Identification of novel resistant sources for ascochyta blight (Ascochyta rabiei) in chickpea

Chickpea (Cicer arietinum L.) is the second largest pulse crop grown worldwide and ascochyta blight caused by Ascochyta rabiei (Pass.) Labr. is the most devastating disease of the crop in all chickpea growing areas across the continents. The pathogen A. rabiei is highly variable. The resistant sources available are not sufficient and new sources needs to be identified from time to time as resistance breakdown in existing chickpea varieties is very frequent due to fast evolution of new pathotypes of the pathogen. Therefore, this work was undertaken to evaluate the existing chickpea germplasm diversity conserved in Indian National Genebank against the disease under artificial epiphytotic conditions. An artificial standard inoculation procedure was followed for uniform spread of the pathogen. During the last five winter seasons from 2014–15 to 2018–19, a total of 1,970 accessions have been screened against the disease and promising accessions were identified and validated. Screening has resulted in identification of some promising chickpea accessions such as IC275447, IC117744, EC267301, IC248147 and EC220109 which have shown the disease resistance (disease severity score ≤3) in multiple seasons and locations. Promising accessions can serve as the potential donors in chickpea improvement programs. The frequency of resistant and moderately resistant type accessions was comparatively higher in accessions originated from Southwest Asian countries particularly Iran and Syria than the accessions originated from Indian sub-continent. Further large scale screening of chickpea germplasm originated from Southwest Asia may result in identifying new resistant sources for the disease.


Introduction
Chickpea (Cicer arietinum L.), is a self-pollinated, diploid (2n = 2X = 16) annual legume which ranks second worldwide after soybean as a food legume crop [1]. It is one of the oldest crops cultivated by man. Archaeological evidences of chickpea dates back to 7,500-6,800 BC were found in the Middle East. More precisely south-eastern Turkey and adjoining Syria are a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 considered as the centre of origin of the crop [2,3]. Chickpea is cultivated mainly in arid and semi-arid areas of more than 50 countries across Asia, Africa, Europe, Australia, North America and South America [4]. Globally 17.2 million tonnes of chickpea is produced from ca. 17.8 m ha land which is ca. 15% of total pulse area well as production. The crop is reported to be susceptible to more than a dozen of well documented pathogens [5]. Among them Ascochyta blight (Ascochyta rabiei Pass. syn. Phoma rabiei Pass., Didymella rabiei Kovatsch.) is the most devastating disease [6]. The pathogen can also infect wild Cicer species like Cicer montbretti, Cicer ervoides, Cicer judaicum, Cicer pinnatifidum, etc [7]. The disease was first observed during 1911 in the North-West Frontier Province region of India, which is now part of Pakistan [8]. Since then the pathogen has spread in almost all chickpea growing regions in the world. The disease has been reported from 34 countries across the six continents and is a major disease of west Asia, northern Africa and southern Europe [5,[9][10][11]. As the pathogen is the seed-borne in nature, the disease might have spread from its origin site to distant continents through chickpea germplasm exchanges. Stem breakage along with girdling and collapse of twigs and pod infection are the two most damaging symptoms of this disease. Pathogen's spores spread through water splashes and wind [12]. Disease is more severe in areas where cool temperature and humid conditions prevail during the chickpea growing season.
Serious outbreaks of the disease had been witnessed from 1981 to 1983 resulted in wiping out of chickpea in northern parts of the country. As a consequence of the frequent epidemics, several prevalent landraces are threatened from the cultivation. During 1920-30 on an average 50% of the crop area in Attock district (now in Pakistan) and adjoining areas failed due to severe outbreak of this disease [13]. In Pakistan, the blight caused losses of nearly 50% of chickpea productions in consecutive three season from 1979-80 to 1981-82 [14]. Despite heavy application of pesticide (azoxystrobin), ascochyta blight caused ca. 20% yield losses in Nebraska, USA in 2001 on almost all the chickpea planted area [15]. Ascochyta blight was found to be the most devastating disease in chickpea growing regions of North China [16] and Ethiopia [17]. As the pathogen is seed borne, the disease has reached to non-traditional chickpea growing counties like Australia, Canada and other parts of the world and has become the major yield limiting factor [10]. This indicates the global importance of the disease. Recently several varieties have been bred with much better resistance to ascochyta blight, but still they require fungicidal application at flowering and pod formation stages [5,18]. Moreover, the A. rabiei keep evolving and so it breaks down the host resistance systems in newly bred chickpea varieties [5,[19][20][21]. Therefore, new sources of resistance are required to sustain chickpea cultivation and production. Owing to the economic importance of the disease, the present study was aimed to evaluate the chickpea germplasm against this disease to identify novel sources and understand the level of ascochyta blight resistance available in chickpea collections.

