Single nucleotide polymorphism (SNP) analysis reveals ancestry and genetic diversity of cultivated and wild grapevines in Croatia

doi:10.21203/rs.3.rs-4617887/v1

Download PDF

Research Article

Single nucleotide polymorphism (SNP) analysis reveals ancestry and genetic diversity of cultivated and wild grapevines in Croatia

https://doi.org/10.21203/rs.3.rs-4617887/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Croatia is a geographically small country, but it has a remarkable genetic diversity of cultivated and wild grapevines. Local germplasm has been characterised by microsatellite markers, but a detailed analysis based on single nucleotide polymorphisms (SNPs) is still lacking. Here we characterize the genetic diversity of 149 accessions from three germplasm repositories and four natural sites using 516,101 SNPs to identify complete parent-offspring trios and their relations with spontaneous populations, offering a proof-of-concept for the use of reduced-representation genome sequencing in population genetics and genome-wide association studies.

Results

Principal component analysis revealed a clear discontinuity between cultivated (V. vinifera) and spontaneous grapevines, supporting the notion that the latter represent local populations of the wild progenitor (V. sylvestris). ADMIXTURE identified three ancestry components. Two vinifera components are alternatively predominant in cultivars grown either in northern Adriatic Croatia and Continental Croatia or in Dalmatia (i.e. central and southern Adriatic Croatia). A sylvestris component, which is predominant in accessions from natural sites, is a minor ancestry component in cultivated accessions. TREEMIX provided evidence of a unidirectional gene flow from the vineyards to natural sites, suggesting that the introduced domesticated germplasm has colonized local wild populations stronger than vice versa. Identity-by-descent analysis indicated an extensive kinship network, including 14 complete parent-offspring trios and involving only cultivated accessions, and 6 full-sibling relationships, and invalidated a presumed pedigree of the most intriguing ones, the one of ‘Plavac Mali’. Despite this strong population structure, significant association was found between 143 SNPs and berry skin colour and between 2 SNPs and leaf hairiness, across two known QTL regions.

Conclusions

The genetic separation between Croatian cultivars and sylvestris excluded the hypothesis that cultivars originated from independent events of local domestication. On the other hand, the evidence of a crop-to-wild gene flow, especially in but not limited to the Modro jezero population, signals the need for an urgent adoption of conservation strategies that preserve the genetic integrity of wild relatives. The SNP panel enables enhanced parentage assignment for grapevine cultivars historically grown in Croatia.

SNP

V. vinifera

V. sylvestris

genealogical relationships

GBS

GWAS

Grapevine (Vitis vinifera subsp. vinifera) is one of the most economically important fruit crops in the Mediterranean Basin and the predominant cultivated species in vine-growing areas worldwide. The wild form V. vinifera subsp. sylvestris naturally occurs only in Europe and Western Asia. The process of domestication, which took place approximately 11,000 years ago [1] and was followed by selection, breeding, migration and wild introgression, has resulted in the cultivated germplasm being more diverse and heterozygous than the present-day populations of its progenitor [2, 3]. Despite being irrelevant for human consumption, V. sylvestris represents an important source of resistance genes against pathogens [4–7], resilience traits against abiotic stresses [8, 9] and beneficial endophytic microbial communities [10].

The exact events that led to grapevine domestication are still controversial [1, 11, 12], in particular whether domestication genes were selected once or twice. What appears to be undisputable is the evidence that domesticated grapevines underwent introgression events from V. sylvestris, as they were moved from the cradle of domestication along migration and trading routes. In particular, introgression from V. sylvestris populations in Western Europe has left a discernible footprint on the genomes of Western wine grapes [1, 11, 13, 14]. This may have helped the grapevines to adapt to new environments and diversify.

The diversity in cultivated grapes and the variation in natural populations originate from a combination of evolutionary driving forces, anthropic activities and historical circumstances [1, 3, 15, 16], at global and local geographical scales. Hermaphroditism, vegetative propagation, desirable flavours, sugar content, berry size and colour were the major targets of artificial selection. Using a large set of 2,096 cultivated genotypes, early genetic analyses defined three distinct groups, corresponding to the three main ecogeographical proles, resulting from selection for either wine making or fresh consumption and from adaptation to different climates in the Mediterranean Sea basin, mainland Europe and Central Asia [15]. The genotyping of national germplasm collections [17–27] and the sequencing of archaeological pips [28] have clarified that a few ancient cultivars, due to either their value of cultivation or founder effects, have generated many descendants within a few generations of sexual reproduction, before the resulting heterozygous genotypes were fixed by vegetative propagation. DNA paternity testing has revealed kinships that would never be predicted upon resemblance and descent groups of highly related cultivars that now dominate the market of varietal wines. In the wild compartment, studies have warned that there is genetic erosion and inbreeding, probably due to anthropogenic environmental impact, limited effective size of natural populations and geographical isolation [2, 3].

Within this global scenario, it is important to preserve and characterize local cultivars and wild populations. At least 127 cultivars are reported in Croatia [17, 18]. Many of them are grown over limited areas of cultivation. Some are only conserved in germplasm repositories. Several populations of V. sylvestris have been documented in natural sites [29]. So far, genetic analysis suggested loose links between European cultivars and local Croatian sylvestris [30, 31]. Phenotypic diversity is quite large countrywide. Grapevine cultivars are predominant in terms of numbers and cultivated surface area, while only a small number of table grape cultivars are grown for fresh consumption. A few complete parent-offspring trios and several parent-offspring relationships were identified using SSR markers [17]. Nonetheless, Mendelian inconsistencies were frequent in genuine pedigrees due to the hidden presence of non-amplifying PCR alleles and the somatic hypervariability of microsatellite DNA. Bacilieri and colleagues found that Croatian cultivars tend to cluster with the group of Balkan and Eastern European wine varieties [15]. However, only a small number of Croatian cultivars were included in their analysis (i.e. those held at the French national collection). The origin of the vast majority of minor Croatian cultivars is still unknown. Several historical cultivars have a tight genetic relation with cultivars grown in neighbouring Italy on the western and northern shores of the Adriatic Sea basin. Numerous synonymies of important genotypes have been discovered in the two countries such as ‘Tribidrag’ and ‘Primitivo’ (also known overseas as ‘Zinfandel’), ‘Maraština’ and ‘Malvasia Bianca Lunga’, and ‘Verdic’ and ‘Glera’ [32]. Common cultivars such as ‘Bombino Bianco’, ‘Heunisch Weiss’ (synonym of 'Belina Starohrvatska') and ‘Blank Blauer’ (synonym of ‘Bljuzgavac’ and ‘Vulpea’) were shown to be important progenitors in both countries [17, 27, 32, 33].

We have recently used a targeted sequencing method known as Single Primer Enrichment Technology (SPET), which was designed to sample 50,000 genic regions, to generate a half million SNPs from traditional Croatian grapevine germplasm [32]. Although whole-genome resequencing is rapidly becoming available at an affordable cost [1], we used this SNP dataset to prove that reduced-representation genome sequencing is a cost-effective alternative for pedigree reconstruction, population genetic analysis and GWAS of qualitative and quantitative traits, revealing with the Croatian case-study that national and transnational grapevine diversity is hierarchically interconnected in Europe.

