Genetic differentiation of morphologically similar polyploid wheat species

Abstract

Background

Wheat is a globally important polyploid crop, with hexaploid bread wheat (Triticum aestivum L.) and tetraploid durum wheat (T. turgidum subsp. durum (Desf.) Husn.) as its main cultivated forms. Despite distinct end-use properties, these species are morphologically similar, making their identification difficult. Traditional phenotypic approaches often fail to resolve closely related polyploid wheats, emphasizing the need for a reliable molecular diagnostic and wheat barcoding strategy.

Method

This study developed and validated a multilocus molecular diagnostic framework for the discrimination of polyploid wheat species. The approach integrates plastid (rbcL, matK), nuclear ribosomal (ITS2, IGS), and nuclear-coding markers (Glu-1 and XDuPw167), all amplified using the Polymerase Chain Reaction. Validation was performed using ten experimental samples and 203 reference sequences retrieved from the NCBI GenBank database. We developed and validated a multilocus molecular diagnostic method for the reliable discrimination of wheat species.

Results

Plastid loci showed limited variation, whereas the IGS region contained a diagnostic 71 bp insertion linked to the D genome, clearly distinguishing hexaploids from tetraploids. The Glu-1 and XDuPw167 loci exhibited genome-specific polymorphisms that further differentiated the two species. The multilocus diagnostic method achieved over 95% amplification success and consistent sequence profiles across replicates, confirming its accuracy and reproducibility.

Conclusions

The proposed molecular diagnostic method provides a reproducible, cost-effective, and high-resolution molecular diagnostic tool for reliable wheat species identification. By combining genome-specific nuclear and expressed sequence tag-simple sequence repeat (EST-SSR) markers, this approach establishes a robust and scalable system applicable to species authentication, seed purity testing, germplasm characterization, and genetic resource management.

Introduction

The Neolithic transition to sedentary agricultural economies, which began approximately 13,000 years ago in the Fertile Crescent, led to the domestication of crops such as wheat and barley (Özdoğan, 2011; Broushaki et al., 2016; Feldman & Levy, 2023). Today, wheat (Triticum spp.) is cultivated worldwide and remains a cornerstone of global food security, with bread wheat (T. aestivum L.) and durum wheat (T. turgidum subsp. durum (Desf.) Husn.) as the two most economically important forms (Shewry & Hey, 2015). Despite their distinct end-use qualities—bread-making versus pasta production—these species are morphologically similar, making their reliable identification challenging.

Bread wheat, a hexaploid species (AABBDD), is characterized by a soft endosperm texture and highly extensible gluten that provides the elasticity and gas-holding capacity necessary for bread production. In contrast, tetraploid durum wheat (AABB) has a harder kernel texture and stronger, non-elastic gluten, which contributes to the firmness and cooking quality of pasta (Mastrangelo & Cattivelli, 2021). These biochemical and technological contrasts are driven largely by differences in glutenin subunit composition—particularly variation at the Glu-1 loci—yet they are not consistently reflected in external morphology (Žilić et al., 2011; Langner et al., 2017; Roncallo et al., 2021). Both species exhibit similar spike and glume structures, and their grains overlap in size and color, leading to frequent misidentification, particularly in mixed, processed, or archaeological material (Ciftci et al., 2019; Feldman & Levy, 2023; Shan & Osborne, 2024). Thus, while phenotypic and agronomic traits underpin functional divergence, they alone are insufficient for precise taxonomic or practical discrimination.

Polyploidy has played a central role in the evolutionary history of wheat, with tetraploid Triticum turgidum L. (AABB) formed through hybridization of Triticum urartu Thumanjan ex Gandilyan (A genome) and Aegilops speltoides Tausch (B genome), and hexaploid T. aestivum (AABBDD) arising from hybridization between tetraploid wheat and Aegilops tauschii Coss. (D genome) (Levy & Feldman, 2022). While polyploidy provides genetic resilience and adaptability, it also results in morphological resemblance among species, complicating taxonomic classification and practical identification.

Traditional diagnostic approaches, including grain morphology or spike traits, often fail to distinguish bread wheat from durum wheat, especially in processed or mixed samples (Goriewa-Duba et al., 2018; Feldman & Levy, 2023). Similarly, widely used plastid barcodes such as rbcL and matK are highly conserved in wheat, limiting their utility for species discrimination (CBOL Plant Working Group, 2009; Hosein et al., 2017). Nuclear ribosomal markers such as ITS2 provide somewhat greater resolution, but are subject to homogenization through concerted evolution, which reduces their discriminatory power in polyploid taxa (Álvarez & Wendel, 2003). Thus, there is a critical need for reliable, reproducible, and cost-effective molecular methods that can discriminate between wheat species.

This study addresses this gap by developing and validating a multilocus molecular approach for species authentication in polyploid wheats. The framework integrates plastid barcodes (rbcL, matK), nuclear ribosomal regions (ITS2, IGS), the Glu-1 glutenin gene, and the expressed sequence tag (EST)-derived simple sequence repeat (SSR) marker XDuPw167. By combining genome-specific and functionally informative loci, the method enhances both universality and diagnostic precision, thereby filling a key methodological gap in the molecular characterization of polyploid crops. The multilocus method was evaluated using ten experimental samples and 203 reference sequences retrieved from GenBank, demonstrating accurate, reproducible, and cost-effective discrimination between T. aestivum and T. turgidum subsp. durum. Beyond improving species-level resolution, the proposed method offers practical advantages over existing molecular approaches: it provides higher discriminatory power than plastid barcodes, lower cost than genome-wide sequencing, and broad applicability to breeding, seed authentication, and germplasm management. Collectively, this study establishes a validated, reproducible, and accessible molecular diagnostic protocol for reliable wheat species identification, suitable for use in agricultural, industrial, and research contexts.

