ORIGINAL ARTICLE Genetic Resources (2024), 5 (9), 72–82 DOI: 10.46265/genresj.WETA7514 https://www.genresj.org ISSN: 2708-3764 Morphological and molecular characterization of ‘Saragolla’ wheats (Triticum turgidum subsp. durum from Abruzzo, Italy) Agata Rascio *,a, Vanessa De Simone a, Lorenzo Goglia b, Silvana Paone a, Maria Pellegrino a and Giuseppe Sorrentino b a Council for Agricultural Research and Economics, Research Centre for Cereal and Industrial Crops S.S, 673 Km 25, 200 71122, Foggia, Italy b Institute for Sustainable Plant Protection - Italian National Research Council (IPSP-CNR), Piazzale Enrico Fermi, 1, 80055, Portici (NA), Italy Abstract: A morphological and genetic characterization of autochthonous ‘Saragolla’ wheats, currently cultivated in Abruzzo Region (Italy), was carried out. Using 15 simple sequence repeat (SSR) markers and 24 UPOV morphological traits we compared: (a) 13 ‘Saragolla’ genotypes with traits of the italicum/apulicum botanical varieties (Saragolla (Sar.) italicum), (b) 26 ‘Saragolla’ genotypes with traits of leucurum/affine botanical varieties (Sar. leucurum), (c) 8 breeding varieties (pure lines), and (d) 5 Italian autochthonous wheats and 1 turanicum line (old wheats). One hundred twenty-six (126) alleles were identified. The number of alleles per locus spanned from 4 to 15 and the number of alleles per genotype varied between 12 and 21. Values of gene diversity (Nei) across the 53 genotypes was 0.17. The groups of Sar. leucurum and Sar. italicum genotypes were morphologically distinguishable from the groups of old wheats and pure lines. Likewise, the analysis of molecular data using the discriminant analysis revealed that genotypes with the Sar. italicum phenotype displayed distinct genetic differences from Sar. leucurum, pure lines and old wheats. These results make Sar. italicum genotypes distinguishable and eligible as a conservation variety. Ward’s clustering analysis of the 53-genotype pool showed that the ‘Saragolla’ landrace is a valuable repository of genetic diversity. Keywords: Abruzzo Region, ’Saragolla’ landrace, durum wheat diversity, genetic characterization Citation: Rascio, A., De Simone, V., Goglia, L., Paone, S., Pellegrino, M., Sorrentino, G. (2024). Morphological and molecular characterization of ‘Saragolla’ wheats (Triticum turgidum subsp. durum from Abruzzo, Italy). Genetic Resources 5 (9), 72–82. doi: 10.46265/genresj.WETA7514. © Copyright 2024 the Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Introduction Traditional crops are generally viewed by consumers as more natural and healthier options (Rascio et al, 2015, 2016). Their reintroduction into cultivation and reinforcing their value chain can play a vital role in bolstering conservation efforts and elevating the value of their cultivation environment. This, in turn, can substantially increase the economic benefits for rural communities (Pallante et al, 2016). Before the onset of ∗Corresponding author: Agata Rascio (email:agata.rascio@crea.gov.it) the breeding era in Italy, initiated by Strampelli in the early twentieth century (Scarascia-Mugnozza, 2005), cultivated wheat primarily consisted of autochthonous varieties. Although there is not a worldwide consensus on this definition, these wheats are often referred to as old/ancient wheats, or landraces (Negri et al, 2009). Autochthonous wheats are regarded as valuable sources of alleles for breeding programmes (Terzi et al, 2005). Indeed, over time, environmental conditions and, to a certain extent, purposeful farmer selection have resulted in the cultivation of plants exhibiting high adaptability and superior performance within their respective cultivation regions (Zeven, 1998). For Received: 18.10.2023 Accepted: 10.04.2024 Published online: 14.05.2024 https://www.genresj.org https://www.doi.org/10.46265/genresj.WETA7514 https://www.genresj.org https://www.doi.org/10.46265/genresj.WETA7514 mailto:e\protect \discretionary {\char \hyphenchar \font }{}{}mail:agata.rascio@crea.gov.it Genetic Resources (2024), 5 (9), 72–82 Characterization of Italian autochthonous ‘Saragolla’ wheats 73 these reasons, the EU (EU, 2009) defines landraces or varieties which are naturally adapted to local and regional conditions as conservation varieties (Spataro and Negri, 2013). They are frequently composed of mixed genotypes (populations) and exhibit a high degree of genetic diversity (Zeven, 1998), in contrast to modern breeding varieties (cultivars or pure lines). Following the ratification of the International Treaty on Plant Genetic Resources for Food and Agriculture by the UN Food and Agriculture Organization (FAO, 2009) in 2001, most Italian regions implemented laws aimed at protecting and promoting indigenous crops. They also provided funding for projects dedicated to studying the conservation varieties’ distinctive characteristics. Their registration involves a formal process where these vari- eties are officially recognized, documented, and often included in seed catalogues or databases. Once regis- tered, conservation varieties are often conserved in com- munity seedbanks or similar facilities where they can be accessed by farmers. Enhancing access to these varieties is important to support agricultural biodiversity, improve resilience in the face of environmental challenges, and contribute to the long-term sustainability