Transcriptome characterization of three wild Chinese Vitis uncovers a large number of distinct disease related genes
Jiao et al. BMC Genomics
Transcriptome characterization of three wild Chinese Vitis uncovers a large number of distinct disease related genes
Chen Jiao 0 1 2
Min Gao 1 2
Xiping Wang 1 2
Zhangjun Fei 0 3
0 Boyce Thompson Institute for Plant Research, Cornell University , Ithaca, NY 14853 , USA
1 State Key Laboratory of Crop Stress Biology in Arid Areas, College of Horticulture, Northwest A&F University , Yangling, Shaanxi 712100 , China
2 Key Laboratory of Horticultural Plant Biology and Germplasm Innovation in Northwest China, Ministry of Agriculture, Northwest A&F University , Yangling, Shaanxi 712100 , China
3 USDA Robert W. Holley Center for Agriculture and Health , Tower Road Ithaca, NY 14853 , USA
Background: Grape is one of the most valuable fruit crops and can serve for both fresh consumption and wine production. Grape cultivars have been selected and evolved to produce high-quality fruits during their domestication over thousands of years. However, current widely planted grape cultivars suffer extensive loss to many diseases while most wild species show resistance to various pathogens. Therefore, a comprehensive evaluation of wild grapes would contribute to the improvement of disease resistance in grape breeding programs. Results: We performed deep transcriptome sequencing of three Chinese wild grapes using the Illumina strand-specific RNA-Seq technology. High quality transcriptomes were assembled de novo and more than 93% transcripts were shared with the reference PN40024 genome. Over 1,600 distinct transcripts, which were absent or highly divergent from sequences in the reference PN40024 genome, were identified in each of the three wild grapes, among which more than 1,000 were potential protein-coding genes. Gene Ontology (GO) and pathway annotations of these distinct genes showed those involved in defense responses and plant secondary metabolisms were highly enriched. More than 87,000 single nucleotide polymorphisms (SNPs) and 2,000 small insertions or deletions (indels) were identified between each genotype and PN40024, and approximately 20% of the SNPs caused nonsynonymous mutations. Finally, we discovered 100 to 200 highly confident cis-natural antisense transcript (cis-NAT) pairs in each genotype. These transcripts were significantly enriched with genes involved in secondary metabolisms and plant responses to abiotic stresses. Conclusion: The three de novo assembled transcriptomes provide a comprehensive sequence resource for molecular genetic research in grape. The newly discovered genes from wild Vitis, as well as SNPs and small indels we identified, may facilitate future studies on the molecular mechanisms related to valuable traits possessed by these wild Vitis and contribute to the grape breeding programs. Furthermore, we identified hundreds of cis-NAT pairs which showed their potential regulatory roles in secondary metabolism and abiotic stress responses.
Grape; Chinese wild Vitis; De novo transcriptome; Disease related genes; Cis-NATs
Grapes are among the most valuable fruit crops, grown
on about 7 million ha with an annual production of
approximately 67 million tonnes worldwide . There
are over 60 species of Vitis around the world  and Vitis
vinifera is the most widely planted grapevine. However,
most cultivars of V. vinifera are highly susceptible to
various economically important diseases such as powdery
mildew (PM) , downy mildew (DM)  and anthracnose
. Enhancing resistance to these diseases is the focus area
in current grape breeding programs. Wild species may
possess valuable genetic variations (e.g. new alleles, SNPs
and indels) in disease resistance genes, particularly, in
nucleotide-binding site leucine-rich repeat (NBS-LRR)
proteins. It has been reported that the species-specific
genes in wild and semi-wild watermelon were highly
enriched with genes involved in disease-related processes
. In addition, the wild eggplant carries nearly 200 extra
disease resistant genes compared to the cultivated eggplant
. Grape, unlike other domesticated crops, has retained
high genetic diversity from wide progenitors ; however,
most European cultivars are susceptible to many fungal
diseases. Moreover, only limited disease resistant loci, such
as Run1 , Ren1 , Ren2  and Ren3  that confer
resistance to PM, and Rpv loci including Rpv1 , Rpv2
 and Rpv3  that confer resistance to DM, have been
identified, mainly from wild grapevine species. Currently,
the widely planted grape cultivars are very sensitive to
diverse pathogens. Thus, genetic engineering of disease
resistance in grape has become an increasing need and
this can be facilitated by the use of wild Vitis resources.
China, as one of the major centers of the origin of
Vitis, has more than 35 native Vitis species . Chinese
wild Vitis are naturally distributed throughout the country
and many of them can survive in regions of high humidity
and moisture , under which the occurrence of fungal
diseases is increased . A number of disease resistant
Chinese wild Vitis have been identified and characterized.
