Fine-mapping and characterisation of genes on barley(Hordeum vulgare)chromosome 2H for salinity stress tolerance during germination

Edward Mwando,Yong Han,Tefera Angessa,Xiao-Qi Zhang,Chengdao Li,*

a Western Crop Genetics Alliance,College of Science,Health,Engineering and Education,Murdoch University,Murdoch,WA 6150,Australia

b Western Australian State Agricultural Biotechnology Centre,Murdoch University,Murdoch,WA 6150,Australia

c Department of Primary Industries and Regional Development,3 Baron-Hay Court,South Perth,WA 6151,Australia

Keywords:Barley Germination Salinity tolerance Diagnostic markers Receptor-like protein kinase 4 Gene expression Sequence analysis

ABSTRACT Salinity causes a detrimental impact on plant growth,particularly when the stress occurs during germination and early development stages.Barley is one of the most salt-tolerant crops;previously we mapped two quantitative trait loci (QTL) for salinity tolerance during germination on the short arm of chromosome 2H using a CM72/Gairdner doubled haploid (DH) population.Here,we narrowed down the major QTL to a region of 0.341 or 0.439 Mb containing 9 or 24 candidate genes belonging to 6 or 20 functional gene families according to barley reference genomes v1 and v3 respectively,using two DH populations of CM72/Gairdner and Skiff/CM72,F2 and F3 generations of CM72/Gairdner/*Spartacus CL.Two Receptorlike kinase 4(RLPK4)v1 or Receptor-like kinase(RLK)v3 could be the candidates for enhanced germination under salinity stress because of their upregulated expression in salt-tolerant variety CM72.Besides,several insertion/deletion polymorphisms were identified within the 3rd exon of the genes between CM72 and Gairdner.The sequence variations resulted in shifted functional protein domains,which may be associated with differences in salinity tolerance.Two molecular markers were designed for selecting the locus with receptor-like protein kinase 4,and one was inside HORVU2Hr1G111760.1 or HORVU.MOREX.r3.2HG0202810.1.The diagnostic markers will allow for pyramiding of 2H locus in barley varieties and facilitate genetic improvement for saline soils.Further,validation of the genes to elucidate the mechanisms involved in enhancing salinity tolerance at germination and designing RLPK4 specific markers is proposed.For this publication,all the analysis was based on barley reference genome of 2017(v1),and it was used throughout for consistence.However,the positions of the markers and genes identified were updated according to new genome (v3) for reference.

1.Introduction

Salinity is a worldwide problem affecting more than 6% of the global land area,which is more than 20% of the total arable and 50% of cropland comprising 33% of irrigated area [1].Saline land is increasing at a disturbing level due to human activities,climate change and seasonal variations [2].Salinity is likely to exacerbate the challenge of meeting the world’s food demands as most cultivated crops are susceptible to salinity stress[3].In Australia,salinity is a serious abiotic stress that is threatening crop production,especially in semi-arid zones that receive less than 450 mm of annual rainfall [4].According to earlier studies by Agarwal [5],Tabatabaei and Ehsanzadeh [6],salinity reduces crop yields by (i)impeding plant access to soil water through increased soil osmotic potential that curbs water and nutrient absorption (osmotic or water-deficit effect),and (ii) causing ionic disproportion and toxicity in plants (salt-specific or ion-excess effect).Crops are more susceptible to salinity during germination and early developmental stages than other growth phases [5,7].

At present,there are no practical solutions for managing salinity on agricultural land [7].However,development of varieties that can withstand salinity stress is a long-term option that requires knowledge of the physiological mechanisms and genetic elements contributing to traits at different plant growth stages [8].Studies have identified various adaptive mechanisms in different plant parts in response to saline conditions or salinity stress that can be summarized into three:osmotic stress tolerance,Na+or Clexclusion,and tissue tolerance to accumulated Na+or Cl-[9].These mechanisms involve multifaceted physiological traits,metabolic alleyways,hormonal pathways,transcriptional responses,gene systems and growth stages working either simultaneously or in seclusion,depending upon the stress period and intensity to enable plants survive salinity stress [10].

Barley(Hordeum vulgare L.)is the fourth most important cereal crop in production and consumption globally [11].It grows in a wide range of environmental conditions,including extreme latitude and altitude,and is one of the hardiest cereal crops[12].Barley is a crop of interest because it can survive saline conditions by accumulating high Na+in the leaves if the ions find their way into the plant [14].The salt-tolerant barley varieties,Numar and ZUG293,exhibited halophytic features to exclude Na+from root uptake,relative to the salt-sensitive varieties,Gairdner and ZUG403 [13].Its capacity to tolerate salinity is attributed to the sequestration of Na+in its intracellular vacuoles and synthesis of compatible solutes in the cytoplasm that balance the osmotic potential of vacuolar Na+[15].As a result,barley is used by breeders as a source of favorable alleles to improve other cereal crops either by conventional or molecular approaches [16].

Similar to other plant species,salinity tolerance in barley varies among genotypes [17] and is controlled by multiple genes that express themselves at various growth stages [2].Salinesusceptible barley varieties will express poor germination,slow growth and development,low survival,and reduced yield and grain quality[7,17,18].Screening barley genotypes for salinity tolerance at various development stages has been continuous.Several quantitative trait loci (QTL) associated with salt tolerance traits in barley have been mapped,including those related to yield and agronomic factors[17],germination and seedling[7,18],plant survival [19],shoot Na+content or Na/K ratio and salt exclusion [20].

The degree of damage depends on the variety,salinity concentration and developmental stage of the crop [21].Germination is the most critical phase in the crop life cycle but the most sensitive to salinity [22] because it determines plant vigour and population which ultimately affect yield [23].Having varieties that can acclimatize to salinity at this stage is essential in salinity-prone cropping regions such as the drier areas of Western Australia that experience hot and dry summers and increased salinity levels just before sowing in autumn [23].Also,germination and initial seedling growth occur in the topsoil where there is high salt accretion due to evapotranspiration[24].The development of improved cultivars at this stage requires an understanding of the gene actions governing salinity tolerance[25].While the physiological and biochemical basis of salinity in crops(barley)are well known,further studies on genetic factors at different growth phases,especially germination,are paramount.QTL for salt tolerance at the germination stage have been screened and mapped [26].The most recent QTNs was reported by Mwando et al.[7],who mapped 19 QTNs on all chromosomes and predicted four genes using 350 diverse barley accessions and 24,000 genetic markers,but there is limited information on the nature and function of the tolerance genes at this stage.

Our previous study mapped two stable quantitative loci on chromosome 2H for salinity tolerance at germination using a doubled haploid(DH)population of CM72/Gairdner[18].Here,we validate the reported QTL and fine-mapped it to a region of 0.341 or 0.439 Mb according to barley reference genomes v1 or v3 respectively,and predicted two receptor-like protein kinase as candidate genes through expression and sequence analysis and designed two molecular markers for selecting the locus.

