Dissection of race 1 anthracnose resistance in a watermelon (Citrullus lanatus var. lanatus) biparental mapping population

Anthracnose, caused by the fungal pathogen Colletotrichum orbiculare (Berk. & Mont.) Arx syn. lagenaria, is one of the most important diseases of watermelon in the United States and worldwide. The study was conducted to identify C. orbiculare race 1 resistance quantitative trait loci (QTL) in a ‘Charleston Gray’, resistant parent, and ‘New Hampshire Midget’, susceptible parent, biparental mapping population. The mapping population consisted of 228 F2 and the validation population consisted of 60 individuals each in BC1P1 and BC1P2. The disease severity was rated using a disease index comprising a rating scale of 0–100%. IciMapping was used to draw the linkage map and R/qtl non-parametric method (‘model = np’) was used to identity QTL. We identified a major disease resistance QTL, Qar1-8, on chromosome 8. The significant SNP marker S8_5149002, part of a putative coiled-coil (CC)–nucleotide-binding site (NBS)–leucine-rich repeat (LRR) (CC-NBS-LRR or CNL; ClCG08G002410), had a LOD of 14.06. The significant marker was validated on mapping populations using R package functions ‘chisq.test’, ‘wilcox.test’, ‘kruskal.test’, and ‘dunn.test’. The significant marker S8_5149002 was also tested for its ability to differentiate race 1 anthracnose resistance on 61 watermelon germplasm including 41 plant introduction (PI) lines. Hence, the diagnostic SNP marker S8_5149002 could be used for marker assisted selection (MAS) for race 1 anthracnose resistance in watermelon breeding programs.


