Marker-assisted screening of resistance to fire blight, powdery mildew, and apple scab in local apple varieties from Uzbekistan Abdushukur I Rakhmatullaev, Muhabbat K Turdieva, Marguba M Rejapova, Abrorjon A Abdurakhimov, Alisher A Abdullaev ABSTRACT Apple (Malus domestica Borkh.) is one of the most economically important fruit crops worldwide; however, its production is severely constrained by major diseases, including fire blight, powdery mildew, and apple scab. Breeding disease-resistant cultivars represents a sustainable alternative to chemical control, particularly through the effective utilization of local germplasm resources from Central Asia. This study aimed to evaluate the presence and distribution of resistance-associated alleles in local apple varieties from Uzbekistan using polymorphic DNA markers. A collection of local apple accessions was screened to identify markers linked to resistance against fire blight, powdery mildew, and apple scab. The analysis revealed substantial genetic variation in resistance gene combinations among the studied varieties. The fire blight-associated marker AE10-375 was detected in 79.8% of the accessions. For powdery mildew resistance, 75.2% of the varieties carried resistance alleles corresponding to both Pl1 and Pl2 genes. Screening for apple scab resistance demonstrated that Vfa2, Vfa1, and Rvi6 were the most prevalent genes, with Vfa2 detected in 95.4% of the accessions. Regional analysis indicated that accessions from Karakalpakstan exhibited the highest proportion of genotypes harboring markers associated with resistance to multiple diseases. Six local varieties—Atlas olma, Turkish, Xuboni, Krasniy jeleznyak, Shoyi olma, and Besh barmoq—were identified as carrying resistance-associated markers for all three diseases. These findings demonstrate that local apple germplasm from Uzbekistan represents a valuable genetic resource for resistance to economically important diseases. The identified genotypes provide promising donor material for breeding programs aimed at developing cultivars with durable, broad-spectrum resistance while reducing reliance on chemical control strategies. Introduction The modern apple was domesticated through trade along the Silk Road from its wild crabapple progenitors in the Tien Shan Mountains of Central Asia [ 1 ]. The genetic diversity found in this region encompasses valuable germplasm that can be exploited in breeding programs targeting agronomically important traits and disease resistance [ 2 ]. Through domestication, alleles conferring disease resistance and response to abiotic stress have been lost because their linkage with unfavorable alleles affects horticultural traits such as fruit size. Climate change and a growing global population heighten the urgency of improving existing crop species or developing new cultivars with disease resistance and environmental resilience. Severe diseases in apples are caused by both bacterial and fungal pathogens. Among them, the most common diseases are fire blight – FB ( Erwinia amylovora ), powdery mildew – PM ( Podosphaera leucotricha ), and apple scab – AS (Venturia inaequalis) [ 3 – 5 ]. Orchards affected by fungal diseases such as AS and PM require intensive plant protection management, with up to 25 fungicide applications required per vegetation period in some regions. The large input of fungicides affects both the environment and human health and increases the price of apple fruits. Furthermore, overuse of fungicides imposes selective pressure on pathogen populations, promoting the development of resistance to chemical agents [ 6 ]. Host genetic resistance is a sustainable alternative management strategy to reduce fungicide inputs [ 7 ]. Many commercial cultivars grown widely today are susceptible to the major pathogens of apple. Early studies identified wild apple species, including M. robusta, M. sublobata, M. atrosanguinea, M. prunifolia , and M. fusca , are carriers of genetic resistance to various diseases [ 8 ]. Despite these findings, there have been limited efforts to incorporate these traits into new cultivars. This is primarily due to the long breeding cycle of apple and the multiple rounds of breeding necessary to reduce the linkage drag of undesirable traits such as poor fruit quality. The goal of many apple breeding programs is to combine important horticultural traits with disease resistance [ 9 ]. To accelerate resistance breeding, molecular markers associated with resistance to AS, PM, and FB have been developed and are increasingly deployed in marker-assisted selection (MAS) programmes [ 4 , 10 – 17 ]. The importance of local and regional germplasm as a source of resistance alleles is increasingly recognized [ 12 ]. Local landraces, shaped by centuries of farmer-driven selection under region-specific disease pressure, represent a particularly relevant source of resistance alleles for breeding programs targeting agroecological conditions similar to those in which they evolved. Unlike exotic wild relatives, local varieties may carry resistance combined with desirable horticultural traits, thereby reducing the challenge of linkage drag in breeding crosses. Uzbekistan, located in Central Asia, has a long history of apple cultivation and breeding. Its close geographic proximity to the primary center of origin of the modern apple suggests the