Plant material

Genotypic data of 149 grapevine accessions, representing unique genotypes, were obtained from previous work [32]. Of these, 108 accessions belong to the traditional grapevine cultivars (Vitis vinifera L. subsp. sativa) in Croatia. The plant material of the 108 grapevine cultivars used in this study is maintained at three Croatian germplasm repositories: Institute of Adriatic Crops and Karst Reclamation, Split, (institute code HRV048); Institute of Agriculture and Tourism, Poreč, (institute code HRV050); University of Zagreb, (institute code HRV041). The remaining 41 accessions belong to the wild grape (Vitis vinifera L. subsp. sylvestris Gmelin) and were growing spontaneously in their natural habitats at three locations in Croatia (Modro jezero, Paklenica, Psunj) and one location past the State border in Bosnia and Herzegovina (Cerovica). Modro jezero is a small lake at the bottom of a sinkhole located in Dalmatian mainland, with ample fluctuations in water depth. Its south-eastern banks, characterized by steep gravel slopes, are adjacent to the residential area of the town of Imotski. Paklenica is also located in Dalmatia, about 170 km North West of Modro jezero. Paklenica National Park, where spontaneous grapes were found, is a coastal mountain area characterized by a deep canyon through which the homonymous river flows, flanked by banks with lush vegetation. Psunj is a mountain in Continental Croatia about 190 km North East of Paklenica and represents a forest habitat. Cerovica is the southernmost location in a karst mountain area characterized by a lack of natural water reservoirs, approximately 100 Km South East of Modro jezero. A list of accessions with their sample and VIVC codes, sampling location, berry colour and main use of the grapes are provided in Supplementary Table S1.

Genetic diversity and identification of population structure

Analyses were conducted using 516,101 SNPs from the dataset of this paper and 224,695 SNPs in common between the dataset of this paper and a WGS dataset from Magris et al. [13], which included a diversity panel of V. vinifera genotypes and samples of V. sylvestris from the ‘Ketsch’ population in Germany (from now on WGS diversity panel). Principal component analysis (PCA) was performed using SNPrelate. Analysis of population structure was performed using the ADMIXTURE software [34]. The best K value was determined by the lowest cross-validation error for each K value (2–15) to determine the most appropriate number of ancestry components. Split and migration events between groups and f3 test were calculated using the TREEMIX software [35]. The groups of cultivated accessions for TREEMIX analysis were generated using the output of ADMIXTURE K = 3 groups. Accessions with one ancestry component higher than 0.85 were assigned to either of the two vinifera groups. The remaining ones were assigned to a vinifera admixed group. The TREEMIX groups of spontaneous grapevines corresponded, the first one, to all individuals from the four sampling locations but one individual and, the other one, to the outlier individual from the Modro jezero population (Sy10). The variance of relatedness between groups explained by each model was calculated using RADpipe [GitHub RADpipe repository, doi: 10.5281/zenodo.17809].

Parentage analysis and kinship network

Only cultivated accessions (108) were considered for parentage analysis. Thresholds of identity‒by‒state (IBS) ratio and genotypic distance were used to estimate pairwise identity-by-descent (IBD) in each genomic window of 100 Kb of non‒repetitive DNA. IBS ratio and genotypic distance was calculated as described in Magris et al. [13]. Parentage analysis was performed based on the IBD cumulative length and segment length distribution of the IBD0, IBD1 and IBD2 windows. The following parameters were used to determine parent-offspring relationship: IBD0 < 15%, IBD1 > 60% of genome length, longest IBD0 segment length < 3.8 Mb.

For each hypothetical trio, we considered biallelic SNPs that were informative in all 3 individuals. If P1×P2→F1 were the individuals under comparison in the reported direction of the relationship and “a” and “b” were the reference and alternative alleles, the following genotypic combinations were compatible with the hypothesis: aa×aa→aa, aa×ab→aa, aa×ab→ab, aa×bb -> ab, ab×aa→aa, ab×aa→ab, ab×ab→aa, ab×ab→ab, ab×ab→bb, ab×bb→ab, ab×bb→bb, bb×aa→ab, bb×ab→ab, bb×ab→bb, bb×bb→bb. On the contrary, the following genotypic combinations were incompatible with the hypothesis: aa×aa→ab, aa×aa→bb, aa×ab→bb, aa×bb→aa, aa×bb→bb, ab×aa→bb, ab×bb→aa, bb×aa→aa, bb×aa→bb, bb×ab→aa, bb×bb→aa, bb×bb→ab. Shortly, alleles in F1 that were not shared with P1 had to be shared with P2, and vice versa. Sites that did not meet this requirement rejected the hypothesis. As hemizygous DNA is present in each parental genome and some heterozygous sites can be erroneously called homozygous due to insufficient read coverage, a background noise of unmatching sites is expected in the offspring, which we modelled in terms of number and chromosomal distribution using known genuine trios (Supplementary Figs. 2–14). The false rate of Mendelian inconsistencies was identified in a percentage of 0.8% of all informative sites. The parental combination for ‘Krstičevica’ showed a slightly higher false rate of 1.17% Mendelian inconsistencies likely due to fewer informative sites in both parents compared to other comparisons, but the distribution of unmatching SNPs across the genome met the requirement of being randomly distributed. Full-sibling relationships were inferred from the comparison of the resolved trios. The information on parent-offspring pairs, complete parent-offspring trios and full-siblings was summarized graphically with a kinship network. The network was generated in R using the “igraph” package [36, 37].

Phenotypic data and GWAS

Phenotyping was performed in a common garden experiment using a reduced panel of 84 accessions, from which highly consanguineous individuals were removed. Two-bud cuttings of 70 cultivated and 14 spontaneous accessions were self-rooted in a perlite-based medium. The rooted cuttings were transplanted into 5-litre potted medium, consisting of a mixture of natural soil, humus, quartz sand and perlite, and grown in a greenhouse at the Institute for Adriatic Crops and Karst Reclamation, Split, Croatia, without supplemental lighting and heating. Each genotype was represented by six replicates. Vines were pruned back to two buds during winter dormancy. Visual scoring of leaf hairiness was performed during the second annual cycle of vegetative growth based on OIV descriptors [38]. The list of accessions with the corresponding OIV descriptors are provided in Supplementary Table S2. The density of prostate hairs between the main veins on the lower side of the leaf blade (OIV 084) was visually assessed using 5 categorical values. Ten mature leaves from the middle third of six shoots per genotype were examined, as described in 2nd Edition of the OIV Descriptor List for Grape Varieties and Vitis Species [38]. In order to have a positive control for GWAS, ensuring that sample size, population structure and kinship were neither limiting factors and nor sources of false positives, we also scored berry colour. GWAS was performed using the Efficient Mixed-Model Association eXpedited (EMMAX) algorithm [39, 40] and 306,965 SNPs with a MAF higher than 0.01. A kinship matrix was used for correcting for kinship. The P-value of individual SNPs was adjusted for multiple testing using the Benjamini-Hochberg correction. Statistically significant thresholds were determined using the Bonferroni method with α = 0.05. Q-Q plots and Manhattan plots were generated using the qqman library in R.