Materials Methods

Overview of the method

This study develops and validates a multilocus molecular method for distinguishing polyploid wheat species (Fig. 1). The workflow consists of:

Figure 1. Methodological workflow for the molecular diagnosis of polyploid wheat species.

Schematic overview of the multilocus PCR and sequencing approach combining plastid, nuclear ribosomal, and nuclear-coding markers (rbcL, matK, IGS, Glu-D1, XDuPw167) to discriminate between hexaploid and tetraploid wheat species.

• i.Plant material preparation and DNA extraction using a modified (cetyltrimethylammonium bromide) (CTAB) protocol.
• ii.Polymerase Chain Reaction (PCR) amplification of standard plastid barcodes (rbcL, matK), nuclear ribosomal regions (ITS2, IGS), the Glu-1 glutenin locus, and the EST-derived SSR marker XDuPw167.
• iii.Sequencing and reference dataset integration by combining experimental data with publicly available GenBank sequences.
• iv.Comparative sequence analysis to assess discriminatory power across loci, validate species-specific polymorphisms, and evaluate reproducibility of the method.

This workflow is designed to be reproducible in standard molecular biology laboratories without requiring high-cost next-generation sequencing platforms.

Plant material and DNA extraction

Two representative wheat genotypes were selected for molecular analysis: T. aestivum cv. Gün 91 and T. turgidum subsp. durum cv. Selçuklu 97. Both cultivars are widely cultivated in Türkiye. For each genotype, five seeds were obtained from the Department of Field Crops, Faculty of Agriculture, Kırşehir Ahi Evran University, and used for DNA extraction and subsequent molecular analyses. The experimental design was aimed at validating the multilocus diagnostic method for species differentiation rather than assessing intra-varietal variability. Wheat seeds were germinated in pots under controlled conditions, and young leaves were collected ten days after germination. Leaf tissues were ground into a fine powder in liquid nitrogen using a sterile mortar and pestle. Genomic DNA was extracted from 15 mg of powdered tissue using a modified CTAB protocol (Doyle & Doyle, 1987) optimized for higher yield and purity. Briefly, samples were incubated in preheated extraction buffer (2% CTAB, 0.1 M Tris–HCl, pH 8.0, 20 mM EDTA, 1.4 M NaCl, and 1% PVP) at 65 °C for 2 h. After centrifugation at 13,000 rpm for 10 min, the supernatant was mixed with an equal volume of chloroform and centrifuged again at 13,000 rpm for 7 min. The aqueous phase was transferred to new tubes, and DNA was precipitated by adding 0.08 volumes of 7.5 M ammonium acetate and 0.54 volumes of isopropanol, followed by incubation at −20 °C for 2 h. The resulting DNA pellet was obtained by centrifugation at 14,000 rpm for 10 min, washed with 70% and 95% ethanol, air-dried, and dissolved in 100 µL of sterile nuclease-free water. DNA quality and concentration were assessed spectrophotometrically using A260/A230 and A260/A280 absorbance ratios. Detailed yield and purity data are presented in Table S1.

Primers and PCR amplification

Specific primers were employed to amplify the plastid barcode loci (rbcL, matK), the nuclear ribosomal regions (ITS2, intergenic spacer (IGS)), and the EST-derived microsatellite locus XDuPw167, which is selected as one of the best EST-SSRs for wheat genotyping (Eujayl et al., 2002). In addition, short fragments of the Glu-1 locus were targeted to assess ploidy-level variation in wheat. Since amplification with the primer pairs 52F and B52F did not yield a D-genome-specific product, additional primers were designed to amplify a region adjacent to the open reading frame of Glu-1 (positions 209–130), as described by Fernandez et al. (2013). In the first-round PCR, fragments of approximately 132 bp were successfully amplified from the X- and Y-type glutenin subunits of the A, B, and D genomes. These products were not genome-specific: primer 52F amplified both A and D genomes, while primer B52F, which differs by a single base from 52F, amplified the B genome. To achieve genome specificity, nested PCR was performed. In these reactions, independent amplification of X- and Y-type subunits was required. Specifically, primers 52F and 156R (5′-ACCATGGCTGCGTGCAC-3′) amplified the Dx subunit (Glu-1x), whereas primers 52F and DY156R (5′-ACAATGGTTGTGTGCAC-3′) amplified the Dy subunit (Glu-1y), each yielding a fragment of 118 bp. The details of all primers and their respective amplification conditions are provided in Tables S2, S3. PCR reactions were carried out in final reaction volumes ranging from 23.5 to 25 µL containing 5 × HOT FIREPol^® Blend Master Mix (Solis BioDyne, Tartu, Estonia), and 20–30 ng of genomic DNA as template. The annealing temperatures and cycling conditions for each locus are summarized in Table S3 and were optimized individually to ensure specificity and amplification efficiency. For the IGS and ITS2 loci, amplification was carried out at an annealing temperature of 60 °C for 30 cycles. The EST-SSR locus XDuPw167 was amplified at 60 °C for 35 cycles, while plastid loci (rbcL, matK,) and Glu-1 loci were amplified at an annealing temperature of 58 °C for 35 cycles. For each species, five biological samples were analyzed, and all PCR assays were performed in duplicate technical replicates to confirm the reproducibility of banding profiles and sequence data. Amplicons were resolved on 3% agarose gels stained with ethidium bromide and visualized under blue light (Vilber Lourmat, Collégien, France).