presence of substantial varietal diversity. In recent years, an increased incidence of FB and PM has been reported in the country, posing a growing threat to apple production [ 18 , 19 ]. Despite the substantial diversity of Uzbek apple germplasm and its proximity to the primary centre of origin of domesticated apple, no comprehensive molecular evaluation of resistance to major pathogens has been conducted in this gene pool. Previous studies have focused on European and North American cultivar collections; local Central Asian landraces remain largely unexplored as sources of resistance alleles for contemporary breeding. The present study fills this gap by applying validated DNA markers to a broad collection of local accessions from three geographically distinct regions of Uzbekistan. The present study aimed to: (i) screen 109 local Uzbek apple accessions for the presence of DNA markers associated with resistance to FB, PM, and AS; (ii) evaluate phenotypic disease resistance under natural field conditions across three geographically distinct regions; and (iii) identify multi-disease-resistant accessions with potential as donor parents in resistance breeding programs. These results provide valuable information for breeders and growers and establish a foundation for further genetic and breeding studies on apple varieties originating from Central Asia. Materials and methods Plant material Young leaves were collected from 109 local apple ( Malus domestica , Malus sieversii , and Malus niedzwetzkyana ) accessions from three geographically distinct regions of Uzbekistan: the northwestern provinces (Karakalpakstan, n = 33; and Khorezm Province, n = 41) and the southern Surxondaryo Province (n = 35). Sampling was designed to capture geographic and varietal diversity across these three distinct agro-ecological zones. Farms were selected in consultation with regional agricultural specialists from the M. Mirzaev Scientific Research Institute of Horticulture, Viticulture, and Winemaking; the Scientific Research Institute of Forestry; and the Uzbek Scientific Research Institute of Plant Growing, to represent historically important apple-growing districts within each province. Within each farm, one tree per variety was sampled as a minimum; where multiple trees of the same variety were present, samples were collected from each tree to account for potential within-variety genetic diversity arising from vegetative propagation over generations. Although accessions shared varietal names, they represent farmer-maintained local landraces propagated over generations, and therefore are not assumed to be genetically identical. Phenotypic characterization and cultivar identification were conducted collaboratively with specialists from the three institutes listed above. All samples were collected with the consent of farm owners and in full compliance with the national regulations of the Republic of Uzbekistan regarding the use of plant genetic resources. Ethics approval and consent to participate: Not applicable. The geographic location of each collection site was recorded using GPS. Each variety was assigned a sample code based on geographic origin, where the first letter(s) correspond to the region; detailed information on variety name, farm owner, and district is provided in Supplementary Table 1 . Leaf samples were placed into individual Ziplock bags containing silica gel and transported to the laboratory for subsequent DNA isolation and molecular analysis. No voucher specimens were deposited, as all samples originated from clearly identified cultivated apple varieties maintained in traditional orchards. Evaluation of disease symptoms The evaluation of apple trees was performed under natural (uncontrolled) conditions on private farms over two growing seasons (2022 and 2023). Disease severity was assessed at two time points per season (early and late in the infection period), and scores were averaged to obtain a representative resistance index for each tree. Farms with no visible disease incidence in either year were excluded from phenotypic analysis, although molecular data were retained for all accessions. Assessments of disease severity were made on each tree, and values of leaf and fruit incidence from the quadrants were averaged to obtain the percentage of diseased leaves and fruit per tree [ 20 ]. Disease severity for FB, PM, AS was evaluated using a standardized five-point ordinal scale (0–4), adapted from methodologies established by van der Zwet [ 21 ], Kim [ 22 ], and Rexhepi [ 23 ]. This scale facilitates consistent phenotypic classification of plant accessions into resistance categories: highly resistant (score 4), moderately resistant (score 3), weakly resistant (score 2), and susceptible (score 1 and 0). Disease-specific criteria were defined as follows: Fire blight: Score 4 = asymptomatic; score 3 = 1–12% of canopy exhibiting blight symptoms; score 2 = 13–50% canopy affected; score 1 = 51–75% canopy affected; score 0 = > 75% canopy affected. Powdery mildew: Score 4 = no visible symptoms; score 3 = 1–5% of leaf area colonized; score 2 = 5.1–20% leaf area affected; score 1 = 20.1–40% leaf area affected; score 0 = > 40.1% leaf area affected. Apple scab: Score 4 = symptomless foliage and fruit; score 3 = 1–10% of leaf and fruit surface exhibiting lesions; score 2 = 10.1–25% surface