Diversity in genetic and downy mildew resistance among wild and mutagenized hops as revealed by single nucleotide polymorphisms and disease rating

Abstract Studies on wild and landrace hops from the Canadian Maritimes are scarce. This study was undertaken to broaden the genetic base of hops and to assess the reaction of the generated variants to downy mildew (DM) disease. A landrace hop (PE Royalty (P-RL)) and a commercial cultivar (Alpharoma) were mutagenized using ethylmethane sulphonate (EMS), and single nucleotide polymorphism (SNP) variations were determined using an amplicon sequencing genetic diversity study. A subset of wild types and a subset of mutagenized hops were inoculated with DM spores and rated for disease symptoms in a controlled environment. The data showed large EMS-induced genetic diversity in the target genes along with natural variations in the wild types. A diversity in DM resistance within the studied collection was also observed. The study showed DM tolerance in some P-RL landrace seedlings, suggesting that these P-RL landraces must have acquired and developed adaptation mechanisms to co-evolve with DM disease in the environment. Further, EMS-induced mutagenesis increased allelic variations that contributed to increased DM resistance in some seedlings. The data recommend the use of true hop seeds for increased genetic variability in breeding programs.


Introduction
Economic value of hops (Humulus lupulus L.) is associated with its cones used as raw ingredients in brewing (DeNoma 2000;Feiner et al. 2021). During their settlement in North America, the first Europeans carried and grew European hops for their own consumption (Tomlan 1992;Morton 2013), and these exotic hops have since then evolved alongside native and endemic north American wild hops (Small 1978). The 19th century has seen the growth of a regional hops' production in the Canadian Maritimes, supplying the local and northeastern brewing industry (McCallum et al. 2019). The disease pressure in these humid climates, mainly by downy and powdery mildew diseases, decimated local hop production in eastern north America in the early 20th century (Morton 2013; Nova Scotia Hops Growers' Guide 2013), shifting the hop production to the pacific northwest of America (Hoerner 1933;Skotland and Johnson 1983). According to available literature, most of the current commercial hop cultivars in North America can be traced back to common ancestries involving Brewers Gold and Northern Brewers (Hampton et al. 2001;Driskill et al. 2022) and are mostly susceptible to common diseases of hops (Salmon and Ware 1931), highlighting their narrow genetic diversity (Patzak et al. 2010). In the past decades, the genetic diversity of many crops have been broadened through artificially-induced mutagenesis, as performed in flax (Chantreau et al. 2013;Fofana et al. 2017), potato (Muth et al. 2008;Somalraju et al. 2018), tomato (Shikata et al. 2015), and eggplant (Xi-ou et al. 2017). No such development has so far been reported in hops, and it could be important for hop breeding programs.
Hops are particularly susceptible to downy mildew (DM) (Pseudoperonospora humuli) and powdery mildew (Gent et al. 2010;Wolfenbarger et al. 2014). In hops, inflorescence and developing cone infections by DM lead to the discoloration of the cones, reduced crop quality, and low levels of bittering acids (Royle and Kremheller 1981). It can also cause crown rot, weak plant growth, and death (Skotland 1961;Skotland and Johnson 1983), and its management is mainly ensured by fungicide application and spring crown pruning (Gent et al. 2010). Currently, only limited hop varietal DM resistance has been reported Rahman et al. 2019), and no resistance genes have so far been described and characterized, although potential candidate genes have been reported (Feiner et al. 2021). Recently, differential production of specialized metabolites, including flavonoids, phenylpropanoids, and terpenoids, was reported in DM-infected and non-infected hop plants (Feiner et al. 2021), suggesting that genes involved in their production may be contributory to DM defense. Among these genes, those involved directly or indirectly in the production of bitter acids have been reported (Li et al. 2015) and include the branched-chain aminotransferases 1 and 2 (BCAT1 and BACT2) (Clark et al. 2013), valerophenone synthase (VPS), humulone prenyltransferases 1 and 2 (HIPT1L and HIPT2), and humulone synthases 1 and 2 (HS1 and HS2) (Fig. 1). We anticipate that new induced allelic variants in some of these genes will lead to broader genetic diversity and DM-resistant phenotypes in hops. Whereas hop diversity using morpho-physiological characteristics (Small 1978;Hampton et al. 2001), molecular phylogeny (Murakami et al. 2006a(Murakami et al. , 2006bBassil et al. 2008;Howard et al. 2011;Driskill et al. 2022), and chemotaxonomy (Hampton et al. 2002;McCallum et al. 2019;Riccioni et al. 2021) has been investigated at regional and global levels, to our current knowledge, no recent study addressed DM resistance in wild hops from the Canadian Maritimes, nor has an induced mutagenesis approach for increased genetic diversity been attempted.
The objectives of the current study were to (i) generate EMS-mutagenized hops for increased genetic diversity and (ii) evaluate a subset of the mutagenized hop clones for DM resistance. We showed induced mutations in hops, which was reflected by variations in DM resistance. The data suggest that the seedlings from the PE Royalty (P-RL) accession may be promising in hops' breeding for DM resistance.

