Resistance and tolerance reactions of winter wheat lines to Heterodera filipjevi in Turkey

Abstract Nematodes attack cereal crops resulting in significant yield losses, estimated at 10%. The plant parasitic nematodes of the genus Heterodera attack cereals, particularly wheat, causing costly financial losses due to impact on yield. The soil borne pathogens program at the International Maize and Wheat Improvement Center (CIMMYT) in Turkey has focused on screening wheat germplasm to identify sources of Heterodera resistant varieties for almost 20 years. The aim of this current study was to validate the finding that resistant lines demonstrate resistant reactions under controlled conditions and to test whether they present tolerant reactions when challenged with cyst nematodes under two different locations in field conditions. The results of this study, including the check lines, indicated that 27 and 28 lines maintained their reactions to H. filipjevi in Eskisehir and Yozgat field, respectively, and 23 lines were the same in both locations. In terms of tolerance, 3 and 13 lines proven to be tolerant and moderately tolerant to H. filipjevi in Yozgat field. In Eskisehir field, 13 and 14 lines were tolerant and moderately tolerant. In both locations, L7 showed tolerance reaction, although it was susceptible. The majority of the resistant germplasm (60%, 14 lines) of screened lines from the Turkey CIMMYT–ICARDA (TCI) nursery were found to be resistant to both H. filipjevi populations including L1, L3, L6, L15, L21, L26, and L34, whereas 17% (four lines) from the USA had the same reaction. L32 showed a high level of resistance and tolerance in both locations and could prove to be promising lines in the breeding programs. The International Winter Wheat Improvement Program (IWWIP) formerly used these resistant lines in the crossing block and subsequently distributed them to more than 150 international collaborators. Regression analysis revealed a negative correlation between yield and RF of H. filipjevi in both nematode populations, which describes the negative impact of this pest on winter wheat. The results of this study are very important for breeding programs especially for the IWWIP, a joint program between the Turkish Ministry of Agriculture and Forestry, CIMMYT, and the International Centre for Agricultural Research in the Dry Areas (ICARDA).

Wheat (Triticum aestivum L.) is a vital staple food crop. An estimated 750 million tons (MT) of the grain was grown on more than 220 million hectares (Mha) worldwide in 2017 (Wuletaw et al., 2016;Dababat and Hendrika, 2018). Turkey contributed more than 22 MT over 7.77 Mha in 2013(FAO, 2018. The world's population is expected to reach 9 to 10 billion in 2050; thus, based on population growth, cereal production must be increased by 50% by 2030 to meet demand (Alexandratos and Bruinsma, 2012).
The average area (in Mha) of production (in MT) and yield (tonnes/ha) of wheat from 2008 to 2012 by country in the Central West Asia and North Africa (CWANA) region is depicted as reviewed by Wuletaw et al. (2016). Their research indicates that low productivity levels of wheat in the CWANA region is due to abiotic stresses (drought, cold, heat, and salinity) and biotic stresses (stripe rust, leaf rust, stem rust, root rots, Russian wheat aphid, barley yellow dwarf virus, sunn pest, and Hessian fly). Recently, Wuletaw et al. (2016) also reviewed wheat production in Sub-Saharan Africa under a changing climate.
Cereal cyst nematode (Heterodera avenae complex; CCN) is found globally and leads to significant economic yield losses, particularly in areas where dryland and cereal monoculture systems are practiced (Nicol et al., 2003;Dababat et al., 2015). After hatching, second-stage juvenile (J2) cyst nematodes start attacking the roots of the plant and begin feeding. The J2s penetrate the root system and after mating, females produce several 100 eggs inside their bodies. The eggs grow, transforming the female into an oblate spheroid. When the host plant begins to die, the cuticle hardens and becomes a cyst, which stores the eggs until the following growing season when they hatch . CCN accelerates synergistic negative effects when combined with other biotic and abiotic factors, such as drought and fungal pathogens (Nicol and Ortiz-Monasterio, 2004;Nicol et al., 2006;. Damage caused by CCN has been reported from different parts of the world and is estimated to have caused grain losses at rates of 20, 90, 50, and 24% in Pakistan, Saudi Arabia, Australia, and the USA, respectively (Nicol, 2002;Dababat and Hendrika, 2018). CCN (mainly H. avenae) is calculated by Barker et al. (1998) to have caused losses of about $78 billion around the globe. Worldwide, cereal production losses are estimated at 10% due to plant-feeding nematodes (Whitehead, 1998).