affected; score 1 = 25.1–50% surface affected; score 0 = > 50.1% surface affected. Molecular analysis The present study does not aim to develop novel markers, but rather to apply validated resistance-associated markers to an underexplored Uzbek local apple germplasm. Genomic DNA from each variety was extracted using the PureLink® Plant Total DNA Purification Kit (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. DNA quality was assessed spectrophotometrically on a BioSpec-nano (Shimadzu Corporation, Kyoto, Japan): only samples with an A260/A280 ratio of 1.8–2.0 and an A260/A230 ratio > 1.5 were used for PCR amplification. All DNA samples were diluted to 20–30 ng/µl, visualized on a 0.8% agarose gel, and stored at −20°C for further analysis. PCR was performed by using primer pairs specific to resistance markers to FB: AE10–375, CH-F7-Fb1, and GE-8019 [ 14 ]; PM: AT20–450 and Pm739 [ 24 ]; and AS: VfC, AM19, and AL07 [ 10 , 11 ]. All primer sequences, annealing temperatures, and expected amplicon sizes are listed in Table 1 . 10.1371/journal.pone.0357988.t001 Table 1 List of the molecular markers for AS, FB and PM resistance loci/genes along with their primer sequences, annealing temperatures and amplicon sizes. № Marker name Resistance gene / QTL Primer pairs Та* (°C) PCR band size** (b.p.) Reference Fire blight 1 AE10–375 FBF7 QTL (LG7) F 5’-ctgaagcgcacgttctcc-3’ R 5’-ctgaagcgcatcatttctgatag-3’ 60 375 / - [14] 2 CH-F7-Fb1 FBF7 QTL (LG7) F 5’-agccagatcacatgttttcatc-3’ F 5’-acaacggccaccagtttatc-3’ 60 210 / 174 3 GE-8019 FBF7 QTL (LG7) F 5’-ttgagaccgattttcgtgtg-3’ R 5’-tctctcccagagcttcattg-3’ 60 397 / - Powdery mildew 4 AT20–450 Pl-1 F 5’-atcagccccacatgaatctcatacc-3’ R 5’-acatcagccctcaaagatgagaagt-3’ 62 450 / 500 [24] 5 Pm739 Pl-2 F 5’-aacgaagatcaagaaacatgg-3’ R 5’-tgccgtcaatggaacacc-3’ 53 269 / - Scab 6 VfC Vfa1 (HcrVf1), Vfa2 (HcrVf2), Vfa4 F 5’-ggtttccaaagtccaattcc-3’ R 5’-cgttagcattttgagttgac-3’ 58 286 / 484 / 646 [ 11 ] 7 AM19 Rvi6 F 5’-cgtagaacggaatttgacagtg-3’ R 5’-gacaaagggcttaagtgctcc-3’ 60 526 / - [ 10 ] 8 AL07 Rvi6 F-5’-tgg aag aga gat cca gaa agt g-3’ R-5’-cat ccc tcc aca aat gcc-3’ 60 466/724 [ 10 ] *Ta: annealing temperature. **PCR band size: fragment sizes corresponding to resistant and susceptible alleles as reported in the original publications; “–” indicates absence of amplification. PCR amplification was performed using 20 ng of genomic DNA in 10 µl volumes containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 100 µM each dNTP, 200 nM each primer and 0.6 U Taq polymerase. PCR amplification was carried out on a Veriti™ Thermal Cycler (Thermo Fisher Scientific) using an initial denaturation step at 95°C for 5 min, followed by 35 cycles of 40 s at 94°C, 1 min at annealing temperatures based on the primers used for each marker ( Table 1 ), 1 min at 72°C, and a final extension at 72°C for 10 min. All PCR amplifications were performed in duplicate; discrepant results were resolved by a third independent reaction. Negative (water) controls were included in every PCR run. Amplified products were separated on 2.5% agarose gels in 0.5 × TBE, stained with ethidium bromide, and photographed under UV illumination. Fragment sizes were scored against a 100 bp DNA ladder (NEB). For dominant markers, a band was scored as present (+) if a fragment of the expected size (±5 bp) was observed; absence of any band was scored as (−). For codominant markers (CH-F7-Fb1, VfC, AL07), individual alleles were identified based on their diagnostic fragment sizes. Scoring was performed independently by two researchers; discrepancies were resolved by a third independent assessment. Statistical analysis PCR band presence/absence data were coded as binary (1/0) for each variety per marker. Statistical analysis was performed in GenAIEx 6.5 [ 25 , 26 ]. Pairwise Nei’s genetic identity between populations was calculated across markers associated with each disease. Genetic relationships between accessions were analyzed by cluster analysis using the neighbor-joining (NJ) method; dendrograms were constructed using MEGA12 software [ 27 ]. Evolutionary distances were computed using the p-distance method. Principal component analysis (PCA) was performed to explore genetic structure using the dplyr, ggplot2, ggrepel, and factoextra libraries in R Studio (version 4.4.2). Spearman rank correlation analysis was performed between resistance index scores and marker presence to assess marker-phenotype associations. Chi-square tests were used to test for significant differences in marker allele frequencies across geographic regions. Statistical significance was set at α = 0.05. Results Fire blight resistance markers The dominant SCAR markers AE10–375 and GE-8019 associated with alleles linked to the FBF7 fire blight resistance QTL [ 14 ] produced clear single bands of 375 bp and 397 bp in 79.81% and 11.0% of samples, respectively ( Table 2 ). Among these accessions, 77 amplified the AE10–375 locus only, whereas 2 amplified GE-8019 alone. Both loci were amplified in 10 accessions, whereas 20 accessions failed to amplify either marker. The FBF7 -resistant allele of GE-8019 was absent in Khorezm accessions, and was distributed in Surkhandaryo and Karakalpak regions in 16% and 12% of accessions, respectively. The dominant marker AE10–375 was found in 92% of Surkhandaryo and 88% of Khorezm samples ( Table 2 ). Overall, 83.48% of the Uzbek apple accessions contained at least one FBF7 -associated locus. The codominant STS marker CH-F7-Fb1 produced amplification products of 174 and 210 bp, the latter associated with resistance, and was present in 91 accessions. Spearman rank correlation between fire blight resistance index and marker presence was significant (rs = 0.51, p < 0.001). Chi-square analysis revealed statistically significant differences in AE10–375 frequency across regions (χ2 = 18.4, df = 2, p < 0.001). 