of food pro- duction systems. Some countries, by officially recogniz- ing and documenting these varieties, provide legal pro- tection to conservation varieties to prevent unautho- rized use or commercial exploitation (EU, 2009). This protection is designed to encourage the continued con- servation and sustainable use of these valuable genetic resources. In Southern Italy, the cultivation of a wheat type known as ‘Saragollo forte’ which is particularly well- suited for pasta production, has been extensively documented in various commercial agreements dating back to the seventeenth century (Fiore, 2013). As introduced by old botanists (Draghetti, 1927; De Cillis, 1927), the use of the plural noun ‘Saragolle,’ highlights the presence of multiple forms of ‘Saragolla’ wheat, all falling under the Saragolla (Sar.) leucurum botanical variety, which was one of the 22 botanical varieties of Triticum turgidum documented at the start of the 1900s (Percival, 1921). In 2004, the ‘Produttori Sementi Bologna’ company registered a variety also named ‘Saragolla’. This variety resulted from crosses between the ‘Iride’ cultivar and the ‘0114’ elite line. Therefore, this ‘Saragolla’ pure line is an enhanced variety and not a local one. Today, a significant number of farmers in Central and Southern Italy are cultivating the old ‘Saragolla’ wheat, either for personal consumption or to establish short food supply chains of autochthonous wheat. The resurgence of interest in ‘Saragolla’ can be attributed to its adaptability to low-fertility soils and its suitabil- ity for cultivation with minimal input methods, mak- ing it especially attractive for agriculturally marginal regions. Moreover, this revived interest in ‘Saragolla’ has been magnified by online sources (Eccellenze D’abruzzo, 2024), which suggest that it can be an Italian alter- native to Khorasan wheat, marketed under the name ‘Kamut’ (Piergiovanni, 2013). Kamut is a registered vari- ety belonging to the tetraploid species Triticum turan- icum Jacubz, initially described by Percival (1921) as T. orientale. This hypothesis is reinforced by the mor- phological resemblance between the elongated seeds of ‘Saragolla’ cultivated in the Abruzzo Region and the seeds of ‘Kamut’ T. orientale Percival. Considering the complexity of the situation, in the years 2018–2020, the local authorities of the Italian Region Abruzzo funded the SARAB project, ‘Characterization of local Saragolla durum wheat populations’, to characterize the ‘Saragolla’ wheat currently cultivated. Through intensive cataloguing based on morphology, the project focused on the main species and botanical varieties of ‘Saragolla’ cultivated in 11 different sites within the Abruzzo Region (Rascio et al, 2021). The results revealed a heterogeneous botanical composition both within and among these sites (Rascio et al, 2022). Nine botanical varieties of durum wheat were observed, the majority belonging to the italicum or apulicum botanical varieties, primarily differing in glume pigmentation intensity. There was also a smaller number of genotypes falling into the leucurum or affine botanical varities (referred to here as Saragolla (Sar.) leucurum), which exhibited variations in seed pigmentation intensity. The main goals of this study were to conduct genetic and morphological characterizations of representative plants belonging to the most prevalent botanical varieties collected in 11 farms that participated in the SARAB project. Additionally, these varieties were compared with a collection of both pure lines (modern) and traditional Italian (old) wheat varieties. The results here shown indicate that, in contrast to the Sar. leucurum, the italicum genotypes shared a close genetic similarity among themselves. They are widespread in the Abruzzo region and genetically distinct from the groups of modern and old genotypes examined in this study, hence they are eligible for registration as a conservation variety. The study also explores the degree of diversity of the four groups of genotypes. Material and methods Plant material Eleven samples of ‘Saragolla’ wheats from 11 loca- tions in Abruzzo (shown in Supplemental Table 1) were grown and characterized at the Council for Agricultural Research and Economics - Research Centre for Cereal and Industrial Crops (CREA-CI) in Foggia (Rascio et al, 2022). About 1,000 individual plants of these heteroge- neous wheats were morphologically examined and 434 durum wheat plants were found. A total of 39 repre- sentative spikes of this subset, selected to be indicative of the prevailing botanical varieties italicum/apulicum and leucurum/affine were used for further genetic and morphological characterization in this study. For com- parison, eight modern varieties/durum cultivars and six samples of old autochthonous durum wheat samples 74 Rascio et al Genetic Resources (2024), 5 (9), 72–82 belonging to the CREA-CI working collection were also included (Supplemental Table 2). They comprised four genotype groups: 1) Sar. italicum: 13 ‘head to row’ of ‘Saragolla’ genotypes exhibiting at least four out of five traits of the italicum/apulicum botanical varieties 2) Sar. leucurum: 22 ‘head to row’ of ‘Saragolla’ genotypes from Abruzzo and four from Puglia displaying at least four out of five traits of the leucurum/affine botanical varieties 3) Modern pure lines : Eight seed samples of durum wheat cultivars: ‘Ciccio’, ‘Cappelli’, ‘Capeiti’, ‘Colosseo’, ‘Duilio’, ‘Simeto’, ‘Svevo’ and ‘Saragolla’ 4) Old wheats: Six seed samples of autochthonous durum wheats, primarily sourced from Sicily (Fiore et al, 2019) and belonging to the CREA-CI working collec- tion. These include: ‘Realforte’, ‘Russello’, ‘Sammarti- nara’, ‘Scorsonera’, ‘Vallelunga pubescent’, and the T. turanicum pure line (PI166959), selected at CREA-CI. Phenotypic assessment In 2021, a total of 53 rows, each 1m in length and spaced 30cm apart, were sown according to the usual agronomic practices (Rascio et al, 2016). Throughout the growth stage, each row was carefully examined to ensure its purity, and one plant was selected for DNA extraction and morphological characterization. The assessment was performed on 24 traits with value scales employed for evaluation in part adhering to the guidelines outlined by the International Union for the Protection of New Varieties of Plants (UPOV, 2012) (Table 1). Molecular marker analysis For each genotype, the extraction of DNA was performed according to the protocol used by Marone et al (2009). Twenty-eight microsatellite single sequence repeat (SSR) markers were selected based on published map data (Marone et al, 2009, 2012), according to the following criteria: locus-specific amplification, high level of polymorphism, and good genome coverage (one marker per chromosome arm). The sequences of the SSR are available in the GrainGenes database (http://wheat.pw.usda.gov). The PCR reactions were performed in 25µl volume in Applied Biosystems 2720 Thermal Cyclers. The reaction mixture contained 60ng of template DNA, 0.2mM of dNTPs, 1X Buffer (10mM Tris–HCl—pH 8.3, 50mM KCl, 1.5mM MgCl2), 0.4µM labelled reverse primer (FAM or HEX or NED or TET), 0.4µM unlabelled forward primer and 0.2U of Taq DNA polymerase (5U/µl) (Kapa). Thermal cycling conditions were as follows: 94◦C for 3min, followed by 45 cycles of 94◦C for 30s, the specific annealing temperature (Ta) for each primer for 30s, 72◦C for 30s, with a final extension at 72◦C for 2min. The amplification products were analyzed by means of capillary electrophoresis (ABI3130), multiplexing different fluorescent dyes. Electropherograms were analyzed with GeneMapper version 4.0. The internal molecular weight standard was 500-ROX (Life Technologies). Statistical analysis Genotypic characterization was performed with 15 SSR markers which gave a clear electrophoretic pattern. To this aim, the genotypic data were transformed into a binomial matrix as present (1) or absent (0) for each marker and this matrix was used to construct Ward’s dendrogram tree to assess genetic diversity. Nei’s gene diversity, percentage of polymorphic loci and Shannon’s information index were determined using the PopGen 1.31 software (Yeh et al, 1999). For assessing marker polymorphism and informa- tiveness, the average polymorphic information content (PIC) was calculated using the following formula intro- duced by Anderson et al (1993): PIC = 1-Σ(Pi)2 where Pi is the number of polymorphic loci/all the number loci. The distances among the four groups of genotypes (Sar. leucurum, Sar. italicum, pure lines and old wheats) were examined by multivariate discriminant analysis and cluster analysis, using the STATISTICA (StatSoft Inc.) software. Results Morphological characterization As shown in Figure 1 and in Supplemental Table 3, the Sar. italicum/apulicum-like genotypes have rather compact, hairy glumes, lightly pigmented spikes, long red or brown-red awns, yellow-amber and elongated grains. The Sar. leucurum/affine genotypes have white and glabrous glumes, elongated, compact spikes, and white or red seeds. The mean values of each morphological and pheno- logical trait for genotypes belonging to the four groups (Sar. leucurum, Sar. italicum, modern cultivars or old wheats) show that the Sar. italicum/apulicum genotypes registered the highest values for glume hairiness and 1,000 seed weight (Table 2). In contrast, the Sar. leu- curum/affine genotypes exhibited the highest values for the shape of the lower glume beak. Modern varieties dis- played smaller height and earlier heading dates, a result of the extensive breeding efforts they underwent. Results of stepwise discriminant analysis performed using the four groups of genotypes as classification categories and visualized through the biplot of canonical variables (Figure 2) showed that the model had a high discriminatory power (Lambda Wilks: 0,0075245; approx. F (48,90) = 7,8312; p < 0,0000), with two discriminant functions that accounted for 92.9% of the explained variance (Table 3). Based on the absolute values of standardized coefficients of the canonical variables (Table 3) the main traits that horizontally contributed to the 4-group separation were the glume hairiness, which was absent in Sar. leucurum and the upper neck glaucosity, which Genetic Resources (2024), 5 (9), 72–82 Characterization of Italian autochthonous ‘Saragolla’ wheats 75 Table 1. The 24 traits used for the morphological characterization of wheat genotypes, and the value scale employed for evaluation (UPOV, 2012). Trait Measure units/score Assessment scale 1 1,000 seed weight g 2 Curvature of lower glume beak 1–7 1 = absent, 7 = strong 3 Lower glume length of beak 1–9 1 = very short, 7 = long 4 Lower