Genetic diversity, population structure and ancestry components

The first two components of the PCA explained 8.8% of the genotypic variation in 149 accessions (Fig. 1A). The first component showed a clear separation between Croatian V. sylvestris and locally grown V. vinifera along the x-axis of the bidimensional plot. One single spontaneous accession (Sy 10) from the Modro jezero population laid in an intermediate position along PC1 nearby widespread cultivars with known high European sylvestris ancestry such as ‘Cabernet Sauvignon’, ‘Chasselas Blanc’ and ‘Welschriesling’, the latter two of which are locally grown in Croatia under the synonymous of ‘Plemenka bijela’ and ‘Graševina’, and the Dalmatian cultivar ‘Divjaka’. It is therefore possible that Sy 10 is either an abandoned local variety or a seedling that were involuntarily introduced into the anthropic environment of Modro jezero or a feral accession deriving from crop-to-wild gene flow in the once wild population of Modro jezero. The cultivated germplasm in Croatia laid in a confined part of the bidimensional space that is delimited by well-known grapevines, which are cultivated in Croatia and beyond in Europe. The few exceptions are represented by three table grapes ‘Krivaja Bijela’, ‘Krivaja Crvena’ and ‘Mijajuša’, the latter of which is a synonym of the popular Greek/Levantine cultivar ‘Xeromachairouda’ [41] also known as ‘Asswad Karech’ [42]. In order to rescale the diversity in Croatia relative to the global grapevine diversity, we generated a PCA using 224,695 common variant sites between the sample set of this paper and a WGS diversity panel. The bidimensional space in Fig. 1B shows that grapevine diversity in Croatia is relatively rich, local sylvestris diversity and diversification are relatively small and low compared to other European wild populations, and Croatian natural sites may contain genuine sylvestris and sylvestris-vinifera admixed individuals. A number of individuals from all but one natural site laid, indeed, midway in the bidimensional space between the sylvestris and sativa distributions and colocalized with KE-06, a known sylvestris-vinifera hybrid from a German natural site.

To substantiate the findings inferred from the PCA, we analysed population structure using a model-based clustering approach that simultaneously estimates population allele frequencies along with ancestry proportions, implemented in the ADMIXTURE software [34]. The cross-validation error indicated that three ancestry components (K = 3) contributed most likely to the genome composition of the analysed germplasm (Supplementary Table S3). According to ADMIXTURE, one ancestry component, which we refer hereafter to as sylvestris, is predominant in natural sites and is present as a minor contributor (< 0.2) in a few local cultivars (Fig. 2). ADMIXTURE confirmed that ‘Cabernet Sauvignon’, ‘Plemenka Bijela’, ‘Graševina’, which is widely grown in Continental Croatia, and ‘Divjaka’, which is grown exclusively in the Pelješac peninsula, contain a relevant fraction of this ancestry component. Natural sites are home to individuals with pure sylvestris ancestry but also to individuals with varying proportions of sylvestris-vinifera admixed ancestry. In all but one of the cases, sylvestris ancestry was largely predominant and the vinifera ancestry component contributed for a fraction from 0.63 to 1 in each individual genome, with a mean of 0.95 and a median of 1 considering the whole sample set. The most extreme case is represented by the individual Sy10 from Modro jezero that showed a predominant 0.65 vinifera ancestry component and is part of the peculiarity of the Modro jezero spontaneous population, wherein all individuals show some degree of admixture. The two ancestry components that, cumulatively, are predominant in the individual genomes of all cultivars, which we refer hereafter to as vinifera, contribute almost equally to the global composition of the cultivated germplasm analysed in this paper (Fig. 2). One vinifera ancestry component is associated with cultivars typically grown in Continental Croatia, northern Adriatic Croatia (Kvarner and Istrian peninsula) and it makes up the entirety of the genomes of the popular cultivars ‘Heunisch Weiss’ (syn. 'Belina Starohrvatska') and ‘Blank Blauer’ (syn. ‘Bljuzgavac’ and ‘Vulpea’). The other vinifera ancestry component is associated with cultivars typically grown in Dalmatia (i.e. central and southern Adriatic Croatia) and it makes up the entirety of the genomes of the cultivars ‘Plavac Mali’ and ‘Bombino Bianco’. The majority of Croatian cultivars showed admixed proportions of these two vinifera ancestry components. Among the most remarkable ones, ‘Malvazija Istarska’ showed approximately three quarters of Continental/Istrian vinifera ancestry component and one quarter of Dalmatian vinifera ancestry component. ‘Tribidrag’ showed the opposite condition.

Origin of admixed ancestries and genealogical relationships

PCA and ADMIXTURE suggested that spontaneous populations in natural sites are in part composed of admixed individuals. The results of ADMIXTURE also conveyed a sense of strong kinship influencing the population structure of the cultivated germplasm. To address these points, we investigated the introgression events that have generated admixed individuals, using the algorithm developed by Pickrell and Pritchard and implemented in the TREEMIX software [35], and searched for complete parent-offspring trios in all possible combinations and permutations of the analysed accessions.

Figure 3 shows the TREEMIX models using a block size of 200 SNPs that explained from 98.3–99.6% of the variance of relatedness among groups, with increasing numbers of migration events. Tree topologies confirm that the spontaneous populations in more remote forest environments in Continental Croatia and in Bosnia Herzegovina are the least related ones to local cultivars, while the spontaneous population in Dalmatia from the site that is most subject to the effects of anthropization on the banks of Modro jezero show signatures of vinifera introgression. The combination of evidence from ADMIXTURE and TREEMIX suggests that at least part of the spontaneous population from the Paklenica National Park have also received some gene flow from vinifera. The f3 test confirmed that the spontaneous populations of Modro jezero and Paklenica National Park are compatible with being the result of admixture between more genuine sylvestris populations, similar to those found at Cerovica and Psunj, and cultivated germplasm.

Fourteen complete parent-offspring trios were identified in the dataset of this paper (Table 1, Supplementary Figs. 1–14). ‘Heunisch Weiss’ (syn. 'Belina Starohrvatska'), ‘Blank Blauer’ (syn. ‘Bljuzgavac’ and ‘Vulpea’), ‘Plavac Mali’ and ‘Bombino Bianco’ were recurrent parents in 11 out of 14 trios. The pair 'Belina Starohrvatska' and 'Bljuzgavac' parented 4 cultivars. One of these full-siblings is the cultivar ‘Surina’, which is reported here for the first time along with 'Svjetljak’, ‘Ranfol’, and ‘Plavec žuti’. The comparison based on IBD segment length of four full-siblings is graphically shown in Supplementary Figs. 15–21. The pair 'Plavac mali' and 'Bombino Bianco' parented two cultivars, i.e. the full-siblings ‘Ninčuša’ and ‘Ljutun’. A total of 72 parent-offspring pairwise relationships were identified in the dataset of this paper (Supplementary Table S4). All these relationships are summarised graphically in the kinship network of Fig. 4.

As a result of the presence of complete trios and many other unoriented parent-offspring pairwise relationships, 23 cultivars are linked to 'Belina Starohrvatska' and 'Bljuzgavac' through a chain of parent-offspring relationships and 36 cultivars are linked to both 'Plavac Mali' and 'Bombino Bianco'. The two kinship groups, one named after its founders and referred to as 'Belina Starohrvatska' and 'Bljuzgavac' and the other one referred to as 'Plavac Mali' and 'Bombino Bianco' represent collectively 56% of the analyzed cultivars. Consanguinity within each group and between groups is furtherly exacerbated by the fact that these four founders are themselves highly consanguineous. 'Belina Starohrvatska', 'Bljuzgavac', 'Plavac Mali' and 'Bombino Bianco' share identical by descent haplotypes across between 23.8% and 31.7% of their diploid genome length in all 6 possible pairwise comparisons. In particular, the members of the pair 'Belina Starohrvatska' and 'Bljuzgavac' and the members of the pair 'Plavac Mali' and 'Bombino Bianco' share by descent 31.7% and 29.3%, respectively, of their diploid genome length.