10.1371/journal.pone.0357988.t002 Table 2 Molecular marker genotypes of 109 local apple varieties showing the presence (+) or absence (–) of resistance-associated alleles for E. amylovora , P. leucotricha , and V. inaequalis. Marker genotypes for resistance to fire blight ( E.amylovora ) Marker genotypes for resistance to powdery mildew ( P.leucotricha ) Marker genotypes for resistance to scab ( V.inaequalis ) GE-8019 AE10–375 CH-F7Fb1 (210 bp) Resistance Index b AT20–450 Pm739 Resitance Index VfC AM19 AL07 № a Variety Pl1 Pl2 Vfa1 (646 bp) Vfa2 (484 bp) Vfa4 (286 bp) Vf (526 bp) Vf (466 bp) Vf (724 bp) Resistance Index SS01 Qirmizak – + – 2 + + 3 – + – – – + 3 SS02 Renet Semerenko – + – 2 + + 4 – + – – – + 3 SS03 Atlas olma – + + 4 + + 4 – + – + – + 4 SS04 Turkish – + + 4 + + 4 – + – + – + 4 SS05 Atlas kechki + + – 4 – + 3 – + – – – + 3 SS06 Renet Simarenko – + + 3 – + 3 – + – – – + 3 SS07 Qirmizak – + + 4 + + 4 – + – – – + 3 SS08 Starkrimson – + + 4 + + 4 – + – – – + 3 SS09 Malus niedzwetzkyana – + + 4 – – 2 – + – – – + 3 SS10 Xuboni – + + 4 + + 4 – + – + – + 4 SS11 Sariq olma + + + 4 + – 3 – + – – – + 3 SD01 Mayskiy – – – 1 – – 1 + + – – – + 2 SD02 Xuboni – + + 4 + – 3 + + – – – + 2 SD03 Mayskiy – + + 4 + – 3 – + – – – + 3 SD04 Xuboni – + – 2 + + 4 – + – – – + 3 SD05 Atlas olma – + – 2 + + 4 – + – – – + 3 SD06 Mayskiy Qizil – + – 2 + – 3 + + – – – + 2 SD07 Eshak olma – – – 1 – + 3 + + – – – + 2 SD08 Qizil kechki olma – + + 4 + – 3 – + – – – + 3 SD09 Oq olma – – – 2 – + 3 – + – – – + 3 KH01 Besh barmoq – + + 3 + + 4 – + – – – + 3 KH02 Renet Semerenko + – – 2 + + 4 – + – + – + 4 KH03 Samarqandskiy ranniy – + + 3 + + 4 – + – + – + 4 KH04 Starkrimson – + – 2 + + 4 – + – – – + 3 KH05 Golden delicious – + – 2 + + 4 – + – – – + 3 KH06 Jonathan – + + 3 – + 3 – + – – – + 3 KH07 Pervenets Samarqanda – + + 3 + + 4 – + – – – + 3 KH08 Besh yulduz – + + 3 – + 3 – + – – – + 3 KH09 Besh barmoq – + – 2 – + 3 – + – – – + 3 KH10 Grimes Golden + + – 4 – + 3 – + – + – + 4 KH11 Qand olma kishgi – + + 3 + + 4 + + – + – + 3 KH12 Mayskiy – + + 3 + + 4 – + – + – + 4 KH13 Red delicious – + + 3 + + 4 – + – + – + 4 KH14 Qizil olma urtagi – + – 2 + + 4 – + – – – + 3 KH15 Qizil olma – + + 3 – + 3 – + – – – + 3 KH16 Shoyi olma – + + 3 + + 4 – + – + – + 4 KH17 Tosh olma – + – 2 – + 3 + + – – – + 2 KH18 Qand olma – + + 3 – + 3 + + – – – + 2 KH19 Krasniy jeleznyak + + – 4 – + 3 + + – + – + 3 XS01 Besh barmoq – + – 2 + + 4 – + – + – + 4 XS02 Grimes Golden – + + 2 + – 3 + – – + – + 2 XS03 So’x go’zali – + + 3 + + 4 – + – – – + 3 XS04 R.Semerenko + + – 4 + + 4 – + – – – + 3 XS05 Qizil sho’r olma – + – 2 + + 4 – + – – – + 3 XS06 Golden delicious – + + 3 + + 4 + – – – – + 2 XS07 Yozgi olma – + + 3 – + 3 – + – – – + 3 XS08 Krasniy jeleznyak + + – 4 + + 4 + + – – – + 2 XS09 Mayskiy – + – 2 + + 4 – + – – – + 3 XS10 Qizil olma + + – 4 + + 4 – + – – – + 3 XS11 Xazaraspskiy letniy – + + 4 + + 4 – + – – – + 3 XS12 Qand olma kishgi – + – 2 + + 4 – + – – – + 3 XS13 Qand olma + + – 4 + + 4 + + – – – + 2 XS14 Muz olma – + – 2 + + 4 + – – – – + 1 XS15 Xivinskiy beshbaroq – + + 4 + + 4 – + – – – + 3 SSh01 Renet Semerenko – – + 1 + + 4 – + – – – + 3 SSh02 Oq olma – – – 1 + + 4 + + – + – + 3 SSh03 Besh yulduz – – – 1 – + 3 + + – + – + 3 SSh04 Qizil olma – – – 1 – + 3 + + – + – + 3 SSh05 Besh yulduz – – – 1 – + 3 + + – + – + 3 SSh06 Renet Semerenko – – – 1 – + 3 + + – + – + 3 SSh07 Oq olma – – – 1 + + 4 + + – + – + 3 SSh08 Besh yulduz – + + 4 + + 4 + + – – – + 2 KB01 Krasniy jeleznyak – + + 4 + + 4 – + – + – + 4 KB02 Jonathan – + + 4 + + 4 – + – – – + 3 KB03 Yozgi olma – – – 1 + + 4 – + – – – + 3 KB04 Starkrimson – + – 2 + + 4 + + – – – + 2 KB05 Xazaraspskiy letniy – – – 1 + + 4 – + – – – + 3 KB06 Xazaraspskiy zimniy – – – 1 + + 4 + + – – – + 2 KB07 Shoyi olma + – – 2 + + 4 + – – – – + 1 KB08 Besh yulduz – + – 2 + + 4 + – – – – + 2 KB09 Golden delicious – + – 2 + + 4 + + – – – + 2 KB10 Malus niedzwetzkyana – + + 4 + + 4 + + – – – + 2 KB11 Rozemary white – + + 4 + + 4 – + – – – + 3 KB12 Besh barmoq – + – 2 + + 4 + + – – – + 2 KB13 Yanar – + + 4 + + 4 + + – – – + 2 KB14 Qizil olma (Red apple) – + + 4 + + 4 – + – – – + 3 SKO01 Qizil olma (Red apple) – + – 2 + + 4 + + – – – + 2 SKO02 Qizil olma (Red apple) – + – 2 + + 4 + + – + – + 3 SKO03 Qizil olma (Red apple) – + – 2 + + 4 – + – + – + 4 SKO04 Qizil olma (Red apple) – + – 2 + + 4 – + – + – + 4 SKO05 Renet Semerenko – + – 2 + + 4 + + – + – + 3 SKO06 Oq olma – + – 2 + + 4 + + – – – + 2 SKO07 Besh yulduz – + – 2 + + 4 + + – + – + 3 SKO08 Oq olma – + – 2 + + 4 – + – + – + 4 SKO09 Oq olma – – – 1 + + 4 – + – + – + 4 SKO10 Oq olma – + – 2 + + 4 – + – + – + 4 SKO11 Besh yulduz – + – 2 + + 4 + + – + – + 3 SKO12 Oq olma – + – 2 + + 4 + + – + – + 3 SKb01 Besh yulduz – + + 4 + + 4 + + – + – + 3 Skb02 Atlas – + – 2 + + 4 + + – – – + 2 SKb03 Qizil olma – – + 2 + – 3 – + – – – + 3 SKb04 