Sequencing and data integration

To complement the experimental dataset and enhance the robustness of the analyses, 203 reference sequences representing the same loci in T. aestivum and T. turgidum subsp. durum were retrieved from the NCBI GenBank database. Sequence selection was based on verified species identification (T. aestivum or T. turgidum subsp. durum), complete coverage and the highest identity for the target locus, and the availability of full metadata, including accession number and source information. Both experimental and reference sequences were aligned and analyzed to assess interspecific and intraspecific variation. The amplified gene regions obtained from five seed samples of each wheat species were sequenced in both forward and reverse directions at BM Labosis (Ankara, Türkiye) using an ABI 310 Genetic Analyzer (PE Applied Biosystems) and an ABI 3730XL 96-capillary automated sequencer. Bidirectional sequencing minimized errors and resolved ambiguities, thereby ensuring high-quality chromatogram data. Sequence visualization, editing, alignment, and phylogenetic analyses were conducted using MEGA v11.0 software (Tamura, Stecher & Kumar, 2021). The NCBI BLASTn algorithm was employed to compare the obtained sequences with reference datasets and to identify highly similar sequences in GenBank (Johnson et al., 2008).

Data analysis and validation of the method

Sequences were visualized, edited, and aligned using MEGA v11.0 (Tamura, Stecher & Kumar, 2021). BLASTn searches were performed against GenBank to confirm sequence identities.

Validation of the method involved:

Assessing discriminatory power of each locus (plastid vs. nuclear vs. EST-SSR).

Testing reproducibility across independent samples.

Comparing experimental results with public reference sequences.

Barcode regions (rbcL, matK, ITS2) served as baseline controls, representing current standard methods. The IGS region, Glu-1 locus, and EST-SSR marker were evaluated as enhanced alternatives, with a focus on identifying loci that yield clear and reproducible discrimination between hexaploid and tetraploid wheat.

Materials accessibility

Wheat seed material is available upon request from collaborating agricultural institutions. All primer sequences and PCR conditions are listed in Tables S2 and S3 to ensure reproducibility. Reference sequences are publicly accessible via GenBank, with accession numbers provided in Table 1. This transparency ensures that the method can be independently replicated and adapted by other researchers.

Table 1. BLAST results of T. aestivum and T. turgidum subsp. durum based on five gene regions.

Samples yielding clean sequences were included in the analysis.

Studied samples	Gene region	Base pair (bp)	Aligned species	GenBank accession code for the studied samples	Query coverage %	Percentage of identity %	Aligned sequences number retrieved from NCBI
T. aestivum		872	T. aestivum (gb—MG958556.1)	PX171930	100	100	38
T. turgidum subsp. durum	matK	872	T. turgidum subsp. durum (gb—KU170120.1)	PX171931	100	100	5
T. aestivum 1, 2, 3, 4, 5	rbcL	603	T. aestivum (gb—MF597070.1)	PX171932–PX171936	100	100	38
T. turgidum subsp. durum 1, 2, 3, 4, 5		872	T. turgidum subsp. durum (gb—KM352501.1)	PX171937–PX171941	100	100	3
T. aestivum 1, 4, 5		195	T. aestivum (gb—KF482082.1)	PX138775	100	100	36
T. aestivum 2	ITS2	195	T. aestivum (gb—MF480405.1)	PX138776	99.49	100
T. aestivum 3		195	T. aestivum (gb—MF480405.1)	PX138777	100	100
T. turgidum subsp. durum 1, 3, 4, 5		195	T. turgidum subsp. durum (dbj—LC377262.1)	PX138780–PX138782, PX138783, PX138784	100	100	6
T. turgidum subsp. durum 2		195	T. turgidum subsp. durum (dbj—LC377262.1)	PX138781	100	99.49
T. aestivum 1, 3		158	T. aestivum (gb—X07841.1)	PX207453, PX207455	100	95.51	5
T. aestivum 2	IGS	158	T. spelta (gb—AF147500)	PX207454	100	96.13
T. turgidum subsp. durum 1,2,3,4		88	Triticum turgidum subsp. durum (emb—AJ009138.1)	PX207458, PX207459	100	91.95	3
T. aestivum	XDupW167	240	T. aestivum (gb—XM_044549473.1)	PX227498	99.17	97.92	1
T. turgidum subsp. durum		226	T. dicoccoides (gb—XM_037593326.1)	PX227499	94.17	92.50	1
T. aestivum	Glu A1	132	T. aestivum (emb—X61009.1)	PX171942	100	100	17
T. turgidum subsp. durum		132	T. turgidum subsp. durum (gb—JQ689002.1)	PX171943	100	100	1
T. aestivum 1, 2	Glu B1	132	T. aestivum (gb—FJ561336.1)	PX207448–PX207449	100	100	26
T. turgidum subsp durum 1, 2		132	T. turgidum subsp durum (gb—JQ689004.1)	PX207450–PX207451	100	100	3
T. aestivum 1,2,3	Glu D1	118	T. aestivum (gb—KM116496.1)	PX171944–PX171945–PX171946	100	98.31	20

Results

DNA extraction and amplification success

High-quality genomic DNA was successfully extracted from all wheat samples using the modified CTAB method, demonstrating the reliability and cost-effectiveness of this widely used protocol. PCR amplification was achieved for all six target loci (rbcL, matK, ITS2, IGS, Glu-1, XDuPw167); however, not all amplicons produced sequences of sufficient quality, and only high-quality reads were retained for subsequent BLASTn and comparative analyses. To enhance analytical robustness, an additional 203 sequences corresponding to the same loci were retrieved from the NCBI GenBank database and incorporated into the dataset. These reference sequences facilitated the detection of conserved motifs, diagnostic indels, single-nucleotide polymorphisms (SNPs), and repeat-number variations distinguishing the two species. Although the multilocus analyses consistently revealed diagnostic polymorphisms, overall sequence variation was extremely low both within and between species. Consequently, population-level statistics such as nucleotide diversity (π), genetic distance (p-distance), or fixation index (Fst) were not calculated, as they would lack biological relevance under such limited polymorphism. Instead, species differentiation was validated through fixed, reproducible sequence differences, confirmed by replicated amplifications showing over 95% success. Collectively, these results underscore the reliability and reproducibility of the multilocus diagnostic framework for distinguishing morphologically similar polyploid wheat species.