Plant materials
The raw starting plant material consisted of four hop germplasms, including one commercial hop cultivar (Alpharoma) and three feral hops. These feral hops were collected as seed lots from cones developed on three different bines harvested from a feral population growing in a protected municipal park in Charlottetown (PE, Canada), referred to as PE Royalty (P-RL). It can be assumed that these seed lots originated from the same clone, but they can also be siblings. These seed lots were named as P-RL-4, P-RL-6, and P-RL-7. The presence of a mixed population of feral male and female plants in a close proximity (∼20 m) at the P-RL site ensured high degree of pollination and seed formation. The commercial Alpharoma clone was purchased as in vitro plantlets from Clean Plant Center, Northwest, Washington State University, WA, USA.

Generation of EMS mutagenized hops
The EMS-induced mutagenized seedlings and clones were obtained through two separate EMS mutagenesis experiments using two hop genetic materials consisting of one Alpharoma and the P-RL genetic stocks.

EMS mutagenesis of Alpharoma
Using Alpharoma as starting material, internodes were collected from in vitro plants and treated with different concentrations (0.25%, 0.5%, or 1% (v/v)) of EMS diluted in sterile water for 1 or 3 h. The treated internodes were rinsed five times with water as described by Somalraju et al. (2018). The explants were incubated in callus induction media (Batista et al. 2000) and plants were regenerated (Roy et al. 2001). Upon the initial plant regeneration, internodes were collected successively from the new plants and re-regenerated into secondary plants. This process was repeated 4-5 times to reduce chimeric-induced somatic variegates.
EMS mutagenesis of P-RL seed Using the P-RL true seed lots as starting material, the seeds were first pre-treated to break dormancy. To achieve this, seeds were incubated in sulfuric acid for 2 min and thoroughly rinsed with sterile water. Seeds were then soaked in saturated baking soda for 2 min and rinsed 3 times with water to remove any trace of sulfuric acid and baking soda. Seeds were air-dried at room temperate and kept at 4 • C for 4 weeks. A pre-germination test was performed on each seed lot by plating 50 seeds on wet filter paper inside a Petri dish as well as in a tray of wet sandy soil. The plated seeds were incubated at room temperature, and seed germination was recorded for two months. After the pre-germination test, 1052 and 1500 seeds from P-RL-6 were treated with 0.5% and 1% EMS for 6 h, respectively, whereas 841 and 1500 seeds from P-RL-7 were treated with the same respective EMS concentrations for 5 h, following the procedures described for true potato seed by Somalraju et al. (2018). A total of 555 seeds from P-RL-4 were treated with water and used as the non-treated control. The seeds were washed thoroughly five times to remove any trace of EMS and dried under fume hood at room temperature. All treated seeds of each lot were planted in sterile sandy soil in germination trays placed in a growth chamber set at 15 • C. Germination was monitored daily for 2 months, from the planting day of 9 May 2016 to 18 July 2016. After two months, the germinated seedlings were transferred into individual pots, assigned an ID number (ID #), and grown in a greenhouse.