glume hairiness 1–9 1 = absent, 9 = very present 5 Straw: pith in cross section 1–7 1 = thin, 7 = thick 6 Grain shape 3–7 3 = slightly, 7 = strongly elongated 7 Grain: length of brush hair 3–7 3 = short, 7 = long 8 Grain weight/plant g 9 Awn colour 1–5 1 = white, 3 = medium dark, 5 = very dark 10 Spike colour 1–5 1 = white, 3 = medium dark, 5 = very dark 11 Awn tip/ear length ratio 1–3 1 = low ratio, 3 = high ratio 12 Spike length cm 13 Spike shape in profile 1–5 1 = tapering, 2 = parallel sided, 3 = slightly clavate, 4 = strongly clavate, 5 = fusiform 14 Ear glaucosity 1–9 1 = absent, 9 = very strong 15 Awn divergence 1–2 1 = adherent, 2 = divergent 16 Spike density 3–7 3 = lax, 5 = medium, 7 = dense 17 Plant height cm 18 Growth habit 1–9 1 = erect, 9 = prostrate (score 9) 19 Recurved flag leaves 1–9 1 = absent, 9 = very high 20 Heading date from April 1st days 21 Flag leaf: glaucosity of sheath 1–9 1 = absent, 9 = very strong 22 Flag leaf lower side glaucosity 1–9 1 = absent, 9 = very strong 23 Upper node hairiness 1–9 1 = absent, 9 = very strong 24 Upper neck glaucosity 1–9 1 = absent, 9 = very strong Figure 1. Comparison of distinguishing characteristics of glume (A), and spike and seed (B) of Saragolla botanical types Sar. italicum and Sar. apulicum: hairy with red glumes and red awns; Sar. leucurum and Sar. affine: glabrous with white glumes and white awns. 76 Rascio et al Genetic Resources (2024), 5 (9), 72–82 Table 2. Mean values of morphological traits of the four groups of durum wheat genotypes. The qualitative traits, for which no specific unit of measurement is provided, were assessed using the value scale established by the International Union for the Protection of New Varieties of Plants (UPOV, 2012). Trait Sar. leucurum Sar . italicum Modern cultivars Old wheats Shape of lower glume beak Mean 6,9 2,0 2,7 2,1 SD 0,6 1,3 2,0 1,6 Glume length of beak Mean 4,0 4,9 4,3 3,6 SD 1,8 1,6 3,0 1,9 Glume hairiness Mean 1,3 8,2 1,3 1,0 SD 1,5 1,7 0,8 0,0 Straw: pith in cross section Mean 5,5 6,7 6,7 6,4 SD 1,9 0,6 0,8 1,0 Ear length (cm) Mean 9,2 9,4 8,5 8,4 SD 1,2 1,0 2,3 2,2 Grain shape Mean 4,6 6,8 3,7 3,6 SD 1,7 0,6 1,0 1,5 Grain: length of hair Mean 3,3 4,1 3,1 3,0 SD 0,9 1,3 0,2 0,0 Awn colour Mean 1,8 2,3 2,2 2,6 SD 1,1 0,9 1,6 1,4 Glume colour Mean 1,7 2,0 1,5 2,0 SD 0,8 0,9 0,5 1,5 Awn tip/ear length ratio Mean 1,6 2,7 1,7 1,6 SD 0,7 0,6 0,8 0,8 Ear shape Mean 2,0 2,2 3,0 2,0 SD 0,9 1,2 1,3 0,8 Awn compactness Mean 1,4 1,5 1,5 1,9 SD 0,6 0,8 0,5 1,5 Ear density Mean 5,2 4,7 4,7 6,1 SD 1,9 1,4 1,5 1,1 Plant height (cm) Mean 108,7 120,0 79,2 101,4 SD 16,2 7,9 17,7 21,7 Grain weight/plant (g) Mean 19,7 15,2 19,0 15,3 SD 6,3 4,2 1,9 3,4 1,000 seed weight (g) Mean 46,5 73,3 48,7 46,4 SD 8,5 3,6 5,3 7,0 Growth habit Mean 3,7 4,4 2,3 4,4 SD 1,4 1,0 2,1 1,0 % recurved flag leaves Mean 6,1 6,7 2,7 5,0 SD 1,6 0,8 1,5 2,0 Heading date Mean 28,3 29,4 18,1 20,7 (days from April 1st) SD 4,1 1,6 6,6 6,3 Flag leaf glaucosity Mean 5,1 5,8 4,7 4,4 SD 1,5 1,3 2,0 2,2 Flag leaf lower glaucosity Mean 3,4 3,0 3,7 3,3 SD 1,4 0,0 1,0 1,4 Upper node hairiness Mean 3,3 3,0 2,3 3,0 SD 1,3 0,0 1,0 1,2 Upper neck glaucosity Mean 5,1 3,3 5,0 4,4 SD 1,5 0,8 1,3 1,5 Spike glaucosity Mean 4,8 6,7 5,7 5,3 SD 1,4 1,1 1,0 1,4 Genetic Resources (2024), 5 (9), 72–82 Characterization of Italian autochthonous ‘Saragolla’ wheats 77 Figure 2. The biplot of canonical variables obtained using a stepwise discriminant analysis based on 24 UPOV morphologi- cal descriptors and four groups of genotypes. The percentages of explained variance by the two roots are shown. was lacking in Sar. italicum (Table 2). The differences in heading date along with the upper neck green glaucosity degree (Table 2) mainly contributed to the vertical separation of Sar. leucurum and italicum from the pool of new and old genotypes. The Mahalanobis distances between groups were all highly significant (Table 4), except between pure lines and old wheats; Sar. leucurum and Sar. italicum were the two most distant and hence morphologically different groups. SSR patterns The molecular analysis utilized a set of 15 SSR markers out of initially 28 tested, each characterized by a distinct electrophoretic pattern for all genotypes. In total, 126 alleles were identified, with the number of alleles per locus spanning from 4 to 15. The Polymorphic Information Content (PIC) values of the SSR markers ranged from 0.52 (for gwm60) to 0.91 (for wmc606 and gwm459), resulting in an average PIC value of 0.77 per locus (Table 5). Excluding from the analysis the alleles that occurred at a low frequency (p < 0.05), the number of alleles per genotype varied between 12 and 21, with the most frequent value being 16 (Table 6). This occurrence was four times higher than what was observed in a study where 104 Ethiopian durum wheat genotypes, representing 13 populations, three regions, and four altitudinal classes, were analyzed using 14 SSR markers (Dagnaw et al, 2023). Diversity The diversity analysis for all cultivars based on SSR markers (Table 7) yielded low mean values (0.28±0.20) of Shannon’s index. The values (0.176±0.5) of Nei’s gene diversity were lower than the minimum observed in 40 winter wheat genotypes (Petrović et al, 2017) coming from European countries (Croatia, Austria, France, Italy, and Russia) and lower than that (0.56) resulting for 124 