2.Materials and methods

2.1.Plant materials

The plant materials used in this study included two barley DH populations (CM72/Gairdner and Skiff/CM72),their parents(CM72,Gairdner,and Skiff),a salt-sensitive cultivar Spartacus CL,and F2and F3generations of CM72/Gairdner/*Spartacus CL and 265 global barley accessions.CM72/Gairdner comprised 102 DH lines,that had been previously used to map salinity tolerance QTL on chromosome 2H at germination[18],and Skiff/CM72 comprised 88 lines.Gairdner with Tas83-587/Onslow pedigree,Skiff with Abed Deba/WI-2235//CD28/WI-2231 pedigree and Spartacus CL with Scope/4*Hindmarsh//HMVB0325-106 pedigree are salinity sensitive Australian varieties.While CM72 is salinity tolerant genotype derived from California Mariout*4/CI1179 (Algerian)//2*California Mariout/Club Mariout/3/CM67.

To develop an F2population of CM72/Gairdner/*Spartacus CL segregating for salinity tolerance at germination,the CM72/Gairdner salt-tolerant DH lines,previously selected for CM72 genotype on chromosome 2H [18],were crossed with Spartacus CL and the F1plants allowed to self-pollinate.The process involved crossing DH lines (WADH13531 and WADH13543) and screening them for salinity tolerance using marker bPb-3858 [18].Lines that had the CM72 genotype on the 2H locus were selected for crossing with Spartacus,which yielded 32 F1seeds.The progeny were confirmed to be heterozygous before crossing them back to Spartacus CL and allowing them to self pollinate.A total of 2020 F2seeds were harvested and cut in half while observing polarity(the embryo and the endosperm sides).One half was used for genotyping with InDel markers from the 2H fine-mapped region and the other half containing the embryo was grown to produce the F3population for phenotyping salinity tolerance at germination.

2.2.Assessment of salinity tolerance during germination

Three replicates of 100 surface-sterilized seeds from the two DH populations were spread on a double layer of Whatman No.1 filter paper placed in a 90 mm Petri dish.Each Petri dish received 4 mm of either (DI) water (control) or 150 mmol L-1NaCl (salt treatment) and was then covered with a lid and labelled.Five Petri dishes were bundled together using cling wrap and placed in a dark oven at 20 °C.After 72 h incubation,the germinated seeds were counted.The tolerance index was calculated as the percentage of seeds germinated in the salt treatment divided by those in deionized water multiplied by 100.While the reduction in germination percent was calculated as the percentage of seed that had germinated in DI water subtract from those under 150 mmol L-1NaCl.

2.2.1.Phenotypic data analysis

The germination percentage of individual lines from the three replicates was used for analysis of variance with SAS version 9.4 [27].Spearman’s rank correlation coefficient analysis was undertaken with IBM SPSS Statistics for Windows,Version 25.0 [28].

2.3.DNA extraction

Genomic DNA was extracted from individual barley seeds of each DH line including the parents using the modified cetyltrimethylammonium bromide (CTAB) method as described by [29].The CTAB method isolates protein from the tissue,separates it from DNA with a salt solution,and purifies the DNA using ethanol.DNA was dissolved in 150 μL TE buffer,and its concentration measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific Inc.,Waltham,MA,USA).

2.4.Genotyping,QTL validation and fine mapping

Insertion-deletion (InDels) markers were designed from an inhouse barley genomic database (http://146.118.64.11/BarleyVar/),between and slightly outside the two flanking markers bPb-3858 and bPb-1103 of the two mapped QTL on chromosome 2H[18,30].Since none of the barley parents of the DH populations was among the database varieties,InDels with polymorphism in more than half of the barley genotypes,with insertion or deletion greater than 10 nucleobases and an amplifying sequence of 100–200 bp amplicon size flanking equally was selected.The primers were blasted on the IPK website http://webblast.ipkgatersleben.de/barley_ibsc/ to ensure that they amplified a single region on the barley genome then tested for polymorphism among the DH parents (CM72,Gairdner and Skiff).The polymorphic InDels were selected and run in the two DH populations(Table S1).A polymer chain reaction (PCR) of 10 μL was used to amplify the template DNA using a Thermal Cycler in an integrated systems of Applied Biosystems (Thermo Fisher Scientific Inc.Waltham,MA,USA) machine with the settings adjusted to an initialization at 95 °C for 3 min,35 cycles of denaturation at 94 °C for 30 s,annealing at 60 °C for 30 s,extension/elongation at 72 °C for 30 s,final elongation at 72 °C for 5 min,and the final hold at 14 °C.The amplification products were visualized on either 2% agarose gel or 6% polyacrylamide gel electrophoresis(PAGE).

The previously mapped diversity array technology (DArT)flanking markers bPb-3858 and bPb-1103 [18] were the benchmarks for placing the InDel markers on chromosome 2H.The new InDels were combined with 315 DArT and 84 simple sequence repeat (SSR) markers (used in initial QTL analysis) to construct a new map using the CM72/Gairdner DH population.JoinMap 3.0 [31] was used for linkage analysis and determination of the likely order of the markers in the DH population and a LOD score threshold of 3.0 at P ＜0.05 used to declare a QTL linked with germination percentage.Recombinant lines from CM72/Gairdner and Skiff/CM72 DH populations were selected between new flanking markers and re-evaluated for response to 150 mmol L-1and 225 mmol L-1NaCl at germination stage.The tolerance index for each recombinant line was related to the polymorphic InDel marker genotype of their respective DH population.All the markers within the new flanking QTL region were used to screen a larger F2population with 2020 individuals derived from M72/Gairdner DH/* Spartacus CL.The resulting recombinant lines were bulked into individual F3populations and tested for germination with 150 mmol L-1and 225 mmol L-1NaCl levels.

2.5.RNA extraction and genes expression profile

Seeds were germinated using DI water and embryos collected at 24 h and 48 h from AC Metcalfe,Morex,Harrington,Stirling and Bass varieties.Germination stage salinity tolerance was previously determined for these five varieties (Table S2).The embryos were snap frozen immediately using liquid nitrogen and stored at-80°C before RNA extraction and gene expression analysis was conducted following [32].

2.6.Molecular marker identification for salinity tolerance loci on chromosome 2H during germination

All the markers within the new flanking region were considered to have potential for use in marker-assisted selection (MAS).To validate them,individual recombinant lines were tested using each marker and their genotype associated with their respective phenotype.Selected markers were tested in a wider collection of 265 barley germplasm from around the world including Australian commercial varieties.The salinity tolerance of the diverse barley panel was previously determined as part of the germplasms used in [7].Hierarchical clustering of phenotype to genotype was done with IBM SPSS Statistics for Windows,Version 25.0[28],after data transformation.