The pioneering work from Wang et al.  and Wan et al.
 identified a large number of Chinese wild grapes that
displayed strong resistance to PM. Wan et al.  also
found about half of the Chinese wild Vitis were resistant
to DM and around one third of them have both PM and
DM resistances. In addition, Wang et al.  and Li et al.
 found that all the investigated Chinese wild Vitis
exhibited much less susceptibility to anthracnose
compared to the two V. vinifera cultivars (Cabernet Sauvignon
and Chardonnay). Moreover, Chinese wild Vitis showed
other specific characteristics, such as high photosynthetic
efficiency in V. quiqangualaris Rehd  and high content
of resveratrol (a phytoalexin that is beneficial to human
health) in V. quiqangualaris Danfeng-2 . However,
the underlying genetic variations contributing to these
phenotypic differences have not yet been explored.
Grapevine is the first fruit crop that had its genome
sequenced. The high quality genome has been served as
the reference for many genetic studies. However, recent
deep sequencing experiments have shown that relying
on a single reference genome may underestimate the
variability among different genotypes . For example,
reconstructing the transcriptomes of different grape
genotypes has revealed substantial heterogeneity in
transcripts that associated with their phenotypes [24,25].
In this study, we aim to characterize Chinese wild Vitis
transcriptomes to explore the potential genetic diversities
such as SNPs and indels. To this end, two Chinese
wild V. pseudoreticulata accessions Baihe-13-1 (BH) and
Hunan-1 (HN), and one V. quinquangularis accession
Shang-24 (S) were selected for deep transcriptome
sequencing. The selected accessions possess valuable
resistances to various fungal pathogens including PM
(BH and S) , DM (BH and HN)  and anthracnose
(BH, HN and S) . We de novo assembled the
transcriptomes and conducted comparative analysis of
SNPs and small indels between wild accessions and
the reference genome, PN40024. Distinct genes, which
represent genes that are absent or highly divergent
from sequences in the reference PN40024 genome,
were then identified from each accession. Using transcript
information from strand-specific RNA-Seq libraries, we
have identified cis-natural antisense transcript (cis-NAT)
pairs, which were known to participate in a broad range of
Transcriptome sequencing, de novo assembly, and
comparison with the reference genome
In order to more broadly capture disease related genes,
we infected young leaves of the three Chinese wild
grapes with PM. Based on the proposed PM infection
cycle [26,27], as well as our previous studies [28-30], we
collected leaves at 0, 6, 12, 24, 48, 72, 96 and 120 hours post
inoculation (hpi). We then prepared eight independent
strand-specific RNA-Seq libraries for each accession. In
total, 24 libraries were constructed and sequenced on the
Illumina HiSeq 2000 platform. After removing adaptors,
low quality sequences, and ribosomal RNA (rRNA) reads
(see method), we obtained a total of 53,890,427, 68,593,803
and 70,718,358 high quality cleaned reads for BH, HN and
S, respectively (Table 1). The transcriptomes of the three
wild Chinese Vitis were constructed de novo separately.
The final assembled transcript sets of BH, HN and S
contained 34,914, 38,528 and 38,204 contigs, respectively,
with N50 lengths of 814 bp, 876 bp and 980 bp and average
lengths of 602 bp, 630 bp and 679 bp (Table 1). The
GC-contents of these three transcript sets were
similar (~44.5%; Table 1), but slightly higher than that of the
PN40024 transcripts (42.0%). The assembled transcriptome
Table 1 Summary of transcriptome sequences and
assemblies of the three Chinese wild Vitis
2,911,785,932 3,703,865,895 3,864,462,091
High-quality cleaned reads 53,890,427
Average contig length
sequences of the three Chinese wild Vitis can be
downloaded and blasted at http://bioinfo.bti.cornell.
To discover the variations between Chinese wild Vitis
and PN40024, a genotype derived from Pinot Noir, all
three transcriptomes were aligned to the PN40024
genome. There were 94% (32,892), 94% (36,151) and 93%
(35,589) of the total transcripts can be uniquely aligned to
the reference genome with 97% sequence identity for
BH, HN and S, respectively (Figure 1). Only 1% of them
were mapped to multiple locations of the PN40024
genome and most of them were mapped to two locations.
The majority of transcripts (83%, 82% and 80% of the
total transcripts from BH, HN and S, respectively)
were mapped to the annotated gene regions (version
12 V0) with overlap fraction 90% and had the same
strand with the corresponding PN40024 transcripts;
whereas around 2% of the transcripts were aligned to gene
regions in antisense orientations. Notably, a small
proportion (5-7%) of the transcripts were aligned entirely to the
intergenic regions. Among all the assembled transcripts,
1,758 (5%, BH), 2,083 (5%, HN) and 2,331 (6%, S) could
not be aligned to the PN40024 genome. We considered
these transcripts as candidates for distinct genes.
Functional annotation of wild Chinese Vitis transcripts
All the transcripts from the three Chinese wild Vitis
were annotated by comparing their sequences against
the TrEMBL and Swiss-Prot protein databases . In
all three genotypes, around 88.5% of the transcripts
had hits in TrEMBL, and around 53.5% had hits in
Swiss-Prot (Table 2). The high percentage of transcripts
that hit to known proteins, combined with the high
mapping rate to the reference genome, indicated the high
quality of the de novo assembled transcripts.