So far, several attempts have been made to control CCN, including through chemicals (Dababat et al., 2014, crop rotation (Smiley et al., 2008), genetic resistance (Dababat et al., 2014(Dababat et al., , 2016, and biological control . Genetic resistance is the most reliable control option as it is cheap, easy to use once identified, and environmentally friendly. Ideally, resistance (the ability of the plant to inhibit nematodes multiplication) should be combined with tolerance (the plant's ability to withstand infection and produce yields despite the CCN attack) (Smiley et al., 2004).
However, on a global scale, there are very few programs working to breed for nematode resistance in wheat. Plant parasitic nematodes are considered to be one of the leading destructive diseases attacking wheat and cause significant yield losses (Dababat and Hendrika, 2018). As yield loss to nematodes presents a clear obstacle to achieving the goal of increased cereal production, a targeted breeding program will be necessary to discover new sources of nematode resistance in wheat.
The soil-borne pathogen program at CIMMYT Turkey annually receives about 1,000 accessions of wheat from the CIMMYT Mexico spring wheat program and IWWIP. They are screened under growth room, greenhouse and field conditions at various locations in Turkey. Cultivars are also screened for multiple disease resistances, such as resistance to different species of root lesion nematodes (e.g. Pratylenchus thornei and P. neglectus), other cyst-forming nematodes (e.g. H. avenae, H. filipjevi, and H. latipons) and two Fusarium species (F. culmorum and F. pseudograminearum). The most resistant lines are then distributed to international collaborators for use in their breeding programs.
However, the habits and nature of CCN is complex due to the existence of pathotypes, which require long-term investment strategies. This is vital to ensure that a selected resistant wheat line can target as many different pathotypes as possible in addition to other Heterodera species. Therefore, the aims of this study were: (i) to screen winter wheat germplasm provided by IWWIP against the cereal cyst nematode H. filipjevi, (ii) validate the resistance of those lines under the infested fields with H. filipjevi, and (iii) study and compare the tolerance reaction of lines under field conditions with the proven resistant ones under controlled conditions.

Germplasm selection
During the annual screening of the winter wheat germplasm provided by IWWIP, a group of 31 entries showed promising resistance potential to H. filipjevi under growth room conditions at the Transitional Zone

Field experiments
Trials were planted in October for both locations and growing seasons, with each entry sown in a plot (1 m wide × 2 m long) consisting of four rows with 20 cm between rows in the TZARI field, whereas plots in ICEA were 5 m long × 1.2 m wide consisting of six rows with 20 cm between rows. Seeds were sown at a rate of 550 to 600 seeds per square meter. A randomized complete block design was used in both locations with three replicates. Trials were fertilized at sowing time with 10-10-10-5 NPKS at 200 kg/ha. Bronate® herbicide (MCPA + bromoxynil) was applied at 470 ml/ha to control broadleaf weeds. Trials were terminated in July. At harvest, spikes from each plot were manually harvested using a sickle and threshed with a small harvester at TZARI and by experimental harvesting machine in ICEA. Grain weight per plot was weighed and recorded.

Nematode sampling
Six soil subsamples of 250 cm 3 each were taken per plot in a zigzag pattern at the beginning of the growing seasons to assess nematode initial populations (P i ). Samples were taken from around seed rows at 20 cm depth using a soil auger (2.5 cm diameter). Extraction of the cysts from the soil was performed as per Dababat et al. (2014) following Fenwick's (1940) can floatation methods for cyst extraction. These trials were replicated for data validation. At maturity or at harvest (early July), similar soil sampling and extraction methods were performed to determine nem-atode final population density (P f ) per plot by taking soil samples from the root niche. The nematode P i and P f were used to calculate the nematode reproduction factor (RF) based on equation: RF = P f /P i . An accession's resistant reaction was classified into one of five distinctive groups based on the reproduction factor (RF): resistant (R) = RF equal or less than 1; moderately resistant (MR) = RF between 1 and 2, a few more cysts than in a resistant check; moderately susceptible (MS) = RF between 2 and 3, distinctly more cysts than in a resistant check, but less than in the susceptible check; susceptible (S) = RF between 3 and 4, more cysts than in the susceptible check; and highly susceptible (HS) = RF more than 4, cyst number higher than in the susceptible check (Dababat et al., 2016) and taking into the account the reaction of the known check lines used in the study. The accession's tolerance reaction was classified into four groups based on the reproduction factor (RF) and yield potential. The groups were: tolerant (T) = plant yield well despite of high nematodes attack; moderately tolerant (MT) = plant yield moderately under moderate nematode attack; intolerant (IT) = plants did not yield well sumptions of normality, Shapiro and Wilk's (1965) test was conducted. Significant differences between lines were detected using protected least significant difference at P < 0.001 using SPSS statistical software V 17.0 (SPSS Inc., Chicago, IL, USA). Linear regression analyses were conducted to describe the relationship between the H. filipjevi reproduction factor (RF) and the grain yield for each line in the two even under low nematode pressure; highly tolerant (HT) = plants yield well though under very high nematode attack (Smiley et al., 2014;Dababat et al., 2019).

