G C A T T A C G genes G C A T Article Phenotyping Mediterranean Durum Wheat Landraces for Resistance to Zymoseptoria tritici in Tunisia Sarrah Ben M’Barek 1,2, Marwa Laribi 2,3 , Hajer Kouki 2, Dalma Castillo 4, Chayma Araar 2,5, Meriem Nefzaoui 2,6, Karim Ammar 7, Carolina Saint-Pierre 7 and Amor Hassine Yahyaoui 2,7,* 1 Regional Field Crops Research Center of Beja (CRRGC), BP 350, Beja 9000, Tunisia; sarrah_bm@msn.com 2 CRP-Wheat Septoria Precision Phenotyping Platform, Tunis 1082, Tunisia; mar.wa199@hotmail.fr (M.L.); koukihajercm@gmail.com (H.K.); chaymaaraar@hotmail.fr (C.A.); meriem.nef@gmail.com (M.N.) 3 National Agronomic Institute of Tunisia, University of Carthage, Tunis 1082, Tunisia 4 CRI-Quilamapu, Instituto de Investigaciones Agropecuaria, Chillán 3780000, Chile; castillo.r.dalma@gmail.com 5 Faculty of Sciences of Bizerte, University of Carthage, Jarzouna, Bizerte 7021, Tunisia 6 Department of Agricultural and Food Sciences, Alma Mater Studiorum, University of Bologna, Via Zamboni 33, 40126 Bologna, Italy 7 International Maize and Wheat Improvement Center (CIMMYT) km, 45 Carretera México-Veracruz El Batan, Texcoco CP56237, Mexico; k.ammar@cgiar.org (K.A.); c.saintpierre@cgiar.org (C.S.-P.) * Correspondence: amor.yahyaoui@gmail.com Abstract: Durum wheat landraces have huge potential for the identification of genetic factors valuable for improving resistance to biotic stresses. Tunisia is known as a hot spot for Septoria tritici blotch   disease (STB), caused by the fungus Zymoseptoria tritici (Z. tritici). In this context, a collection Citation: Ben M’Barek, S.; Laribi, M.; of 3166 Mediterranean durum wheat landraces were evaluated at the seedling and adult stages Kouki, H.; Castillo, D.; Araar, C.; for STB resistance in the 2016–2017 cropping season under field conditions in Kodia (Tunisia). Nefzaoui, M.; Ammar, K.; Unadapted/susceptible accessions were eliminated to reach the final set of 1059 accessions; this was Saint-Pierre, C.; Yahyaoui, A.H. termed the Med-collection, which comprised accessions from 13 countries and was also screened in Phenotyping Mediterranean Durum the 2018–2019 cropping season. The Med-collection showed high frequency of resistance reactions, Wheat Landraces for Resistance to among which over 50% showed an immune reaction (HR) at both seedling and adult growth stages. Zymoseptoria tritici in Tunisia. Genes Interestingly, 92% of HR and R accessions maintained their resistance levels across the two years, 2022, 13, 355. https://doi.org/ confirming the highly significant correlation found between seedling- and adult-stage reactions. Plant 10.3390/genes13020355 Height was found to have a negative significant effect on adult-stage resistance, suggesting that either Academic Editors: Anna this trait can influence disease severity, or that it can be due to environmental/epidemiological factors. M. Mastrangelo and Accessions from Italy showed the highest variability, while those from Portugal, Spain and Tunisia Elisabetta Mazzucotelli showed the highest levels of resistance at both growth stages, suggesting that the latter accessions Received: 20 December 2021 may harbor novel QTLs effective for STB resistance. Accepted: 9 February 2022 Published: 16 February 2022 Keywords: Mediterranean landraces; durum wheat; Zymoseptoria tritici; phenotyping; sources of resistance; diversity; seedling; adult; agronomic traits Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. 