Evaluation of barcode markers

The plastid regions rbcL and matK, as well as the nuclear ITS2 locus, were amplified to establish a baseline for comparison with alternative markers. Three barcode regions, matK, rbcL, and ITS2, were amplified, yielding fragments of approximately 603 bp, 842 bp, and 195 bp in both wheat species, respectively (Table 1). Sequence analysis revealed that the rbcL region in five T. aestivum and five T. turgidum subsp. durum samples, as well as the matK region in one T. aestivum and one T. turgidum subsp. durum sample, were identical within each species. Furthermore, the studied T. aestivum and T. turgidum subsp. durum accessions showed 100% sequence similarity with 43 reference rbcL and 41 reference matK sequences of T. aestivum and T. turgidum subsp durum accessions previously deposited in the NCBI GenBank database (Table S4, Figs. S1, S2). These findings indicate that the barcode regions rbcL and matK showed complete conservation across T. aestivum and T. turgidum subsp. durum, and no polymorphism was detected within the dataset analyzed.

Amplification and sequencing of the ITS2 region demonstrated a generally high level of sequence conservation between the studied T. aestivum and T. turgidum subsp. durum accessions. In studied T. aestivum, a limited number of nucleotide substitutions were observed at positions 164 and 175. These polymorphisms were not consistently detected across all individuals. By contrast, T. turgidum subsp. durum exhibited a highly conserved sequence profile, with only a single substitution identified at position 175 in one accession (T. durum 2). Comparative alignment with 40 reference ITS2 sequences from GenBank confirmed the high sequence similarity of both wheat species with previously reported accessions, with only two nucleotide substitutions observed at positions 164 and 175 (Table 2, Fig. S3). Overall, the ITS2 region exhibited low discriminatory power for distinguishing between T. aestivum and T. turgidum subsp. durum, although slight intra-specific variation was detected within T. aestivum.

Table 2. Comparison of the ITS2 sequences obtained from samples of T. aestivum, T. turgidum subsp. durum and aligned sequences.

Species/Base Locations	1	.	.	.	21	.	.	.	77	114	.	.	.	148	.	.	.	164	.	.	.	175	.	.	.	195
T. aestivum 1	T	.	.	.	C	.	.	.	A	C	.	.	.	A	.	.	.	T	.	.	.	A	.	.	.	G
T. aestivum 2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	A	.	.	.	G	.	.	.	.
T. aestivum 3	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	G	.	.	.	.
T. aestivum 4	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
T. aestivum 5	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
gb—KF482082.1—T. aestivum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
gb—KC589709.1—T. aestivum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
emb—FM998918.1—T. aestivum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
gb—MF480405.1—T. aestivum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	G	.	.	.	.
gb—OQ883696.1—T. aestivum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	G	.	.	.	.
gb—PP814623.1—T. aestivum	.	.	.	.	T	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	G	.	.	.	.
T. turgidum subsp. durum 1	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
T. turgidum subsp. durum 2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	G	.	.	.	.
T. turgidum subsp. durum 3	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
T. turgidum subsp. durum 4	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
T. turgidum subsp. durum 5	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
dbj—LC377262.1—T. turgidum subsp. durum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
gb—KM352501.1—T. turgidum subsp .durum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
gb—MZ230674.1—T. turgidum subsp..durum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.

Validation of the IGS region as a ploidy diagnostic marker

Consistent amplification of the 87 bp A/B-genome fragment was obtained in all replicates, whereas the 155 bp D-genome–specific fragment appeared only under optimized PCR conditions and was detected in approximately 30% of reactions. Amplification and sequence alignment of the IGS region clearly distinguished hexaploid T. aestivum from tetraploid T. turgidum subsp. durum, yielding ∼155 bp products for T. aestivum (A, B, and D genomes) and ∼87 bp products for T. durum (A and B genomes only) (Fig. 2). Within the three studied T. aestivum samples, nucleotide polymorphisms were detected at one position (base 89) (Table S5). Alignment with reference sequences from A. tauschii (gb— KF482112.1)—the donor of the D genome—confirmed that the inserted region is homologous to D genome–specific sequences, further supporting the association of this insertion with the hexaploid genome composition. The IGS sequences of Triticum spelta L. (AF147500.1, OM289954.1), T. aestivum (X07841.1, M37269.1) and A. tauschii (KF482112.1), which possess the D genomes, were retrieved from the GenBank database and compared with the sequence data of the studied T. aestivum (Fig. 3). The T. aestivum sequences that matched the D genome from the wheat samples showed differentiation at 16 base positions. Four T. turgidum subsp. durum samples that have identical sequences exhibited the shorter structure without the insertion, consistent with their tetraploid genome composition. In the region shared with T. aestivum and T. turgidum subsp. durum, differentiation was observed at eight base positions in the 87 bp IGS region. Also, four deletions were detected in this shared region of the IGS gene. Only three alignments with reference sequences from T. turgidum subsp. durum (AJ009138.1, PX207458.1, PX207459.1) indicated seven nucleotide polymophisms (Fig. 4).

Figure 2. Amplification pattern of the IGS region in hexaploid and tetraploid wheat species.

Amplification pattern of the IGS region in hexaploid T. aestivum and tetraploid T. turgidum subsp. durum. T.a, T. aestivum; T.d, T. turgidum subsp. durum; L, Low molecular weight DNA ladder.

Figure 3. Multiple sequence alignment of the IGS region among the studied T. aestivum and T. turgidum subsp. durum samples and reference sequences of D-genome–containing species.