Oq olma – + – 2 + + 4 – + – – – + 3 SKb05 Qizil olma – – – 1 + + 4 – + – – – + 3 SKb06 Qizil olma – + – 2 + + 4 – + – – – + 3 SKb07 Oq olma – + – 2 + + 4 – + – – – + 3 SKb08 Renet Semerenko – – – 1 + + 4 – + – + – + 4 SKb09 Qizil olma – – – 1 + + 4 – + – + – + 4 SKb10 Qizil olma – + – 1 + + 4 – + – – – + 3 XK01 Karvak qutir olma – + – 1 + + 4 – + – – – + 3 XK02 Avgustovskiy – – – 1 – – 1 – + – – – + 4 XK03 Ruxi janon – + – 1 + + 4 – + – – – + 4 XK04 Qand olma – + – 1 + + 4 – + – – – + 3 XK05 Karvak olma + + – 2 + + 4 – + – – – + 3 XK06 Qutir qand olma – + + 3 + + 4 – + – – – + 3 XK07 Jeleznyak – – – 1 + + 4 + + – + – + 3 XK08 Muz olma – + – 2 + + 4 – + – + – + 3 XK09 Renet Semerenko – + + 3 + + 4 + + – – – + 4 XK10 Besh barmoq + + + 4 + + 4 – + – + – + 2 XK11 Red delicious – + – 2 + + 4 + + – + – + 4 a-The letters in the accession number were assigned depending on their location in the geographical region of Uzbekistan. S -Surkhandarya, K- Karakalpakstan, X – Khorezm. b-The resistance index was assigned depending on the degree of pathogen infestation; the disease resistance indices were as follows: 1-susceptible, 2-low resistance, 3-moderate resistance, and 4-high resistance. Apple scab resistance markers The dominant marker AM19 and the codominant AL07 marker, both associated with Rvi6 [formerly HcrVf2 ] apple scab resistance, were amplified in 33.94% and 100% of accessions, respectively. The codominant AL07 marker amplifies two alleles: a 466 bp fragment associated with resistance and a 724 bp susceptibility-associated fragment. The 724 bp allele was found in 100% of accessions, whereas the 466 bp resistance allele was not amplified in any accession. The VfC marker detected two loci, Vfa1 (646 bp) and Vfa2 (484 bp); no Vfa4 (286 bp)-specific amplification was identified. The Vfa2 allele was identified in 95.41% of accessions, whereas Vfa1 was present in 36.69% of accessions. The distribution of loci was uneven among populations; Vfa1 was found primarily in Karakalpak accessions (9.1%), followed by Surkhandaryo (4%), and was not detected in the Khorezm region ( Table 2 ). Pairwise Nei’s genetic identity via VfC loci revealed the highest similarity between Karakalpak and Surkhandaryo accessions (97.5%), and the lowest similarity between Khorezm and Surkhandaryo accessions (84.3%) ( Table 3 ). 10.1371/journal.pone.0357988.t003 Table 3 Pairwise Nei’s genetic identity among apple accessions from different geographic regions on the basis of loci associated with pathogen resistance. Surkhandarya Karakalpak Fire blight ( E.amylovora ) Scab ( V.inaequalis ) Powdery mildew ( P.leucotricha ) Fire blight ( E.amylovora ) Scab ( V.inaequalis ) Powdery mildew ( P.leucotricha ) Surkhandarya 1 1 1 – – – Karakalpak 0.997 0.975 0.995 1 1 1 Khorezm 0.967 0.843 0.967 0.971 0.865 0.988 Powdery mildew resistance markers The Pl-1 resistance allele was found in 71.3% of varieties in the homozygous state and in 14.8% in the heterozygous state. The Khorezm region contained the most accessions with the Pl-1 allele (90.4%), followed by Karakalpak (78.8%) and Surkhandaryo (72.4%). The dominant Pm739 marker for Pl-2 resulted in amplification of a single 269 bp product in 33 Karakalpak, 24 Khorezm, and 42 Surkhandaryo accessions ( Table 2 ). Pairwise Nei’s genetic identity was highest between Surkhandaryo and Karakalpak accessions (99.5%); Khorezm and Surkhandaryo presented the lowest identity of 96.7% ( Table 3 ). Spearman correlation between PM resistance index and combined Pl-1/Pl-2 marker presence was significant (rs = 0.44, p < 0.001). Chi-square analysis confirmed significant regional differences in Pl-1/Pl-2 allele distribution (χ2 = 12.7, df = 2, p = 0.002). Phylogenetic analysis and cluster structure To visualize genetic relationships among accessions based on their combined resistance marker profiles, a neighbor-joining (NJ) phylogenetic tree was constructed for all 109 accessions using eight markers simultaneously ( Fig 1 ). The analysis grouped accessions into five biologically meaningful clusters that reflect distinct combinations of resistance alleles across the three diseases. 10.1371/journal.pone.0357988.g001 Fig 1 Neighbor-joining phylogenetic tree of 109 Uzbek apple cultivars based on resistance marker data from all eight markers (AE10-375, CH-F7-Fb1, GE-8019, AT20-450, Pm739, VfC, AM19, AL07). Evolutionary distances were computed using the p-distance method in MEGA12. Five biologically distinct clusters are identified: (i) Blue cluster — accessions carrying AE10-375 without Vfa1 , with a single PM resistance allele, corresponding to moderate PM resistance and low-to-moderate FB and AS resistance; (ii) Green cluster — accessions lacking both AE10-375 and GE-8019 but carrying both Pl-1 and Pl-2 , corresponding to high PM resistance combined with low FB resistance; (iii) Red cluster — accessions with multiple FB marker combinations (AE10-375 + CH-F7-Fb1 210 bp or GE-8019 + CH-F7-Fb1) together with both PM alleles, corresponding to moderate–high FB and PM resistance; (iv) Light blue cluster — accessions with Vfa2 and both PM alleles but without AE10-375, corresponding to strong PM and AS resistance with reduced fire blight protection; (v) Purple cluster — accessions distributed across all three regions sharing an intermediate marker genotype (single-allele FB markers, Vfa2 only, one PM