2.7.Estimation of gene expressions under salt stress by real-time quantitative PCR

Seeds of CM72 and Gairdner were used for this experiment.First,the seeds were cut into half to reduce starch content and then germinated in 150 mmol L-1NaCl and DI water separately after sterilization.Embryos were collected by separating them from the endosperm at 4 time points (TP) where TP1 was 16 h,TP2 40 h,TP3 64 h and TP4 88 h after germination.The embryos were snap-frozen using liquid nitrogen and stored at -80 °C.Tissue samples were crushed to a smooth powder in frozen liquid nitrogen by a mortar and pestle.Total RNAs was isolated using TRIsure reagent (Bioline,Eveleigh,New South Wales,Australia).RNA was extracted with chloroform,precipitated by isopropyl alcohol before the pellet was washed in 75%ethanol,air-dried and dissolved in DEPC-treated water.Eventually total RNA was purified using RNeasy Plant Mini kit (Qiagen,Chadstone,Victoria,Australia) by being treated with DNase I to digest DNA and chromatins.DNA free RNA was recovered by precipitating using sodium acetate in 100% ethanol washed with 75% ethanol and dissolved in DEPC-treated water.The integrity of RNAs was determined using formaldehyde-agarose gel electrophoresis and the concentration measured using UV/VIS spectrophotometry (NanoDrop 1000,Thermo Fisher Scientific Inc.,Waltham,MA,USA).

The first-strand cDNA was synthesized through reverse transcription using SensiFAST cDNA Synthesis Kit following manufactures procedure (Bioline Australia).Quantitative polymerase chain reaction (qPCR) was performed using gene-specific primers (Table S3),in the reaction system of SensiFAST SYBR L0-ROX Kit on a Roche 480 real-time PCR machine.HvGAPDH was employed as our inside reference control [33].Real-time quantitative polymerase chain reaction (RT-qPCR) was done in three biological and technical repetitions.The comparative quantification 2-ΔΔCT method was used to estimate the relative expression levels and evaluate quantitative variation of genes following the equations (1) to (5) below as modified from Livak and Schmittgen [34],where embryos harvested at specific time point was related to the respective time point from the same variety between target gene and housekeeping HvGAPDH for salinity treated and control separately (fold change) as exemplified in equations i and ii.The fold change in treated were correlated to control of individuals and individuals to averages(comparative fold change) as shown in equations iii and iv and gene expression (relative fold change) was considered as given in equation v.The expression levels and quantitative variation of genes data were calculated from independent replications and subjected to the analysis of variance (ANOVA) done to determine the significant differences between the replicates before testing the means by Duncan’s Multiple Range tests(DMRT) at P ＜0.05.SPSS statistical computer software program[28] was used to plot the graphs.

2.8.Amplification and full-length sequencing of genes from genomic DNA of barley

Genomic DNA extraction that followed the method by[29]was used to extract DNA from both CM72 and Gairdner seedlings and subjected to polymerase chain reaction (PCR) as described earlier but using gene-specific markers (Table S4) in a volume of 25 μL using 1 PCR buffer.The primers were designed from gene sequences obtained from the NCBI databases that amplified sections of each gene into several fragments.Purified fragments for sequencing reaction were obtained from gel fragment by Clean Up,where excised gel was centrifuged for 30–60 s at 14,000 r min-1.The sequencing reaction contained 4 μL dGTP:BDV3.1 in a ratio of 1:3,3 uL template and 3 uL of forward primer in a 10 uL reaction volume.The sequencing reaction was processed under the following conditions:96 °C for 2 min,35 cycles of (96 °C for 10 s,50 °C for 5 s,60 °C for 4 min) and 14 °C hold.This was followed by ethanol precipitation where 1 μL of 125 mmol L-1EDTA(disodium salt),1 μL of 3 mol L-1sodium acetate and 25 μL of 100%ethanol were added into each tube succeeding the order after spinning down.This was mixed by pipetting and incubated at room temperature for 20 min,then centrifuged using maximum speed of 14,000 r min-1at 4 °C for 30 min.The supernatants were removed by inverting the tubes on filter paper and centrifuging at low speed of 200 r min-1for 1 min.The DNA pellets were rinsed by adding 125 μL of 75%ethanol into each tube and centrifuging at 14,000 r min-1for 5 min.The cleaning solution was removed for the pellet to dry at room temperature.Finally,the samples were sequenced in Australian State Agricultural Biotechnology Centre(SABC),Murdoch University.Computer software MEGA-X [35]and Geneious 6.0 [36] were used to align and analyse the sequences,while the online tool Multalin version 5.4.1 was used to generate the images [37].

2.9.Candidate genes analysis

Predicted protein sequences of the two RLPK4′s were extracted from barlex (https://apex.ipk-gatersleben.de/apex/f?p=284:10)[38] and blasted on NCBI website (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE=Proteins).A representative sequence from each of the different species including wheat,rice,maize and sorghum with a percentage resemblance greater than 70% were selected and extracted before being aligned at https://blast.ncbi.nlm.nih.-gov/Blast.cgi?PAGE=Proteins.The positions of the conserved domains in the middle of both genes were done using the respective protein sequences (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi).

3.Results

3.1.Germplasm response to salinity stress at germination

Germination of the three parental lines (tolerant CM72,and sensitive Skiff and Gairdner) in 150 mmol L-1NaCl was 96.7%,78.0%,and 68.7%,respectively(Figs.S1,2).The two DH populations differed significantly for germplasm,treatment and their interaction (Table S5).Both DH populations had an average germination of the parental lines in DI water of 97.58%,while in 150 mmol L-1NaCl it was 82.72% for CM72/Gairdner and 82.02% for Skiff/CM72 population,respectively.Germination ranged from 46.7%to 99.3% in the CM72/Gairdner population and 42.7%–96.0% in the Skiff/CM72 population exposed to salinity stress.

The tolerance index of CM72 (96.7%) indicated that it was less affected by 150 mmol L-1NaCl at germination followed by Skiff(78.0%).However,Gairdner was the most affected variety at 68.67%.Across the two DH populations,a transgressive tolerance index was exhibited by individual lines where some lines were below and above the parents (Fig.S3).There was a negative relationship between tolerance index and germination reduction due to salinity stress with a correlation coefficient of r=-0.97(Fig.1A) for CM72/Gairdner and -0.95 for Skiff/CM72 (Fig.1B)indicating that either tolerance index or reduction in germination can explain the tolerance capacity of each DH line.

3.2.Validation and refining genomic interval of chromosome 2H QTL

Among the 200 tested InDel markers in the two DH populations,41 in CM72/Gairdner and 12 in Skiff/CM72 were polymorphic(Table S1).InDel markers that were polymorphic in either of the two DH populations were evaluated in their respective DH populations and a new genetic map was constructed using the CM72/Gairdner population.A total of 82 markers were scanned on chromosome 2H that included 42 new InDel,33 DArT,and 7 SSR markers.The new linkage analysis confirmed the two QTL previously mapped on 2H for germination percentage at 150 mmol L-1and 300 mmol L-1NaCl associated with markers bPb-3858,bPb-1103[18] and InDel 74–48 (Table 1).Thirteen recombinant lines from CM72/Gairdner and 16 from Skiff/CM72 were selected for their clear-cut phenotypic responses and marker information.

Table 1 Chromosome 2H QTL linked to salinity tolerance during germination.

Table 2 Gene annotation on v1 and v3 for region within chromosome 2H QTL for salinity tolerance at germination.