1. Introduction Durum wheat is an important crop in the Mediterranean basin that has been cultivated over centuries and under widely variable climatic conditions. The crop originated and was Copyright: © 2022 by the authors. domesticated in the Fertile Crescent (10,000 BP), and spread from the east to the west of the Licensee MDPI, Basel, Switzerland. Mediterranean basin [1], reaching the Iberian Peninsula around 7000 years BP [2]. The di- This article is an open access article versification of the durum wheat genome and the development of a large collection of local distributed under the terms and populations in this region [3] are therefore mainly due to the contrasting environmental conditions of the Creative Commons conditions. Moreover, multiple invasions that have occurred in the region, the migration of Attribution (CC BY) license (https:// wheat from the east to the west of the Mediterranean basin, wheat imports, and natural creativecommons.org/licenses/by/ and human selection are other important factors that contributed to its diversification 4.0/). Genes 2022, 13, 355. https://doi.org/10.3390/genes13020355 https://www.mdpi.com/journal/genes Genes 2022, 13, 355 2 of 20 (Mercer and Perales, 2010). This vast biodiversity within the species [4] also extends to the vast array of homemade foods derived from durum grains. Moreover, due to their high protein content and gluten strength, cultivars of durum wheat are preferred to produce semolina for use as pasta products, couscous, and bulgur [1,5]. Mediterranean durum wheat landraces have been, and continue to be, a great source of novel useful genes that could be further exploited by breeders. Several studies have revealed the usefulness of landrace genetic resources, as they do offer a key element in breeding due to their inherent adaptability to respective agroeco- logical niches [6–8] while maintaining considerable diversity between and within popula- tions [9]. However, it is well known to breeders that useful genes are often linked to some undesirable traits that lead to longer cycles of selection to remove them. Landraces have been extensively characterized in terms of genetic diversity and population structure [3,10], and have great potential for the identification of novel sources of resistance to biotic and/or abiotic stresses [3,11–14]. Climate change does not only affect abiotic stresses such as drought, heat and cold, but also affects various fungal pathogens (rusts, leaf and head blights) [15] that have the potential to rapidly adapt to climate change; hence, it presents great limiting factors for wheat production in the Mediterranean region, which hosts some of the most damaging and virulent races of diseases and pests [16]. Septoria tritici blotch (STB), caused by the foliar fungal pathogen Z. tritici, is one of the most important threats to productivity in the Mediterranean basin, and particularly in Tunisia [17–19]. Ever since its emergence in 1970—which coincided with the introduction of the commonly high-yielding durum wheat varieties that became susceptible to Septoria over a few years [20]—Tunisia, and possibly other countries, experienced serious recurrent epidemics of STB, with yield losses reaching up to 40% [21]. Currently, in many countries, disease management relies on the use of fungicides and/or resistant cultivars. However, both fungicides and resistant cultivars are likely to lose efficacy against Z. tritici due to its high evolutionary potential [22]. Other factors—such as high seeding rate, early sowing, the excessive use of fertilizers, the slow release of resistant varieties, a lack of variety replacement, scarce crop rotations, a shift to reduced/no tillage agriculture, and the excessive use of fungicides—are major constraints limiting more efficient approaches to managing STB in Tunisia. In