DNA extraction
Leaf discs were collected from the 40 selected clones listed in Table 1, arranged into a 96-well plate, and freeze-dried. Leaf samples were ground using a Geno grinder 2000 (Spex Sample Prep, Spex CertiPrep, Metuchen, NJ, USA) and the genomic DNA extracted using a DNeasy 96 Plant Kit (Qiagen, Mississauga, ON, Canada). DNA was quantified using a Quant-iT PicoGreen dsDNA Assay Kit (Thermo Fisher Scientific, CA, USA) as per the manufacturer's specifications.

Targeted custom AmpliSeq gene panel design
A custom gene panel consisting of seven targeted bitter acid biosynthesis genes was designed using Ion AmpliSeq  and BACT2,VPS,HIPT1L and HIPT2,and HS1 and HS2 are of interest to the study. BCAT1 and BACT2, branched-chain aminotransferas es 1 and 2; VPS, valerophenone synthase; HIPT1L and HIPT2, humulone prenyltransferases 1 and 2; and HS1 and HS2, humulone synthases 1 and 2.
Designer software v.5.2 (Thermo Fisher Scientific, CA, USA) as previously reported (Somalraju et al. 2020). The custom gene panel included HS1 and HS2, HIPT1L and HIPT2, VPS, and BACT1 and BACT2, as described in Table 2, and was used as target for AmpliSeq sequencing. The custom AmpliSeq panel was designed to cover 45 642 of the 49 720 bp pseudomolecule DNA target regions, achieving 92% coverage. The custom panel included 193 primer pairs divided into 3 pools of 67, 65, and 61 primer pairs, respectively, and targeting 193 amplicons (Table S1). The 49 720 bp target sequence was uploaded as a BED file and used as a reference for calling single nucleotide polymorphism (SNP) variants from the germplasm panel.

AmpliSeq library preparation, sequencing, and detection of single nucleotide polymorphisms
The AmpliSeq library preparation and sequencing were performed using individual DNA samples as previously described (Fofana et al. 2017), and the sequencing was performed on an Ion Torrent Personal Genome Machine (PGM) equipped with an Ion Chef (Thermo Fisher Scientific) using an Ion PGM HiQ Sequencing kit and 850 flows (Thermo Fisher Scientific) as described in Fofana et al. (2017). The reads were mapped against the assembled 49 720 bp pseudo molecule DNA sequence from the hop draft genome uploaded as a BED file, and SNP variants were called in each germplasm accession using the Ion Suite Software v 5.04 and the Ion Server and with parameters previously described (Fofana et al. (2017). To detect EMS-induced mutations in each genetic stock (Alpharoma and P-RL), the pseudo molecule sequence obtained from the cultivar Alpharoma was used as a reference to filter out all natural variations in the EMS-treated clones derived from Alpharoma. Similarly, for all the other P-RL clones (EMS-treated and wild), the consensus of the pseudo molecule sequence from the non-EMS-treated P-RL was used as a reference to filter out all natural variations. All SNP variants shared with any of the reference sequences were not considered, and only unique SNP variants in hop accessions were kept. Default variant caller plugin setup and default parameters were used as previously described (Fofana et al. 2017;Somalraju et al. 2020). The position of the SNP variants in the

Phylogenetic tree construction
To determine the genetic structure in the studied germplasm panel, sequence alignments of gene-specific sequence (including intron and exons) obtained from individual accessions and the HIPT1L reference sequence were performed using MEGA X (Kumar et al. 2018). Because of its central role in the bitter acids' biosynthetic pathways and reduced gap (Table S1), HIPT1L gene sequence was used to construct a phylogenetic tree using the Neighbor-joining method (Saitou and Nei 1987) with 500 bootstrap replicates (Felsenstein 1985). The optimal tree with the sum of branch length = 37.89 is shown. The evolutionary distances were computed using the Maximum Composite Likelihood method (Tamura et al. 2004) and are in the units of the number of base substitutions per site. The analysis involved 84 nucleotide sequences. Evolutionary analyses were conducted in MEGA X (Kumar et al. 2018).