Source of germplasm and selection of experimental sites
The chickpea germplasm lines were randomly selected and obtained from the National Genebank, ICAR-NBPGR, New Delhi. Accessions used in this study originated from 17 countries. Maximum number of the chickpea germplasm screened are of Indian origin (1,567 acc.) followed by Iran (155 acc.), Syria (75 acc.), Ethiopia (71 acc.), Mexico (26 acc.), Turkey (21 acc.) and other countries (55 acc.) ( Table 1 & Fig 1). The detailed accession wise passport information is given in the S1 Data. The experiments were conducted at two locations i.e. Punjab Agricultural University, Ludhiana, India (30˚54' N and 75˚48' E) and HAREC, CSKHPKV, Dhaulakuan, India (30˚30' N and 77˚28' E). Both the locations are recognised as natural endemic regions for the Ascochyta rabiei due to optimal weather conditions for the disease spread during the season.  Table 2. Screening was done utilizing Augmented Block Design (ABD) in which susceptible checks viz. L550 and JG62 were repeated after each fifth row alternately. Resistant check used was PB 5. These are commonly used chickpea varieties used as checks for ascochyta screening [22,23]. Sowing was done on appropriate time in each winter season using recommended agronomic practices.

Ascochyta rabiei inoculum preparation
At PAU, Ludhiana, isolate 8 of race 6(3968) was used for mass multiplication, whereas, at HAREC, HPKV, Dhaulakuan, local prevalent isolates were used for creating artificial epidemics [24]. As per the recent studies on revealing pathogenic variability and diversity existing in A. rabiei, the local prevalent isolate at Dhaulakuan are AR5, AR6, AR7 [25,26]. The inoculums was mass-multiplied on Kabuli chickpea seed media. For preparing inoculums, seeds of Kabuli chickpea were soaked overnight in water and about 50 g of the soaked seeds were transferred in 250 ml flasks. These were sterilized by autoclaving twice at 121˚C (15 psi) for 25 min. Sporulated inoculum of the A. rabiei isolate grown on dextrose agar was transferred aseptically onto the seeds in the flask. The inoculated flasks were incubated at 20 ± 0.5˚C with a 12 h alternate light and dark period. The flasks were frequently shaken to avoid clumping of inoculum. Abundant conidial production was obtained by harvesting in water after 6-8 days.

Seed planting, artificial pathogen inoculation and disease assessment
The planting was done during first week of November in each winter season. Each accession was grown in 3 m row length spaced at 40 cm and replicated twice at all the locations. Indicator-cum-infector rows of susceptible checks L550 and JG62 were planted after every five test rows. To keep at least minimum threshold level of the disease and uniform spread throughout the experimental field, artificial inoculation of the pathogen was taken to avoid any chance of scape of the disease. Inoculation was done at the pre-flowering stage i.e. by January last week. The field was irrigated in the morning hours on the day of inoculation. The inoculation was

Statistical analysis
The Wilcoxon Signed Rank test, a non-parametric test was done using IBM SPSS Statistics software. The test was done for paired samples with α = 0.05 to test the H 0 hypothesis. The H 0 hypothesis of the test was that there was no difference in AB disease severity level between two experimental locations. A common set of accessions (1,230), which were screened in three seasons i.e. rabi 2014-15, 2015-16 and 2016-17 were used to test the hypothesis. The appropriate corrections was applied to overcome the impact of ties and continuity in the data set. The asymptotic p-value was computed using exact method.