Statistical analysis
The CCN data were transformed and then analyzed using analysis of variance (ANOVA). To meet as-

Results
Results indicated that 23 lines kept their resistant reaction under both field conditions. Thus, only four (L7, L16, L24, and L35) showed susceptibility reaction under both locations (Table 1). Three lines were proven to be tolerant and 13 lines moderately tolerant to H. filipjevi population in Yozgat, while 9 and 13 lines were tolerant and moderately tolerant in Eskisehir region, respectively. Around 60% (14 lines) of the resistant lines from TCI were found resistant to both H. filipjevi populations including L1, L3, L6, L15, L21, L26, and L34, whereas 17% (four lines) from the USA had the same reaction including L9, L29, L32, and L33. Line 32 gave tolerance and resistance reactions to both H. filipjevi populations in both regions. The two susceptible check cultivars "Bezostaya" and "Kutluk" confirmed their susceptibility and intolerance reactions in both locations. Additionally, the two resistant cvs., 'Katya' and 'Sonmez' conserved their resistance in both locations. Analysis of population structure based on grain (resistance reaction) displayed two distinct groups among the evaluated 35 lines (including checks) in Eskisehir and three distinct groups in Yozgat experiments (Fig. 1A). The same analysis describes the population structure based on obtained tolerance reaction and displayed four distinct groups among the evaluated 35 lines (including checks) in Eskisehir and Yozgat experiments (Fig. 1B).
In Yozgat (ICAE), we obtained 28 resistant lines (R), including L1, L5, L14, L15, and L32, alongside six susceptible lines (S), including L4, L7, L16, and L20 and one moderately susceptible line (MS), L10 (Fig.  1A), whereas the same analysis displayed four distinct groups in terms of tolerance reaction among the 35 evaluated lines (Fig. 1B). The first group indicated three tolerant lines (T), including L1 and L32. Group 2 comprised 13 moderately tolerant lines (MT), including L5, L14, L22, and L31. Group 3 comprised 10 moderately intolerant lines (MIT), including L2, L3, L8, and L24. The final group comprised nine intolerant lines (IT), including L6, L10, L26, and L33. L32 and L1 in the first group have the best tolerant performance in Yozgat field with grain values of 5187 kg/ha and 3908 kg/ha and RF values of 0.4 and 0.2, respectively. Additionally, the L5 in the second group has a grain value of 3,064 kg/ha and RF of 0.7. L28 in the last group has the lowest grain yield (1,291 kg/ha) although under low nematode population of 0.9 RF, indicating this line was resistant and intolerant (Fig. 1). In this location, only two lines (L1 and L32) were shown to be resistant and tolerant, whereas several lines demonstrated a combination of susceptibility and moderate intolerant reactions including L16, L20, and L24.
As shown in Figure 1, the result from ICEA of lines L7, L12, and L21 showed both higher yields, and higher RF which confirms some tolerance of the lines in the TZARI experiment against H. filipjevi. Those lines could be accepted as tolerant lines in the present study. For lines L6, L23, L24, L26, and L29 in ICEA experiment, grain yield was low even at low initial densities of nematodes (P i ), indicating intolerance and susceptibility of those lines to H. filipjevi. Line L7 was the only line that combined susceptibility and tolerance to this nematode in both locations, providing a high yield despite high RF.
We detected significant differences (P < 0.001) in the host status of the 31 IWWIP lines tested to H. filipjevi based on their yield and reproduction factor (RF) in field conditions in TZARI (RF) (Fig. 2). The highest yield was obtained from the L32 (5,363 Kg/ha), which showed tolerant and resistant reactions to both nematode populations. On the other hand, the L35 (check line) gave the lowest yield (2,785 Kg/ha) with an intolerant reaction. Of the tested 31 lines, 5 lines (L4, L11, L29, L33, and L34) were as resistant as the controls with known resistance to H. filipjevi (L27) and significantly lower RF compared to the other IWWIP lines evaluated in this study. The lowest RF was achieved by L3 (0.4), while L12 gave the highest RF in the Eskisehir field experiment (2.2) (Fig. 2).