Downy mildew disease evaluation
The plant material used consisted of 40 hop clones, including 10 non-EMS-treated and 30 EMS-treated clones, with 12 clones derived from Alpharoma and 28 from P-RL seed stocks as listed in Table 1. The hop clones were pot-grown to full vegetative stage in a greenhouse at 20 • C in August 2017. To generate back up clones, explants were collected from each clone and inserted into test tubes containing water and sucrose. The test tubes were placed in a growth chamber and maintained under an 8 h dark/16 h light regime at 20 • C and 70% RH to induce rooting. As a destructive experiment, a single plant per clone was used in each experiment, and two independent experiments were performed. The original potted clones served as plant material for the first disease evaluation experiment in a greenhouse, whereas the rooted test tube clones were used for a second disease evaluation experiment in a growth chamber.

Plant inoculation and hop downy mildew disease rating
The hop downy mildew (P. humuli) inoculum was obtained from field-grown hop cultivars at AAFC's Harrington Farm (Charlottetown, PE). DM symptoms were ascertained by an experienced plant pathologist prior to collecting symptomatic living leaves displaying active sporulation. Leaves were taken to the laboratory, and spores were collected by brushing and rinsing the leaf surface using a sterile brush and water. The solution was collected in a sterile container. The sporangia were counted using a hemocytometer. The 40 hop clones grown in pots were moved to a pathology greenhouse compartment for inoculation and disease evaluation. Each plant was sprayed with 16 mL of spore suspension (39 000 sporangia/mL) and bagged to maintain high humidity. Disease symptoms were monitored daily. Upon the appearance of the first symptoms after 12 days post inoculation, the disease progression on the whole plant was rated weekly, for the number of infected leaves, surface area covered, and disease severity determined as % of infection at the end-point after 4 weeks. The experiment was repeated using the same clones in test tubes. In this case, each clone was sprayed with 2 mL of spore suspension (10 000 sporangia/mL) at the 3-5 leaf stage and bagged individually. Disease development was monitored as in the first experiment.