Table 1

List of complete parent-offspring trios reporting the percentage of Mendelian inconsistencies, the number of informative sites and previous references reporting the parentage.
Offspring	Parent 1	Parent 2	Mendelian error	Informative sites	Previous reference
Ljutun (HR24)	Bombino bianco (HR44)	Plavac mali (HR33)	0.66%	387,242	[17]
Ninčuša (HR32)	Bombino bianco (HR44)	Plavac mali (HR33)	0.67%	378,126	[22]
Gegić (HR41)	Bombino bianco (HR44)	Bilina privlačka (AF15)	0.66%	430,842	[17]
Krstičevica (HR67)	Bombino bianco (HR44)	Plavina (HR28)	1.17%	266,891	[17]
Debit (HR64)	Bombino bianco (HR44)	Lasina (HR23)	0.64%	427,447	[22]
Kurtelaška (AF24)	Bombino bianco (HR44)	Maraština omiš (HR19)	0.53%	411,318	[22]
Ranfol (AF18)	Belina Starohrvatska (AF13) (Heunisch Weiss)	Bljuzgavac (AF21) (Blank blauer)	0.70%	285,861	[22]
Surina (IPTPO08)	Belina Starohrvatska (AF13) (Heunisch Weiss)	Bljuzgavac (AF21) (Blank blauer)	0.66%	287,812	this paper
Plavec žuti (AF12)	Belina Starohrvatska (AF13) (Heunisch Weiss)	Bljuzgavac (AF21) (Blank blauer)	0.71%	287,823	[17]
Svjetljak (AF32)	Belina Starohrvatska (AF13) (Heunisch Weiss)	Bljuzgavac (AF21) (Blank blauer)	0.71%	287,881	[17]
Mejsko bijelo (AF34)	Duranija (IPTPO07)	Žumić (AF05)	0.73%	225,938	[17]
Glavinuša (HR18)	Vugava vis (HR37)	Plavac mali (HR33)	0.55%	397,147	[22]
Dolcin (IPTPO10)	Rukatac (HR01) (Malvasia bianca lunga)	Verdić (AF19)	0.51%	481,631	[57, 58]
Pošip obični (HR05)	Zlatarica blatska (HR61)	Bratkovina blatska (HR60)	0.66%	428,187	[59]

While Mendelian inconsistencies have low frequencies and are randomly distributed across the genome in bona fide parent-offspring trios, their incidence is higher and their chromosomal distribution is uneven for the proposed parent-offspring trio that was claimed for explaining the origin of ‘Plavac Mali’ [43]. ‘Plavac Mali’ is the most widely planted red cultivar in Dalmatia and its pedigree was particularly intriguing because it seemed to result from a cross between the two local cultivars ‘Tribidrag’ and ‘Dobričić’ based on microsatellite DNA allele sizes. In the present study, IBD analysis of the proposed parentage of ‘Plavac Mali’ showed 1.7% Mendelian inconsistencies. Unmatching SNPs densely clustered on several chromosomal regions, thereby representing intervals of IBD0 that were incompatible with their origin from either ‘Tribidrag’ or ‘Dobričić’ (Fig. 5A). We validated this finding using WGS data that became publicly available recently (Fig. 5B). A percentage of 1.2% of the 472,828 SNPs that were informative in all three individuals with respect of the reference genome revealed Mendelian inconsistencies across the same chromosomal regions shown by reduced-representation genome sequencing data. A higher saturation of variant sites and a lower background noise from WGS data compared to reduced-representation genome sequencing data, allowed us to estimate data that 19 chromosomal regions in ‘Plavac Mali’, involving 12 out of 19 chromosomes and covering cumulatively 23.3% of haploid genome, are incompatible with being inherited from either ‘Tribidrag’ or ‘Dobričić’. While we could confirm that ‘Plavac Mali’ has a parent-offspring relationship with both ‘Tribidrag’ and ‘Dobričić’, we can exclude that both ‘Tribidrag’ and ‘Dobričić’ are the parents of ‘Plavac Mali’.

‘Plavac Mali’ showed 8 unoriented parent-offspring relationships in our dataset (Fig. 4), beside those with ‘Tribidrag’, ‘Dobričić’, and three cultivars for which the direction of the relationship was resolved. We therefore tested each individual of those 8 ones (Fig. 4) and their compatibility with being the parent of ‘Plavac Mali’ in a parental combination with either ‘Tribidrag’ or ‘Dobričić’. None of these combinations could explain the origin of ‘Plavac Mali’ (Supplementary Table S5). In a similar way, we tested each individual of the 5 ones that showed an unoriented parent-offspring relationships with ‘Tribidrag’ (Fig. 4), their compatibility with being the parent of ‘Tribidrag’ in a parental combination with ‘Plavac Mali’. None of these combinations could explain the origin of ‘Tribidrag’.

Genome wide association study results

Berry colour

The Manhattan plot in Fig. 6A shows 143 SNPs above the significance threshold that are associated with berry colour, 140 of which are located on chromosome 2. Of these (Supplementary Table S6), 139 SNPs span an interval of approximately 8.4 Mbp, which is centered on the array of MybA genes that is known to encode two transcription factors (MybA1 and MybA2, Fig. 6C) controlling the expression of a key structural gene for anthocyanin biosynthesis [44]. One SNP outside that interval is located at chr2:7,159,705 in the VIT_00013173001 gene sequence, which has been annotated as a BSK2-like serine/threonine protein kinase. Three associated SNPs are located on chromosome 12 of the reference genome in a short interval that is highly similar in nucleotide sequence to an interval on chromosome 2 that is located at 16.7 Mbp, within the region shown in Fig. 6C, where several associated SNPs are in linkage disequilibrium with the MybA genes. The latter may be considered as a kind of false association, because a true positive SNP-trait association revealed by the reads is displaced from the correct genomic location due to misalignments on the reference genome.

Leaf hairiness

Two SNPs on chromosome 5 were significantly associated with the density of prostate hairs between the main veins on the abaxial leaf lamina (Fig. 6B). This trait is also known in viticulture as ampelographic descriptor OIV 084 [38]. The first SNP (A/T) is located at chr5:1,424,308, in an exon of the gene VIT_05s0077g01800 that translates into an amino acid sequence with no predicted function and no known protein domains, according to NCBI and InterPro databases [45]. The second SNP (T/A) is located at chr5:1,525,871 in an exon of the gene VIT_05s0077g01940 that is predicted to encode an IBR3 acyl-CoA dehydrogenase, which oxidizes Indole-3-butyric acid (IBA) into indole-3-acetic acid (IAA) in auxin biosynthesis [46]. However, given the extent of linkage disequilibrium in our dataset, it is unlikely that exactly those genes may have a role in explaining the observed phenotypic variation. Figure 6D shows, indeed, an interval of approximately 1 Mbp that is likely to contain the causal factor. No other OIV descriptor showed statistically significant associations.

as leaf hairiness in the cultivated compartment might be an adaptive trait, we sought to investigate possible relations between leaf hairiness and ancestry components. When we sorted cultivars in categorical phenotypes (i.e. the only one assigned to OIV 084 class 9 was merged to those assigned to class 7 in the following analysis), we did not detect differences in the distribution of vinifera ancestry components among hairiness categories, but we found an inverse trend of decreasing sylvestris ancestry component as hairiness increased (Fig. 7). In particular, the distributions of the sylvestris ancestry component were significantly different (p = 0.041) at a two-sided Wilcoxon test between cultivars with hairiness class 3 and cultivars with hairiness classes 7–9.