addition, cereal crops are dominated by a monoculture of genetically uniform wheat cultivars over very large areas. In Tunisia, durum wheat variety “Karim” (Jori“S”/Anhinga“S”//Flamingo“S”), which is highly susceptible to Z. tritici, covers more than 60% of the durum wheat acreage [23–25]. Despite all of the research undertaken, STB is still a major durum disease in the Mediterranean region, and with the pressure of Septoria becoming insensitive to some fungicides [26], demand for new Septoria resistant wheat varieties by farmers has increased. Moreover, the high genetic variability of Z. tritici that is mainly driven by sexual repro- duction [27–31] allows not only the natural increase in inoculum density, but also new combinations of virulence alleles. Consequently, it seems likely that most resistance genes will not last long, and there will be a continual need to identify new sources of resistance. Breeding for resistance is therefore an ongoing process that requires new resistance alleles to be incorporated into crop varieties, which must be managed strategically. This underscores the need for crop genetic-resource conservation and the need to systematically test wild relatives, landraces, and other germplasms, to identify new genetic sources of resistance to major diseases [32–34]. Recent studies have shown that Tunisian durum wheat landraces carry effective sources for resistance to Z. tritici [35–37]. The objectives of this study were to evaluate a sub-set of Mediterranean durum wheat landraces, mainly originating from Algeria, France, Italy, Portugal, Spain and Tunisia, for resistance to Z. tritici under field conditions at seedling and adult stages. The relationship between disease development and plant height (PH) was also evaluated. The integrated field seedling and adult plant phenotyping method reported in this study provides a great tool for identifying novel and durable resistance sources. Our method has the advantage of assessing the populations over a long growing period (GS11–87) [38] under the same field Genes 2022, 13, 355 3 of 20 conditions for two years. Assessing plants at early and various growth stages can enable selection for desirable gene combinations. Moreover, the outcome of the present study will help in the identification of landraces that can be exploited for improving resistance to Z. tritici in Tunisia. 2. Materials and Methods 2.1. Plant Material, Experimental Design and Inoculation Method A collection of 3166 durum wheat accessions, provided by the USDA Aberdeen Gene Bank, including accessions from 17 different countries (Algeria (214), Palestine (3), Cyprus (44), Egypt (163), France (79), Greece (67), Israel (28), Italy (275), Jordan (56), Lebanon (11), Morocco (174), Portugal (389), Spain (128), Switzerland (2), Syria (45), Tunisia (299) and Turkey (1189)) were tested for resistance to Z. tritici under field conditions at the CRP Wheat Septoria Precision-Phenotyping Platform of the experimental station located at Kodia (36◦32′51.89 N, 9◦0′40.73 E)-INGC (Bou Salem, Tunisia), during the cropping season of 2016–2017. These materials were at different improvement statuses, including landraces (2352), cultivars (234), genetic material (15), breeding material (220) and unknown improve- ment status (345). Only 1059 (33.5%) of these accessions were retained and tested during the 2017–2019 cropping seasons for Septoria tritici blotch (STB) resistance (Supplementary Table S1). The remaining 2107 accessions (66.5%) were unadapted, as they were either winter type, highly susceptible to yellow rust, or failed to germinate, and were thus not considered for further