Downy mildew pathogen load determination in inoculated hops
During DM disease assessment in the first experiment, leaf samples from 27 individual plants were collected at 4 time points and stored at −80 • C. Plant genomic DNA was extracted from the leaves using a Mag-Bind Plant DNA Plus Kit (Omega Bio-tek, Inc.). As a positive control, a 20 mL DM sporangia suspension solution was pelleted, and the DNA was extracted from the sporangia using the MoBio Power Soil DNA Isolation Kit (MO BIO Laboratories, Inc.). The DM's DNA was used as a positive control in the experiment. DNA samples were quantified using a NanoDrop One Microvolume UV-Vis Spectrophotometers (ThermoFisher Scientific) and were used to confirm the presence of DM and to quantify the pathogen load in the leaf samples at time points 1 and 4 using quantitative PCR (qPCR) and digital droplet PCR (ddPCR). qPCR reactions were carried out on a BioRad CFX96 thermal cycler using 2X PerfeCTa SYBR Green FastMix. The reaction mixture consisted of 20 μL, including 9 ng genomic DNA, and 0.4 μmol/L of each forward (ITS1F, CTTGGTCATTTAGAGGAAGTAA) and reverse (ITS2, GCTGCGTTCTTCATCGATGC) primers. Gene-specific primers were designed from Humulus lupulus GADPH (HL-GAPDH-F: AC-CGGAGCCGACTTTGTTGTTGAA, HL-GAPDH-R: TCGTACTCTG-GCTTGTATTCCTTC) and used as a housekeeping gene and for normalization for gene quantification. Thermal cycling was carried out as follows: initial denaturation at 95 • C for 3 min followed by 45 cycles of 95 • C for 15 s and 55 • C for 30 s. Melt curves were generated from 65 to 95 • C to ensure specificity of the amplification, and the data were analyzed using CFX Manager Software (version 2.0; Bio-Rad) according to the 2 − Ct method (Livak and Schmittgen 2001). The output data were expressed as log 10 fold change values in each sample. The pathogen load in the leaf tissue was further assessed by ddPCR. The optimized 20 μL ddPCR reaction mix contained 1× QX200™ ddPCR™ EvaGreen Supermix (Bio-Rad), 0.1 nmol/L of forward and reverse primers, and 0.13 ng of DNA. The 20 μL of ddPCR reaction product and 65 μL of Droplet Generation Oil for Evagreen were used to generate droplets following the manufacturer's instructions (Bio-Rad). The droplet emulsion was used in PCR reactions performed on a C1000 Touch Thermal Cycler (Bio-Rad). PCR thermal cycling conditions consisted of the activation of DNA polymerase at 95 • C for 10 min, followed by 40 cycles of denaturation at 94 • C for 30 s, and primer annealing and elongation at 59 • C for 60 s. Droplets were stabilized by a final step of 98 • C for 10 min. Each thermal cycle was performed using a temperature ramp of 2 • C/s, and the temperature of the thermal cycler was kept at 105 • C. After amplification, the PCR plate was directly transferred to a QX200™ Droplet Digi-tal™ System (Bio-Rad) droplet reader, and data acquisition and analysis were performed using QuantaSoft 1.7.4.0917 software (Bio-Rad). Data were reported as copy number per ng DNA.

Statistical analysis
Statistical analysis was conducted using GenStat (Release 21 for Windows) for principle component analysis (PCA) using Euclidian distances to infer the relationships between SNP variations observed in individual genes and DM disease reaction. The inference was based on individual DM reads as well as on the mean data.

Effect of EMS on seed germination
From a total of 555 non-EMS-treated P-RL-4 seeds, only 5% germinated over 2 months of germination time, highlighting some viability and (or) dormancy issues with many of the seeds plated for germination and confirming the trend (4/50 = 8%) found in the pilot germination test. The low germination rate of hop seeds is not necessarily likely related to only EMS treatment, as hop seeds are notoriously difficult to germinate (E. Small, personal communication). After seed treatment with EMS, a dose-dependent effect was observed, with 1% EMS causing lower germination rates compared to 0.5% EMS for the same exposure time, although slight variations were observed between seed lots (Table 3).