The transcripts were further annotated by assigning
them human-readable functional terms extracted from
the functional descriptions of their homologous proteins
using AHRD . Approximately 80-81% of the transcripts
from the three species could be assigned with functional
terms. The Gene Ontology (GO) terms for each transcript
were then extracted according to the annotations in
Swiss-Prot and TrEMBL (Table 2). In total, about 77%
transcripts from each accession could be assigned with GO
terms from at least one of the three GO categories,
biological process, cellular component and molecular
function. The top five subcategories in biological
process were cellular process, biosynthetic process,
response to stress, cellular component organization
and nucleobase-containing compound metabolic process
(Additional file 1). Pathway analysis  revealed
approximately 3,000 genes from each accession that involved in
505513 biochemical pathways.
Distinct genes identified in the three wild Chinese Vitis
The assembled transcripts included redundant sequences,
mainly due to alternative splicing. We found that 4.80%
(BH), 6.81% (HN) and 8.94% (S) transcripts could be
clustered with other transcripts (sequence identity 97%
and overlap length 100 bp). Only one representative
transcript from each cluster was used in the downstream
functional enrichment/annotation analysis, to avoid the
repetitive counting of the same genes. Finally, we obtained
a total of 1,650-2,000 distinct unique transcripts in
each of the three accessions. The coding potential of
these transcripts was then assessed by Coding Potential
Calculator (CPC) . Consequently, we got 1,058, 1,296
and 1,315 distinct genes with high coding potential from
BH, HN and S, respectively, as well as 596, 609, and 732
potential non-coding transcripts (Figure 2A). The list of
these transcripts is provided in Additional file 2. We then
performed the GO term enrichment analyses (corrected
P-value 0.05) on these three distinct protein-coding gene
datasets. A total of 19 GO terms under the category of
biological process were enriched in distinct genes in all
three accessions (Table 3 and Figure 2B), and the most
Figure 1 Mapping of de novo assembled transcripts of the three Chinese wild Vitis, BH (A), HN (B), and S (C), to the reference PN40024
genome. No mapping: Contigs not mapped; Multiple hits: Contigs mapped to multiple genomic locations; Unique hit: Contigs mapped to
unique genomic locations; Mapping in intergenic regions: Contigs mapped to intergenic regions; Small overlap with gene regions: Contigs
mapped to gene regions with low overlapping (<90% of contig length); Large overlap with gene regions (+): Contigs mapped to gene regions
in sense directions with high overlapping (90% of contig length) ; Large overlap with gene regions (): Contigs mapped to gene regions in
antisense directions with high overlapping (90% of contig length). BH, V. pseudoreticulata accession Baihe-13-1; HN, V. pseudoreticulata
accession Hunan-1; S, V. quinquangularis accession Shang-24.
Table 2 Annotations of assembled transcriptomes of the
three Chinese wild Vitis
18,704 (53.57%) 20,491 (53.18%) 20,703 (54.19%)
26,936 (77.15%) 29,491 (76.54%) 29,195 (76.42%)
MetaCyc pathway 3,114 (8.92%)
BH, V. pseudoreticulata accession Baihe-13-1; HN, V. pseudoreticulata
accession Hunan-1; S, V. quinquangularis accession Shang-24.
representative GO term was response to stimulus
(GO:0050896), which included nearly half of the distinct
genes. Among the child terms of GO:0050896, the
most significantly enriched one was defense response
(GO:0006952; Table 3). We also identified 15, 18 and 30
enriched child GO terms associated with
resistancerelated biological processes in BH, HN and S, respectively
(Additional file 3).
Plants produce three main secondary metabolites,
terpenes, phenolics and nitrogen- and sulfur-containing
compounds, which provide a major barrier against the
attack of pathogens and herbivores [35,36]. Here, we found
that several biological processes related to biosynthesis of
phenolic compounds such as coumarin, were enriched
in the distinct genes in all three accessions (Table 3).
Biosynthesis of flavonoids, a class of phenolic
compounds, were enriched in distinct genes of both BH
and S. It is worth noting that callose deposition in
cell wall and cell wall thickening biological processes
and two other processes involved in the metabolism
of anthocyanin-containing compounds were enriched
in the distinct genes of the S accession. In addition,
terpenoid compound metabolism and glucosinolate
metabolic related processes were only enriched in the
HN distinct genes (Additional file 3).
We found an important category of enriched biological
processes that were related to the metabolism and action
of plant hormones, a group of small molecules which
function as versatile regulators of plant growth, development,
reproduction and response to abiotic or biotic stresses
[37,38]. Specifically, jasmonic acid related processes were
enriched in distinct genes of BH and S, whereas ethylene
biosynthetic and metabolic processes were overrepresented
in distinct genes of both HN and S. Moreover, abscisic acid
related process was only enriched in HN while processes
related to salicylic acid and brassinosteroid were enriched
in S distinct genes (Additional file 3).