Ethiopian genotypes (Dagnaw et al, 2023). The among-groups comparison indicated that old wheats and Sar. leucurum exhibited the highest percent- age of polymorphic loci, followed by pure lines and old wheats. The measurement of gene diversity, estimated by both Nei’s gene diversity and Shannon’s information index, yielded similar values for Sar. leucurum, breed- ing lines, and old wheats, and the lowest values for Sar. italicum. In the case of Sar. leucurum, Sar. italicum, old wheats and pure lines, the highest average number of amplified alleles per locus was observed in 1B (long arm), 6B (short arm) and 6B (long arm), respectively, with average values of 9.0, 6.5, 2.5 and 2.3, respectively. It’s worth noting that old wheats and modern pure lines exhibited the highest percentage of polymorphisms detected by SSR markers in the B genome (Table 8), likely originating from a species, or several species closely related to Aegilops speltoides Tausch, a cross-pollinating species; while the Sar. leucurum and Sar. italicum sets had the highest percentage of polymorphisms in the A genome, which can be traced back to diploids like T. urartu Thumanjan ex Gandilyan (Wang et al, 2007). A similar clustering pattern was observed when analyzing both the morphological traits (Figure 2) and molecular marker profiles (Figure 3) of all 53 genotypes. This analysis employed a hierarchical grouping method, without missing data in the dataset. The resulting dendrogram revealed four major clusters (Figure 3). Cluster 1A comprised all 13 Sar. italicum genotypes, 9 out of 22 Sar. leucurum genotypes of Abruzzo and 1 Sar. leucurum genotype from Puglia. The second major cluster, 1B, could be further subdivided into two subclusters: 1B1 and 1B2. The 1B1 cluster included two subgroups: the first subgroup contained four closely related breeding lines (‘Colosseo’, ‘Simeto’, ‘Ciccio’ and ‘Capeiti’), two Sar. leucurum genotypes, and the old wheat ‘Vallelunga pubescent’; the second subgroup was larger, consisting of modern varieties (‘Duilio’, ‘Realforte’, ‘Svevo’ and the ‘Saragolla’ pure line), some old wheats (‘Russello’, ‘Scorsonera’, ‘Sammartinara’, ‘Cappelli’ and the T. turanicum line), along with eight Sar. leucurum genotypes from Puglia or Abruzzo. Cluster 1B2 included six strongly related Sar. leucurum genotypes: one was from Puglia and five from Abruzzo. Discussion The morphological and genetic characterization of autochthonous wheats serves the dual purpose of safeguarding the economic interests of farmers and increasing consumers’ trust in the origin and quality of food products entering the market (Terzi et al, 2005). A recent morphological analysis conducted by the SARAB project on wheat crops in 11 farms across 78 Rascio et al Genetic Resources (2024), 5 (9), 72–82 Table 3. Values of the standardized coefficients for the canonical variables included in the discriminant functions, obtained using the four groups of genotypes (Sar. leucurum, Sar. italicum, pure lines and old wheats) as classification categories. root 1 root 2 Glume hairiness -0,99 -0,26 Heading date (days from 1/4) 0,05 -1,39 Upper neck glaucosity 0,73 -0,98 Growth habit -0,48 0,64 Grain shape -0,06 -0,50 1,000 seed weight 0,03 0,44 Spike glaucosity -0,43 -0,05 Spike shape -0,27 -0,08 Glume colour 0,18 -0,57 Awn colour -0,40 0,38 Plant height 0,48 -0,17 Flag leaf lower side glaucosity 0,19 -0,49 Grain weight/plant 0,09 -0,36 Straw: pith in cross section -0,23 0,20 Lower glume: length of beak -0,36 -0,29 Spike density 0,21 0,31 Eigenvalue 12.4 3.4 Explained cumulative variance (%) 72.6 92.9 Table 4. Pairwise square Mahalanobis distances (plain text) and probability values (italics) for the contrasts between the four groups of genotypes. Sar. leucurum Sar. italicum Pure lines Old wheats Sar. leucurum 0,0000 0,00036 0,00003 Sar. italicum 69,48 0,00000 0,00000 Pure lines 23,76 78,44 0,01962ns Old wheats 27,19 68,77 20,95 Table 5. List of SSR markers used for molecular analysis, number of alleles and Polymorphic Information Content (PIC) obtained for each marker in the 39 Saragolla wheat lines. A, A genome; B, B genome; L, long arm; S, short arm. Marker Chromosome No. of alleles PIC 1 gwm311 2A(L) 11 0,83 2 gwm1042 3A(L) 6 0,55 3 gwm299 3B(L) 7 0,74 4 barc45 3A(S) 4 0,77 5 gwm495 4B(S) 8 0,82 6 gwm1093 4A(S) 14 0,83 7 gwm1084 4B(L) 8 0,75 8 gwm865 5A(L) 8 0,80 9 gwm154 5A(S) 9 0,73 10 gwm499 5B(L) 11 0,80 11 gwm1017 6A(L) 6 0,80 12 gwm459 6A(S) 12 0,91 13 gwm193 6B(L) 4 0,76 14 gwm60 7A(S) 7 0,52 15 wmc606 7B(S) 11 0,91 Genetic Resources (2024), 5 (9), 72–82 Characterization of Italian autochthonous ‘Saragolla’ wheats 79 Table 6. Average number of SSR alleles per genotype. Only alleles occurring with a frequency higher than 11 (p < 0.05) are included. The codes from S1 to S11 indicate the cultivation sites (see Supplemental Table 1) of Saragolla wheats from Abruzzo and S12 indicates the cultivation site in Puglia. The extra letters and numbers differentiate the genetically characterized plants within each site. Old wheats and pure lines are described in Supplemental Table 2. Group Genotypes Average allele no. Group Genotypes Average allele no. Sar. leucurum S1F3 13 Sar. italicum S1H3 16 S2E 13 S2P2A 15 S2F 14 S3P27F 20 S3P27A 16 S4P23B 14 S3P3B 15 S5P49B 21 S3P3I 16 S6P6A 16 S4P4B 16 S7P43C 16 S4P4A 16 S8P8D 16 S4Z 15 S8P9A 16 S5P5C 16 S10P50C 