EMS-induced variations in seven target genes
Following EMS treatment and plant regeneration, we sought to determine the EMS-induced genetic variations through SNP variant analysis. The 7 target genes were successfully amplified and sequenced from 27 germplasm accessions, while 13 accessions had truncated amplification in at least 1 target gene or showed poor sequencing reads that prevented accurate and reliable SNP calls in all genes. These 13 samples were excluded from further SNP variant analysis. After variant calls against the pseudomolecule references, a total of 310 SNP variants were detected in the 12 686 bp exonic regions of the 7 target genes (Table 2), for a frequency of 1 SNP variant per kb (12 686 × 27/310 = 1105 bp) in the studied germplasm. The SNP variants observed in the Alpharomaderived clones treated with EMS varied from 2 to 63 across the 7 genes, with BCAT1 and VPS showing the lowest and highest induced mutations, respectively. The EMS-induced mutations in these clones ranged from 11 to 28 SNPs, for an average of 23 mutations per clone in the 7 genes. The clones Alpha-1%EMS-11 and Aplha-0.25%EMS-1 showed the lowest and highest induced-mutation rates, respectively (Table 4). Using P-RL-4 as a reference, a total of 128 SNP variants were observed in the exonic regions of 19 mutagenized seedlings, with HS1 showing the most variations (Table 5). The number of variants per gene ranged from 0 to 69, with HIPT1L and VPS showing the lowest and higher variations, respectively. EMSinduced variations in P-RL-derived seedlings ranged from 0 to 22, for an average of 7 mutations per clone in the 7 genes (Table 5). No EMS-induced mutations were observed in HIPT1L, and the induced variations in BCAT2 and HS1 were 1 and 69 SNPs, respectively (Table 5). The phylogenetic tree depicting the genetic diversity among the 32 Alpharoma-and P-RLderived clones built using HIPT1L sequence is shown in Fig. 2. P-RL wild types and mutagenized Alpharoma and P-RL clones clearly separated, with the sequences from Alpharoma and the HIPT1L reference gene from hops' cultivar Nugget found to be more closely related. The same trend was observed with HS1 (not shown), and the diversity among the P-RL-derived mutagenized clones is depicted.

Diversity in downy mildew resistance
To assess whether the EMS-induced variations were also reflected in diversity to DM disease reaction, a subset of hops' clones was challenged with DM. A total of 40 clones, including 19 mutagenized clones derived from the P-RL seed stock, 11 mutagenized clones derived from Alpharoma, 9 P-RL wild feral clones, and 1 Alpharoma clone, were inoculated and rated under high DM pressure. After infection and disease rating, 10 clones showed consistently lower DM symptoms over time ( Fig. 3; Fig. S1). Three and five of these clones were Alpharoma-and P-RL-derived mutagenized clones, respectively, and showed ∼2-4X fewer symptoms compared to their controls at each rating time point (Fig. 3). Two P-RL wild feral clones (P-RL-4-CTL13 and P-RL-7-CTL2) showed less symptoms compared with the other P-RL wild feral counterparts albeit having moderately higher infection levels than the five EMS-treated P-RL resistant seedlings. Nine of the 40 clones showed similar to higher DM disease ratings compared with the wild Alpharoma (Alpha-CTL), whereas 20 other clones were moderately tolerant when compared to the wild Alpharoma (Fig. 3). The mutagenized clone P-RL-7-1%EMS-7 was the most susceptible, whereas P-RL-7-0.5%EMS-4 was the most resistant. The same disease infection trends were observed in the second experiment using the rooted clones in test tubes.