The results indicated that there was a significant difference (P < 0.001) in the host status of the IWWIP lines to H. filipjevi, ranging from resistant to susceptible in terms of the reproduction factor, which fluctuated from 0.2 to 1.4 in ICEA field (Fig. 2). Among the studied lines, four (L5, L3, L15, and L29) were as resistant as the controls with known resistance to H. filipjevi (L18 and L27) and significantly lower RF compared to the other IWWIP lines evaluated in this study. The lowest RF was achieved by L1 (0.2), while both L20 and L35 gave the highest RF values (1.3 and 1.4, respectively). A total of 27 lines were shown to maintain resistance in Yozgat region (Fig. 2). The highest yield was obtained by L32 (5,187 kg/ha), indicating that this line is both resistant and tolerant to H. filipjevi. While the lowest yield was obtained by L28 (1,291 Kg/ha) with a resistance capacity and intolerant reaction even with a RF value of 0.9 (Fig. 2).

Discussion
The current study reports the first quantifiable and comprehensive evidence of the host status of different winter wheat accessions under H. filipjevi infested fields in the provinces of Yozgat and Eskisehir in Turkey. The results of this study indicate that H. filipjevi is considered an important factor which causes severe damage to different wheat cultivars and has potential to decrease growth and yield of wheat in Turkey Toktay et al., 2015;Imren et al., 2016Imren et al., , 2017Yıldız et al., 2017). The resistance/tolerance reactions of the 31 winter wheat lines evaluated in this study were identified with a range of responses, from resistant to susceptible and tolerant to intolerant to H. filipjevi.
Resistant or tolerant wheat cultivars are the most successful and preferable method to manage nematodes in cereals (Dababat et al., 2014). Tolerant varieties suffer from the infection with little yield reduction even when their roots are invaded by nematodes, while resistant varieties reduce the rate of nematode multiplication in the roots (Roberts, 2002). Therefore, resistance should be combined with tolerance, the ability of the host plant to maintain yield potential in the presence of the nematode (Cook and Rivoal, 1998). Moreover, wheat varieties with resistance or tolerance have been shown to provide resistance against a wide variety of both biotic and abiotic stresses (Kimber and Feldman, 1987). Wheat yield depends on the interaction between ecological and edaphic factors, but it is highly responsive to Heterodera species infection, even when other stresses restrict yield (Smiley et al., 2005). Therefore, the use of resistant or tolerant lines requires a sound knowledge of the virulence spectrum and pathotypes of the targeted species of nematode (Majnik et al., 2003;Smiley et al., 2005). As a result, wheat accessions resistant or tolerant to H. filipjevi populations in one region might be fully susceptible to populations in other regions (Smiley et al., 2005). Findings of this study indicated that around 23 lines kept their resistant reaction under field conditions; three lines were proven tolerant and 13 lines moderately tolerant to H. filipjevi (Yozgat), while 13 lines were tolerant, and 14 lines were moderately tolerant in Eskisehir region. This location indicated four intolerant lines, fewer in comparison to Yozgat region, as it comprised nine intolerant lines, which can explain the reduction of tolerance can be caused by high nematodes densities.
Nowadays, wheat landraces and domesticated wheat have been used to identify many important agronomical traits (Kimber and Feldman, 1987;Van-Slageren, 1994). Although considerable research has been performed to identify resistance sources in wheat, no durable resistant cultivar to Heterodera species is currently available. Therefore, there is an urgent need to identify new resistance sources and to pyramid and incorporate them into high-yielding cultivars. Many studies have attempted to seek new genotypes of cyst nematode resistance (Smiley et al., 2005;Hajihasani et al., 2010;Fard et al., 2018). Nematode resistance or tolerance in wheat has been reported from pot (glasshouse and growth chamber) and field experiments (Hajihasani et al., 2010;Şahin et al., 2010;Moustafa et al., 2018). Hajihasani et al. (2010) reported that several growth parameters (plant height, root dry weight, aerial shoot dry weight, and grain yield) of the wheat cultivar "Sardari" were significantly decreased when H. filipjevi populations increased. Their results showed that H. filipjevi caused significant reduction in grain yield when grown under high nematode initial population density. Yuan et al. (2011) investigated the resistance of 75 wheat cultivars or lines from CIMMYT under greenhouse and field conditions using a relative resistance index (RRI) and P f /P i ratios of H. filipjevi population from Xuchang, Henan province, China and found no cultivar was immune to H. filipjevi. However, a few entries displayed resistance or high resistance in both environments, such as 6R(6D), MACKELLER, CPI 133842, CPI 133814, and CROC_1/AE.SQUARROSA (224) // OPA-TA*1. Moustafa et al. (2018) screened 17 local and international wheat genotypes against Hal population of H. avenae in Saudi Arabia and their results showed that the 10 studied wheat genotypes were significant-ly different in terms of their resistance and tolerance to nematode. On the other hand, CIMMYT spring wheat genotypes (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31) plus the cvs. AUS-30851 and Yecora Rojo were found to be the most susceptible genotypes to the tested Saudi population of H. avenae. Fard et al. (2018) reported that grain yield was significantly affected by H. filipjevi as well as other growth parameters in all Back-cross Rowshan, Pishtaz, and Parsi cultivars in the province of Isfahan in Iran.