Introduction
Watermelon (Citrullus lanatus var. lanatus) occupies 7% of global vegetable production acreage and is among the top five most consumed fresh fruits in the world (Yong and Guo 2017). In 2020, watermelon was grown on 100,000 acres and worth $574 million in the U.S. (USDA-NASS 2021). The major watermelon growing states in the U.S. are Florida, Texas, Georgia, and California. Anthracnose is one of the major diseases of watermelon and other cucurbits and is caused by the fungal pathogen Colletotrichum orbiculare (Berk. & Mont.) Arx syn. lagenaria. The fungus is a hemibiotroph ascomycete that occurs intracellularly in the plant hosts (Perfect et al. 1999;Dickman 2000;Xuei et al. 1988). Seven races of C. orbiculare have been described based on differential host reaction (Goode 1958;Dutta et al. 1960;Jenkins 1964). Wasilwa et al. (1993) grouped a total of 92 isolates of C. orbiculare into ten vegetative compatibility groups (VCGs), of which three (VCG 1, 2, and 3) were pathogenic in cucurbit differentials. Two distinct virulent phenotypes were observed and isolates in VCG 1/VCG 3 had disease reactions similar to previously described race 1, whereas isolates in VCG 2 gave disease reactions similar to race 2 (Wasilwa et al. 1993). The three races of the fungus (races 1, 2, and 3) belong to the pathogenic VCGs and have received particular attention in watermelon (Boyhan et al. 1994). A large number of watermelon germplasm are resistant to Colletotrichum orbiculare race 1 and 3, while others are susceptible (Wasilwa et al. 1993;Maynard and Hopkins 1999). The disease affects all above ground parts and symptoms include angular, brown to black leaf spots; tan, oval-shaped lesions in stems; sunken, and water-soaked spots on fruits (Elwakil et al. 2013;Dutta 1958;Layton 1937). Wet weather conditions such as rain and high humidity provide a favorable environment for dispersion and germination of conidia, and subsequent infection in plant (Maynard and Hopkins 1999).
Several accounts of anthracnose as a major disease in cucurbits can be traced back to the late nineteenth century and early twentieth century (Gardner 1918;Parris 1949). The most severe reports of this disease was mainly in south, southeast, northeast, and midwest regions of the U.S. (Wasilwa et al. 1993), with up to 30% yield loss reported in watermelon (Parris 1949) and 60% yield loss reported in other cucurbits (Thompson and Jenkins 1985). A significant negative impact on plants due to anthracnose is on fruit quality, as this disease influences grading standards of watermelon outlined by the United States Department of Agriculture (USDA-AMS 2021). Research on anthracnose disease management in watermelon has been prioritized in the past (King and Davis 2007), and is still considered a major research priority (Kousik et al. 2016).
Several efforts focused on breeding watermelon varieties for anthracnose resistance have been reported (Huh et al. 2010a, b;Crall et al. 1994;Norton et al. 1993;Crall 1990). Resistance to race 1 anthracnose in watermelons has been shown to be governed by a single dominant locus, Ar-1, and resistance was dominant to susceptibility (Layton 1937;Wehner 2012). Utilizing molecular markers closely associated with underlying genes can increase efficiency of the breeding programs (Xu and Crouch 2008). Single nucleotide polymorphism (SNP) markers are the latest of the molecular markers, succeeding restriction fragment length polymorphisms (RFLP) markers (Beckmann and Soller 1986), random amplified polymorphic DNA (RAPD) markers (Williams et al. 1990), simple sequence repeats (SSRs) or microsatellite markers (Litt and Luty 1989;Akkaya et al. 1992), and amplified fragment length polymorphisms (AFLP) markers (Vos et al. 1995). The popularity of SNP markers stems from the fact that they are commonly occurring DNA sequence variations, the basis of most genetic variation (Ganal and Röder 2007;Chagné et al. 2008), high density, cost effective and efficient compared to previous types of markers (Xu and Crouch 2008), and may affect protein function if present in the coding sequences (Yuan et al. 2006). The quantitative trait loci (QTL) discovery in watermelon facilitates marker-assisted selection (MAS) and breeding for several traits such as Fusarium wilt resistance (Lambel et al. 2014;Branham et al. 2018Branham et al. , 2020Meru and McGregor 2016), gummy stem blight resistance (Lee et al. 2021;Gimode et al. 2021), and fruit/flesh quality traits (Fall et al. 2019;Ren et al. 2018;Yang et al. 2021).
In the current study, we identified a major QTL for C. orbiculare race 1 resistance from 'Charleston Gray' in a F 2 population and validated on BC 1 populations. We further delineated a putative race 1 anthracnose resistant gene in the QTL region and used a previously reported SNP marker (Jang et al.

Page 3 of 12 157
Vol.: (0123456789) 2019) to differentiate race 1 anthracnose resistant and susceptible individuals from the mapping population, as well as the broader watermelon germplasm pool.

Developing biparental mapping populations
The watermelon mapping populations were developed at North Carolina State University. Two parental lines, 'Charleston Gray'(resistant, female parent, P 1 ) developed by C. F. Andrus in 1954(Andrus 1955, and 'New Hampshire Midget' (susceptible, male parent, P 2 ) were used to generate F 1 , F 2 , BC 1 P 1 , and BC 1 P 2 mapping populations. The mapping populations consisted of 228 F 2 individuals as well as 60 individuals each in BC 1 P 1 and BC 1 P 2 .