Correlations between SNP variations and DM disease severity
To assess whether induced mutations in specific genes are more related to DM disease incidence or severity, PCA and correlation analyses were performed. From the 40 EMStreated hop clones evaluated for DM reaction, only 10 had undergone through both the Amplicon sequencing for SNP variant detection and DM disease assessment (Table 6). Correlation studies showed weak and negative but non-significant correlations between SNP variations observed in the individual gene and DM reaction. The correlations between each of the BCAT2, HIPT1L, HS1, HS2, and VPS genes and DM reactions were −0.033, −0.181, −0.175, −0.205, and −0.076, respectively, showing a trend that DM infection decreases as the SNP variants increase (Fig. S2). In contrast, weak and positive correlations were observed among BCAT1 (R 2 = +0.061), HIPT2 (R 2 = +0.20), and DM, highlighting a tendency for increased DM disease with more mutations in these genes. The PCA shows that HS1 and HS2 have a strong correlation with each other and grouped together (Group 1) and were less correlated to DM (Fig. 4). In contrast, increased DM incidence is strongly associated with BCAT1 (Group 2). BCAT2, VPS, and HIPT2 were clustered as one group (Group 3) and were less correlated to DM incidence. HIPT1L appeared as a singlet between Groups 2 and 3. Nonetheless, the correlation relationship among the three groups was not significant. P-RL-6-1%EMS-7 (31% infection) and the mean DM rating were closely associated with BCAT1 in Group 2, while the two clones P-RL-6-1%EMS-5 and P-RL-6-1%EMS-6 with 20.25% and 23% infection, respectively, were closely associated in Group 1. Four Alpharoma clones (Alpha-0.25%EMS-1, Alpha-0.5%EMS-8, Alpha-1%EMS-9, and Alpha-1%EMS-10) with infection rates ranging from 8.25% to 33.75% (22% average) were scattered between Groups 2 and 3, with Alpha-0.5%EMS-8 (8.75% infection) and Alpha-1%EMS-9 (33.75% infection) more associated in Group 3. Alpha-0.5%EMS-7 (37.5% infection) was closely associated to HIPT1L.  In planta downy mildew quantification To confirm the presence of powdery mildew in infected leaves, qPCR analysis was performed in the 27 selected hops' clones based on DM disease profile. DM DNA amplification was observed in all the 27 hop clones, with a large variation among clones (Fig. 5). The level of DM DNA was found to increase with time as infection progressed. To further confirm the qPCR data, ddPCR analysis indicated differential pathogen load in the samples (Fig. 6). Over time, the DM DNA count detected by ddPCR was found to be unchanged in 3 clones, reduced in 9 clones, and increased in 15 clones of the 27 samples. Thus, the ddPCR data for 15 clones were in agreement with those from the qPCR. The slight discrepancies between the two methods has been reported in previous studies (Arvia et al. 2017) and have been attributed to an inaccurate quatification of the starting DNA template by Nanodrop and (or) to the saturation effect by ddPCR when the template per reaction is ≥10 5 copies/reaction (Pinheiro et al. 2012;Verhaegen et al. 2016;Arvia et al. 2017). The low load of PM DNA observed in some samples over time despite the high DM infection seems in agreement with the saturation effect by the high copy number/reaction as starting template in the ddPCR reaction. Of the 10 clones that expressed low DM infection rate (Fig. 3), 7 clones showed low pathogen load (Fig. 6), highlighting the consistency of DM disease evaluation.

