of the character grain yield /plant. Two markers PSR 6512 and GWM 526 were moderately associated for heat tolerance for the character harvest index. (Yang et al. 2002) identified two markers linked to grain filling duration of heat tolerance in hexaploid wheat. Recently, bulked segregant analysis to detect QTL related to heat tolerance in rice (Oryza sativa L.) using SSR markers have been reported (Zhang et al. 2009). Three SSR markers (Xgwm132, Xgwm577 and Xgwm617) were linked to GFR as an indicator for heat tolerance in wheat by quantitative trait loci analysis of the F 2 population.

Conflict of interest:
Authors declare no conflict of interest.
those mentioned characters would be effective owing to considerable presence of additive gene action.
The seventeen primer pairs exhibited different levels of polymorphism amongst the nineteen wheat varieties under evaluation in the present investigation. Many studies have also reported remarkable differences in allelic diversity among various microsatellite loci (Ram et al. 2007). The alleles revealed by markers showed a higher degree of polymorphism. The highest number of alleles per locus was observed in the B genome while in the D genome, lower polymorphic loci were obtained. Similar observation for higher polymorphism level of genome B was also reported by (Wang et al. 2007). Further analysis for genetic diversity among A, B, and D genomes in common wheat showed that the lowest genetic variation in the D genome was a major reason leading to fragile genetic basis of modern cultivated wheats (Zhang et al. 2002).
The highest PIC value of 0.83 was recorded for the primer DUPW 217, which were recorded in the B genome. (Roussel et al. 2004) reported highest PIC value (0.686) in the B genome as compared to A and D genomes. Occurrence of null alleles was also noticed in various varieties for a particular locus. For 16 loci null alleles were detected. Null alleles were also detected by (Teklu et al. 2007). Larger the repeat number involved in the SSR locus, the larger was the number of identified alleles. The observed levels of polymorphism are positively correlated with the length of the microsatellite repeats. (Huang et al. 2002) observed an increase in number of allele with the repeat number of the microsatellites used and their relative distance from the centromere, and they were not dependent on the motif of microsatellites.
The first di-genotypic cluster consists of variety PBW 343 and C 306 and second multi-genotypic cluster included the rest of seventeen wheat varieties. A perusal of similarity coefficients clearly reflected that a very high degree of similarity exists between wheat varieties Mon's Ald's and SAWSN 3101 (0.70). On the other hand, the two most distantly related cultivars were AKAW 4008 and PBW 343. Similar studies were conducted by different investigators using SSR markers (Chakravarthi et al. 2006). It was assumed that such a high level of genetic similarity may be the result of biased selection of the material in the previous breeding programs, which ultimately narrowed the genetic base of the wheat germplasm.
The result of our study clearly indicated that utilization of seventeen SSR markers were sufficient for discrimination and unambiguous identification of all the nineteen wheat varieties included in the analysis. Seven markers, namely BARC 321, PSR 6469, PSP 3058 WHE024.