There is a long-lasting debate in viticulture as to what extent local grapevines are really autochthonous and how much they differ in ancestry and genealogy from those in neighboring countries. The results of the present study add food for through about these points.

As to the first point, taking for granted that grapevine domestication has not occurred in Europe and domesticated germplasm was introduced from elsewhere, which is not challenged by the results of this study, the term autochthonous in a strict sense implies that indigenous ancestry in European cultivars may only derive from introgression from indigenous populations of sylvestris. In this study, Croatian cultivars did not show evident genomic signatures of local sylvestris introgression, in agreement with earlier studies [30]. Western Europe cultivars, such as ‘Plemenka bijela’ (syn. ‘Chasselas Blanc’) and ‘Cabernet Sauvignon’, and Central-Eastern Europe cultivars, such as ‘Graševina’ (syn. ‘Welschriesling’), which were introduced in Croatia from elsewhere, have higher sylvestris ancestry than local germplasm, corroborating previous findings [16, 31]. The only exception is represented by the cultivar ‘Divjaka’, which could have, however, either inherited some sylvestris ancestry through Western Europe cultivars or being the direct result of a recent hybridization with local sylvestris. Our analysis could not resolve this level of detail. In a broader sense, the term autochthonous is also used to recognize that a cultivar (i.e. precisely the seedling) originated in the geographical area where it is currently grown, no matter the ancestry and the origin of their parents. In this respect, we found two large descent groups among local cultivars, one originating from ‘Belina Starohrvatska’ (syn. 'Heunisch Weiss') and/or ‘Bljuzgavac’ (syn. ‘Blank Blauer’ and ‘Vulpea’), the other one originating from ‘Plavac Mali’ and ‘Bombino Bianco’. The fact the so many seedlings of these progenitors are present only locally makes it more likely that at least part of them originated in this geographical area rather than they were all introduced from other areas from which they later disappeared. This hypothesis is reinforced by the evidence that descendants of ‘Belina Starohrvatska’ and/or ‘Bljuzgavac’ are typically grown in Continental Croatia, northern Adriatic Croatia (Kvarner and Istrian peninsula), where their parents are also found, and descendants of ‘Plavac Mali’ are typically grown in Dalmatia, along with their parent. An exception is represented by descendants of ‘Bombino Bianco’ that are typically grown in Dalmatia, but the parent is not cultivated in Dalmatia and is only known in the Istrian peninsula under the synonym of ‘Trevolina Istriana’ [47].

This brings us into the second point of discussion: how grapevine diversity is hierarchically interconnected country-wise and Europe-wise. It comes as no surprise that 'Heunisch Weiss' has several descendants and relatives in Croatia, as this genotype has by far the highest number of parent-offspring relationships across the Balkans, Northeastern Italy, Eastern and Central Europe. No wonder either that ‘Blank Blauer’/’Vulpea’ is a common parent in Continental Croatia, northern Adriatic Croatia (Kvarner and Istrian peninsula), as it has several parent-offspring relationships with grapevines in Austria, in Hungary and in the northern Adriatic shores of Northeastern Italy [27], including the cultivar ‘Glera’ that is used therein for the production of the sparkling Prosecco wine. It is also noteworthy that one of the parent-offspring relationships of ‘Blank Blauer’/’Vulpea’, ‘Visparola’, is itself a major progenitor of cultivars in Central Italy. ‘Bombino Bianco’ deserves more attention. In Dalmatia, there are no historical records of ‘Bombino Bianco' [48], where it was once mistaken for the cultivar ‘Debit’ (syn. ‘Puljižanac’). ‘Bombino Bianco’ is extensively grown on the Adriatic shores of Central Italy, where is also known under the synonym of ‘Passerina’. D'Onofrio et al. [27] discovered that 'Bombino Bianco' has a parent-offspring relationship with 11 Italian cultivars, in addition to the 11 Croatian cultivars identified in our study. It remains therefore impossible to infer from historical and genetic data whether ‘Bombino Bianco’ was a popular cultivar once grown across all Adriatic shores and it gave birth to descendants everywhere, or its descendants were introduced across the Adriatic from a smaller area where they originated, which is not necessarily corresponding to the present-day distribution of ‘Bombino Bianco’.

Geographically, there is a common and narrow genetic basis of the germplasm today found in the Adriatic Sea Basin, which extends northwards into Central Europe and eastwards into the Danube Basin. Nonetheless and in contrast with the general rule of high consanguinity, it was striking to notice that no parent-offspring relationship was found between the two most widely grown Croatian white cultivars, ‘Graševina’ and ‘Malvazija Istarska’, and the rest of Croatian germplasm, although these genotypes share the same ancestry components with the rest of the germplasm. As for ‘Graševina’, also known as ‘Welschriesling’, no parent-offspring relationship was found either in the panel of this article or in WGS panels from [13] and [1], including collectively more than 3,500 accessions. So far, only one parent-offspring relationship has been reported with the Italian cultivar ‘Orsolina’ [27]. As for ‘Malvazija Istarska’, the only known parent-offspring relationship is represented by ‘Vega’, which was selected by the breeder Giovanni Dalmasso from an intentional cross between ‘Malvazija Istarska’ and ‘Furmint’ [27, 49]. These unique genotypic combinations, which may derive from extinct parents, enrich diversity by reassorting common ancestries into novel allelic combinations. PCA results are indeed coherent with the assumptions that cultivars in the Western Balkans are a relevant part of the existing diversity in Europe [13, 15, 17, 26], some of them played an important historical role in the generation of descent groups in other parts of Europe, and a few of them are singular interesting genotypic combinations.

The origin of the most widely grown Croatian red cultivar, ‘Plavac Mali’ deserves a separate discussion. While we conclusively exclude that ‘Plavac Mali’ is the offspring of the cross between 'Tribidrag' and 'Dobričić' [17, 22] and the result of other combinations reported in the results section, all other scenarios and speculations are possible. Considering the historical records that mention 'Tribidrag' as early as the 15th century [50, 51], it is more likely that 'Tribidrag' predates ‘Plavac Mali’ rather than vice versa. If that were true, it would be consequent to assume that 'Tribidrag' is one of the parents of ‘Plavac Mali’, and 'Dobričić' is an offspring of 'Plavac Mali'. The fact that 'Dobričić' is only present on the small island of Šolta, off the Central Dalmatian coast, and the fact that it has no parent-offspring relation with known cultivars other than 'Plavac Mali' are in favour of a distal position occupied by 'Dobričić' on the family tree. The opposite hypothesis should invoke that 'Dobričić' is ancestral to both 'Tribidrag' and ‘Plavac Mali’ and has progressively disappeared from viticulture, while 'Plavac Mali' emerged long after its presumed offspring 'Tribidrag' became widely known. However, caution should be used with these interpretations. Although the century-old cultivation of ‘Tribidrag’ was confirmed with microsatellite analysis of a 90-year-old herbarium [52], more ancient historical records may have associated the name 'Tribidrag' with one or more different genotypes that were not corresponding to 'Tribidrag' as we know it nowadays. Similar cases of uncertain direction in parent-offspring relationships for important regional cultivars has led to considerable debate in the past, which is well illustrated by the case of 'Sangiovese'. Initially, two hypotheses were proposed based on SSR alleles [53, 54] that consistently indicated ‘Ciliegiolo’ as one of the biological parents, until trio analysis based on SNP markers [27] clarified that ‘Ciliegiolo’ is an offspring and not a parent of ‘Sangiovese’.