We identified 145 (BH), 215 (HN) and 196 (S) distinct
genes that were predicted to encode disease-related
proteins (Figure 2C). Interestingly, a large number of them
were NBS-LRR genes (81 in BH, 114 in HN and 107 in S),
which play important roles in plant effector-triggered
immunity (ETI) as they can directly or indirectly detect
pathogen-associated proteins . In addition, a large
Figure 2 Distinct genes in the three Chinese wild Vitis, BH, HN and S. (A) Number of distinct protein-coding genes and non-coding
transcripts from de novo transcriptome assemblies of the three Chinese Vitis leaf tissues, which were collected at 0, 6, 12, 24, 48, 72, 96, and
120 hours post inoculation with PM, respectively. (B) Venn diagram of GO terms enriched in the distinct protein-coding genes. (C) Number of
distinct genes encoding disease resistance proteins, receptor like kinases and transcription factors. BH, V. pseudoreticulata accession Baihe-13-1;
HN, V. pseudoreticulata accession Hunan-1; S, V. quinquangularis accession Shang-24.
Table 3 GO terms enriched in the distinct gene sets of all three Chinese wild Vitis
GO:0098542 defense response to other organism
GO:0050896 response to stimulus
GO:0006950 response to stress
GO:0006952 defense response
GO:0009617 response to bacterium
GO:0045087 innate immune response
GO:0006955 immune response
GO:0042742 defense response to bacterium
GO:0009814 defense response, incompatible interaction
GO:0009816 defense response to bacterium, incompatible 6
GO:0009626 plant-type hypersensitive response
GO:0098581 detection of external biotic stimulus
GO:0009804 coumarin metabolic process
GO:0009805 coumarin biosynthetic process
GO:0015074 DNA integration
GO:0015916 fatty-acyl-CoA transport
GO:1901337 thioester transport
GO:0080001 mucilage extrusion from seed coat
GO:0009835 fruit ripening
BH HN C* B Corrected C P-value 47% 35% 3.73E-13
38% 24% 1.65E-22
42% 24% 2.41E-45
41% 24% 9.38E-39
26% 11% 4.26E-37
31% 12% 2.01E-78
31% 12% 2.34E-75
BH, V. pseudoreticulata accession Baihe-13-1; HN, V. pseudoreticulata accession Hunan-1; S, V. quinquangularis accession Shang-24.
*C: frequency of distinct genes belonging to the corresponding GO terms; B: frequency of all transcripts in each wild Vitis species belonging to the corresponding
number of genes contained known protein domains
related to disease resistance, such as NB-ARC ,
LRR, TIR and Dirigent . Another group of distinct
disease-related genes were those encoding receptor
like protein kinases (66, 92 and 133 in BH, HN and S,
respectively) (Figure 2C). In addition to the genes described
above, we discovered several genes that are involved in
plant-pathogen interaction, such as Mlo,
phytoalexindeficient 4 (PAD4), enhanced disease susceptibility 1
(EDS1), RPW8.2 and genes encoding lipoxygenases [42-45].
Transcription factors play a central role in mediating
biological activity in plant cells. We discovered 19
transcription factors in each of the three Vitis genotypes. More
than 70% of these transcription factors belonged to the
ethylene-responsive factor family, which has been reported
to be involved in the control of primary and secondary
metabolism, growth and developmental programs, as well
as responses to environmental stresses .
SNPs and small indels between wild Chinese Vitis and
Genomic variations, such as SNPs and small insertions
and deletions (indels) are important driving force of genetic
diversity. We identified SNPs and small indels through
mapping the RNA-Seq reads from each accession to the
reference PN40024 genome. We obtained a total of 110,450
SNPs and 2,354 small indels between BH and PN40024,
87,583 SNPs and 2,079 small indels between HN and
PN40024, and 89,024 SNPs and 2,085 small indels between
S and PN40024. Among these variations, 52% (BH), 48%
(HN) and 49% (S) of SNPs, and 3.19% (BH), 3.56%
(HN) and 4.22% (S) of small indels were located in
the annotated coding regions (Figure 3A and 3B). In
BH, we identified ~24,000 nonsynonymous substitutions,
which potentially affect approximately 10,000 genes. In
the other two accessions, we detected more than 18,000
nonsynonymous mutations that may alter the function
of ~9,000 genes. The overall ratio of nonsynonymous
to synonymous sites was ~0.7 in all three accessions.
Interestingly, this ratio in NBS-LRR genes was
substantially higher (1.3 in BH, 2.1 in HN and 3.0 in S).
SNPs located at specific regions might have large effects
on corresponding genes, such as gain or loss of start/stop
codons, and disruption of splice site acceptors or donors
[47,48]. We found 32, 24 and 24 genes in BH, HN and S,
respectively, contained loss of start codon SNPs which
would change the length of protein products (Figure 3C).
About 40 genes with loss of stop codons and 70 genes
Figure 3 SNPs and small indels between the three Chinese wild Vitis, BH, HN and S, and PN40024. (A) Number of SNPs at different
annotated regions of the reference PN40024 genome. (B) Number of small indels at different annotated regions of the reference PN40024
genome. (C) Number of genes affected by SNPs and small indels. SSA/D, splice site acceptor or splice site donor; FS, frame shift; SpG, stop codon
gained; SpL, stop codon lost; StL, start codon lost. BH, V. pseudoreticulata accession Baihe-13-1; HN, V. pseudoreticulata accession Hunan-1;
S, V. quinquangularis accession Shang-24.
with gain of stop codons were identified in each accession.