Multiple sequence alignment of the IGS gene region comparing the studied T. aestivum sample, reference sequences of D-genome–containing species (T. aestivum, T. spelta, A. tauschii), and the studied T. turgidum subsp. durum sample.

Figure 4. Pairwise sequence alignment of the IGS gene region between the studied T. turgidum subsp. durum sample and the T. turgidum subsp. durum reference sequence retrieved from NCBI.

Glu-1 gene region: genome-specific polymorphism

In the present study, Glu-1 subunits from the A, B, and D genomes were successfully amplified and analyzed in hexaploid bread wheat (T. aestivum) and tetraploid durum wheat (T. turgidum subsp. durum). First-round PCR generated 132 bp fragments corresponding to the x- and y-type glutenin subunits of the A and B genomes (Tables 1 and 3). These amplicons were not genome-specific and aligned with both Ax and Bx subunits. In contrast, nested PCR produced a 118 bp fragment specific to the Dx subunit, while the Dy subunit was not amplified under identical conditions (Tables 1 and 3). Comparative analysis of the Glu-1 loci across subgenomes revealed that A and B genome sequences were identical between T. aestivum and T. turgidum subsp. durum, indicating intraspecific conservation. In contrast, the D genome, which is unique to hexaploid wheat, displayed clear sequence polymorphisms among T. aestivum accessions. Alignment with reference datasets (18 A, 29 B, and 20 D genome sequences from GenBank) confirmed the presence of both conserved and polymorphic sites. At the Glu-A1.1x locus, three variable sites were identified among 18 T. aestivum accessions, including one indel at position 29 and two SNPs. Additional substitutions were detected, such as a G insertion at position 48 and a T/C transition at position 71, confirmed by alignment with reference sequences (LT626214.1, DQ533690.1). In contrast, two T. turgidum subsp. durum accessions showed complete conservation at the same sites (Table 3, Fig. S4). Similarly, the Glu-B1.1x locus was highly conserved across both species. Among 28 T. aestivum and five T. turgidum subsp. durum accessions, only one variant was detected—a C substitution at position 44 in T. aestivum (AJ567973.1)—while all T. durum accessions were identical (Table 3, Fig. S5). By contrast, the Glu-D1.1x locus, found only in T. aestivum, exhibited significant sequence variability. Among 23 T. aestivum accessions, several SNPs were detected. Three accessions aligned with the reference sequence T. aestivum (KM116496.1) with 98.31% identity. Additional polymorphisms were identified, including a G substitution at position 44 (AB485591.1) and a T substitution at position 48 (KM116496.1), along with further accession-specific SNPs (Table 3, Fig. S6).

Table 3. Comparison of the Glu 1 sequences obtained from samples of T. aestivum, T. turgidum subsp. durum and aligned sequences.

Common sequence part of HMW- GS ( Glu 1 )	A, B, and D Genome																		A and B Genome
HMW- GS (Glu A1.1)	1	11	29	33	44	48	65	70	71	73	82	87	93	100	109	112	118	119		.	126	.	132
HMW- GS (Glu A1.1)
T. aestivum	A	G	G	T	A	T	C	C	A	A	T	C	A	A	G	G	G	T		.	T	.	C
emb—X61009.1—T. aestivum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
emb—LT626214.1—T. aestivum	.	.	.	.	.	.	.	.	.	G	.	.	.	.	.	.	.	.		.	C	.	.
emb—LT626213.1—T. aestivum	.	.	.	.	.	.	.	.	.	G	.	.	.	.	.	.	.	.		.	C	.	.
gb—DQ533690.1—T. aestivum	.	.	-	.	.	.	.	T	.	.	.	.	.	.	.	.	.	.		.	.	.	.
T. turgidum subsp. durum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
gb—JQ689002.1—T. turgidum subsp. durum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
HMW- GS (Glu B1.1)	A	A	G	T	C	T	C	C	A	G	A	C	T	G	A	A	T	C		.	C	.	C
T. aestivum 1	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
T. aestivum 2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
emb—AJ567973.1—T. aestivum	.	.	.	.	.	C	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
gb—KC254854.1—T. aestivum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
T. turgidum subsp. durum 1	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
T. turgidum subsp. durum 2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
gb—JQ689004.1—T. turgidum subsp. durum	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.		.	.	.	.
HMW- GS (Glu D1.1)	A	G	G	T	A	T	C	C	A	G	T	C	A	A	G	G	G
T. aestivum 1	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
T. aestivum 2	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
T. aestivum 3	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.	.
dbj—AB485591.1—T. aestivum	.	.	.	.	G	.	.	.	G	.	.	.	.	.	.	.	.
gb—KM116496.1—T. aestivum	.	.	.	.	G	.	T	.	G	.	.	.	.	.	.	.	.
emb—AJ577815.1—T. aestivum	.	.	.	.	G	.	.	.	G	.	.	T	.	.	.	.	.

Notes.

The “_” and “.” symbols indicate the deletion and the identical bases, respectively.

Variation in repeat number and sequence of XDuPw167 Locus

The callose synthase gene spans 6,455 base pairs, within which the XDuPw167 EST-SSR locus is located between positions 6,171 and 6,408 bp. Comparative sequence analysis of this gene between T. aestivum and T. turgidum subsp. durum revealed 14 single-nucleotide polymorphisms (SNPs) across the 6,455 bp region (Table 4). Notably, four SNPs were located within the EST-SSR locus (Fig. 5). Further sequence characterization of the XDuPw167 EST-SSR region demonstrated clear genetic divergence between the studied hexaploid T. aestivum and tetraploid T. turgidum subsp. durum. The amplicon length of this locus was 240 bp in T. aestivum and 226 bp in T. durum. A distinct difference in the number of AT tandem repeats was also observed: T. aestivum harbored 15 repeats spanning positions 92–121, whereas T. durum contained only eight repeats distributed between positions 106–121 (Fig. 5). Alignment of the amplified XDuPw167 sequences with GenBank references confirmed its association with the callose synthase gene. The T. aestivum sequence exhibited 100% identity and 97.92% similarity with the reference T. aestivum sequence (XM_044549473.1). Although no annotated callose synthase sequence is available for T. turgidum subsp. durum, its sequence aligned with tetraploid T. dicoccoides, showing 100% identity and 98.23% similarity (XM_037593326.1).