allele), corresponding to broad but moderate resistance across all three diseases. The blue cluster (predominantly Surkhandaryo accessions) comprised varieties carrying the AE10−375 fire blight marker but lacking the Vfa1 scab allele and carrying only one PM resistance allele. This marker profile corresponds to the phenotypic pattern of moderate PM resistance combined with low-to-moderate resistance to FB and AS observed in this region. The green cluster (mixed regional origin) contained accessions characterized by the absence of both FB-associated SCAR markers (AE10−375 and GE-8019) combined with strong Pl-1 and Pl-2 PM allele presence, explaining the phenotypic pattern of high PM resistance despite low fire blight resistance. The co-localization of Pl-1 and Pl-2 on chromosome 12 likely maintains their coupling in this cluster [ 28 ]. The red cluster contained accessions with the most complete fire blight marker combinations (AE10–375 + CH-F7-Fb1 210 bp, or GE-8019 + CH-F7-Fb1), reflecting high-to-moderate FB resistance scores, combined with both Pl-1 and Pl-2 PM alleles. The simultaneous presence of multiple FB and PM resistance alleles in this cluster explains the consistently elevated resistance indices for both diseases. The light blue cluster comprised accessions with AE10–375 absent but carrying Vfa2 and both PM alleles, corresponding to the phenotype of strong PM and AS resistance with reduced fire blight protection. Finally, the purple cluster contained accessions distributed across all three regions sharing an intermediate marker genotype: single-allele FB markers, Vfa2 only (no Vfa1 ), and at least one PM allele — a combination corresponding to broad but moderate resistance across all three diseases. Overall, clustering revealed distinct molecular differentiation between regional apple varieties consistent with independent, region-specific selection histories. Karakalpak accessions were overrepresented in the red and light blue clusters (higher combined resistance), while Surkhandaryo accessions were more evenly distributed across all clusters, reflecting the broader phenotypic diversity observed in that region ( Fig 1 ). Geographic distribution of phenotypic resistance The geographic distribution of apple accessions and their phenotypic resistance ratings are illustrated in Fig 2 . A small proportion of accessions in all regions were susceptible to scab and powdery mildew (0–4% across regions). Among Karakalpakstan varieties, at least one PM resistance allele was present in all accessions ( Fig 2 ). High resistance to powdery mildew was detected in old varieties from all three regions: 68% in Surkhandarya, 76% in Karakalpakstan, and 88% in Khorezm — consistent with reports of low PM susceptibility in old German varieties [ 13 ]. 10.1371/journal.pone.0357988.g002 Fig 2 Geographic distribution of apple varieties and their phenotypic resistance to major diseases across sampling regions of Uzbekistan. Pie charts show the proportions of varieties exhibiting high, moderate, low resistance, and susceptibility to apple scab (AS), fire blight (FB), and powdery mildew (PM) in Karakalpakstan, Khorezm, and Surkhandarya. Map boundaries generated using publicly available geographic data from SimpleMaps ( https://simplemaps.com ), used under the Creative Commons Attribution 4.0 License. Resistance levels were determined from phenotypic disease scoring conducted over two growing seasons (2022–2023). Multi-disease resistant accessions Principal component analysis visualized the multi-dimensional resistance structure across all accessions ( Fig 3 ). PC1 explained 36.3% and PC2 explained 33.7% of total variance, together capturing most of the inter-accession resistance variation. The PCA revealed clear separation of susceptible and resistant varieties, with most local germplasm falling into low- to moderate-resistance groups. Six local varieties — Atlas olma (SS03), Turkish (SS04), Xuboni (SS10), Krasniy jeleznyak (KB01), Shoyi olma (KH16), and Besh barmoq (XK10) — were identified as carrying resistance-associated markers for all three diseases, making them especially valuable as multi-disease-resistant donor parents in integrated breeding strategies ( Fig 3 ). 10.1371/journal.pone.0357988.g003 Fig 3 Principal component analysis (PCA) of apple accessions based on resistance indices to fire blight, powdery mildew, and apple scab. Each point represents an individual apple variety positioned along the first two principal components (PC1 = 36.3%; PC2 = 33.7% of total variance). Colors indicate resistance categories: red = susceptible; blue = low resistance; green = moderate resistance; purple = resistant. Ellipses represent 95% confidence intervals for each resistance category. Six multi-disease-resistant accessions identified as carrying resistance-associated markers for all three diseases are labeled. Discussion This study aimed to characterize local Uzbek apple germplasm using marker-assisted and phenotypic approaches to identify genotypes with combined resistance to major apple diseases. The markers used serve as indicators of the presence of resistance-associated alleles; marker presence indicates potential resistance and should be complemented by phenotypic validation before definitive selection decisions are made. Most local varieties have been cultivated for many years across