15 S5P31D 16 S11P11A 17 S6P21D 15 S7A 16 S6P41C 15 S7C 14 S8A1 14 Old wheats ‘Sammartinara’ 15 S8R1 14 ‘Realforte’ 12 S8 15 ‘Scorsonera’ 17 S9P22D 15 ‘Russello’ 16 S9P9D 23 ‘Vallelunga Pubescent’ 13 S9P22A 15 ‘Turanicum’ 17 S10P32A 14 Modern pure lines ‘Capeiti’ 14 S11P24A 15 ‘Ciccio’ 15 S12P12A 14 ‘Simeto’ 15 S12P54B 16 ‘Colosseo’ 15 S12P35A 14 ‘Duilio’ 16 S12P54A 14 ‘Saragolla’ 15 S1E4 15 ‘Cappelli’ 13 ‘Svevo’ 16 Table 7. Genetic diversity indices over 15 SSR loci for all 53 Italian genotypes tested in the study, as well as for the four groups categorized by ‘Saragolla’ botanical variety or control group (Means±SD). PIC, Polymorphic Information Content. Shannon’s information index Percentage of polymorphic loci Nei’s gene diversity PIC Sar. leucurum 0.19±0.19 71.4 0.10±0.13 0.902±0.08 Sar. italicum 0.08±0.18 22.2 0.05±0.13 0.097±0.08 Old wheats 0.17±0.22 39.7 0.10±0.14 0.929±0.10 Modern pure lines 0.19±0.19 61.1 0.11±0.12 0.764±0.11 All genotypes 0.28±0.20 97.63 0.17±0.15 0.673±0.35 Table 8. Percentage of polymorphism detected by SSR in A and B genomes of four groups of genotypes ‘Saragolla’ leucurum ‘Saragolla’ italicum Old Italian wheats Modern pure lines Genome A 54,1 54,5 46,2 48,5 Genome B 45,9 45,5 53,8 51,5 80 Rascio et al Genetic Resources (2024), 5 (9), 72–82 Figure 3. SSR-based genetic distances among the 53-genotype pool, through Ward’s clustering. The percentage of the prevalent type of genotypes included in each cluster is indicated. Characters: red = Sar. leucurum/affine genotypes from Abruzzo; blue = Sar. leucurum from Puglia; black = Sar. italicum/apulicum genotypes from Abruzzo; violet = modern and old varieties. Codes are as in Table 6. the Abruzzo Region revealed significant morphological diversity within the cultivated ‘Saragolla’ variety (Rascio et al, 2022). This diversity poses a challenge in accurately defining their distinct traits. The morphotypes that are both quantitatively and widely spread, across most of the 11 wheat farms included in the SARAB project, belong to the Sar. italicum/apulicum or the Sar. leucurum/affine botanical varieties (Rascio et al, 2022). The genetic character- ization described here aimed to validate whether the observed morphological similarity among the genotypes from Abruzzo corresponds to genetic similarity and to develop a tool to differentiate them. The results presented here confirm the efficacy of SSR markers to characterize wheat genotypes (Wang et al, 2007; Dagnaw et al, 2023). In fact, the 15 SSR primers used in this experiment showed detectable polymorphisms in all the 53 genotypes and their mean polymorphic information content (PIC = 0.77) makes their use very informative. ‘Saragolla’ genotypes, belonging to the italicum/apulicum botanical varieties can be eligible as conservation varieties. These genotypes are widely cultivated across the Abruzzo Region, and they also exhibit a noteworthy genetic similarity, as indicated by the low values of Nei’s gene diversity and Shannon’s information index. Additionally, Sar. italicum genotypes display both a distinct phenotype and genotype in comparison to Sar. leucurum, older wheats and pure lines. It is worth noting that Ward’s clustering analysis of the 53-genotype pool revealed significant genetic diversity between Sar. italicum and Sar. leucurum genotypes. Out of the 26 Sar. leucurum genotypes examined, only 9 displayed a significant genetic resemblance to Sar. italicum. Six were categorized within the broader groups of pure lines and old wheats, while five formed a distinct group of genotypes very closely related genetically, but distinct from all others. The clustering analysis also revealed a stronger genetic similarity between most Sar. leucurum genotypes and three out of four ‘Saragolla’ genotypes from Puglia and a somewhat lesser degree of affinity with the oldest Genetic Resources (2024), 5 (9), 72–82 Characterization of Italian autochthonous ‘Saragolla’ wheats 81 genotypes. Expanding on the hypothesis (Zeven, 1998), that factors such as geographic distance, environmental conditions and the selection made by farmers can shape the genetic composition of local wheats, it is plausible to infer that the migration of wheat commenced from Sicily. In fact, the leucurum genotypes were documented in Southern Italy as early as the beginning of the 1900s (Percival, 1921; De Cillis, 1927; Draghetti, 1927) and were likely among the oldest cultivated in Sicily (Porceddu et al, 1981). From Sicily, it is plausible that these wheats initially spread to the nearby region of Puglia and then reached Abruzzo where the cross with indigenous wheat occurred as well as the selection of alleles improving adaptability, productivity and quality. In terms of affinities with ’Kamut’, the results suggest that the Sar. italicum genotypes, despite having elongated and large seeds similar to T. turanicum, formed a distinct cluster and showed a closer genetic relationship to Sar. leucurum and old durum wheats. Conclusions The ‘Saragolla’ wheat presently grown in the Abruzzo Region is characterized by its rich diversity, predomi- nantly comprising Sar. italicum/apulicum and Sar. leu- curum-like durum wheats, some of which share close morphological and genetic traits. Recently, these geno- types have been officially registered as ‘Saragolla’ con- servation varieties from Abruzzo. The genetic