Table 4. Comparison of the β-1,3-glucan synthase sequence, including XDuPw167 EST-SSR sequences of T. aestivum and T. turgidum subsp. durum retrieved from NCBI.

β-1,3-glucan synthase															XDuPw167 EST-SSR
Sequence position	63	84-89	171	443	1826	2465	2468	2469	2471	2472	2473	2474	2489	5786	6268 - 6280	6281	6282	6283	6284	6285	6286	6287	6288	6289	6290	6291	6292	6293	6294	6295	6416
T. aestivum XM 044549473.1	T	–	C	A	A	T	C	T	T	G	A	T	T	T	ATATAT…	A	T	A	T	A	T	A	T	A	T	A	T	A	T	A	G
T. turgidum subsp. durum XM 037593326.1	C	C	G	G	G	C	A	C	A	C	C	A	C	C	—————	A	T	A	T	A	T	A	T	A	T	A	T	A	T	A	G

Notes.

The “_” and “.” symbols indicate the deletion and the identical bases, respectively.

Figure 5. Sequence alignment of the XDuPw167 EST-SSR region comparing the studied T. aestivum and T. turgidum subsp. durum samples with their corresponding reference sequences retrieved from NCBI.

Discussion

Evaluation of barcode markers

This study utilized a focused experimental dataset consisting of ten samples (five per species). The primary objective was to establish and validate a reproducible multilocus molecular diagnostic rather than to explore intra-specific genetic diversity. While the dataset was adequate for methodological development and diagnostic testing, it was not intended to support population-level or geographic inferences. Sequence analyses of the chloroplast barcode regions (rbcL and matK) revealed complete sequence identity between T. aestivum and T. turgidum subsp. durum, consistent with previous reports highlighting the limited resolution of plastid barcodes in closely related polyploid crops (CBOL Plant Working Group, 2009; Chen et al., 2021). Consistent with our findings, Lubna-Asaf et al. (2022) reported that synteny analysis revealed high sequence similarity among Triticum plastomes, particularly within protein-coding regions such as rbcL and matK. The strong sequence conservation observed across accessions reflects both the slow evolutionary rate and functional constraint of chloroplast genes (Daniell et al., 2016), as well as the relatively recent divergence of hexaploid and tetraploid wheats (Wang et al., 2024). Similarly, Awad et al. (2017) demonstrated that the standard plastid barcodes rbcL and matK possessed limited discriminatory power among Triticum accessions in Egypt. Taken together, these studies and the present findings confirm that plastid loci such as rbcL and matK remain valuable as baseline markers for amplification consistency and higher-level phylogenetic inference but lack sufficient interspecific polymorphism for reliable species-level identification in polyploid wheat. Therefore, the inclusion of complementary nuclear markers is essential to achieve diagnostic resolution.

Analysis of the ITS2 region revealed a high degree of sequence conservation between T. aestivum and T. turgidum subsp. durum. Although ITS2 is one of the most widely used nuclear barcodes in plants due to its high copy number and strong discriminatory potential at the genus level (Álvarez & Wendel, 2003; CBOL Plant Working Group, 2009), it provided limited resolution in this study. Only minor polymorphisms were detected within T. aestivum (positions 164 and 175) and a single base substitution in one T. turgidum subsp. durum accession, which were insufficient for reliable species-level discrimination. The limited variability of ITS2 in polyploid wheats likely reflects concerted evolution, a homogenizing process that reduces polymorphism among multiple ribosomal DNA repeats across subgenomes (Schoch et al., 2012; Wang et al., 2023). Additionally, the complex polyploid structure of wheat further constrains ITS2 divergence by maintaining rDNA uniformity. Consistent with these findings, GenBank alignments also confirmed the conserved nature of ITS2 sequences across accessions of both species. Overall, while ITS2 remains valuable for broader phylogenetic and evolutionary studies, its limited sequence divergence restricts its diagnostic utility for distinguishing closely related polyploid wheat taxa.

The ribosomal DNA intergenic spacer (rDNA IGS) is a highly variable non-coding region located between the 26S and 18S rRNA genes and contains repetitive elements that evolve rapidly (Carvalho, Guedes-Pinto & Lima-Brito, 2011; Bendich & Rogers, 2023). Because of its high variability, the IGS region has been widely used to determine ploidy levels in wheat. Early studies demonstrated its diagnostic utility. Sallares, Allaby & Brown (1995) identified a 71 bp insertion specific to the D genome, allowing the distinction between tetraploid and hexaploid wheats in charred archaeological samples. Subsequent work has confirmed that IGS-derived polymorphisms accurately reflect genome composition, with expected PCR fragment sizes of approximately 158 bp for the D genome and 87 bp for the A/B genomes (Brown et al., 1998; Li et al., 2011; Oliveira et al., 2012; Değirmenci et al., 2022). In the present study, IGS analysis clearly demonstrated its diagnostic potential by differentiating tetraploid T. turgidum subsp. durum (A and B genomes) from hexaploid T. aestivum (A, B, and D genomes). The observed fragment sizes—approximately 87 bp for T. turgidum subsp. durum and 155 bp for T. aestivum—correspond to the absence or presence of the D-genome–specific insertion, respectively. This size variation provides a simple, reproducible, and cost-effective means for ploidy-level identification, making IGS particularly valuable when genome-wide sequencing is not feasible, such as in seed authentication, trade control, or archaeobotanical investigations.