Inoculum preparation and pathogen inoculation
Colletotrichum orbiculare race 1, collected in North Carolina in 1998, was used to inoculate seedlings. The inoculum preparation and inoculation were conducted as described by Patel (2019). In brief, the fungus was grown on green bean agar (GBA) media for three-weeks. Spores were harvested by adding 10-15 mL distilled water to each agar plate, rubbing the surface of the agar with a sterile metal spreader, pouring the spore suspension into a sterile conical flask, and passing it through four layers of cheesecloth. Concentration of the inoculum was measured using a hemocytometer and adjusted to 100,000 spores mL −1 prior to inoculation. One drop of Tween-20 was added to every 500 mL of the spore inoculum. The three-week-old watermelon seedlings grown in the greenhouse were inoculated with the spore inoculum. After inoculation, seedlings were kept in a humidity chamber, in the greenhouse, for 48 h in darkness at 80-100% relative humidity, and at a temperature of 22-24 °C. Then, seedlings were moved to the natural light, and rated at 8, 11, and 14 days post inoculation (dpi).
DNA isolation, ddRADseq library construction, and genotyping by sequencing A total of 360 watermelon leaf samples (three P 1 , three P 2 , six F 1 , 60 BC 1 P 1 , 60 BC 1 P 2 , and 228 F 2 individuals) were collected from three-week-old seedlings. Samples were freeze-dried immediately, and genomic DNA was extracted from lyophilized samples using E.Z.N.A. Plant DNA Kit (Omega Biotek, GA, USA) following manufacturer's protocol. The DNA were quantified using Quant-iT-PicoGreen (Invitrogen, Thermo Fisher Scientific, USA) following the manufacturer's instructions. Due to some samples yielding low amounts of DNA, a total of 188 watermelon samples (three P 1 , two P 2 , six F 1 , 48 BC 1 P 1 and 129 F 2 individuals) were sent to Texas A&M AgriLife Genomics and Bioinformatics Service, College Station, TX (https:// www. txgen. tamu. edu/) for double digest restriction-site associated DNA sequencing (ddRADseq) as described previously (Yang et al. 2020) with the following changes. The restriction enzymes EcoRI and NlaIII were used for library prep and inserts from 400 to 600 bp were selected on a Pippin prep (Sage Science, Boston, MA, USA). The ddRADseq libraries were sequenced using 40% of a NovaSeq S4 X lane (2 × 150 bp paired-end run; Illumina, Inc., San Diego, CA, USA).
Raw sequences were demultiplexed using Illumina bcl2fastq, allowing for 1 base error in the barcode sequences. Sequences were first quality-filtered using the program FASTX-Toolkit (http:// hanno nlab. cshl. edu/ fastx-toolk it). Raw sequencing reads were first trimmed to remove low quality bases with quality score less than 20 on the ends of reads and reads with 30% or more bases showing low quality score (Q < 15) were removed. The reference genome for watermelon was downloaded from NCBI website (GCA_000238415.2). Bowtie2 [http:// bowtiebio. sourc eforge. net/ bowti e2/ index. shtml] was used 157 Page 4 of 12 Vol:. (1234567890) to align quality-filtered reads to the reference with the default parameters. Aligned reads were then processed with SAMtools v1.19 to generate coordinate sorted binary SAM files (BAM). Reads with mapping quality (MQ) less than 5 were removed. The local re-alignment tool in the Genome Analysis Toolkit (GATK, https:// softw are. broad insti tute. org/ gatk/) was used to perform re-alignment in Insertion/Deletion regions as previously described. Finally, the processed alignment files were fed to the tool Haplotype-Caller, which is part of the GATK, to call variations and perform genotyping for each sample. Once the SNP calling process was completed, individual SNPs with more than 20% missing data and Minor Allele Frequency (MAF) less than 0.05 in each population group were removed.

QTL mapping
Genotypic data were assigned to A (P 1 type, homozygous resistant), B (P 2 type, homozygous susceptible), H (heterozygous), and X (missing) types. Since the phenotypic disease rating data for F 2 population were found to be in a non-normal distribution, QTL analysis was done on 'qtl' package (Broman et al. 2003) with a non-parametric method ('model = np') on R software (R Core Team 2014; version 3.6.2) with RStudio GUI (RStudio-Team 2021). The logarithm of odds (LOD) threshold of 4.11 for QTL detection was estimated with 1000 permutations. As genotyping-bysequencing (GBS) genotypic data had higher missing values, QTL analysis was also performed after imputing missing genotypic data using a multiple imputation method ('method = imp') (Sen and Churchill 2001) on R 'qtl'. A graphical display of allele effects was done using "Effect Plot" function on R 'qtl'. A genetic linkage map was constructed using IciMapping V4.1 (Meng et al., 2015), whereas the linkage map was displayed using MapChart version 2.32 (Voorrips 2002).