Moreover, small indels mapped to coding regions
(Figure 3B) led to the frame shifts of 74, 72 and 87 genes
in BH, HN and S, respectively (Figure 3C). The identified
SNPs and small indels were listed in Additional file 4.
Cis-NATs in wild Chinese Vitis
Natural antisense transcripts (NATs) are endogenous
RNA sequences partially or entirely complemented to
other transcripts and cis-NAT pairs are two transcripts
from the same genomic locus but on opposite strands
. We explored cis-NATs in our strand specific RNA-Seq
data and identified 112, 145 and 196 cis-NAT pairs in BH,
HN and S, respectively (Additional file 5). The classification
of cis-NAT pairs based on their overlapping patterns
was shown in Table 4. In addition, 84 (75.0%; BH),
115 (79.3%; HN) and 156 (79.6%; S) cis-NAT pairs
were designated as coding-noncoding pairs depending
on their coding potential calculated by CPC, while 25
(22.3%), 22 (15.2%) and 33 (16.8%) were classified as
coding-coding pairs. Cis-NATs were previously reported
to be involved in modulating various abiotic and biotic
stresses [50,51]. Through functional analysis of the
cis-NAT pairs we identified in the three Chinese wild
grapes, we also found those related to abiotic stress
responses were significantly enriched. In addition, cis-NATs
involved in secondary metabolite biosynthetic
processes were highly enriched; specifically, among the
top five enriched biological processes, four (GO:0009698,
GO:0009813, GO:0009812 and GO:0009699), three
(GO:0009813, GO:0009812 and GO:0044550) and two
(GO:0009813 and GO:0009812) enriched biological
processes in BH, HN and S, respectively, were associated with
flavonoid metabolic processes (Additional file 6).
In this study, we de novo assembled transcriptomes of
three wild Chinese Vitis accessions. Distinct genes
and genomic variations between Chinese Vitis and the
reference PN40024 were extensively investigated. We
Table 4 Structure analysis of cis-NAT pairs in the three
Chinese wild Vitis
Tail to tail (3 to 3)
Head to head (5 to 5)
identified a total of more than 1,000 distinct protein-coding
genes, as well as 600700 non-coding transcripts from each
of the three Chinese wild Vitis. Almost half of the distinct
protein-coding genes were functionally related to stimulus
responses, and ~30% were involved in defense responses.
The enrichment of stress-related processes in distinct genes
is consistent with the hypotheses that wild plants, which
are often exposed to an adverse environment, such as
extreme temperatures, aberrant pH, toxic chemicals and
pathogens, tend to have accelerated adaptive evolution of
stress genes , while during domestication, a large
number of stress and defense-related genes are lost in the
cultivated species, which might be due to the many years of
cultivation and selection that have focused on desirable
fruit qualities at the expense of disease resistance .
Genomic variations, such as SNPs and indels, are
important sources of genetic diversity. They are functionally
significant and can cause phenotypic changes. Genetic
analysis of plant disease resistance has shown that
resistance is dominated by multiple loci and alleles at
each locus are often highly polymorphic [53-55]. In the
present study, we obtained 87,000-110,000 homozygous
SNPs and ~2,000 small indels between each of the three
wild Chinese Vitis and V. vinifera, genotype PN40024.
This number is distinguished from a previous report on
the identification of many fewer SNPs (~59,000) between
V. vinifera cv. Corvina and PN40024 transcriptomes ,
probably due to both species belong to V. vinifera.
Cis-NATs are derived from the same genomic loci as
their sense counterparts, but from the opposite strand.
Cis-NAT pairs can be classified into three groups based
on their overlapping patterns, including 3 end overlap,
5 end overlap and one entirely contained within the
other . In our study, across all three Vitis spp., the
majority of cis-NAT pairs were those entirely contained
within the other transcripts. This is consistent with the
finding in soybean  but different from that in Arabidopsis
, where most cis-NAT pairs have 3 end overlap.
Cis-NATs were reported to participate in plant responses
to a wide range of abiotic stresses, such as cold  and
salt  stresses. Consistent with these findings, we found
that cis-NATs identified from wild Chinese grapes were
significantly enriched with transcripts related to stress
responses. In addition, we also observed high enrichment
of secondary metabolism-related genes in the wild grape
cis-NATs, implying the functional diversity of cis-NATs in
mediating important biological processes in plants.
In the present study, we found 6-10% distinct genes
were surface-localized receptor like kinases (RLKs). Some
RLKs, as well as receptor like proteins (RLPs), are
patternrecognition receptors (PRRs) . They contain various
ligand-binding ectodomains that can perceive
pathogenassociated molecular patterns (PAMPs), e.g. flagellin,
peptidoglycans (PGNs) and chitins, or damage-associated
molecular patterns (DAMPs). RLKs (e.g. LYK1) or RLPs
(e.g. LYM2) can function independently or cohesively .