analysis. The 1059 accessions, mostly composed of landraces (66%), were then screened for STB resistance/tolerance at both seedling and adult growth stages, at the location that is known as a hot-spot region for Septoria. These accessions were from Algeria (190), Egypt (7), France (58), Greece (2), Israel (3), Italy (199), Jordan (6), Morocco (8), Portugal (304), Spain (68), Syria (1), Tunisia (208) and Turkey (5) (Figure 1). All accessions were planted on 16 November 2016 and 13 November 2018, in a wheat-after-wheat produc- tion system. An augmented experimental design with unreplicated entries and replicated checks was implemented during both 2-year trials. The plots consisted of two rows of 1 m in length. The spacing between the plots and blocks was 0.5 m and 1 m, respectively. Each block contained three local checks—‘Karim’, ‘Nasr’ and ‘Salim’—known for their susceptibility, moderate susceptibility, and resistance to STB disease, respectively. These checks were sown in two rows and replicated among blocks, with a total of 30 replicates per check. The average of these replicates was used to classify the different accessions based on their levels of resistance/susceptibility (Supplementary Table S2). The replicated checks were also used as a tool to verify the uniformity of the infection among and within plots. The susceptible variety, ‘Karim’, was also planted in the middle of the block, and served as a disease spreader to further induce infection and ensure optimal disease distribution among and within plots. Straw from the previous cropping seasons was incorporated into the soil with a rotary harrow. The inoculum density was approximately 200 g·m−2. Additional, straw inocula- tions were performed when the tested material reached the growth stage GS10 [38]. These inoculations were conducted by evenly spreading freshly cut infected wheat-straw over the experimental plots and disease spreader rows, using the susceptible cultivar ‘Karim’ to ensure optimal disease development. Moreover, to maintain a high level of disease pres- sure, artificial infection was performed. Five Z. tritici strains (TU16318, TU16323, TU16344, TU16363, TU16364), which were collected from the same region in 2016 and are known for their virulence, were chosen to prepare the inoculum. The inoculum concentration was adjusted to 1 × 106 spores·mL−1 amended with 0.1% Tween 20 surfactant (polyoxyethyle- nesorbitan monolaurate) (Merck, Watford, UK), to reduce surface tension. Wheat plants were inoculated after sunset using a sprayer (Efco AT800, Reggio Emilia, Italy) three to four times prior to stem elongation at the tillering stage (from GS11 to GS29) [38]. Irrigation was applied as needed to ensure favorable conditions for STB development and standard wheat agronomic practices were applied during the 2016–2017 and 2018–2019 crop seasons. Genes 2022, 13, 355 4 of 20 Genes 2022, 13, x FOR PEER REVIEW 4 of 22 Moreover, plant height was measured for all 1059 accessions during the two-year trials at maturity, for each plot, from ground level up to the extreme of the spike (including awns). Figure 1. Origin of 1059 accessions tested over 2-year trials (2016–2017 and 2018–2019) at the CRP WhFeiagtu Sreep1t.orOiar iPghinenoof t1y0p5i9nga cPcleastsfioornms, teexspteedriomveenrt2a-l ysetaatriotrni aolfs K(2o0d1i6a– (2B0o1u7 Saanldem20, 1T8u–n2i0s1ia9)). at the CRP Wheat Septoria Phenotyping Platform, experimental station of Kodia (Bou Salem, Tunisia). 