Fig.1.Correlation coefficient of tolerance index and germination reduction due to 150 mmol L-1 NaCl in (A) CM72/Gairdner and (B) Skiff/CM72 DH populations.

3.3.Physical mapping and comparative analysis on 2H QTL interval

Flanking markers InDel 74–47 and InDel 20–44 were used as the starting point to narrow the region likely to contain the tolerant genes using the new recombinant lines.The two markers were chosen for their polymorphism in both DH populations.The region had 21 polymorphic markers in both DH populations with 16 in CM72/Gairdner,two in Skiff/CM72 and three overlapping.The genotype data indicated that six lines from CM72/Gairdner[WADH13529,WADH13534,WADH13538 (sensitive) and WADH13536,WADH13537,and WADH13561(tolerant)]and three lines from Skiff/CM72 [WADH13772 and WADH13820 (sensitive)and WADH13806 (tolerant)] were recombinant within the region(Fig.S4).

Among all the markers,InDel 15–008 was considered the left boundary for the gene region and InDel 74–63 was best suited for the right boundary because the same lines above showed similar genotype-phenotype combinations (Fig.S4).Three lines WADH13529,WADH13534,and WADH13538 of CM72/Gairdner DH population had Gairdner genotype for InDel 15–008 and were salt-sensitive,while three other lines WADH13536,WADH13537,and WADH13561 had CM72 genotype and were salinity tolerant.The genotype of these six recombinant lines in between the left and right boundary markers region matched with their phenotype response to salinity stress.Therefore,the candidate gene(s) were likely located close to and/ or in between InDel 13–010 and InDel 15–013 on chromosome 2H.While there were fewer polymorphic markers(5)and lines(3),the Skiff/CM72 DH population indicated a likelihood of candidate gene(s) around the region.Lines WADH13772 and WADH13820 had Skiff genotype for InDel 74–56,which were sensitive,while line WADH13806 had CM72 genotype and it was tolerant (Fig.S4).

Four recombinant lines with similar allele were identified after screening 2020 F2lines of CM72/Gairdner/*Spartacus CL using 4 markers(InDel 15–008,15–009,74–61,and 74–63).The genotype of the lines was 2 markers for CM72 type (InDel 15–008 and15–009)and 2(InDel 74–61 and 74–63)for Spartacus CL.Germination stage tolerance index of the F3progeny lines derived from these four lines was similar to that of Spartacus CL (sensitive) in 150 mmol L-1and 225 mmol L-1NaCl treatment.The genotype of the lines for markers InDel 15–008 and InDel 15–009 was the opposite of phenotypic expression(Fig.S4)indicating that the gene responsible for increased germination under salinity stress was likely to be located between markers InDel 15–008 and InDel 15–009 (723886744 and 724,170,810 v1 or 630,120,437 and 630,396,991 v3) and InDel 15–013 (724,511,661 v1 or 630,559,551 v3).

3.4.Gene annotation and identification

Overall,according to reference genome of 2017 v1 thirteen candidate genes were identified between markers InDel 15–008 and InDel 15–013 (Table 2),the function of which has been found to belong to eight different families and two undescribed proteins[30].Of the thirteen genes in the region,four belonged to the receptor-like protein kinase 4 family with three in a repeat sequence.The salinity tolerance gene co-segregated with the two InDel markers 74–61 and 74–63 (Fig.2),because CM72 allele of the markers was associated with tolerance while Gairdner with sensitivity.InDel 74–61 is within receptor-like kinase 4 (HORVU2Hr1G111760.1),indicating the likelihood of the gene to be enhancing salinity tolerance during germination.When the InDel markers (15–008 and 15–013) were aligned to the new reference genome v3,where twenty-four candidate genes belonging to twenty functional families were identified as in Table 2[75].Three receptor-like kinase were still recorded,and three of which were following each other in a series.Again,InDel 74–61 was within receptor-like kinase (HORVU.MOREX.r3.2HG0202810.1) which is the same gene as receptor-like kinase 4 (HORVU2Hr1G111760.1)v1.On that note therefore,further analysis in the following parts of this paper was based on barley reference genome v1 [30].

3.4.1.Candidate gene profiles and potential expression prediction

To find out if the genes are possible candidates,their level of expression was explored in tolerant (AC Metcalfe and Bass) and sensitive(Morex,Harrington and Stirling)varieties(Table S6)during the early stages of germination at two time points(24 h and 48 h).The expression levels of HORVU2Hr1G111840.5 (Glutathione-S-transferase (GST) family protein) was generally high across all the varieties with Bass recording the highest expression (1046 RPKM) at 48 h followed by HORVU2Hr1G111780.3 receptor-like protein kinase 4–2 (RLPK4-2) (73 RPKM) at 48 h in Stirling.HORVU2Hr1G111760.1,a receptor-like protein kinase 4–1 (RLPK4-1)was consistently expressed in all varieties at 24 h and 48 h,but at moderately low levels than GST family protein) and RLPK4-2with its best expression of 36 RPKM in AC Metcalfe at 48 h.Expression of three genes,namely HORVU2Hr1G111750 disease resistance protein(DRP),HORVU2Hr1G111790.7 receptor-like protein kinase 4(RLPK4),and HORVU2Hr1G111880.1 endoglucanase 3(EDG3) recorded very low or no expression level at all,in some of its time points (Fig.S4;Table S6).

A search for each of the gene was undertaken in barley genome explorer (https://apex.ipk-gatersleben.de/apex/f?p=284:10) [38]to identify the expression profile at different growth stages in fragments per kilobase million (FPKM) (Table S7).Three genes (HORVU2Hr1G111760.1,HORVU2Hr1G111780.3,and HORVU2Hr1G111840.5) expressed themselves at early growth stages and are most likely to be candidate genes for salinity tolerance at germination (Fig.S6).This finding paved the way for identifying candidate genes for salinity tolerance in barley on the chromosome 2H locus mapped by [18].

Fig.2.Fine-mapping results and gene marker alignment on chromosome 2H.(A)QTL of DArT and SSR markers initially detected as flanking between bPb-3858 and bPb-1103.(B)Genetic map created using InDel,DArT and SSR markers within the QTL region.(C) Fine-mapped 2H QTL region flanking between markers InDel 15–008 and 74–63.(D)Genes close to(red)and between(blue)InDel markers 74–61 and 74–63.CFP,cyclin family protein;DRP,disease resistance protein;RLPK4-1,receptor-like protein kinase 4;RLPK4-2,receptor-like protein kinase 4;RLPK4-3,receptor-like protein kinase 4;GST,glutathione-S-transferase family protein;EDG3,endoglucanase 3;UP,undescribed protein;TEDCP3,transmembrane emp24 domain-containing protein 3.