In this study, we discovered five distinct genes (four in S
including one LYK1 and one LYM2 gene, and one in HN)
containing the LysM motif, a common unit in both RLKs
and RLPs that are responsible for binding to various types
of PGNs and chitins. Unlike typical PRRs, Pep1 receptor
(PEPR2) acts as a receptor for PEP defense peptides
and senses an endogenous elicitor that potentiates
PAMP-inducible plant responses . One PEPR2 was
found in HN distinct genes. PEPR2 and its closest
homolog PEPR1 can interact with BAK1 , which
is a co-receptor for several PAMPs and regulates their
function . In addition, other RLKs that do not
belong to PRRs, such as wall-associated kinase (WAK)
also play important roles in plant immunity signaling
pathway. In this study, we found 710 WAK family
members in each distinct gene set. WAKs have an
extracellular domain and they can interact with pectin
and other proteins located in the cell wall. Induction
of WAK1 is required for plants to survive during P.
syringae infection. Furthermore, the increase of WAK1
mRNA levels is part of the defense response caused by
exposure to jasmonic acid (JA), ethylene, or fungi . A
previous study has unraveled adaptive evolution of
extracellular domains of RLKs , which may explain why
many RLKs and RLPs were observed in distinct genes.
The discovery of these genes may also imply the functional
conservation in response to pathogen invasion.
Pathogens can overcome PAMP-triggered immunity
(PTI) by deploying effectors to interfere with the PTI
activated signaling pathways. These effectors can be
perceived by plant R genes (mainly NBS-LRR) and activate
plant immune system, also known as effector-triggered
immunity. These pathogens are usually highly specialized
for specific host plants, and the interaction at the molecular
level is often complicated because of the co-evolution of
the host and pathogens . Wild and domesticated plant
species have been exposed to the natural selection forces,
and show divergent R genes as they have to initiate an arms
race with the pathogen effector. However, compared to
clonally propagated grape cultivars, wild Vitis may have
evolved new disease resistance genes during sexual
propagation. Several studies have demonstrated that
in plant R-gene products, LRR domains are the major
determinants of recognition specificity for effectors
 and these domains were under diversifying selection
to increase amino acid variability. Mechanisms for the
evolution of new specificities are flexible, such as gene
conversion and unequal recombination, as well as accumulation
of amino acid codon exchanges in members of anciently
duplicated gene families . In our distinct genes, ~8%
were NBS-LRR genes. Notably, from our SNP analysis, we
found an R gene, GSVIVT01032161001, which has longer
protein product in the three wild Vitis compared with that
in PN40024 caused by a SNP in its stop codon. The longer
R gene might confer increased specificity for pathogen
recognition in the three Chinese wild Vitis species.
Interestingly, in a wild potato, a homolog of the Vitis
longer R gene has been identified, which is heterozygous
(six amino acids loss in the 18th LRR repeat for one allele)
and confers broad spectrum resistance to late blight .
In this study, a number of flavonoid and isoprenoid
biosynthesis genes were identified in the distinct gene
sets. Flavonoid and isoprenoid biosynthesis was found to be
up-regulated in both V. vinifera and V. pseudoreticulata
that were inoculated with powdery mildew [27,68].
We also found cis-NATs were enriched with flavonoid
biosynthesis-related genes in all three wild grapes,
which may suggest that flavonoid metabolism is regulated
by cis-NATs. In our distinct gene sets, several genes
related to coumarin biosynthesis and metabolism were
discovered. Coumarin, a known phytoalexin, was found to
accumulate in parsley cells treated with a fungal glucan
elicitor . Several other phytoalexin-related genes were
also found in our distinct gene sets, including stilbene
synthase (STS) which is a key gene in the biosynthesis of
stilbenes, and phytoalexin-deficient 4 (PAD4), a well-known
resistance gene in Arabidopsis. A recent study showed that
Arabidopsis transformed with Vitis enhanced disease
susceptibility 1 (EDS1) and PAD4 did not display rescued
resistance to powdery mildew, even though these two
proteins interacted when transiently expressed in Nicotiana
benthamiana. Therefore, involvement of additional
interacting proteins might be necessary for resistance
to occur .
Transcriptional regulation of stress responsive genes
plays a central role in abiotic/biotic stress responses.
ERF, a subfamily of AP2 transcription factor, has been
reported to be involved in many stress responses .
Pti4, an ERF in tomato, can be phosphorylated by a
disease resistance protein (Pto) and regulate GCC-box PR
genes . McGrath et al.  found that AP2/ERF
genes were the predominant transcription factor family
genes responsive to both JA and a fungal pathogen
Alternaria brassicicola in Arabidopsis. Specifically, AtERF2
acted as a positive regulator of JA-responsive defense gene
expression and resistance to A. brassicico while AtERF4
showed the opposite functions. Their results suggest that
plants coordinately express multiple repressor- and
activator-type AP2/ERFs during pathogen challenge to
modulate defense gene expression and disease resistance
. Interestingly, in our distinct gene sets from each
of three wild species, 70% of the transcription factors
belonged to the AP2/ERF family, which suggest that
in Chinese wild Vitis the distinct AP2/ERF genes
might co-evolve with R genes and have a high degree
of sequence variations.