Comparative sequence analyses further confirmed that the inserted IGS fragment in T. aestivum is homologous to A.s tauschii-derived sequences, supporting its established role as the D-genome donor (Mosulishvili et al., 2021; Dubcovsky & Dvořák, 2007). The consistent presence of this insertion in hexaploid species (T. aestivum, T. spelta) and its absence in tetraploid taxa (T. durum, T. dicoccoides (Asch. & Graebn.) Schweinf.) highlight the robustness of this marker for identifying ploidy levels. Base substitutions observed at different positions within the IGS region also reveal clear genomic differentiation between A, B, and D genomes. A total of 18 polymorphisms were detected within D-genome sequences and 19 substitutions between the A and B genomes, emphasizing the evolutionary divergence between tetraploid and hexaploid wheats (Levy & Feldman, 2022; Cavalet-Giorsa et al., 2024). These findings align with reports that structural differences in rDNA loci remain stable and diagnostic across wheat lineages (Huang et al., 2023; Cavalet-Giorsa et al., 2024). Collectively, these results confirm that the IGS region serves as a robust and informative molecular marker for both phylogenetic and diagnostic purposes. The presence or absence of the D-genome-specific insertion provides a straightforward sequence-based criterion that complements other nuclear and plastid markers in wheat systematics. Beyond its taxonomic value, the IGS marker also captures the genomic consequences of allopolyploidization, offering insights into the evolutionary history of wheat while providing a practical tool for breeding, germplasm management, and evolutionary studies.

Nuclear marker performance

The Glu-1 locus in wheat encodes high molecular weight (HMW) glutenin subunits, which are key determinants of dough elasticity and viscosity and, therefore play a central role in bread-making quality (Kroupina et al., 2023). This locus is located on the long arm of chromosome 1 in the A, B, and D genomes of wheat, with each genome harboring paralogous genes that encode two types of subunits: x and y. The x (132 bp) and y (132 bp) subunits form the major coding regions of glutenin proteins (Gu et al., 2006; Rathan et al., 2020). PCR amplification and sequencing of orthologous Glu-1 regions across the A, B, and D genomes of hexaploid wheat (T. aestivum) and the A and B genomes of tetraploid wheat (T. turgidum subsp. durum) provided new insights into the genetic differentiation of polyploid wheat species. Among the three loci, Glu-D1, present only in hexaploid wheat, emerged as a key diagnostic marker whose sequence polymorphisms allow clear discrimination between tetraploid and hexaploid species. In contrast, Glu-A1 and Glu-B1 were highly conserved, confirming their limited discriminatory value and emphasizing the complementarity of conserved and variable loci within the multilocus framework. Amplification of Ax and Bx subunits from both species, and the Dx subunit exclusively from T. aestivum, confirmed the genome specificity of the markers and aligns with earlier studies describing the distinctive role of the D genome in hexaploid wheat (Fernandez et al., 2013; Gu et al., 2006). Comparative sequence analyses revealed strong conservation of Glu-A1 and Glu-B1 across both ploidy levels, with no nucleotide substitutions detected in T. turgidum subsp. durum and only minor indels and SNPs in T. aestivum. This stability reflects the evolutionary constraint and functional importance of these loci and explains the morphological similarity observed between bread and durum wheats despite genomic differences (Kroupina et al., 2023; Al-Khayri et al., 2023). By contrast, Glu-D1 exhibited multiple polymorphisms, including clustered SNPs and accession-specific substitutions, consistent with the D genome’s origin from Aegilops tauschii and the introduction of novel allelic variation during allopolyploidization (Dvořák & Akhunov, 2005; Wang et al., 2024). Previous phylogenetic studies indicate that multiple tetraploidization events, involving divergent A genomes and less divergent B genomes, contributed to the complex genomic architecture of bread wheat (Gu et al., 2006; Orlovskaya et al., 2020; Levy & Feldman, 2022). The polymorphic nature of the Dx subunit therefore represents a valuable diagnostic feature, useful both for distinguishing ploidy levels and for breeding applications aimed at improving gluten quality (Zhang & Appels, 2024). Moreover, genome-wide SNP studies have reinforced the diagnostic and applied significance of such markers in wheat taxonomy and cultivar differentiation (Hyun et al., 2020; Winfield et al., 2016). From an evolutionary perspective, the contrasting patterns of conservation and polymorphism among Glu-1 loci mirror the divergence history of the wheat genomes, which separated approximately 5–7 million years ago (Allaby, Banerjee & Brown, 1999; Zohary, Hopf & Weiss, 2012). The relatively slow evolutionary rate of the B genome (Feldman & Levy, 2023; Dvořák & Akhunov, 2005) explains its low variability, whereas the more recently incorporated D genome retains greater allelic diversity, providing a molecular signature of species differentiation. The Glu-1 loci highlight both evolutionary stability and diagnostic potential in polyploid wheats. The conserved A and B genome loci reflect functional constraint, while the polymorphic Glu-D1 locus represents a genome-specific marker of high diagnostic value and a reservoir of allelic diversity with direct implications for wheat breeding and improvement (Lee et al., 2025).