3.5.Identification of potential molecular markers for marker-assisted selection

Twenty-one InDel markers within the region delimited by InDel 74–47 (723,218,508) and InDel 20–44 (726,080,570) that is 2.9 Mb (2,862,062) were considered with the potential for markerassisted selection (MAS) of salinity tolerance at germination in barley (Figs.2,S4).The 19 markers identified within the salinity tolerance region of 2H in the CM72/Gairdner DH population were categorized into seven groups that explained tolerance index variation of 9.06%–11.12%(Table S8).The InDel markers in salinity tolerance region of 2H in the six recombinant lines of CM72/Gairdner DH population selected for their association between markers and phenotypic expression in response to salinity stress had similar grouping compared with the whole lines but explaining wider range of phenotypic variations of 1.84%–19.61%(Table S8).

In Skiff/CM72 DH population,the five markers associated with salinity tolerance in the region were divided into three groups explaining phenotypic variation of 10.63% to 14.52%.The three recombinant lines selected from this population within the region,belonging to three different groups of markers explaining phenotypic variation of 6.65% to 20.03% (Table S8).The Skiff/CM72 DH population recombinant lines had a genotype-phenotype similarity of 66.7%for 4 markers and 100%for one(Fig.S4).Only markers that were showing 100% genotype-phenotype consistencies in their respective recombinant populations and explaining the highest phenotypic variations were selected for further analysis.Therefore,markers InDels 15–008,15–009,74–61,and 74–63 in the CM72/Gairdner DH population explaining 19.62% and InDel 74–56 explaining 20.03%of the phenotypic variations in the recombinant lines and 9.89%in the whole population were considered to be possible diagnostic markers (Table S9).

The fine-mapping results from the CM72/Gairdner DH population located InDel 74–56 outside the boundary region likely to have tolerant genes.Therefore,only InDels 15–008,15–009,74–61 and 74–63 were considered for further evaluation.These four InDel markers were genotyped in a panel of 265 barley germplasm from across the globe including Australian commercial varieties.Markers InDel 74–61 and 74–63 were consistent with each other for both the A (CM72) and B (Gairdner) allele while,InDel 15–009 was consistent but uniformly opposite and InDel 15–008 was neither stable nor consistent hence not considered for further analysis.Genotyping of InDels 15–008,15–009,74–61 and 74–63 on the F2generation of CM72/Gairdner DH/*Spartacus CL resulted in four recombinant lines whose phenotypic expression indicated that the two markers (InDel 74–61 and 74–63) were the closest to the gene(s)of salinity tolerance at germination on the 2H locus.Therefore,two markers namely,InDels 74–61 and 74–63 explaining phenotypic variation of 3.04% and 2.97% respectively,in the 265 diverse barley germplasm were considered candidate Diagnostic markers on chromosome 2H containing gene(s)for salinity tolerance at germination(Fig.S7).Cluster analysis of these two InDel markers and the tolerance index of the 265 diverse barley germplasm formed two major clusters,CM7 (C-type) and Gairdner (Gtype) groups with average tolerance indices of 81.38% and 78.02%,respectively.The G-type had higher number of genotypes(210) than the C-type (55).Furthermore,the G-type was distinguished into two minor groups of 201 and 9 entries,with respective average salinity tolerance indices of 77.35% and 78.69% each.Tolerance index of the whole population of 265 accessions was 79.01% which was above the G-group but below C-group(Fig.S8),and a standard deviation of ±1.23 was calculated.Therefore,the average tolerance index of C-type was significantly different from G-type and the whole population.Further the distribution of G and C-type groups were analysed in the 265 diverse barley germplasm based on their origin and head type.Germplasms from Asia had 23% &77%,Australian 26% &74%,Europe 19% &81%,North America 20% &80%,Middle East 21% &79%,South America 18%&82%and Africa 27%&73%of C-type and G-type respectively.On the other hand,six-row barley germplasms had 30%of C-type&70% of G-type while two-row head type with 25% &75% in their respective order.

3.6.Expressions of 4 genes in CM72 and Gairdner under salinity stress by real-time qPCR

Seven genes from the fine mapped region had a perceptible expression during germination under deionised (DI) water (control)in the two varieties CM72 and Gairdner.Under DI water generally,the expression levels of the genes were higher in Gairdner than in CM72.Disease resistance protein (DRP) (HORVU2Hr1G111750.5)recorded the highest fold changes while Cyclin family protein (CFP) (HORVU2Hr1G111740.1) had the lowest in Gairdner.(Fig.S9).In the 150 mmol L-1NaCl treatment,the relative expression level of the genes made a shift in the two varieties.Apart from CFP that had lower values for CM72,the fold changes were generally higher in CM72 in the rest of the genes.The expression levels of Receptor-like protein kinase 4–1 (RLPK4-1) (HORVU2Hr1G111760.1),RLPK4-2 (HORVU2Hr1G111780.3) and RLPK4-3 (HORVU2Hr1G111790.7) were relatively higher in CM72,but not significant at all time points.GST family protein(HORVU2Hr1G111840.5) and EDG3 (HORVU2Hr1G111880.1)recorded higher and significant fold change under stress at 40,64 and 88 h in CM72.CFP expression was inhibited in both CM72 and Gairdner at 16 h and 40 h of salt treatment,but Gairdner showed increased expression at 64 h.A possible reason may be that the salt-sensitive variety has reached a critical point for growth and development under stress,so the gene is triggered to increase the rate of cell division.This pattern maybe in response to salinity stress rather than a tolerance mechanism.DRP expression was induced in Gairdner at the late stage of salinity treatment,however,it’s generally inhibited by salt stress in both varieties(Fig.S9).

Comparative analysis of fold changes under salinity and control treatments of the seven genes indicated that RLPK4-1,RLPK4-2,RLPK4-3,GST family protein and EDG3 were upregulated in CM72 and suppressed in Gairdner under salinity stress.Gene expression was induced at 40,64 and 88 h in RLPK4-1,RLPK4-2,RLPK4-3 and EDG3 and at 16,64 and 88 h in GST family protein and EDG3.Interestingly,salt stress upregulated the comparative expressions of RLPK4-1,RLPK4-2 and RLPK4-3 in Gairdner at 16 h.The upregulation at 16 h was just momentous in RLPK4-2 as it was quickly suppressed at 40 h and started to increase in the subsequent times of 64 and 88 h even though in all cases Gairdner’s levels were less than that of CM72.RLPK4-3 expression was higher in Gairdner at 16 h and dropped below CM72 at 40,64 and 88 h,but noticeably at 40 h and 88 h the comparative expression level of Gairdner was the same.Relative expression of RLPK4-1 in Gairdner remained unchanged in all the hours and relatively below CM72 at 40,64,and 88 h (Fig.3).

Fig.3.Quantitative variation of genes in proportional expression levels of seven genes in the embryo during germination of CM72 and Gairdner under 150 mmol L-1 NaCl(treated)relative to deionized(DI)water(control).HORVU2Hr1G111740.1,cyclin family protein;HORVU2Hr1G111750.5,disease resistance protein;HORVU2Hr1G111760.1,receptor-like protein kinase 4–1;HORVU2Hr1G111780.3,receptor-like protein kinase 4–2;HORVU2Hr1G111790.7,receptor-like protein kinase 4–3;HORVU2Hr1G111840.5,glutathione-S-transferase family protein;HORVU2Hr1G111880.1,endoglucanase 3.Expression of seven genes were analysed from RT-qPCR results using comparative quantification 2-ΔΔCT method with HvGAPDH as internal control under 150 mmol L-1 NaCl stress and DI water at 16,40,64,and 88 h.Data was shown as means±SD.Error bars which are not overlapping differ significantly.*,significant difference at P ＜0.05.