multiple regions of Uzbekistan. We found that varieties with the same names but grown in different regions differ in their combination of DNA markers and degree of disease resistance. This suggests that they might not be true to type, even though they continue to be labeled with their original varietal names. It is possible that growers in different regions have favored varieties with greater disease resistance, leading to parallel selective breeding. The high frequency of resistance-associated marker alleles in the studied germplasm does not always translate to correspondingly high phenotypic resistance scores. This discrepancy reflects the polygenic and quantitative nature of disease resistance in apple. Individual marker loci represent only components of the broader resistance architecture; expression of resistance is modulated by the genetic background, specific pathogen race, and environmental conditions. Epistatic interactions between resistance loci may further contribute to phenotypic variation; complementary gene action between Rvi6 and additional Rvi loci has been reported to enhance resistance beyond that conferred by any single locus [ 29 ]. Cases of markers present but low phenotypic resistance may reflect gene silencing, incomplete penetrance, or the presence of pathogen races capable of overcoming specific resistance genes. The most well characterized resistance gene to apple scab is Rvi6 (formerly HcrVf2 ), derived from the Japanese crab apple Malus floribunda 821 . Screening for scab resistance genotypes via the AL07 marker [ 10 ] resulted in amplification of a 724 bp susceptibility-associated fragment in all samples; the Rvi6 -linked 466 bp allele was not detected in any accession. This is striking given that the dominant AM19 was amplified in 33.94% of the varieties and was found to be 0.2 cM to the right of the Rvi6 locus in the same position as AL07. These data suggest that recombination occurs between the two markers in all Uzbek varieties, which is highly unlikely given the small distance between the markers, or that there is a large nucleotide difference in the region AL07 that is amplified between the M. floribunda 821 and Uzbek varieties. The screening of the varietal population for specific paralogs at the Rvi6 locus was performed via molecular markers to amplify Vfa1 , Vfa2 , and Vfa4 [ 11 ]. Malnoy et al. (2008) transformed susceptible ‘Gala’ with each paralog and reported that Vfa1 and Vfa2 confer resistance, whereas lines expressing Vfa4 were more susceptible; later work by Joshi et al. (2011) characterized Vfa2 as the only functional resistance gene [ 30 , 31 ]. Our work revealed no amplification of the Vfa4 paralog in any variety. This could indicate that this paralog has diverged in sequence drastically enough in Uzbek varieties to not be detected. Vfa1 and Vfa2 were detected in 36.69% and 95.41%, respectively. These results are broadly consistent with findings in Iranian apple germplasm [ 11 ]. Varieties rated as susceptible contained the Vfa1 paralog alone or in combination with Vfa2, suggesting that there are races of Venturia inaequalis present in Uzbekistan capable of overcoming Rvi6 -mediated resistance; there are reports of Rvi6 overcoming races in Europe and North America [ 32 ]. Two non-mutually exclusive explanations are proposed for the AL07 discrepancy: (i) sequence divergence in the AL07 primer-binding region in Uzbek varieties relative to the M. floribunda 821 source from which AL07 was designed; (ii) altered linkage phase between AL07 and AM19 in this germplasm. We recommend future sequencing of the AL07 target region to discriminate between these hypotheses. The genetic assessment of apple varieties for fire blight resistance revealed that only 21.1% of the studied accessions presented high resistance, whereas 35.78% presented low resistance, and 18.35% were susceptible ( Table 2 ). These results indicate that genetic resistance to this disease remains limited among cultivated local apple varieties. Molecular screening for fire blight resistance via SCAR and SSR markers associated with the AE10–375, GE-8019, and CH-F7-Fb1 loci provides essential insights into the genetic architecture of resistance within the studied germplasm. The dominant SCAR marker AE10–375, which is linked to the FBF7 locus [ 14 ], was detected at a high frequency (79.8%) among the analyzed apple varieties. However, the lack of correlation between molecular genetic results and phenotypic resistance data suggests that this locus alone confers only partial or conditional resistance, which is consistent with previous reports [ 33 , 34 ]. In contrast, the GE-8019 marker was observed at a low frequency (11.0%), indicating its limited distribution in the studied gene pool, which may reflect either its rarity in local germplasmor its potential loss during breeding selection for other agronomic traits. The codominant CH-F7-Fb1 SSR marker, particularly its 210-bp allele, was found to be associated with fire blight resistance and was detected in 37.6% of the studied apple accessions. Notably, only 35.78% of the accessions carried both AE10–375 and CH-F7-Fb1–210, and—contrary to earlier reports suggesting genetic linkage between these two loci [ 14 ] —no evidence of cosegregation was observed. The independent segregation of AE10–375 and CH-F7-Fb1 likely reflects