In Turkey, breeders have selected wheat varieties for Heterodera resistance for decades due to high infection pressure. Şahin et al. (2010) also screened 150 international and national wheat varieties against H. filipjevi under in vitro conditions and found that 5 international (Hartog, Katea, HN7/OROFEN/BJN8/3/ SERI, IWA8604765, and IWA8608077) and 10 national wheat genotypes (Yakar 99, Sönmez, Kırmızı Mısri, Altındane 12, Kunduru 1149, Yelken 2000, Üveyik, Sorgül, Germir, and Tosunbey) were the most resistant wheat variety and pedigrees found against H. filipjevi Haymana population. Imren et al. (2012) investigated 82 wheat genotypes for their responses to H. avenae under in vitro conditions. They reported that 4 national varieties (including Adana 99 (PFAU/ SERI82//BOG"S"), 23 international germplasm, and 17 wild genotypes were found moderately resistant to nematode.  screened 42 CIM-MYT wheat lines originating from a cross between the Middle Eastern resistant landrace 'AUS4930 7.2' and the widely adapted high yielding but susceptible CIMMYT line 'Pastor' and found that 5 and 9 of these lines were resistant and moderately resistant to the Haymana population of H. filipjevi, respectively.  investigated six common bread wheat varieties grown in Turkey for their responses to H. avenae under field conditions. They found no varieties with complete resistance to H. avenae. However, they reported Adana 99, Ceyhan 99, and Silverstar were moderately resistant and that Karatopak, Osmaniyem, and Seri 82 were moderately susceptible against H. avenae. Yavuzaslanoglu et al. (2016) reported that among the screened Iranian wheat landrace accessions against H. filipjevi, one germplasm [PI628144 (syn. AUS28321)] was resistant and five other accessions were moderately resistant under greenhouse conditions. Dababat et al. (2019) reported 484 of CIMMYT's spring wheat accessions for resistance to P. thornei of which 56 lines were pre-identified as resistant under controlled growth room conditions. These lines were further evaluated for their resistance and tolerance reactions under field conditions, where 14 accessions maintained their resistance, and 16 were moderately resistant against P. thornei. Four lines gave excellent resistant and tolerance reactions to P. thornei. The relationship between the nematode reproduction factor (P f /P i ) and wheat grain yield in field experiments fit a linear regression model. In this study, the reaction of the 31 lines plus the four checks showed that around 10 lines were resistant and tolerant in Eskisehir, whereas just two lines gave the same reaction in Yozgat. The line 'L32' provided resistant reaction and tolerance capacity to both H. filipjevi populations.
The low reproduction factor of the tested lines is normal, as those lines were screened many times under the controlled conditions for H. filipjevi as they were resistant against it. The RF normally goes to a value that exceeds six in nurseries screened for the first time. Our results indicated significant negative relationship between RF and wheat yields. The reduction in grain yield increased with both increasing P i and P f . Similar results reported by Imren and Elekcioğlu (2014) who described a negative relationship between the RF of H. avenae and grain yield of three cultivars Seri 82, Osmaniyem, and Karatopak in the Eastern Mediterranean region of Turkey. The negative relationship between nematode P i densities and grain yields for the tested germplasm in Eskisehir are similar to those of microplot trials conducted in Iran on Sardari cultivar by Hajihasani et al. (2010) and Fard et al. (2018) and to the field experiments conducted by Smiley et al. (2005).
Wheat germplasm with acceptable levels of resistance and tolerance are subsequently crossed with high-yielding susceptible cultivars. This is an essential step as most of the locally adapted wheat varieties are susceptible to H. filipjevi and identifying a new resistant wheat germplasm will allow breeders to develop new crosses with their local varieties, improving their genetic resistance to the targeted nematode species. So far, hundreds of CCN-resistant genotypes have been identified from the IWWIP materials, but the genetic nature of the resistance and genetic diversity of these germplasms are not yet known. Therefore, studying the genetic background of those lines is important to understand the resistance novelty of those lines and use those lines with different genetic backgrounds in crosses and for pyramiding a new resistance and high yield that increases grain yield to support food security.