Analysis of ddRADseq data
Generated DNA fragments (400-600 bp inserts) were selected on the Pippin Prep platform. After construction of ddRADseq libraries, they were sequenced using 40% of a NovaSeq S4 X lane (2 × 150 bp pairedend run), and an average of 4.79 million (M) reads/ sample or 1.44 giga base (Gb) per sample were generated. A 4X genome coverage (depth) was obtained on average. Approximately 50% of reads were chloroplast or mitochondria based on the basic local alignment search tool (BLAST). However, upon manually checking several reads to the reference genome (https:/www. ncbi. nlm. nih. gov/ assem bly/ GCA_ 00023 8415.2), samples aligned from 89.06 to 99.45% to the reference genome. This attests to a recent finding that there is exchange of genetic material between nuclear and organelle genome, and the mitochondrial and chloroplast genomes in watermelon share about 33% and 47% homology, respectively with the nuclear genome (Cui et al. 2021). The reference genome obtained from the National Center for Biotechnology Information (NCBI) website (GenBank assembly accession: GCA_000238415.2) corresponds to the watermelon cultivar '97103' v2 Genome in the Cucurbit Genetics Database (http:// cucur bitge nomics. org/ ftp/ genome/ water melon/ 97103/ v2/). At median 3 and mean 11 coverage depth, a total of 147,600 raw, unfiltered single nucleotide polymorphisms (SNPs) were obtained. After removing SNPs with depth > 20, a total of 134,136 SNPs were remaining. After filtering SNPs with minor allele frequency (MAF) < 0.05 and more than 20% missing data, a total of 653 SNP markers were left.

QTL mapping, genetic linkage map and resistant gene
After aligning the SNP regions between '97103' and 'Charleston Gray' genomes, the physical coordinates of markers were updated to represent 'Charleston Gray' and used in the linkage map construction for Chromosome 8 (Fig. 2). The genetic linkage maps were also drawn for remaining ten chromosomes using physical coordinates of '97103' watermelon genome ( Supplementary Fig. S2). The rank based non-parametric QTL analysis was done on R 'qtl' and a significant SNP marker S8_5149002 was observed in the major QTL region (LOD = 14.06) (Table 1 and Fig. 2). The effect plot for the marker showed that the disease index was low and similar for the homozygous resistant (Ar-1Ar-1) and heterozygous individuals (Ar-1ar-1) as compared to the homozygous susceptible (ar-1ar-1) (Fig. 3). Since GBS genotypic data resulted in higher missing value, QTL analysis was re-analyzed after multiple imputation in R/qtl. The LOD score for the significant marker increased from 14.06 to 44.42 after imputation. The QTL was validated on BC 1 P 1 and BC 1 P 2 populations, where only the latter population showed a significant QTL with S8_5149002 being the significant marker (LOD = 8.32; Supplementary Fig. S3). The physical coordinate of S8_5149002 marker did not align with the physical positions of adjacent markers on F 2 population probably due to inversion or crossover in this genome segment on the mapping population or due to the small population size of the mapping population. Such discrepancy in the order of marker locations was also observed earlier in the same region of Chromosome 8 in watermelon (Shang et al. 2016;Jang et al. 2019). The results from this study showed that a significant QTL, Qar1-8, from 'Charleston Gray' contributed to race 1 anthracnose resistance.