There is also a lot of uncertainty about the status of the present-day populations of European sylvestris and their relations with locally grown cultivars. This indeterminateness is due to the fact that, when signs of admixture are found in both compartments, the direction of the gene flow is hard to ascertain in presence of very small population sizes on one side and lack of random mating on the other side. Data from the present article suggest a crop-to-wild gene flow, in particular into the spontaneous population at Modro jezero. This is compatible with the presence of neighbouring vineyards and the long history of grape cultivation in the area, as well as with more anthropization at around that site compared to the other sites analysed in this study. While other spontaneous populations, in particular those at Cerovica and Psunj, seem to be less affected by this issue, we cannot prove that they are completely exempt from some level of recent or ancient gene flow. Out of the 41 accessions sampled from natural sites, 35 have more than 0.85 sylvestris, but when samples of German sylvestris are included in the analysis only 14 Croatian samples maintain a predicted sylvestris ancestry proportion greater than 0.85, and none of them has a predicted sylvestris ancestry proportion greater than 0.95. In the most recent comprehensive study by Dong et al. [1], Croatian sylvestris was predominantly categorised as admixed vinifera-sylvestris, with the exception of three individuals.

Reduced-representation genome sequencing was an effective approach in finding significant SNP-trait associations in a GWAS application. The advantage of the SPET method, compared to other GBS approaches, resides in the fact that it can be targeted to the gene space, investigating portions of the transcriptional unit from all predicted genes. In this study, we successfully identified 2 SNPs associated with the density of prostate hairs on the abaxial side of the leaf blade on the upper subtelomeric end of chromosome 5. Two QTL mapping studies consistently identified a QTL for leaf hair density across the same region [55, 56]. The latter study also proposed a functional candidate gene in the QTL region (VIT_05s0077g01390), based on its predicted homology with GLABROUS INFLORESCENCE STEMS transcription factors that control trichome cell development in Arabidopsis. VIT_05s0077g01390 is located approximately 295 Kb upstream of our associated SNPs (Fig. 6D). The GWAS results suggest that it is possible to identify reliable associations between sparse SNPs across large regions and traits, thanks to high linkage disequilibrium in the cultivated compartment. A higher SNP density, which can be generated only through whole-genome sequencing, is, however, needed in germplasm panels in which linkage disequilibrium decays more rapidly and for narrowing genetic intervals down to single gene candidates.

The genetic separation between Croatian cultivars and sylvestris excludes that cultivars originated from independent events of local domestication. On the other hand, the evidence of a crop-to-wild gene flow, especially in but not limited to the Modro jezero population, raises an alarm on the genetic integrity of natural populations of the wild progenitor. The results of study show country-wise and Europe-wise interconnections of the Croatian grapevine germplasm and stimulate the incorporation of local spontaneous genetic resources into breeding programs for introduction adaptation and expanding diversity.

SNP – single nucleotide polymorphism; QTL – quantitative trait loci; SSR – simple sequence repeat; NGS – next generation sequencing; GBS – genotyping by sequencing; SPET – single primer enrichment technology; WGS – whole genome sequencing; GWAS – genome wide association study; IBS – identity by state; IBD – identical by decent; PCA – principal component analysis; EMMAX – Efficient Mixed-Model Association eXpedited

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Availability of data and materials

Raw reads are deposited in the NCBI Sequence Read Archive under the BioProject number PRJNA1126049. WGS raw reads were retrieved from the BioProject numbers PRJNA373967, PRJNA390884, PRJNA385116, PRJNA321480 and from the Genome Warehouse in the National Genomics Data Center, China National Center for Bioinformation accession number CRA006917. The plant material of the 108 cultivars used in this study is held at three Croatian germplasm repositories: Institute of Adriatic Crops and Karst Reclamation, Split, institute code HRV048; Institute of Agriculture and Tourism, Poreč, institute code HRV050; University of Zagreb, institute code HRV041.

Additional file 1: Supplementary tables S1 to S6

Additional file 2: Supplementary figures 1 to 21

Competing interests

The authors declare that they have no competing interests

Funding

This publication was financially supported by the project “Assessment of drought tolerance of Croatian grapevine germplasm-TOLVIN” (grant number: KK.05.1.1.02.0010) funded by the European Regional Development Fund (ERDF) and by the Croatian Science Foundation under the project “Young researchers’ career development—training new doctoral students”.

Authors' contributions

L.M. collected phenotypic data and together with G.M., G.D.G. and M.M. performed genetic analyses. L.M., G.D.G, and G.Z. drafted the manuscript. E.M., M.B., I.P. and G.Z. conceived the study. All authors participated in the interpretation of the results and contributed to the final version of the manuscript.

Acknowledgements

Authors' information (optional)