Results
The main objective of this study was to identify new sources of resistance against the A. rabiei disease in Indian National Genebank chickpea collections. Therefore, wide range of cultivated chickpea genetic diversity was explored against the disease and it was found that chickpea accessions with robust resistance are very rare. A total of 1,970 chickpea accessions comprised of Indigenous accessions (1,567) and exotic accessions (403 representing 20 countries were screened against this disease ( Table 1). The disease severity was very high in both the locations and only very few chickpea accessions having resistance against the disease were identified (Table 2 and Fig 3). Based on the mean value of disease severity score (0-9) across the locations and seasons, only 0.5% of total accessions were scored as resistant, followed by moderately resistant (1.6%), susceptible (10.6%) and rest were highly susceptible (87.4%). When resistant and moderately resistant types of accessions were traced back to their origin country, it was observed that frequency of such accessions was comparatively higher in accessions originated from Southwest Asian countries particularly Iran and Syria than the accessions originated from Indian sub-continent. Total 338 accessions originated from Southwest Asia. Out of which five accessions (1.5%) were found resistant, 22 moderately resistant (6.5%), 45 susceptible (13.3%) and 266 highly susceptible (78.7%). Whereas, only four resistant (0.26%), nine moderately resistant (0.57%) and 161 susceptible (10.27%) were found out a comparatively large population (1,567 accessions) originated from India (Table 1). Screening across both the locations indicated that the chickpea germplasm is highly susceptible for the disease and there was no highly resistant/immune type of accession against the pathogen. However, some promising accessions in the category of resistant (score 1-3) and moderately resistant (score 3.1-5) were identified and also validated in multiple seasons ( Table 2). During winter 2014-15, at Ludhiana centre four accession i.e. EC220109, EC267154, EC267240 and EC528345B showed moderately resistant reaction with a score of 5.0. During the same season at Dhaulakuan location, 40 accessions were found with disease severity score less than 3.0 of which 12 accessions i.e. IC83811, EC267265, EC267272, EC267293, EC489910, EC554996, IC83389, IC83390, IC83453, IC83774, EC220109 and EC267240 were found resistant (severity score � 2). Screening during winter 2015-16 at Ludhiana centre resulted in identification of eight accessions i.e. IC114477, IC275447, IC209641, IC209317, IC209355, IC209375, IC244185 and IC269305 with their respective score of 2, 3, 4 and rest with 5 ( Table 2) (Table 1). A Kabuli grain type accession EC267301 was found resistant at Dhaulakuan in winter 2015-16 which was further  seven new accessions  viz. IC297322, IC275501, IC275448, IC244328, IC41651, IC267114, ICC4061 with moderate level of resistance were identified. There are few other promising accessions which are still being further validated in coming seasons. However, we could narrow down five promising accessions in the category of resistant (score 1-3) which were found constantly resistant across the locations and in multiple seasons. These accessions are IC275447, EC267301, IC117744, IC248147 and EC220109. (Table 2 and Fig 2). Promising accessions which were observed resistant or moderately resistant are highlighted in S2 Data.
It was observed that the disease severity over the locations was not consistent with respect to individual accessions. Graphical representation of germplasm based on their severity scale indicated highly skewed distribution. Most of the accessions belong to highly susceptible category, followed by susceptible type and it was also observed that the disease severity at Ludhiana location was comparatively higher than the Dhaulakuan location (Fig 3).
The experimental sites i.e. Ludhiana and Dhaulakuan are hotspot region to the pathogen, but the screening results shows that the severity of the disease was different. It has been observed that the severity of the disease was relatively higher at the Ludhiana compared to Dhaulakuan. Statistical analysis was done to understand whether the results obtained from both the locations are significantly different from each other or not in terms of their disease severity. The Wilcoxon Signed Rank test on paired set of accessions (1,230) indicated that the distribution for the disease severity was not the same ( Table 3). The null hypothesis (H 0 ) was rejected and alternate hypothesis (Ha) i.e. 'median rank of the two dependent samples is not the same' was accepted. Therefore, it was speculated that the variation in disease severity between two locations may be due to two reasons i.e. varied environmental conditions or pathogen racial differences. However, results obtained from the disease screening for paired set of accessions also indicate that the disease severity varies for accessions between locations, i.e. accessions showing resistance in one location may be susceptible in another locations and vice versa. Therefore, it was established that the variation in disease severity might be due to racial difference between two locations.