Discussion
Sustainable worldwide hop production relies on the availability of germplasms that are resistant to major diseases of hops, which include downy and powdery mildew (Henning et al. 2015). By treating two hop germplasms with EMS and evaluating a set of clones for DM reaction, we show large genetic variation in terms of natural and EMS-induced diversity. This diversity was reflected and substantiated by a diverse DM disease response following artificial inoculation. To our knowledge, this study is the first to report EMS mutagenesis in hops in generating germplasm with enhanced DM resistance. Fig. 2. Phylogenetic tree constructed using HIPT1L gene-specific sequence depicting the diversity among wild and mutagenized hops. The evolutionary history was inferred using the Neighbor-joining method (Saitou and Nei 1987). The bootstrap consensus tree inferred from 500 replicates (Felsentein 1985) was taken to represent the evolutionary history of the taxa analyzed (Felsentein 1985). Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed. The evolutionary distances were computed using the Maximum Composite Likelihood method (Tamura et al. 2004) and are in units of the number of base substitutions per site. This analysis involved 32 nucleotide sequences. All ambiguous positions were removed for each sequence pair (pairwise deletion option). There were a total of 3080 positions in the final data set. Evolutionary analyses were conducted in MEGA X (Kumar et al. 2018). The reference sequence for HIPT1L (accession KM222441.1) from hop cultivar Nugget in the NCBI gene bank (Li et al. 2015) is marked by the red asterisk.
Chemometric studies in hops reported contrasting proportions of co-humulone between H. lupulus subspecies lupulus accessions and those of H. lupulus subspecies lupuloides, with lupulus producing a low co-humulone ratio compared to lupuloides (McCallum et al. 2019). Using SNP polymorphism data derived from seven genes involved in bittering acid production (Paniego et al. 1999;Clark et al. 2013;Li et al. 2015), the current study revealed a large variation (natural and EMS-induced) in hop seedlings, true hop seeds, and mutagenized explant clones derived from Alpharoma. Similar EMS-induced SNP variations have been reported in flax, potato, tomato, and eggplants (Muth et al. 2008;Fig. 3. Downy mildew symptom progression in 40 non-and EMS-mutagenized hops seedlings after 4 weeks of rating. R1-R4 indicate the four DM rating time points. The red bar indicates a 12% threshold DM infection set as a comparison baseline from the Alpha-CTL following the first rating (R1); the blue stars indicate the clones showing low DM infection and are referred to as resistant clones; the green stars indicate the clones showing moderate DM infection when compared to the wild Alpharoma (Alpha-CTL); the red stars indicate the clones showing similar to higher DM infection when compared to the wild Alpharoma (Alpha-CTL).   Chantreau et al. 2013;Shikata et al. 2015;Fofana et al. 2017;Xiou et al. 2017;Somalraju et al. 2018). By challenging wild and EMS-treated clones with DM, a diversity in disease response was observed in both types, suggesting that wild feral types had persisted and co-evolved in the environment with epiphytotic DM epidemics and might have developed adaptation mechanisms through natural mutations during the nearly 200 years of co-evolution with different local DM strains. This assumption was further corroborated by EMSinduced mutations that showed a correlation, albeit weakly, with DM reaction. In this study, a very small set of hop clones was used in a correlation analysis, and these correlations should be taken with caution. Further study using a large mutant population could lead to stronger and more valid conclusions in this regards. Nonetheless, the current data show that EMS-induced mutations increased hop segregant clones   for DM resistance in the two genetic stocks studied and further support the need for enlarged genetic diversity of true hop seeds for phenotypic characterization in hops. Prenylation of secondary metabolites in the terpenophenolic biosynthesis pathway is critical for bitter acid biosynthesis in hops (Li et al. 2015). SNP variations in the prenylation gene HIPT1L, along with other genes HS1, HS2, and VPS, were found to be negatively correlated, albeit weakly, with DM disease development, suggesting an induction of favorable allelic variants for DM resistance. HIPT1L showed a single EMSinduced mutation in the Alpharoma clone, Alpha-0.5%EMS-7, a clone with 37.5% infection. This observation suggests that not all induced mutations are favorable alleles for disease resistance. Recently, Feiner et al. (2021) reported a correlation between DM resistance and phenylpropanoid metabolites, and a quantitative inheritance of DM resistance has also been reported (Henning et al. 2015). The current study appears to be in agreement with these findings in that mutations in multiple genes may be required for quantitative resistance traits as reported for powdery mildew (Asgarinia et al. 2013;Marone et al. 2013). To date, however, no specific gene associated with DM resistance is known to our knowledge. A detailed study using genome-wide association studies of a germplasm panel phenotyped for DM along with phenylpropanoid metabolite profiling would contribute in the identification of QTNs and genes associated with DM resistance. This study highlights the efficiency of EMS-induced mutations in hops' seedlings for increased genetic variability and efficient selection of breeding lines. In conclusion, the current study showed that some ancient European H. lupulus var. lupulus hop ferals collected from the Canadian Maritimes have co-evolved, acquired new allelic variants to adapt to the environment, and have been expressing some resistance to epiphytotic DM strains. EMS mutagenesis led to increased genetic diversity in two hops' germplasm, and this diversity was reflected by diverse DM reactions following artificial inoculation. Because of their biology, this study suggests that true hop seeds (generative reproduction) may be a more appropriate starting material for breeding programs and for generating new allelic variants faster compared to rootstock or rhizomes (clonal reproduction) for assessing both disease resistance and metabolite diversity. Further studies will be required to evaluate larger germplasm populations, including the mutagenized seedlings for their DM disease phenotypes and alpha and beta acid profiles for use in association studies leading to gene discovery.