RLPK4-1,RLPK4-2,RLPK4-3,GST family protein and EDG3 were highly induced in Gairdner but not CM72 under control(consistent with Figs.S5,6;Tables S6,7).They were gradually upregulated in CM72 under treatment but not substantial in Gairdner,since the expression level was much higher than CM72 under control condition.Based on the relative and comparative expression patterns,we selected genes focusing on:(i) genes that were expressed in both CM72 and Gardner under control treatment in all time points;(ii) genes that displayed a continuous up-regulation pattern in both varieties with little variation among them or Gairdner being higher initially and eventually CM72 progresses to outshine in the subsequent time points under salinity treatment;(iii) genes whose comparative expression (salinity treatment less control treatment),were up-regulated at time point 1 more than or equivalent to that in Gairdner than CM72 but were down-regulated or remained unchanged in the subsequent time points;(iv) genes with initial low proportional expression (treated less control) at time point 1 in CM72 but eventually increased at time points 2 onwards to more than that of Gairdner.

The three receptor-like protein kinase 4 were fitting most of the four groups outlined earlier.While RLPK4-3 displayed a continuous up-regulation pattern in both varieties,there was no variation among them at 16 h and 40 h,but eventually CM72 progresses to outshine Gairdner at 64 h and the two varieties were equal again at 88 h,therefore it was not likely to be the candidate gene.RLPK4-1 was fitting the description,but notable,was the up regulation of Gairdner to levels higher than CM72 at time point 2 under treated conditions making it prospective to salinity stress at germination(Fig.S9).RLPK4-2 was fulfilling all the descriptions above almost impeccably and hence forthcoming as a possible gene.Therefore,RLPK4-1 and 2 (HORVU2Hr1G111760.1 and HORVU2Hr1G111780.3) are proposed to be the most likely genes contributing to salinity stress tolerance at germination by enhancing germination in barley.

3.7.Candidate gene’s structure analysis

Three receptor-like protein kinase 4 together with GST family protein were amplified and their sequences between the two varieties(CM72 and Gairdner)compared.Though it was not fulfilling the threshold outlined in 4 points earlier,GST family protein was included because of its high expression level during germination.Based on barley genome explorer (https://apex.ipk-gatersleben.de/apex/f?p=284:10) [38] and genomic database (http://146.118.64.11/BarleyVar/) [30] three receptor-like protein kinase 4 were initially selected as possible candidate genes and analysed(Table 3).RLPK4-2 with 4633 bp full length was the shortest compared with RLPK4-1 with 6614 bp and RLPK4-3 with 6754 bp.RLPK4-2 and RLPK4-1 had 3 exons each with almost similar coding sequences of 1551 and 1548 bp translating into 516 and 515 amino acids,respectively.GST family protein was the largest gene with 7999 bp full length,2 exons and 804 bp coding sequence translatable to 267 amino acids[30].When these genes were compared in the current v3 genome reference,RLK 1,2 &3 and GST had 1422,1284 &1278 and 732 coding sequences,that were translating to 473,427 &424 and 243 amino acids respectively [75].

Table 3 Analysis of the four genes in the fine mapped region based on barley genome explorer of 2017 and genomic database.

The sequences comparison between genomic DNA of the two varieties (CM72 and Gairdner) in the four genes revealed similarities between the parents.There were not many variations in the promoter regions and the exons of RLPK4-3 and GST family protein(between CM72 and Gairdner that may be associated with dissimilar expression levels.However,in exon 3 of RLPK4-1,724182861–724185561 on 2H of barley genome we observed deletions at 6 different sites of Gairdner sequence totalling to 12 bp,1 bp in CM72 and 1 bp insertion in both varieties separately in comparison with Morex.Further,the alteration caused a probable loss of 6 amino acids (3 Ser,1 Arg,1 Pro,and 1 Ala) and some substitutions in Gairdner (Ala to arg,2 leu to 2 ala,ser to glu,ala to ser,and thr to ser) (Fig.S10).An insertion of 18 bp back-to-back in exon 3,in CM72 and 1 bp in Gairdner of RLPK4-2,724201515–724203996 on 2H was observed.More deletions totalling to 22 bp at different locations within CM72 sequence was detected in RLPK4-2.Theinserted bases in CM72 sequence were projected to translate into 6 extra amino acids(1 Ala,2 Pro,1 Glu,and 2 Thr)and the deletions causing a likely loss of 12 amino acids (2 Arg,1 Tyr,1 Glu,1Pro,3 Ala,1 Asp,1 Phe,1 Asp,and 1 Thr) (Fig.S11).The insertions and deletions in the exons of the two genes may be important for the difference in salinity tolerance levels between the two varieties during germination.

3.7.1.Relationship between the two RLPK4 of barley and homolog genes from other species

The two RLPK4 (HORVU2Hr1G111760.1 and HORVU2Hr1G111780.3) had similar and identical 10 hits that were aligned using MEGA-X [35] as shown in Fig.S12.G-type lectin Sreceptor-like serine/threonine-protein kinase SD2-5 of Triticum urartu had the highest similarity of 86.83% while G-type lectin Sreceptor-like serine/threonine-protein kinase SD2-5 of Setaria viridis was the least at 71.71%.Phylogenetic tree was constructed by means of maximum likelihood (ML) process executed in MEGA-X and it indicated that barley was related to wheat with a slight distance from it.However,it was more distinct from Setaria(Fig.S13).

3.7.2.Comparison of the conserved domains in the two RLPK4

Fig.4.Conserved domains for receptor-like protein kinase 4.Receptor-like protein kinase 4–1 (HORVU2Hr1G111760.1.(B) Receptor-like protein kinase 4–2(HORVU2Hr1G111780.3).PKc,protein kinases,catalytic domain;STKc,serine/threonine kinases;IRAK,interleukin-1 receptor associated kinases.The red dotted line is the boundary between the two genes while the 2 blue are for the conserved domain.The amino acids sequences bolded are reserved domains for the ATP binding,active,polypeptide substrate binding and activation loop (A-loop) sites.