Dong Y, Duan S, Xia Q, Liang Z, Dong X, Margaryan K, et al. Dual domestications and origin of traits in grapevine evolution. Science. 2023. Vol. 379 (6635):892-901. doi:10.1126/science.add8655
Laucou V, Lacombe T, Dechesne F, Siret R, Bruno JP, Dessup M, et al. High throughput analysis of grape genetic diversity as a tool for germplasm collection management. Theor.and App. Genet. 2011. 122(6), 1233–1245. doi:10.1007/s00122-010-1527-y
Aradhya MK, Dangl GS, Prins BH, Boursiquot JM, Walker MA, Meredith CP, et al. Genetic structure and differentiation in cultivated grape, Vitis vinifera L. Genet. Res. 2003. 81(3):179-192. doi:10.1017/S0016672303006177
Riaz S, Boursiquot JM, Dangl GS, Lacombe T, Laucou V, Tenscher AC, et al. Identification of mildew resistance in wild and cultivated Central Asian grape germplasm. BMC Plant Biol. 2013. 13(1):149. doi:10.1186/1471-2229-13-149
Coleman C, Copetti D, Cipriani G, Hoffmann S, Kozma P, Kovács L, et al. The powdery mildew resistance gene REN1 co-segregates with an NBS-LRR gene cluster in two Central Asian grapevines. BMC Genet. 2009. 10:89 doi:10.1186/1471-2156-10-89
Lukšić K, Zdunić G, Hančević K, Žulj Mihaljević M, Mucalo A, Maul E, et al. Identification of powdery mildew resistance in wild grapevine (Vitis vinifera subsp. sylvestris Gmel Hegi) from Croatia and Bosnia and Herzegovina. Sci. Rep. 2022. 12:2128. https://doi.org/10.1038/s41598-022-06037-6
Di Gaspero G, Cattonaro F. Application of genomics to grapevine improvement. Aust. J. Grape Wine Res. 2010. 16:122–130. doi:10.1111/j.1755-0238.2009.00072.x
Daldoul S, Hanzouli F, Hamdi Z, Chenenaoui S, Wetzel T, Nick P, et al. The root transcriptome dynamics reveals new valuable insights in the salt-resilience mechanism of wild grapevine (Vitis vinifera subsp. sylvestris). Front. Plant Sci. 2022. 13: 1664-462X. doi: 10.3389/fpls.2022.1077710
Marrano A, Micheletti D, Lorenzi S, Neale D, Grando MS. Genomic signatures of different adaptations to environmental stimuli between wild and cultivated Vitis vinifera L. Hortic. Res. 2018. 5(1):34 doi: 10.1038/s41438-018-0041-2
Campisano A, Pancher M, Puopolo G, Puddu A, Lòpez-Fernàndez S, Biagini B, et al. Diversity in endophyte populations reveals functional and taxonomic diversity between wild and domesticated grapevines. Am. J Enol. Vitic. 2015. 66(1):12–21. doi:10.5344/ajev.2014.14046
Xiao H, Liu Z, Wang N, Long Q, Cao S, Huang G, et al. Adaptive and maladaptive introgression in grapevine domestication. Proc. Natl. Acad. Sci. U S A. 2023. 120(24): e2222041120. doi: 10.1073/pnas.2222041120
Xiao H, Liu Z, Wang N, Gaut BS, Zhou Y. Reply to Blanco-Pastor: Introgression and heterozygosity complicated grapevine domestication. Proc. Natl. Acad. Sci. U S A. 2023. 120(49): e2314162120. doi: 10.1073/pnas.2314162120
Magris G, Jurman I, Fornasiero A, Paparelli E, Schwope R, Marroni F, et al. The genomes of 204 Vitis vinifera accessions reveal the origin of European wine grapes. Nat. Commun. 2021. 12, 7240. https://doi.org/10.1038/s41467-021-27487-y
Arroyo-García R, Ruiz-García L, Bolling L, Ocete R, López MA, Arnold C, et al. Multiple origins of cultivated grapevine (Vitis vinifera L. ssp. sativa) based on chloroplast DNA polymorphisms. Mol. Ecol. 2006. 15(12):3707–3714. doi:10.1111/j.1365-294x.2006.03049.x
Bacilieri R, Lacombe T, Le Cunff L, Di Vecchi-Staraz M, Laucou V, Genna B, et al. Genetic structure in cultivated grapevines is linked to geography and human selection. BMC Plant Biol. 2013. 13(1):25. doi:10.1186/1471-2229-13-25
Myles S, Boyko AR, Owens CL, Brown PJ, Grassi F, Aradhya MK, et al. Genetic structure and domestication history of the grape. Proc. Natl. Acad. Sci. U S A. 2011. 108(9):3530–3535. doi:10.1073/pnas.1009363108
Žulj Mihaljević M, Maletić E, Preiner D, Zdunić G, Bubola M, Zyprian E, et al. Genetic diversity, population structure, and parentage analysis of Croatian grapevine germplasm. Genes. 2020. 11(7):1–35. doi:10.3390/genes11070737
Maletić E, Pejić I, Karoglan Kontić J, Zdunić G, Preiner D, Šimon S, et al. Ampelographic and genetic characterization of Croatian grapevine varieties. Vitis. 2015. 54:93–98. doi: https://doi.org/10.5073/vitis.2015.54.special-issue.93-98
Marsal G, Mateo-Sanz JM, Canals JM, Zamora F, Fort F. SSR analysis of 338 accessions planted in Penedès (Spain) reveals 28 unreported molecular profiles of Vitis vinifera L. Am. J. Enol. Vitic. 2016. 67(4):466–70. doi:10.5344/ajev.2016.16013
Oualkadi A El, Ater M, Messaoudi Z, Laucou V, Boursiquot JM, Lacombe T, et al. Molecular characterization of Moroccan grapevine germplasm using SSR markers for the establishment of a reference collection. Jour. Int. Sci. Vigne Vin. 2009. 43(3):135-148. doi:10.20870/oeno-one.2009.43.3.797
Lopes MS, Sefc KM, Eiras Dias E, Steinkellner H, Laimer da Câmara Machado M, da Câmara Machado A. The use of microsatellites for germplasm management in a Portuguese grapevine collection. Theor. Appl. Genet. 1999. 99:733–739. doi:10.1007/s001220051291
Lacombe T, Boursiquot JM, Laucou V, Di Vecchi-Staraz M, Péros JP, This P. Large-scale parentage analysis in an extended set of grapevine cultivars (Vitis vinifera L.). Theor. App. Genet. 2013. 126(2):401–414. doi:10.1007/s00122-012-1988-2
Jaillon O, Aury JM, Noel B, Policriti A, Clepet C, Casagrande A, et al. The grapevine genome sequence suggests ancestral hexaploidization in major angiosperm phyla. Nature. 2007. 449(7161):463–467. doi:10.1038/nature06148
Myles S, Chia JM, Hurwitz B, Simon C, Zhong GY, Buckler E, et al. Rapid genomic characterization of the genus Vitis. PLoS One. 2010. 5(1):e8219. doi:10.1371/journal.pone.0008219
Le Paslier MC, Choisne N, Scalabrin S, Bacilieri R, Berard A, Bounon R, et al. The GrapeReSeq 18 k Vitis genotyping chip. 2013. Conference: IX International Symposium on Grapevine Physiology and Biotechnology, Santiago, Chile.
Laucou V, Launay A, Bacilieri R, Lacombe T, Adam-Blondon AF, Bérard A, et al. Extended diversity analysis of cultivated grapevine Vitis vinifera with 10K genome-wide SNPs. PLoS One. 2018. 13(2):e0192540. doi:10.1371/journal.pone.0192540
D’Onofrio C, Tumino G, Gardiman M, Crespan M, Bignami C, de Palma L, et al. Parentage atlas of Italian grapevine varieties as inferred from SNP Genotyping. Front. Plant Sci. 2021. 11:605934. doi: 10.3389/fpls.2020.605934
Ramos-Madrigal J, Runge AKW, Bouby L, Lacombe T, Samaniego Castruita JA, Adam-Blondon AF, et al. Palaeogenomic insights into the origins of French grapevine diversity. Nat Plants. 2019. 5(6):595–603. doi: 10.1038/s41477-019-0437-5
Zdunić G, Maul E, Hančević K, Leko M, Butorac L, Mucalo A, et al. Genetic diversity of wild grapevine [Vitis vinifera L. subsp. sylvestris (Gmel.) Hegi] in the eastern Adriatic region. Am. J. Enol. Vitic. 2017. 68(2):252–257. doi:10.5344/ajev.2016.16072
Zdunić G, Lukšić K, Nagy ZA, Mucalo A, Hančević K, Radić T, et al. Genetic structure and relationships among wild and cultivated grapevines from central Europe and part of the western Balkan peninsula. Genes. 2020. 11(9):962. doi:10.3390/genes11090962