2.2S.trDaiwse afsreomRa tthineg previous cropping seasons was incorporated into the soil with a ro- tary harrow. The inoculum density was approximately 200 g·m−2. Additional, straw inoc- ulationsT wheerMe epdeirtfeorrrmaneeda nwchoelnle ctthieo nteosfte1d0 5m9 aatcecreisasl iroenasc,hheedr etahfete grrcoawllethd sthtaegMe eGdS-1c0o l[l3ec8t]i.o n, Thwe as subsequently evaluated for STB resistance under field conditions. To better discriminate ressei sintaonccuelaattiosenesd wlinegrea cnodnadduuclttedgr boyw ethvesntalgye sspurenaddeirnfige flrdecsohnlyd cituiot nins,feacctceedss wiohnesawt-esrterarwat ed oveart tGhSe e(1x1p–e2r0im) aenndtaGl pSlo(3t7s –a8n7d) dstiasgeaess,e rsepsrpeeacdtievre rloy.wAs,t uthsiensge tehdel isnugscsetapgtieb, lae 0cutolti5vascra ‘lKeaw- as rimim’ top leenmsuenrete odp(tTimaballe d1i)s,ewashee dreevsecloorpems beenttw. Meeonre0o–v1eirn, dtoic mataeianntaiimn ma huingeh tloevheilg ohfl ydirseesaisstea nt pre(sHsuRr)er, easrptoifnicsiea.l Sincfoercetsioenq uwaalst ope2rfaonrdm3edin. dFiicvaet eZr. etsriisttiacin tst(rRai)nasn (dTUm1o6d3e1r8a, teTlUy 1r6e3si2s3t,a nt TU(1M63R4)4, ,r eTsUp1e6ct3i6v3e,l yT,Uw1h6i3le64s)c,o wrehsicohf 4waenred c5oclloercrteesdp ofrnodmt othme osadmerea treelgyiosuns icne p20ti1b6l ea(nMd Sa)rea nd knosuwsnc efpotri bthleei(rS )v, irreuslpeenccteiv, ewlyer(eT acbhloes1ena ntod pTraebplearSe1 )t.he inoculum. The inoculum concen- tration was adjusted to 1 × 106 spores·mL−1 amended with 0.1% Tween 20 surfactant (pol- yoxTyabetlhe y1.leRneeascotirobnittaynp ems oofnthoelaMureadt-eco) ll(eMcteiorncka,t tWheastefoedrdli,n, gUstKag),e taog ariendstuZc.et rsituicrif.ace tension. Wheat plants were inoculated after sunset using a sprayer (Efco AT800, Reggio Emilia, Reaction Type at GS Scale Italy) three t(o1 1fo–2u0r) times prior to s0t–e5m elongation at the tiSllyemripntgo mstaDgees c(frripotmio nGsS11 to GS29) [38]. Irrigation was applied as needed to ensure favorable conditions for STB development and standSaeredd lwingheHaRt agronomic p0–r1actices were appLleiesdio ndnuortinagp ptahree n2t0o1r6v–e2r0y1s7m aanlld 2018– 2019 crop sSeeaesdolinnsg. R 2 Apparent small lesion MoreSoevedelri,n gplManRt height was 3measured fToyrp iaclall 1Z0. 5tr9it iacci cl essisoinosn(sp ydcunridiniag) othnel etawveos-y2–e3ar trials at mSaeteudrliitnyg, fMorS each plot, fro4m groundW leevlle-dl euvpe ltoop tehdel eesxitornemupe toof tthhired spanikdef o(iunrcthluldeainvegs awns). Seedling S 5 Clear, susceptible reaction present on all leaves 2.2H. DR:isheigahsely Rreastiisntagn t; R: resistant; MR: moderately resistant; MS: moderately susceptible; S: susceptible. TheA Mt tehdeitaedruraltnsetaang ec,oSllTecBtiporno gofr e1s0s5io9n acwceasseiovnaslu, hateerdeabftyerm ceaallseudr itnhge SMTeBdi-nccoildleecnticoena, nd wasse svuebristeyqbuaesnetdlyo envtahlueadteodu bfoler- dSTigBit rsecsaisleta(n0c0e– 9u9n)d[e2r0 ]fiwelhde croenthdeitifiornsst.d Tigo ibt einttdeirc datiescsrdimisei-ase natien criedseisntcaencoen atth eseiendfelicntged apndla natdsu, latn gdrothwethse sctoangdesd uignidt erer fefiresldto ctohnedsiteivoenrsi,t yacocfeisnsifoenctsi on we(rTea rbaltee2d) abty GeSv a(1lu1a–t2i0n)g atnhde pGySc n(3id7–ia87c)o vsteargaegse,. rAescpceescstiiovnelsyw. Aerte thraet esdeeodvlienrgm sutaltgipe,l ea c0o tnos ec- 5 sucatilvee wtiams eism, sptlaermtinegntwedit h(TGaSb3le7 ,1a)t, 1w0-hdearye isnctoerrevsa lbse[t3w8]e.en 0–1 indicate an immune to highly resistant (HR) response. Scores equal to 2 and 3 indicate resistant (R) and moder- ately resistant (MR), respectively, while scores of 4 and 5 correspond to moderately sus- ceptible (MS) and susceptible (S), respectively (Tables 1 and S1). Genes 2022, 13, 355 5 of 20 Table 2. Reaction types of the Med-collection at the adult stage against Z. tritici. Reaction Types at GS (37–87) rAUDPC Range Symptom