distance observed among 39 ‘Saragolla’ genotypes, representa- tive of only two out of the nine previously identified botanical varieties of durum wheat, exceeded that found among the other 13 modern, or old durum wheats used in this study, which differ for age of cultivation and origin. Consequently, at the Maiella National Park seedbank, targeted ex situ conservation measures will be implemented to preserve the currently cultivated populations. A more extensive genetic characterization will enable the assessment of existing variability within the ‘Saragolla’ landrace, for adaptive and agronomically valuable traits, useful for breeding improved varieties. Supplemental data Supplemental Table 1. Geographic coordinates of the cultivation sites for the 12 Saragolla wheats. Supplemental Table 2. Passport details of old and modern wheats used in the present work. Supplemental Table 3. Morphological traits of Saragolla leucurum, italicum, and modern and old wheats. Acknowledgements The authors wish to thank the farm owners who pro- vided the studied materials, Dr Maurizio Odoardi and Dr Daniela Codoni (Department of Rural Development and Fisheries Policies – Promotion of Knowledge and Innovation in Agriculture – DPD022) of the Abruzzo Regional Authorities, for their valuable contribution to the ‘SARAB project: Characterization of ancient ‘Saragolla’ populations from the Abruzzo Region’. The authors also wish to thank Mr Leonardo Morrone and Mr Vito De Gregorio for their valuable assistance in con- ducting the experimental trials. Funding This work was in part supported by the Abruzzo Region. Author contributions AR, study conception and manuscript draft; VDS, molecular analysis; LG, analysis and interpretation of results; SP, data collection; MT, field trials; GS, manuscript revision. Conflict of interest statement The authors declare no conflict of interest. References Anderson, J. A., Churchill, G. A., Sutrique, J. E., Tanksley, S. D., and Sorrells, M. E. (1993). Optimizing parental selection for genetic linkage maps. Genome 36, 181–186. doi: https://doi.org/10.1139/g93-024 Dagnaw, T., Mulugeta, B., Haileselassie, T., Geleta, M., Ortiz, R., and Tesfaye, K. (2023). Genetic Diversity of Durum Wheat (Triticum turgidum L. ssp. durum, Desf). Genes 14(1155). doi: https://doi.org/10.3390/ genes14061155 De Cillis, E. (1927). I grani d’Italia volume 173. (Roma: Tipografia della Camera dei deputati). Draghetti, A. (1927). Forme e limiti dello xerofitismo nel frumento (Forl̀ı: Tipografie Valbonesi). Eccellenze D’abruzzo (2024). Saragolla: il grano dei faraoni in Abruzzo. url: https: //www.eccellenzedabruzzo.it/grano-saragolla/. EU (2009). Commission Directive 2009/145/EC of 26 November 2009 providing for certain derogations, for acceptance of vegetable landraces and varieties which have been traditionally grown in particular localities and regions and are threatened by genetic erosion and of vegetable varieties with no intrin- sic value for commercial crop production but devel- oped for growing under particular conditions and for marketing of seed of those landraces and vari- eties. url: https://eur-lex.europa.eu/legal-content/ EN/TXT/?uri=CELEX%3A02009L0145-20130828. FAO (2009). The International Treaty on Plant Genetic Resources for Food and Agriculture (Rome, Italy: Food and Agricultural Organization of the United Nations). url: https://www.fao.org/3/a-i0510e.pdf. Fiore, M. (2013). ‘Saragollo Triticum Apulum - La produzione cerealicola nel territorio di Torremaggiore nei secoli XVI e XVII. Cooperativa Agricola Fortore, Torremaggiore. Fiore, M. C., Mercati, F., Spina, A., Blangiforti, S., Venora, G., Dell’acqua, M., and Sunseri, F. (2019). High-throughput genotype, morphology, and quality traits evaluation for the assessment of genetic https://doi.org/10.1139/g93-024 https://doi.org/10.3390/genes14061155 https://doi.org/10.3390/genes14061155 https://www.eccellenzedabruzzo.it/grano-saragolla/ https://www.eccellenzedabruzzo.it/grano-saragolla/ https://www.fao.org/3/a-i0510e.pdf https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A02009L0145-20130828 https://www.genresj.org/index.php/grj/article/view/genresj.WETA7514/suppdata147 https://www.genresj.org/index.php/grj/article/view/genresj.WETA7514/suppdata147 https://www.genresj.org/index.php/grj/article/view/genresj.WETA7514/suppdata147 82 Rascio et al Genetic Resources (2024), 5 (9), 72–82 diversity of wheat landraces from sicily. Plants 8(116). doi: https://doi.org/10.3390/plants8050116 Marone, D., Laido, G., Gadaleta, A., Colasuonno, P., Ficco, D. B., Giancaspro, A., Giove, S., Panio, G., Russo, M. A., De Vita, P., and Cattivelli, L. (2012). A high-density consensus map of A and B wheat genomes. Theoretical and Applied Genetics 125, 1619–1638. doi: https://doi.org/10.1007/s00122- 012-1939-y Marone, D., Olmo, A. I. D., Laidò, G., Sillero, J. C., Emeran, A. A., Russo, M. A., Ferragonio, P., Giovanniello, V., Mazzucotelli, E., De Leonardis, A. M., and De Vita, P. (2009). Genetic analysis of durable resistance against leaf rust in durum wheat. Molecular Breeding 24, 25–39. doi: https://doi.org/10.1007/ s11032-009-9268-9 Negri, V., Maxted, N., and Veteläinen, M. (2009). Euro- pean landrace conservation, an introduction. In Euro- pean landraces: on-farm conservation, management and use, ed. Veteläinen, M., Negri, V., and Maxted, N., (Rome, Italy: Bioversity International), volume 15 of Bioversity International Technical Bulletin. Pallante, G., Drucker, A. G., and Sthapit, S. (2016). Assessing the potential for niche market develop- ment to contribute to farmers livelihoods and agro- biodiversity conservation: Insights from the finger millet case study in Nepal. Ecological Economics 130, 92105–92105. doi: https://doi.org/10.1016/j. ecolecon.2016.06.017 Percival, J. (1921). The Wheat Plant: a Monograph (London: Duckworth and Co). Petrović, S., Marić, S., Čupić, T., Rebekić, A., and Rukavina, I. (2017). Assessment of molecular and phenotypic diversity among winter wheat cultivars. Genetika 49, 583–598. doi: https://doi.org/10.2298/ GENSR1702583P Piergiovanni, A. R. (2013). Capillary electrophoresis: a useful tool for the management of plant genetic resources. Analytical and Bioanalytical Chemistry 405, 481–491. doi: https://doi.org/10.1007/s00216-012- 6127-z Porceddu, E., Vannella, S., and Perrino, P. (1981). Character analysis and numerical classification in a wheat collection from Sicily. Die Kulturpflanze 29, 251–266. doi: https://www.cabidigitallibrary.org/ doi/full/10.5555/19831622160 Rascio, A., Beleggia, R., Platani, C., Nigro, F., Codianni, P., De Santis, and Fragasso, M. (2016). Metabolomic diversity for biochemical traits of Triticum sub-species. Journal of Cereal Science 71, 224–229. doi: https:// doi.org/10.1016/j.jcs.2016.08.009 Rascio, A., Codianni, P., Paone, S., Fiorillo, F., and Marone, D. (2021). SARAB project, characterization of ancient Saragolla populations from Abruzzo Region. Tecnica Molitoria 72, 37–42. Rascio, A., Fiorillo, F., Paone, S., De Santis, G., and Sorrentino, G. (2022). Quantitative botanical characterization of Saragolla wheat landraces from Abruzzo and Puglia Regions of Italy. Plant Genetic Resources 20, 434–441. doi: https://doi.org/10.1017/ S1479262123000345 Rascio, A., Picchi, V., Naldi, J. P., Colecchia, S., De Santis, G., Gallo, A., and De Gara, L. (2015). Effects of temperature increase., through spring sowing, on antioxidant power and health-beneficial substances of old and new wheat varieties. Journal of Cereal Science 61, 111–118. doi: https://doi.org/10.1016/j. jcs.2014.09.010 Scarascia-Mugnozza, T. G. (2005). The contribution of Italian wheat geneticists, from Nazareno Strampelli to Francesco D’Amato. In Tuberosa, R., Phillips, R. L., and Gale, M., Proceedings of the International Congress on: The Wake of the Double Helix, 53-75. Spataro, G. and Negri, V. (2013). The European seed legislation on conservation varieties: focus, implemen- tation, present and future impact on landrace on farm conservation. Genetic resources and Crop Evo- lution 60, 2421–2430. doi: https://doi.org/10.1007/ s10722-013-0009-x Terzi, V., Morcia, C., Gorrini, A., Stanca, A. M., Shewry, P. R., and Faccioli, P. (2005). DNA-based methods for identification and quantification of small grain cereal mixtures and fingerprinting of varieties. Journal of Cereal Science 41, 213–220. doi: https://doi.org/10. 1016/j.jcs.2004.08.003 UPOV (2012). Test Guideline . url: https://www.upov. int/edocs/tgdocs/en/tg120.pdf. Wang, H., Chen, P., and Liu, D. (2007). Assessment of genetic diversity of Yunnan, Tibetan, and Xinjiang wheat using SSR markers. Journal of Genetics and Genomics 34, 623–633. doi: https://doi.org/10.1016/ s1673-8527(07)60071-x Yeh, F. C., Yang, R. C., and Boyle, T. (1999). Microsoft Window-based freeware for population genetic analysis (POPGENE). Version 1.31. url: https: //sites.ualberta.ca/∼fyeh/popgene.html. Zeven, A. C. (1998). Landraces, a review of definitions and classifications. Euphytica 104, 127–139. doi: https://doi.org/10.1023/A:1018683119237 https://doi.org/10.3390/plants8050116 https://doi.org/10.1007/s00122-012-1939-y https://doi.org/10.1007/s00122-012-1939-y https://doi.org/10.1007/s11032-009-9268-9 https://doi.org/10.1007/s11032-009-9268-9 https://doi.org/10.1016/j.ecolecon.2016.06.017 https://doi.org/10.1016/j.ecolecon.2016.06.017 https://doi.org/10.2298/GENSR1702583P https://doi.org/10.2298/GENSR1702583P https://doi.org/10.1007/s00216-012-6127-z https://doi.org/10.1007/s00216-012-6127-z https://www.cabidigitallibrary.org/doi/full/10.5555/19831622160 https://www.cabidigitallibrary.org/doi/full/10.5555/19831622160 https://doi.org/10.1016/j.jcs.2016.08.009 https://doi.org/10.1016/j.jcs.2016.08.009 https://doi.org/10.1017/S1479262123000345 https://doi.org/10.1017/S1479262123000345 https://doi.org/10.1016/j.jcs.2014.09.010 https://doi.org/10.1016/j.jcs.2014.09.010 https://doi.org/10.1007/s10722-013-0009-x https://doi.org/10.1007/s10722-013-0009-x https://doi.org/10.1016/j.jcs.2004.08.003 https://doi.org/10.1016/j.jcs.2004.08.003 https://www.upov.int/edocs/tgdocs/en/tg120.pdf https://www.upov.int/edocs/tgdocs/en/tg120.pdf https://doi.org/10.1016/s1673-8527(07)60071-x https://doi.org/10.1016/s1673-8527(07)60071-x https://sites.ualberta.ca/~ fyeh/popgene.html https://sites.ualberta.ca/~ fyeh/popgene.html https://doi.org/10.1023/A:1018683119237 https://sites.ualberta.ca/~fyeh/popgene.html Introduction Material and methods Plant material Phenotypic assessment Molecular marker analysis Statistical analysis Results Morphological characterization SSR patterns Diversity Discussion Conclusions Supplemental data Acknowledgements Funding Author contributions Conflict of interest statement