recombination events or distinct haplotype structures in the diverse Uzbek germplasm pool with M.sieversii and M. niedzwetzkyana ancestry. Accessions harboring combinations of multiple fire blight markers consistently presented moderate to high resistance, supporting allele pyramiding as a strategy for durable resistance [ 16 ]. Phenotypic resistance data confirmed that apple accessions harboring combinations of AE10–375 and CH-F7-Fb1–210, or GE-8019 and CH-F7-Fb1–210, consistently presented moderate to high resistance. Notably, only the local cultivars “Sariq olma” (SS11) and “Besh barmoq” (XK10) harbored all three resistance loci, and phenotypic evaluation likewise revealed these accessions as the most highly resistant to fire blight. This finding clearly demonstrates the cumulative effect of multiple resistance loci, supporting the notion that pyramiding favorable alleles represents a promising strategy to achieve durable and broad-spectrum resistance against fire blight. Similar trends have been reported in previous studies, where the combination of multiple resistance loci conferred stronger and more stable resistance than single-locus effects [ 14 , 33 ]. Comparing results with other germplasm collections, the high PM resistance frequency in Uzbek old varieties (68–88% across regions) is consistent with findings in old German varieties [ 13 ]. The co-localization of Pl-1 and Pl-2 on chromosome 12 [ 28 ] likely explains the high co-frequency of both alleles in Karakalpakstan accessions, as linked inheritance maintains their coupling phase across generations of open pollination. From a practical breeding perspective, the six accessions (Atlas olma (SS03), Turkish (SS04), Xuboni (SS10), Krasniy jeleznyak (KB01), Shoyi olma (KH16), Besh barmoq (XK10)) identified as carrying markers for all three diseases represent high-priority candidates for use as donor parents in resistance breeding crosses. A marker-assisted backcross (MAB) strategy could be employed to introgress multiple resistance loci into elite cultivar backgrounds while minimising linkage drag using background selection with neutral markers. Several limitations of this study should be acknowledged: (i) phenotypic evaluations were conducted under natural field conditions with variable disease pressure across farms and seasons; (ii) standard resistant and susceptible control genotypes were not included for marker validation; (iii) the marker panel covers a subset of known resistance loci (additional loci such as Rvi4, Rvi15 , and PM QTLs on LG2 and LG3 may have gone undetected); (iv) The absence of the AL07 resistance allele in all accessions may reflect primer-template mismatch arising from sequence divergence between the primer design template ( M. floribunda 821) and Uzbek apple varieties. AL07 primer specificity in Uzbek germplasm requires sequencing confirmation; (v) Standard reference genotypes (e.g., M. floribunda 821 as positive Rvi6 control, known susceptible cultivars as negative controls) were not included in the present study, since there are not available in local collections. This represents a methodological limitation; validated reference controls will be included in future studies. Conclusions This study presents the first comprehensive molecular evaluation of resistance-associated alleles for fire blight, powdery mildew, and apple scab across 109 indigenous Uzbek apple varieties. Key findings are: (i) substantial genetic variation in resistance-associated loci exists across Uzbek apple germplasm, with regional differentiation reflecting geographic origin and historical selection; (ii) six multi-disease-resistant accessions (Atlas olma, Turkish, Xuboni, Krasniy jeleznyak, Shoyi olma, Besh barmoq) were identified as priority breeding donors; (iii) the AL07/AM19 discrepancy warrants follow-up sequencing; (iv) fire blight resistance remains limited in frequency, emphasizing the need for allele pyramiding strategies. The resistance-associated markers used in this study indicate the presence of resistance alleles and should be regarded as tools for prioritizing accessions for more detailed phenotypic evaluation, not as standalone criteria for resistance designation. Durable resistance requires validated functional confirmation through controlled inoculation experiments. Future research directions include: (i) whole-genome sequencing or dense SNP genotyping for comprehensive resistance locus characterization; (ii) controlled inoculation experiments to validate resistance of identified accessions; (iii) inclusion of genome-wide SSR markers for pedigree and population structure analysis; (iv) marker pyramiding in breeding crosses using identified multi-resistant donors. Supporting information S1 Table List of apple genotypes ( Malus domestica , M. sieversii , and M. niedzwetzkyana ) sampled in Uzbekistan, including variety name, geographic origin, farm/owner. (DOCX) S1 Fig PCR amplification profile of the AE10–375 marker associated with fire blight resistance in local apple varieties. M100: 100 bp DNA ladder. The expected amplicon size is 375 bp. (TIF) S2 Fig PCR amplification profile of the AT20–450 marker associated with field resistance to powdery mildew in local apple varieties. KH01–19 represent plant DNA samples. M100: 100 bp DNA ladder. The expected amplicon size of the AT20–450 marker is 450–500 bp. (TIF)