SSRs are well-established molecular markers due to their high informativeness and co-dominant inheritance. Expressed Sequence Tag–derived SSRs (EST-SSRs), in particular, offer the added advantage of being located within expressed genes, which makes them both evolutionarily conserved and transferable across related species (Eujayl et al., 2002; Yu et al., 2004; Wang et al., 2007). Despite their generally lower polymorphism compared with genomic SSRs, certain EST-SSR loci—such as XDuPw167—can display exceptional allelic variation and thus hold great diagnostic potential (Kong, Bai & Li, 2014; Manco et al., 2020). The XDuPw167 locus, mapped to chromosome 6AL of wheat, is closely linked to the Sr13 resistance gene region. This locus corresponds to a genomic region annotated as encoding β-1,3-glucan synthase (callose synthase), an enzyme involved in the biosynthesis of callose—a β-1,3-glucan polysaccharide that serves as a structural component of the plant cell wall and accumulates during development and stress responses (Eujayl et al., 2002; Wang et al., 2015). In this study, the XDuPw167 locus revealed distinct allelic differences between hexaploid T. aestivum and tetraploid T. turgidum subsp. durum. The hexaploid species produced a longer amplicon containing 15 repeat motifs, whereas the tetraploid carried only eight, indicating pronounced repeat-number variation. This sharp contrast aligns with the known dynamic nature of SSR evolution, likely driven by polymerase slippage within a non-coding segment of the β-1,3-glucan synthase gene (Sureshkumar et al., 2025). Because this gene encodes a vital enzyme for callose biosynthesis—a key component of plant defense—its flanking conserved coding regions maintain functionality while internal SSR regions accumulate neutral variation. This duality makes XDuPw167 an ideal functional marker: highly polymorphic yet biologically meaningful (Nahas et al., 2020). Sequence comparisons further confirmed that the β-1,3-glucan synthase gene itself is highly conserved between the two wheat species, consistent with its essential role in cell wall integrity and stress response. However, the XDuPw167 locus exhibited both repeat-number variation and SNPs, creating a dual polymorphic signature that enhances its diagnostic resolution. These polymorphisms likely occur in untranslated or intronic regions, or represent synonymous substitutions within the coding sequence, preserving protein function while reflecting genomic divergence (Castle, 2011). Functionally, β-1,3-glucan synthase mediates callose deposition during responses to biotic and abiotic stress (Seo et al., 2022). Therefore, variation at the XDuPw167 locus may subtly influence defense-related gene expression, positioning this locus not only as a diagnostic marker but also as a functional marker with potential relevance for marker-assisted selection (MAS). Among all loci examined, XDuPw167 demonstrated the strongest discriminatory power. Its combined repeat and SNP polymorphisms within a functional gene provide a unique molecular signature distinguishing tetraploid and hexaploid wheat while linking genetic variation to potential adaptive traits.

Applied implications

The multilocus molecular framework developed in this study offers a reproducible, low-cost, and scalable strategy for discriminating morphologically similar polyploid wheat species. By integrating plastid, nuclear ribosomal, and nuclear-coding markers, the method maximizes both universality and diagnostic precision. The combined analysis of IGS, Glu-1, and XDuPw167 loci provides complementary genetic information: IGS identifies ploidy-level differences (Huang et al., 2023), Glu-1 reveals genome-specific polymorphisms (Lee et al., 2025), and XDuPw167 delivers functional resolution within expressed genes (Nahas et al., 2020). This integrative approach ensures accurate species identification even when morphological characters or ecological data are inconclusive. Practically, the multilocus method holds significant potential for seed authentication, germplasm curation, and quality control in breeding programs. In applied settings, it can be adopted for rapid verification of wheat cultivar identity, detection of unintentional admixtures in seed lots, and monitoring of genetic integrity in breeding lines. Because the method relies on conventional PCR and Sanger sequencing rather than genome-scale analyses, it remains accessible to standard molecular laboratories without requiring high-cost sequencing platforms (Vasile et al., 2020; Satam et al., 2023). Recent genomic and taxonomic studies have emphasized the continued importance of classical locus-based assays for their reproducibility, cost efficiency, and compatibility with small or degraded DNA samples (Mahelka et al., 2021; Cavalet-Giorsa et al., 2024). Determining ploidy levels through genome-specific loci such as IGS and Glu-D1 provides a faster and more economical alternative to flow cytometry or nuclear DNA quantification, which demand specialized equipment and fresh, high-quality tissue (Chang, Davis & Kwon-Chung, 2024; Nowicka et al., 2021). Beyond agricultural contexts, the framework has broader relevance for archaeobotanical and evolutionary studies, where ancient or degraded DNA often limits analytical options. The diagnostic power of the IGS region, in particular, enables reliable determination of wheat ploidy levels in archaeobotanical remains, contributing to reconstructions of early crop domestication and trade. Likewise, the genic SSR marker XDuPw167 links molecular diversity to functional traits, offering opportunities for marker-assisted selection (MAS) and adaptive breeding.

Conclusions

This study establishes a robust multilocus molecular framework for distinguishing morphologically similar polyploid wheat species. The integration of plastid, nuclear ribosomal, and nuclear-coding loci provides reliable species-level discrimination, with the IGS, Glu-D, and XDuPw167 markers collectively capturing both structural and functional genomic variation. The multilocus approach offers high diagnostic accuracy, reproducibility, and cost efficiency, making it suitable for routine applications in seed authentication, germplasm management, and evolutionary research. These findings reinforce the continued relevance of targeted locus-based assays as accessible and practical tools complementing high-throughput genomic methods.

Methodological conclusion

The validated multilocus method—combining IGS, Glu-D, and XDuPw167 loci—constitutes a reproducible and scalable workflow for molecular identification of wheat species. Each marker provides complementary diagnostic information: IGS distinguishes ploidy levels, Glu-D reveals genome-specific variation, and XDuPw167 detects functional polymorphisms within expressed genes. The protocol relies solely on standard PCR and Sanger sequencing, offering a rapid and low-cost alternative to genome-wide or cytogenetic analyses. Its modular design allows straightforward adaptation to other cereal taxa, supporting the broader development of standardized, locus-based diagnostic systems for complex polyploid crops.

Funding Statement

The authors received no funding for this work.