Descriptions Adult HR <0.2 No symptoms or small lesions at lower leaves Adult R 0.2–0.4 Apparent infection at lower leaves but small lesions Adult MR 0.4–0.6 Mid-height infection, and apparent lesions not well developed Adult MS 0.6–0.8 Well defined lesions up to flag leaf, and well-developed lesions up to F−1 * Adult S >0.8 Well-developed lesions up to flag leaf HR: highly resistant; R: resistant; MR: moderately resistant; MS: moderately susceptible; S: susceptible. * F−1 corresponds to the leaf below the flag leaf. Hence, the symptoms and lesion development over the assessment period were summarized by the area under disease progress curve (AUDPC), which allowed the identi- fication of different classes of resistance. The area under disease-progress curve (AUDPC) and the relative area under diseaseprogress curve (rAUDPC) were determined according to [39]; n−1 y + y AUDPC = ∑ i i+1 × (ti+1 − t i=1 2 i) where: yi: STB severity at time ti, t(i+1) − ti = time interval (days) between two disease scores, n = number of times when STB was recorded. AUDPC (genotype) rAUDPC = AUDPC (Karim) where: Karim is the susceptible check of the corresponding trial. Based on the levels of resistance and susceptibility of the local checks Salim, Nasr and Karim of each trial, five classes were established at adult growth stage (Table 2). The relative area under disease progress curve (rAUDPC) was calculated for all 1059 accessions over the 2-year trials using the AUDPC of the susceptible check ‘Karim’ for each corresponding cropping season. The rAUDPC of both ‘Salim’ and ‘Karim’ (Table 2 and Table S2), were ranked as highly resistant (HR), resistant (R), moderately resistant (MR), moderately susceptible (MS) and susceptible (S) (Table 2 and Table S1). 2.3. Statistical Analysis R software version 4.1.2 (R Foundation for Statistical Computing (R Core Team (2021)) [40] was used for all data analysis. Principal Component Analysis (PCA) was performed on the parameters PH, AUDPC and rAUDPC over the two testing seasons using the R package ‘MASS’ [41]. The determination and visualization of clusters was performed using R packages ‘factoextra’, ‘cluster’ and ‘stats’ [40,42,43]. The coefficient of the correlation between variables (seedling and adult reaction, PH) was determined with ‘cor.test’ function from the R package ‘stats’. The analysis of variance (ANOVA) was performed with the ‘aov’ function from the R package ‘stats’ [40]. 3. Results 3.1. Disease Response of the USDA Mediterreanean Collection The USDA Mediterranean collection showed a high frequency of resistance (HR and R = 83.9%), among which 58.5% showed an immune reaction (HR) at adult growth stage (Table 3). The absence of typical symptoms of Z. tritici in these accessions could be attributed to a major-gene resistance or disease-escape traits, notably height. Genes 2022, 13, 355 6 of 20 Table 3. Disease reaction at adult growth stage of the 3166 Mediterranean landraces used in this study. 2016 HR R MR MS S Missing TOTAL Subset 1853 800 268 129 97 19 3166 Algeria 58 119 28 3 6 0 214 France 28 24 10 10 7 0 79 Italy 122 62 41 30 20 0 275 Portugal 317 59 10 1 1 1 389 Spain 104 18 6 0 0 0 128 Tunisia 155 111 21 7 5 0 299 Other 1069 407 152 78 58 18 1782 Total 1853 800 268 129 97 19 3166 HR: highly resistant; R: resistant; MR: moderately resistant; MS: moderately susceptible; S: susceptible. A set of the USDA Mediterranean accessions (33.5%) was kept for the 2018–2019 cropping season, and comprised 1059 accessions that we named the Med-collection. The latter was composed of landraces (66%), breeding lines (12%), unknown improvement status (11%), cultivars (9%) and