Phenotypic and genotypic ratios
Colletotrichum orbiculare race 1 affects watermelon and cucumber in which a single dominant resistance gene was reported (Layton 1936;Hall et al. 1960; Barnes and Epps 1952). In this study, the Chisquare analysis showed a goodness of fit for a single dominant gene controlling race 1 anthracnose resistance both phenotypically and genotypically. Similar Mendelian phenotypic segregation ratios were reported and suggested that a single dominant gene was involved for race 1 anthracnose resistance in watermelon-'Africa 8' (Layton 1937), and 'Charleston Gray', 'Congo' and 'Fairfax'  ( Table 3). Resistance to anthracnose in beans was also found to be dominant in crosses of resistant × tolerant and resistant × susceptible varieties (Andrus and Wade 1942). A single dominant gene for anthracnose resistance was also reported in cucumber (Barnes and Epps 1952). Robinson et al. (1976) assigned Ar gene symbol for the anthracnose resistance gene in watermelon and cucumber. Winstead et al. (1959) also reported that race 1 anthracnose resistance gene also conferred race 3 anthracnose resistance in watermelon by superimposing race 3 inoculum on race 1 inoculated plants and vice-versa. The pedigree of 'Charleston Gray' had 'Africa 8', whereas the pedigree of 'Congo' and 'Fairfax' had 'African' (Table 3). It is most likely that Ar-1 in 'Charleston Gray' might had been inherited from 'Africa 8', a race 1 anthracnose resistance founder parent.

Resistance QTL and putative genes
In the study, the preliminary analysis using GBS markers identified race 1 anthracnose resistance QTL on chromosome 8 (in between coordinates 4,847,957 and 6,294,791). The QTL on chromosome 8 was on the similar region to the previous study (Jang et al. 2019). One of the genes, CC-NBS-LRR (CNL; Cla001017 or ClCG08G002410), in the QTL region was reported for race 1 anthracnose resistance on breeding line 'DrHS7250' (Jang et al. 2019). We converted the high-resolution melting (HRM) SNP marker, CL14-27-9, for ClCG08G002410 onto the PACE marker and designated it as S8_5149002. We reanalyzed F 2 mapping population data by including genotypic data for marker S8_5149002. The result showed that S8_5149002 was the significant marker R Levi et al. 2001b) New Hampshire Midget (Favorite Honey × Dakota Sweet) S (Yeager 1950;Rhodes et al. 1992) 157 Page 10 of 12 Vol:. (1234567890) with LOD 14.6 and up to 44 (with imputation) and could indicate that ClCG08G002410 could also be the race 1 anthracnose resistance gene in 'Charleston Gray'.
Variable genotype of watermelon germplasm The significant SNP marker S8_5149002 clearly differentiated the disease response of the individuals of the mapping population as well as the germplasm present in the watermelon breeding program. Genotype of several watermelon germplasm and hybrids showed homozygous resistance to race 1 anthracnose (Supplementary Table S4). These include 'Crimson Sweet', 'TASTIGOLD', 'AU-Sweet Scarlet', 'AU-Golden Producer', 'Perola', 'Crimson Diamond', 'Graybelle', 'Verona', 'SUNSHADE', 'Sugarlee', 'Dixielee', 'Jubilee', 'Big Stripe', 'Pronto', 'Pathfinder F 1 ', and 'Fascination'. For most of them the source of race 1 anthracnose resistance might had inherited from the founder parent-'Africa 8'. It is intriguing that the Ar-1 gene is exhibiting resistance for more than 50 years. Interestingly, the germplasm PI 189225 which is resistant to race 2 anthracnose (Levi et al. 2001a), showed susceptible genotype (ar-1ar-1) for race 1 anthracnose suggesting race-specific resistance provided by the R-genes.

Conclusion
The study delineates a major QTL region on chromosome 8 governing race 1 anthracnose resistance and putative CC-NBS-LRR (CNL; ClCG08G002410) could be a potential resistance gene in Charleston Gray. Further study is needed to validate that the CNL is the Ar-1 gene. The S8_5149002 is a diagnostic marker for race 1 anthracnose resistance and could be used in MAS in watermelon breeding programs.