All the hits above 70%(Fig.S14)of the two genes indicated that they were more diverse at the beginning and at the end,but more conserved in the middle.The preserved area for RLPK4-1 was between amino acids 150–435 (285 amino acids) and RLPK4-2 was 165–450(285 amino acids)both of which are within the third exon(Fig.4).The preserved domains for RLPK4-1 contains protein kinases,catalytic domain (PKc)-like superfamily that is composed of catalytic domains of serine/threonine and tyrosine-specific protein kinases,RIO kinases,(typical serine protein kinases),aminoglycoside phosphotransferases,and choline kinases.The ATP binding site on conserved domain of RLPK4-1 have been mapped in Fig.4A as follows:i.STELGSGGFGVVYKGELPNGLPVAVKVL ii.HLVRLYGFCFDPDTKALVYEYL ENG and iii.VHYDIKPPNIL LTADFTPKVADFG (the amino acid sequences underlined and bold are conserved domains for ATP binding site).Conserved domain of RLPK4-2 has serine/threonine kinases (STKc),interleukin-1 receptor associated kinases(IRAK),related STKs and PKc_like superfamily(Fig.4B).The ATP binding,active,polypeptide substrate binding and activation loop (A-loop) sites on conserved domain have been mapped in Fig.4B as follows:i.STELGSGGFGVVYKGELPNRLSVAVKLL ii.HVHLVRLYGFCFDPDTKALVYEYLENGSLEKY and iii.HYDIKPANILLTADFTPKVADFGLARL GERENTHMSSLT GGGRGTPGYA(the amino acid sequences underlined and bold are conserved domains for the sites).The occurrence of conserved area in exon 3 of the genes (that are hypothetically starting on amino acid 61 and 76 to the end for RLPK4-1 and RLPK4-2 respectively)is important because any variations in these regions as reported here is likely to cause changes in amino acids that can affect the gene function.

4.Discussion

4.1.Fine-mapping and gene annotation

The double haploid (DH) lines from the CM72/Gairdner and Skiff/CM72 populations showed continuous distribution but varied germination percentage.This demonstrates the quantitative nature of salinity stress tolerance in barley.Some lines showed transgressive tolerance index in both directions,indicating the likelihood of favourable and unfavourable allelic combinations in parents.Two studies conducted on barley reported different QTL for salinity tolerance at germination stage and confirmed that they differed from those at the seedling stage [18].

Saturating markers in a specific chromosomal region is the first step in map-based gene isolation and to improve the accuracy of MAS [39].In our previous study,the QTL conferring salinity tolerance at germination were mapped to a big interval on the short arm of chromosome 2H [18].In the present study,we increased the marker density on the QTL flanking region before identifying gene(s) and developing molecular markers.Screening the parents with InDel markers identified 42 of polymorphic markers in CM72/Gairdner compared with 12 Skiff/CM72 DH Populations.This confirms the fact that CM72/Gairdner DH population is derived from varieties that vary more between susceptible(Gairdner) and resistant (CM72) parents compared to Skiff/CM72 DH population from Skiff (susceptible) and CM72.

Using InDel markers is advantageous because they give an accurate nature of allelic variation of haplotypes and provide high density near the locus [40].The inclusion of new markers within the QTL region increased their density enabling the construction of a new genetic map and validation of the previous flanking markers.The screening of recombinant lines and phenotypic matching allowed us to narrow the interval to a region of 0.341 Mb.Within this region,we identified 9 genes belonging to five families including cyclin family protein,disease resistance protein,glutathione Stransferase family protein,endoglucanase 3,undescribed protein,and transmembrane emp24 domain-containing protein 3.The receptor-like protein kinase 4 was repeated,three in a series following each other.The receptor-like protein kinase 4 subfamily enhances the reverse inhibitory effect of salinity during germination in rice [41] and has been characterized for salinity tolerance in maize [42],wheat [43],soybean [44],rice [45],cotton [46] and model Arabidopsis [47].Overexpression of receptor-like kinases improves salinity,oxidative stress and ABA tolerance in their seeds during germination and enhances early root growth in Arabidopsis[48],soybean[49],boosts shoot Na+elimination and improves biomass in both Arabidopsis and barley [50].

4.2.Identification of potential molecular markers

Selecting for a trait based on phenotyping is slow,laborious,environment-dependent and needs lots of space.However,cultivar release could be accelerated by identifying parental tolerant germplasm and choosing progeny through MAS using dependable molecular markers,such as PCR-based,that are easy to run [51].With dependable molecular markers already developed,MAS would be able to quickly identify salt-tolerant germplasm at the germination stage for crossing and selection in barley breeding programs [9].In this study,we designed two molecular markers from the fine-mapped region with a 100% phenotype-genotype match in the recombinant lines of CM7/Gairdner DH likely to be associated with the salt-tolerant gene(s) located in the chromosome 2H locus.Furthermore,our results indicate that the gene(s)responsible for the QTL effect is/are likely to be between or close to markers InDel 74–61 and 74–63.InDel 74–61 is inside one of our target genes (HORVU2Hr1G111780.3) that belongs to the receptor-like protein kinase 4 family.In the longer term,other markers (SNP-based) within InDel 74–61 and 74–63 will need to be designed to allow for a more informative marker haplotype that will remove issues of marker-gene recombination.It is likely that the above markers could be converted for this purpose or do SNP screening as per the methods in this study using F2/F4populations.

A collection of 265 barley accessions from across the globe offered a retrospective analysis for validation of the identified markers and tested their usability as a selection tool in a commercial barley breeding program.The cluster analysis divided germplasm into subpopulations that exhibit a mean tolerance index that corresponds with that of the DH or F2populations.However,there was some variation in the means of the subpopulations in the worldwide barley germplasm,most likely due to factors such as the linkage between markers broken by recombination,other genes playing a role since the trait is polygenic,and genetic variation at other loci.Four parents,constituting the two DH and F2backcross populations,were included in the genotyping analysis of barley germplasm;notably,Gairdner and Spartacus CL fell into the same group,distinct from CM72,while Skiff was in a distinct smaller subgroup but the same major group as CM72,as predicted from our fine-mapping results.The above results suggest that the markers could be confidently used to practice negative selection to remove genotypes without markers because it is rare for the most tolerant or sensitive germplasm to be erroneously predicted by the marker genotype.

It is clear from this study and other reports that no single gene can offer complete salinity tolerance at germination or any other growth stage,which confirms the polygenic nature of this trait[9].MAS to combine multiple tolerant loci with divergent functions could widen the tolerance range of varieties and result in additive effects on tolerance levels.However,combining different tolerant loci does not necessarily produce an additive effect [52].Markerassisted selection is the best way to identify salinity tolerance gene/loci pyramiding from different sources and the most effective loci combinations [53].The markers developed in this study will enable the 2H locus to be combined with others to breed barley varieties with a high level of mixed salinity tolerance,used for germplasm selection targeting QTL,and provide a base for the development of gene-specific markers.

4.3.Expression studies of tolerant and sensitive verities revealed the involvement of receptor-like protein kinase 4 in salinity stress tolerance during germination

Without overemphasizing salinity affects most plant growth aspects including germination through osmotic stress,ionic imbalance,nutrition imbalance and oxidation [54],as well as,changing gene expression levels that affect plant response indirectly [55].Germination involves a process where a quiescent,dry seed imbibes water to facilitate embryonic axis elongation that is enhanced by several internal (like gene expression) and external factors including those with potential resistance like salinity stress[56].The processes by which barley seeds can maintain high germination under salt stress is not well investigated and little is known about the genes involved.To circumvent the adverse impact of environmental stress,plants either evade or sidestep the harmful effects by varying the expression of genes associated with stress management signalling [57].Key categories of genes displaying upregulation under stress conditions are linked to cellular activities like metabolite synthesis for osmoregulation,transportation of ions,hormone secretion,signal recognition and signal transmission [57].The essential high expression levels of genes under stress are hypothetically associated with tolerance capacity of a variety.