genetic material (<2%). The accessions were from ma- jor durum-wheat-growing countries where Septoria is a major disease of durum wheat, including Algeria, France, Egypt, Italy, Portugal, Spain, Tunisia, Greece, Israel, Morocco, Syria, Jordan, and Turkey. Countries represented by low numbers of accessions (≤8) were grouped together and named Other. Data on disease evaluation based on seedling and adult scores (AUDPC and rAUDPC) Genes 2022, 13, x FOR PEER REVIEW 7 of 22 under field conditions of the Med-collection, as well as phenology data such as plant height, over two seasons of the experiment are shown in Supplementary Table S1. 33.2.2. .RReaecatciotniosn osf GofeGnoetnyopteys paecrsoascs rtohses Ttwheo TwriaolsT rials TThhe ecocmompapraisroisno onf otfhteh me emanea rnArUADUPDCP aCnda nstdanstdaanrdd adredvdiaetvioina toiof nalol foaf ltlhoef cthheecckhs ecks used uisnedth iins tshtiusd styuadrye arreep rreepsreensetnedtedin inS uSupppplelemmeennttaarryy TTaabbllee SS22 aanndd FiFgiugruer 2e. 2. FFigiguurer e2.2 B. oBxo pxloptslo otfs tohfe trheelatrievlea tairveea aurnedaeru tnhdee dristehaesed-pisreoagsres-ps rcougrvre s(srAcuUrDvPeC()r AofU thDeP MCe)do-f the Med- ccoolllelcetciotino annadn thde tchheecckhs eincokcsuilnatoecdu wlaitthe dZ. wtriittihci uZn.dterrit fiiceildu cnodnedritfiioenlsd inc othned 2i0t1io6n–2s01in7 atnhde 22001186––2017 and 22001198 c–ro2p01p9incgr osepapsoinngs. seasons. OOptpimtimal arlairnafianllf aalllonalgo wngithw tihteh atrhteifiaciratli fiincoiacluilantoiocnusl aatliloownseda lgloowode dengvoiroodnmeennvtiarlo nmental conditions that were conducive to the development of Septoria among the genotypes and tchoe ncdheitcikosn dsutrhinatg wboetrhe sceoansodnusc. iNveevteortthheeledsse,v ae hloigphmere dnitseoafsSe epprteossruiarea fmoro tnhge ethvealguaetneodt ypes and accessions was observed in 2017 (Figure 2). The combined analysis of variance showed no significant difference at seeding stage (p < 1) and a moderate significant difference at the adult stage (p < 0.01) between years 2017 and 2019, with regard to the disease progress (Figure 2, Table 4). At the seedling and adult stages, the genotype term in the ANOVA analysis was highly significant at p ≤ 0.001, which confirms that the observed variation was mainly due to the contribution of the variable genetic background of the tested germplasm. Conversely, there was no significant genotype–year interaction, indicating that genotypes behaved similarly between years (Table 4). Genes 2022, 13, 355 7 of 20 the checks during both seasons. Nevertheless, a higher disease pressure for the evaluated accessions was observed in 2017 (Figure 2). The combined analysis of variance showed no significant difference at seeding stage (p < 1) and a moderate significant difference at the adult stage (p < 0.01) between years 2017 and 2019, with regard to the disease progress (Figure 2, Table 4). At the seedling and adult stages, the genotype term in the ANOVA analysis was highly significant at p≤ 0.001, which confirms that the observed variation was mainly due to the contribution of the variable genetic background of the tested germplasm. Conversely, there was no significant genotype–year interaction, indicating that genotypes behaved similarly between years (Table 4). Table 4. Analysis of variance (ANOVA) of the relative area under the disease progress curve on the Med-collection inoculated by Z. tritici, at seedling and adult plant stages, during two cropping seasons (2016–2017 and 2018–2019). Physiological Source of Sum of Mean of Stage Variation Squares Squares F Value Pr (>F) Genotype 3344 3.160 2.518 <2 × 10−16 *** Residuals 