In the present study we de novo constructed transcriptomes
of three Chinese wild grapes, which showed resistances to
various fungal pathogens. A comprehensive comparison
between these transcriptomes and the reference grape
genome unraveled a large number of distinct genes and a
rich resource of genetic variations such as SNPs and small
indels. Interestingly, many genetic divergences between
wild and cultivated grapes were found to be highly related
to many important biological processes, particularly defense
associated processes, suggesting that the accelerated
evolution of these genes may contribute to plant
adaptation to different environments. Furthermore,
the significant enrichment of cis-NAT pairs related to
secondary metabolism and abiotic stress responses
may shed lights on the potential regulatory roles of
cis-NATs in Chinese wild Vitis.
Plant material, PM inoculation and RNA-Seq library
Two Chinese wild V. pseudoreticulata accessions
Baihe13-1 and Hunan-1, and one V. quinquangularis
accession Shang-24, were maintained in the grape germplasm
resource orchard at Northwest A&F University, Yangling,
China (34 20 N, 10824 E). Young leaves from three
separate vines of each accession were inoculated with PM
[Erysiphe necator (Schw.) Burr.] as previously described
. E. necator as an obligate biotrophic fungus, grows
and reproduces only on living grapes. The isolate we used
was obtained from grape leaves showing fully developed
PM symptom. To collect samples that had a good
represntation for the PM infection process, which would
more broadly capture disease related genes responsive
to PM infection, we harvested leaves at 0, 6, 12, 24,
48, 72, 96 and 120 hpi based on the proposed PM
infection cycle [26,27] and our previous studies [28-30].
The collected leaves were immediately frozen in liquid
nitrogen and stored at 80C till use. Total RNA was
extracted following the method described in Guo et al.
. The quality and quantity of RNA were assessed by
electrophoresis on 1% agarose gels and by a NanoDrop
1000 spectrophotometer (Thermo Scientific, Wilmington,
DE, USA), respectively. Strand-specific RNA-Seq libraries
were constructed using the protocol described in
Zhong et al.  and sequenced on the Illumina HiSeq
2000 platform using the single-end mode.
Data processing, de novo assembly and comparison with
the reference genome
Raw reads from each of the three Chinese wild Vitis
accessions were processed using Trimmomatic  to
remove adaptor and low quality sequences. Reads
shorter than 40 bp were discarded. The resulting reads
were aligned to the ribosomal RNA database  using
bowtie  and those aligned were discarded. The
resulting high-quality cleaned reads (final reads) were
subjected to Trinity  for de novo assembly with the
minimum kmer coverage set to two. The final reads
were then aligned to the assembled contigs using bowtie
. To remove false transcripts with antisense direction
which was due to the incomplete digestion of the 2nd strand
during the strand-specific RNA-Seq library construction
, contigs with the number of reads aligned in sense
direction less than 1/10 of the number of reads aligned in
antisense direction were discarded. The assembled contigs
were then compared against the GenBank nt database ,
and those having hits from viruses, bacteria, archaea and
fungi, but not from plant species, were removed. The final
set of assembled contigs were further clustered to remove
redundancies using iAssembler  with sequence identity
cutoff set to 97%. Finally, we used seqclean  to trim
polyA tails and to remove rRNA sequences at the contig
level. The final assembled transcripts were aligned to
the 12 PN40024 genome assembly  using BLAT
 with sequence identity no less than 97%.
Functional annotation of assembled transcripts
To annotate the assembled transcripts, their sequences
were searched against the TrEMBL and Swiss-Prot 
protein databases using the blastx program, with the
E-value cutoff of 1e-10. GO terms were assigned to
the assembled transcripts based on the GO terms assigned
to their hits in TrEMBL and Swiss-Prot databases .
The functional terms were assigned to each transcript
using automated assignment of human readable
descriptions (AHRD) . Enzyme-encoding genes were extracted
based on the AHRD result, and used to predict
biochemical pathways using the Pathway Tools software .
Identification of distinct genes
We first clustered the assembled transcripts under the
criteria, sequence identity 97% and overlap length 100 bp,
to remove redundant transcripts that could be derived
from the same gene locus, mainly due to alternative
splicing. The longest representative transcript from each
cluster was kept and then aligned to the PN40024 genome
 using BLAT  with sequence identity 97%. The
unmapped transcripts, which were considered as distinct
transcripts, were then checked for their coding potential
using Coding Potential Calculator (CPC) . Transcripts
with positive coding potential scores were identified as
distinct protein-coding genes and those with negative
scores were non-coding transcripts. GO term enrichment
analysis of the distinct protein-coding genes was performed
using GO::TermFinder . The putative functional
domains from Pfam database  were identified for
these distinct genes using HMMER3.0 .
Identification of SNPs and small indels
To identify SNPs and small indels between each of the
three Chinese wild accessions and the grape reference,
raw RNA-Seq reads were aligned to the grape reference
genome using the Burrows-Wheeler Aligner . Only
one of the duplicated RNA-Seq reads was kept to
minimize the artifacts of PCR amplification, and only
reads uniquely mapped to the genome were kept. Following
mapping, SNPs and small indels were identified based
on the mpileup files generated by SAMtools . The
identified SNPs and small indels were supported by at
least four distinct RNA-Seq reads and had an allele
frequency > 70%. The effects of detected SNPs and small
indels were analyzed based on the 12 V0 annotation of
the PN40024 genome.