Discussion
The Ascochyta blight disease has become a major limiting factor for yield enhancement in chickpea in all chickpea growing areas. Ascochyta rabiei is the causal organism for the blight, which is known to be highly variable fungus [5,26]. The heterothallic nature of the fungus and sexual mating between two mating types i.e. MAT-1 and MAT-2 ensures sexual recombination and new allelic combination [21]. The rapid evolving genome of the fungus outpace the existing genic/allelic combinations responsible for defence mechanism in host plant. Therefore, the rapid breakdown of resistance in earlier identified donors as well as in varieties are reported [5,[19][20][21]. Understanding the host resistance mechanism against the disease is the perquisite for breeding resistant varieties. However, numerous studies done to understand the inheritance of the disease resistance indicated quite a few types of inheritance, which further complicate breeding resistance. There were some studies which showed that the host resistance is governed by single dominant gene [28,29], two dominant complementary genes [30]. A study based on two interspecific RIL populations shows that the three recessive and complementary major genes with several modifiers are responsible for the resistance [31]. They further mapped the major QTLs (QTL-1 and QTL-2) in linkage group 6 and 1 [30]. There are several other studies indicating that the loci governing the disease resistance are quantitative in nature [32][33][34][35][36]. A study showed that the inheritance is genotype specific [37]. Out of six resistant genotypes used in crossing, five showed independent single dominant gene action while one genotype showed single resistant gene action. Further a study on fifteen resistant genotypes in combination of screening against four races indicated that host resistance is genotype specific [11]. Therefore, it is imperative to find new sources of the disease resistance which may have novel defence mechanisms. The new sources will play the crucial role in developing genotypes with long-lasting resistance amid fast evolving pathogen. Therefore, this study was undertaken with the objective to identify new sources of resistance against the A. rabiei in Indian chickpea collections which is poorly explored. Both the experimental locations are designated hotspots for the disease where frequent severe disease incidences are observed under natural field conditions. However, artificial optimal conditions given for disease development were helpful in uniform spread of the disease in field and to avoid any escape. The screening and validation results indicate that the chickpea accessions with robust resistance are very rare in cultivated chickpea gene pool. However, we have identified five accessions viz. IC275447, IC117744, EC267301, IC248147 and EC220109 displaying constant resistance against the disease across the locations in multiple seasons ( Table 2 & Fig 2). From the literature survey we could also find out that the promising accessions identified in this study have not been reported earlier thus making them as novel and valuable genetic resources which can be used as donors for the chickpea improvement programs. However, the resistant germplasm lines may be further tested in other epidemic locations of the A. rabiei representing the other races in order to identify lines having broader resistance. It was observed that disease reaction of an accession varies from one screening location to another. Therefore, it appears that accessions such as IC275447, IC117744, EC267301, IC248147 and EC220109 which were found resistant in both the locations have broader resistance, which might be governed by multiple factors/genes. Similar differential pathogenicity results were reported in an extensive chickpea screening done against the disease from 1978 to 1982 in more than 11 countries [38]. This indicates the presence of variability in endemic pathogen races of the experimental locations. Comparatively higher frequency of R and MR type accessions originated from Southwest Asian region might be due to the variation in pathogen pressure between these regions. Chickpea cultivation in some areas are more affected by frequent epidemics of the disease due to more favourable weather conditions (cool, wet and cloudy) for pathogen spread during the growing season. The more stringent natural selection pressure of the pathogen over the years might have resulted in enhanced frequency of resistant alleles/QTLs in the chickpea germplasm from this region. The ascochyta blight is one of the major constraint of chickpea in WANA (Western Asia and North Africa) and Southern Europe [10,17,39]. In India the disease is a major problem of northern states particularly Punjab Uttarakhand and parts of Himachal Pradesh [40]. These are also the regions where the major chunk of the chickpea diversity evolved. Therefore, we speculate that the chickpea germplasm originated from this region may be better source of identifying novel sources of the pathogen resistance.
Though it has been a period of a century since the disease was first appeared and has spread across the continents, very limited success has been achieved in identifying chickpea genotypes with robust and broader resistance. Several small scale chickpea germplasm screenings have been reported [41][42][43] which resulted in identification of resistant or tolerant chickpea accessions. In a study three chickpea accessions i.e. PI 559361, PI 559363 and W6 22589 were identified as resistant from total 44 accessions screened [19]. But another report showed that the resistant sources are rare [44]. In a study 29 resistant lines were reported [45]. These were identified from 150 elite chickpea breeding lines which were screened using combination of methods i.e. cut-twig based lab screening method and commonly used field screening method. Screening of 19,343 global chickpea germplasm collection (12,749 desi and 6,594 Kabuli types) against six races of A. rabiei at Tel Hadya, Syria, between 1979 and 1991 could only result in finding three desi type (ICC4475, ICC6328, and ICC12004) and two Kabuli type (ILC200 and ILC6482) chickpea accessions resistant against all six races. This work was a joint venture of International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Patancheru, India and International Center for Agricultural Research in the Dry Areas (ICARDA), Syria. The identified resistant chickpea lines are the only main source used in current national and international breeding programs.
Recent past experiences of A. rabiei epidemics and large scale germplasm evaluation indicated that the disease resistance sources in chickpea genepool are rare and finding robust sources of resistance against all the prevalent isolates of the pathogen would be the best sustainable option. As a result of climate change increased frequency of non-seasonal erratic winter rains and medium to high speed winds are becoming more common and the trend may remain the same in near future, which are likely to create more congenial environment for the disease spread. Therefore, identifying promising resistant sources within cultivated chickpea primary genepool will be of strategic importance. Further, identification of genes/QTLs and pyramiding them in elite chickpea cultivars will be the most economical and sustainable way forward. The resistant germplasm identified in this study will have direct utilization to combat the problem to sustain chickpea production.