Riaz S, De Lorenzis G, Velasco D, Koehmstedt A, Maghradze D, Bobokashvili Z, et al. Genetic diversity analysis of cultivated and wild grapevine (Vitis vinifera L.) accessions around the Mediterranean basin and Central Asia. BMC Plant Biol. 2018. 18(1):137. doi:10.1186/s12870-018-1351-0
Marinov L, Zdunić G, Di Gaspero G, Magris G, Morgante M, Lukšić K, et al. Analysis of Croatian wild and cultivated grapevine diversity by genotyping by sequencing. Vitis. 2023. 62:23–29. doi: 10.5073/vitis.2023.62.special-issue.23-29
Crespan M, Mercati F, De Lorenzis G, D’Onofrio C, Sunseri F. Two main distinct evolutionary stories describe the Italian grapevine assortment. Vitis. 2023. 62:30–36. doi: 10.5073/vitis.2023.62.special-issue.30-36
Alexander DH, Novembre J, Lange K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 2009. 19(9):1655–1664. doi:10.1101/gr.094052.109
Pickrell JK, Pritchard JK. Inference of population splits and mixtures from genome-wide allele frequency data. PLoS Genetics, 8(11):e1002967. doi:10.1371/journal.pgen.1002967
Csárdi G, Nepusz T. The igraph software package for complex network research. InterJournal, Complex Systems. 2006. 1695(5):1–9.
Csárdi G, Nepusz T, Traag V, Horvát S, Zanini F, Noom D, Müller K. igraph: Network Analysis and Visualization in R. 2024. doi:10.5281/zenodo.7682609
OIV. 2nd Edition of the OIV Descriptor List for Grape Varieties and Vitis Species. 2001. https://www.oiv.int/node/2830
Guo DL, Zhao HL, Li Q, Zhang GH, Jiang JF, Liu CH, et al. Genome-wide association study of berry-related traits in grape [Vitis vinifera L.] based on genotyping-by-sequencing markers. Hortic. Res. 2019. 6:11. doi:10.1038/s41438-018-0089-z
Kang HM, Sul JH, Service SK, Zaitlen NA, Kong SY, Freimer NB, et al. Variance component model to account for sample structure in genome-wide association studies. Nat. Genet. 2010. 42(4):348–354. doi:10.1038/ng.548
Zdunic G, Preece JE, Dangl GS, Koehmstedt A, Mucalo A, Maletic E, Pejic I. Genetic Characterization of Grapevine Cultivars Collected throughout the Dalmatian Region. Am. J. Enol. Vitic. 2013. 64(2):285–290. doi:10.5344/ajev.2012.12085
Cohen P, Bacilieri R, Ramos-Madrigal J, Privman E, Boaretto E, Weber A, et al. Ancient DNA from a lost Negev Highlands desert grape reveals a Late Antiquity wine lineage. Proc. Natl. Acad. Sci. U S A. 2023. 120(17): e2213563120. doi: 10.1073/pnas.2213563120
Maletić E, Pejić I, Karoglan Kontić J, Piljac J, Dangl GS, Vokurka A, et al. Zinfandel, Dobričić, and Plavac mali: The genetic relationship among three cultivars of the Dalmatian Coast of Croatia. Am. J. Enol. Vitic. 2004. 55(2):174–180. doi: 10.5344/ajev.2004.55.2.174
Walker AR, Lee E, Bogs J, McDavid DAJ, Thomas MR, Robinson SP. White grapes arose through the mutation of two similar and adjacent regulatory genes. Plant Journal. 2007. 49(5):772–785. doi: 10.1111/j.1365-313X.2006.02997.x
Paysan-Lafosse T, Blum M, Chuguransky S, Grego T, Pinto BL, Salazar GA, et al. InterPro in 2022. Nucleic Acids Res. 2023. 51(D1):D418–427. doi: 10.1093/nar/gkac993
Zolman BK, Nyberg M, Bartel B. IBR3, a novel peroxisomal acyl-CoA dehydrogenase-like protein required for indole-3-butyric acid response. Plant Mol. Biol. 2007. 64(1–2):59–72. doi:10.1007/s11103-007-9134-2
Vitolović V. (1960). Vinogradarstvo Istre. Arhiv za poljoprivredne nauke, sveska 15, Beograd
Bulić S. 1949: Dalmatinska Ampelografija. Poljoprivredni Nakladni Zavod, Zagreb, Croatia.
Raimondi S, Carlomagno A, Ruffa P, Oglietti S, Novello V, Schneider A. Pedigree reconstruction of wine and table grape crossbreeds created in Italy by Giovanni Dalmasso. Sci. Hortic. 2017. 219:125–130. doi:10.1016/j.scienta.2017.02.044
Tudor A. Nekoliko povijesnih podataka o tribidragu (zinfandel, crljenak kaštelanski, tribidrag). Mogućnosti: književnost, umjetnost, kulturni problemi. 2010. 1(7-9):103-109
Juran K. Arhivski zapisi o sortama vinove loze u Dalmaciji u 15. i 16. stoljeću. Vjesnik dalmatinskih arhiva. 2020. 1(1):141-151. https://doi.org/10.58565/vda
Malenica N, Šimon S, Besendorfer V, Maletić E, Karoglan Kontić J, Pejić I. Whole genome amplification and microsatellite genotyping of herbarium DNA revealed the identity of an ancient grapevine cultivar. Naturwissenschaften. 2011. 98(9):763–772. doi:10.1007/s00114-011-0826-8
Vouillamoz JF, Monaco A, Costantini L, Stefanini M, Scienza A, Grando MS. The parentage of “Sangiovese”, the most important Italian wine grape. Vitis. 2007. 46(1):19-22
Bergamini C, Caputo AR, Gasparro M, Perniola R, Cardone MF, Antonacci D. Evidences for an alternative genealogy of “Sangiovese.” Mol. Biotechnol. 2013. 53(3):278–288. doi:10.1007/s12033-012-9524-9
Kono A, Ban Y, Mitani N, Fujii H, Sato S, Suzaki K, et al. Development of SSR markers linked to QTL reducing leaf hair density and grapevine downy mildew resistance in Vitis vinifera. Molecular Breeding. 2018. 38(11):138. doi:10.1007/s11032-018-0889-8
Barba P, Loughner R, Wentworth K, Nyrop JP, Loeb GM, Reisch BI. A QTL associated with leaf trichome traits has a major influence on the abundance of the predatory mite Typhlodromus pyri in a hybrid grapevine population. Hortic Res. 2019. 6(1):87. doi: 10.1038/s41438-019-0169-8
Crespan M, Crespan G, Giannetto S, Meneghetti S, Costacurta A. “Vitouska” is the progeny of “Prosecco tondo” and “Malvasia bianca lunga”. Vitis. 2007. 46(4):192-194
Cipriani G, Spadotto A, Jurman I, Gaspero G Di, Crespan M, Meneghetti S, et al. The SSR-based molecular profile of 1005 grapevine (Vitis vinifera L.) accessions uncovers new synonymy and parentages, and reveals a large admixture amongst varieties of different geographic origin. Theor. and App. Genet. 2010. 121(8):1569–1585. doi:10.1007/s00122-010-1411-9
Piljac J, Maletić E, Karoglan Kontić J, Dangl GS, Pejić I, Mirošević N, et al. The parentage of Pošip bijeli, a major white wine cultivar of Croatia. Vitis. 2002. 41(2):83-87.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
29 Jul, 2024
Reviews received at journal
28 Jul, 2024
Reviews received at journal
22 Jul, 2024
Reviews received at journal
16 Jul, 2024
Reviewers agreed at journal
01 Jul, 2024
Reviewers agreed at journal
29 Jun, 2024
Reviewers agreed at journal
27 Jun, 2024
Reviewers invited by journal
25 Jun, 2024
Editor invited by journal
25 Jun, 2024
Editor assigned by journal
25 Jun, 2024
Submission checks completed at journal
25 Jun, 2024
First submitted to journal
21 Jun, 2024

You are reading this latest preprint version

Single nucleotide polymorphism (SNP) analysis reveals ancestry and genetic diversity of cultivated and wild grapevines in Croatia

Status:

Version 1

Abstract

Background

Results

Conclusions

Figures

Background

Materials and methods

Results

Discussion

Conclusions

Abbreviations

Declarations

Ethics approval and consent to participate

Consent for publication

Availability of data and materials

Competing interests

Funding

Authors' contributions

Acknowledgements

Authors' information (optional)

References

Additional Declarations

Supplementary Files

Status:

Version 1