In this study,we compared a salt tolerance variety(CM72)and a salt-sensitive variety (Gairdner) in 150 mmol L-1NaCl to investigate the variation in genetic responses,which is predictable to be associated to salinity tolerance.Based on the real time-qPCR analysis,we observed that 2 RLPK4(HORVU2Hr1G111780.3 and HORVU2Hr1G111760.1)were significantly upregulated in salt-tolerant variety(CM72)under salt stress than sensitive Gairdner.In consistence with our results,Real-time PCR results showed that the expression levels of receptor-like protein kinases(RLKs)in soybean were principally induced by salt stresses and their overexpression in Arabidopsis promoted seed germination,as well as primary root and rosette leaf growth during the early stages of salinity stress[49].

Nanda et al.[58],reported a combination of two receptor-like kinases working together synergistically to regulate the timing of germination and were responsive to salt and osmotic stress in Arabidopsis seeds.Salinity susceptible mutants seeds also showed a hypersensitive reaction to ABA at germination,at the same time,exhibiting heightened upregulation of germination repressors and dormancy inducers (ABA-insensitive-3,ABA-insensitive-5,DELLA encoding RGL2 and delay-of-germination-1) under salinity.Receptor-like kinases (RLKs) are plasma membrane receptors/proteins molecules involved in detection of external and endogenous cell signals or stimuli both biotic and abiotic [59].Embryogenesis receptor-like kinases (SERK) genes,a subgroup of RLKs play a role in several signalling that are vital for plant,development and immune system [60].In specific,they are vastly expressed in the early stages of somatic embryogenesis [61],and hence,the study of their involvement in salinity tolerance during germination is of supreme standing to advance the commercial production of barley.In the coding sequence,of exon three in the two receptor-like protein kinase’s 4 (HORVU2Hr1G111780.3 and HORVU2Hr1G111760.1) we reported variation in bases sequences that could be associated with the variation in the expression levels of the genes in CM72 and Gairdner varieties.Any variation in the coding sequences of genes is likely to change the amino acids that will be spliced and eventually modify its expression.Similar to our study,Do et al.[62]reported a variation of 4 bases in coding region of OsHKT1;1 gene of rice leading to substitution of 4 amino acids that had no potential effects to resulting protein structure,but caused variations in post-translational modifications.

Homolog G-type lectin S-receptor-like serine/threonine-protein kinase SD2-5 from Triticum urartu and Setaria viridis had the highest and lowest percentage (87% and 72%) similarity for both receptor-like protein kinase’s 4 (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE=Proteins).The genes may be closely related in structure and function however,there was none with a 90%–100%resemblance in amino acids meaning that RLP4 is somewhat unique.Receptor-like protein kinase’s superfamily have been reported to duplicate themselves in organisms with variation in the copy numbers and size due to variance in expansion in species genome size differences and have redundant function within a cluster[63].G-type lectin S-receptor-like serine/threonine protein kinase (GsSRK) have been reported to positively regulate salinity tolerance in plants,especially in wild soybean.Real-time PCR has shown GsSRK is upregulated by ABA,salt,and drought stresses.Its overexpression enhanced seed germination,main root and growth of rosette leaf during the initial stages of salt stress and eventually higher chlorophyll content,taller plants,better yields,lower ion leakage and more siliques at the adult growing phase in Arabidopsis.In alfalfa,overexpression of GsSRK resulted to more twigs but petite shoots,healthier growth,reduced ion leakage(low Na+and high K+)and MDA content,increased SOD activity,proline content (plays a role in ROS scavenging,ion homeostasis,and osmotic regulation) and salt stress tolerance [64].

The highly conserved sites with binding abilities in plants suggest that any alteration in their structure units can eliminate protein function,and consequently evolutionary less favoured [65].Transcription factor domain capability to confer protein binding and arbitrating structural communications is drawn from many biological activities like signal transduction,biological molecule modification and cellular biosynthesis [66].Protein kinases function in big numbers of diverse signalling pathways,with their catalytic activity being very critical in regulating growth and protection of the organisms.Different active kinases assume patently comparable structure of catalytic domains,but inactive ones are flexible to allow for adoption of different conformations to a precise controlling protein in response to communications[67].

The 2 protein kinases are composed of catalytic domains of serine/threonine and tyrosine-specific protein kinases (STKs),RIO kinases,(typical serine protein kinases),aminoglycoside phosphotransferases,and choline kinases that are involved in catalysing the transfer of gamma-phosphoryl from ATP to hydroxyl groups in substrates of proteins [68].The conserved domains of ATP binding site GXGXXGX14K (X being any amino acid) [69],active site,polypeptide substrate binding site and activation loop in the two receptor-like protein kinase’s 4 (Fig.4) were highly conserved.IRAKs plays a role in toll-like receptor (TLR) and interleukin-1 (IL-1) signalling pathways,thus critical in regulating innate immune responses and inflammation [70].Different types of IRAKs have been reported (IRAK-1,-2,-3 (or -M),and-4) that exhibit different functions and levels of expression,subcellular distribution and dissimilarly arbitrate TLR signalling [71].Generally,-1,-2,and -4 are universally expressed as active kinases,while IRAK-M is only induced in monocytes (barley)and macrophages and usually is an inactive kinase [72].IRAKM contains a central kinase domain (a pseudokinase domain)on top of an N-terminal death domain (DD),a proST region (rich in serines,prolines,and threonines),and a C-terminal domain like other IRAKs [73].They are hormonal regulated pathways for resisting attack and binds to signalling peptides to limit stem cell proliferation,maintenance of shoot and root apical meristem in embryos [74].The presence of inactive kinase IRAK-3 or M(Fig.4B) in the conserved domain of RLPK4-2,is an indication that it’s likely to show more plasticity to external stimulus like salinity stress.

CRediT authorship contribution statement

Edward Mwando:Investigation,Formal analysis,Writing–original draft.Yong Han:Methodology,Formal analysis,Supervision,Writing– review &editing.Tefera Angessa:Resources,Supervision,Writing– review &editing.Xiao-Qi Zhang:Methodology,Resources.Chengdao Li:Conceptualization,Funding acquisition,Project administration,Writing– review &editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported by Australian Grains Research and Development Corporation (GRDC) grant IDUmu00046 and Graduate Research Funds from Murdoch University.Also,expertise assistance from the institutions and staff of Murdoch University,Western Crop Genetics Alliance,Western Australian State Agricultural Biotechnology Centre and The Department of Primary Industries and Regional Development is much appreciated.The authors would like to express their sincere gratitude to Lee-Anne McFawn,Sharon Westcott and Gaofeng Zhou for the assistance at various stages of the material preparation,experiment and data analysis for this manuscript.

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.10.008.