1218 1.255 Seedling Year 1 0.635 0.282 0.595 Residuals 4561 2.249 Genotype × Year 1236.302 0.317 0.9238 0.9238 Residuals 4.000 4.00 Year 0.34 0.335 6.046 0.014 * Residuals 117.17 0.055 Genotype 88.15 0.083 2.988 <2 × 10−16 *** Residuals 29.36 0.027 Adult Genotype × Year 29.164 0.027 0.524 0.832537 Residuals 0.529 0.529 Seedling 36.1 36.17 958.6 <2 × 10−16 *** Residuals 76.49 0.04 PH 23.45 0.868 19.48 <2 × 10−16 *** Residuals 89.22 0.044 Level of improvement 14.96 3.739 77.46 <2 × 10−16 *** Residuals 97.71 0.048 Significance codes: *** highly significant as p ≤ 0.001; * significant as p ≤ 0.05. The Med-collection showed a diverse response to STB (Supplementary Table S1) under field conditions at both seedling and adult stages, exhibiting reactions that ranged from susceptible (S) to highly resistant (HR) (Figure 3). Results showed that over 50% of the accessions had immune to highly resistant reactions (Figure 3), and around 30% had good resistance levels. Accessions in this category could be used as source of resistance in breeding programs. About 7% of the accessions in 2019 were susceptible types that should not be considered for exploitation by breeding programs. Interestingly, 92% of HR and R accessions maintained their resistance levels across the two years. The Pearson’s correlation coefficient (r) values between two repeated experiments was highly significant, with 0.51 and 0.43 for adult–adult and seedling–seedling reactions for both years, respectively (Table 5). Similarly, the correlation between seedling and adult reactions was highly significant for both years (Pearson’s correlation r = 0.597, p ≤ 0.001; r = 0.455, p ≤ 0.001, respectively). GGeenneess 22002222,, 1133,, x3 5F5OR PEER REVIEW 9 of 22 8 of 20 FFigiguurere 33. B. Bara gr rgarpahp hshsohwowinign gthteh reersepsopnosnes oefo tfhteh ceocroe rceoclloelclteicotnio angaaginasint Zst. Ztr.ittirciit iacti saetesdeleidngli nagnda naddualdt ult ggrroowwthth ssttaaggeess uunnddeerr ffiielld conditions for two croppiing sseeaassoonnss ((22001166––22001177a anndd2 2001188––22001199).).T ThheeX X-a-xis arxeips respernetssenthtse tthyep teyopfe roefs irsetsainstcaenfcoeu fnoudnidn ianc caecscseisosniosncso cmopmrpisriinsignhg ihgihglhylyre rseisitsatnatn(tH (HRR),)r, erseissitsatnant t( R), (Rm)o, mdeordaeterlayterleys risetsaisntta(nMt (RM),Rm), omdoedraetrealtyelsyu ssucsecpetpibtilbel(eM (MS)S,)a, nadnds ususcsecpeptitbiblele( S(S))a acccceessssioionnss.. VVaalluueess on otnh ethbea brarre prerpesreensetnfrt efqreuqeunecnycoyf olfa nladnrdarcaecse(s% (%) b) absaesdedo nonth tehierirle lveevleol fofr erseissitsatnancec/e/ssuusscceeppttiibbiilliittyy.. TablTe h5.eP Peearasrosno’ns’cso crroerlraetiloantiocone fcfioceieffnitcfieonr tt h(er)d ivsaelauseesp abreatmweeteerns tawndo argerpoenaotmedic etrxapitesraimmoenngtst he was highly significant, with 0.51 and 0.43 for adult–adult and seedling–seedling reactions wheat genotypes across seasons (2016–2017 and 2018–2019). for both years, respectively (Table 5). Similarly, the correlation between seedling and adult reactions was highly significSaenetd floirn gbo2t0h1 9years (PearsoPnH’s2 c0o1r7relation r = 0.r5A97U, Dp P≤C 02.001091; r = 0.455, p ≤ 0.001, respectively). Seedling 2017 r = 0.431 p < 2.2 × 10−16 *** - - Table 5. Pearson’s correlation coefficient for the disease parameters and agronomic traits among the wheat genPoHty2p0e1s9 across seasons (2016–-2017 and 2018–2019).r = 0.679 r = −0.407 p < 2.2 × 10−16 *** p < 2.2 × 10−16 *** rAUDPC Seedling 2019 PH2017 2017 - r = −0.264 rAUDPC2r0=190 .512 r = 0.431 Seedling 2017 p < 2.2 × 10−16 *** p < 2.2 × 10−16 *** AUDPC = area undpe