Identification of cis-NATs
To identify cis-NAT pairs, we first compared the sequences
of assembled transcripts from each wild Vitis accession
against themselves. Transcript pairs which were aligned in
reverse direction and had the overlap length greater than
50 bp were kept. The resulting transcript pairs were then
aligned to the reference grape genome and those aligned to
the same genome regions and showing distinct splicing
patterns were identified as cis-NAT pairs.
Availability of supporting data
The raw sequencing data has been deposited in
NBCI SRA under the accession numbers SRP051051,
SRP051054 and SRP051078: http://www.ncbi.nlm.nih.
Additional file 1: GO classification of assembled transcriptomes of
the three Chinese wild Vitis. The figure shows GO term classification of
assembled transcripts of the three Chinese wild Vitis, V. pseudoreticulata
accession Baihe-13-1 (BH), V. pseudoreticulata accession Hunan-1 (HN)
and V. quinquangularis accession Shang-24 (S).
Additional file 2: Distinct genes identified in the three Chinese wild
Vitis. The table provides the lists of distinct genes identified in the three
Chinese wild Vitis, V. pseudoreticulata accession Baihe-13-1 (BH), V.
pseudoreticulata accession Hunan-1 (HN) and V. quinquangularis
accession Shang-24 (S).
Additional file 3: GO terms enriched in the distinct genes from the
three Chinese wild Vitis. The table provides the lists of GO terms that
were enriched in the distinct genes from the three Chinese wild Vitis, V.
pseudoreticulata accession Baihe-13-1 (BH), V. pseudoreticulata accession
Hunan-1 (HN) and V. quinquangularis accession Shang-24 (S).
Additional file 4: SNPs and small indels between each of the three
Chinese wild Vitis. The table provides the lists of SNPs and small indels
between the three Chinese wild Vitis, V. pseudoreticulata accession
Baihe-13-1 (BH), V. pseudoreticulata accession Hunan-1 (HN), V.
quinquangularis accession Shang-24 (S), and V. vinifera PN40024.
Additional file 5: cis-NAT pairs identified in the three Chinese wild
Vitis. The table provides the lists of cis-NAT pairs identified in the three
Chinese wild Vitis, V. pseudoreticulata accession Baihe-13-1 (BH), V.
pseudoreticulata accession Hunan-1 (HN) and V. quinquangularis
accession Shang-24 (S).
Additional file 6: GO terms enriched in cis-NATs from the three
Chinese wild Vitis. The table provides the lists of GO terms that were
enriched in cis-NATs from the three Chinese wild Vitis, V. pseudoreticulata
accession Baihe-13-1 (BH), V. pseudoreticulata accession Hunan-1 (HN)
and V. quinquangularis accession Shang-24 (S).
CJ and MG prepared samples for Illumina sequencing. CJ performed data
analysis. CJ and ZF wrote the manuscript. ZF and XW designed the
experiment and provided guidance. All authors have read and approved
The authors would like to thank Monica Franciscus for proofreading.
This work was supported by grants from National Science Foundation
(IOS-0923312 and IOS-1313887) to ZF, National Natural Science Foundation
of China (grant no. 31272136) and the Program for Innovative Research Team
of Grape Germplasm Resources and Breeding (grant no. 2013KCT-25) to XW.
CJ was partially supported by a fellowship from China Scholarship Council.
1. FAOSTAT. http://faostat3.fao.org/
2. Owens CL . Grapes . In: Hancock J, editor. Temperate Fruit Crop Breeding. Netherlands: Springer; 2008 . p. 197 - 233 .
3. Doster MA , Schnathorst WC . Comparative susceptibility of various grapevine cultivars to the powdery mildew fungus Uncinula necator . Am J Enol Viticult . 1985 ; 36 : 101 - 4 .
4. Langcake P , Lovell PA. Light and electron microscopical studies of the infection of Vitis spp by Plasmopara viticola, the downy mildew pathogen . Vitis . 1980 ; 19 : 321 - 37 .
5. Mortensen JA . Sources and inheritance of resistance to anthracnose in Vitis . J Hered . 1981 ; 72 : 423 - 6 .
6. Guo S , Zhang J , Sun H , Salse J , Lucas WJ , Zhang H , et al. The draft genome of watermelon (Citrullus lanatus) and resequencing of 20 diverse accessions . Nat Genet . 2013 ; 45 : 51 - 8 .
7. Yang X , Cheng YF , Deng C , Ma Y , Wang ZW , Chen XH , et al. Comparative transcriptome analysis of eggplant (Solanum melongena L) and turkey berry (Solanum torvum Sw): phylogenomics and disease resistance analysis . BMC Genomics . 2014 ; 15 : 412 .
8. Myles S , Boyko AR , Owens CL , Brown PJ , Grassi F , Aradhya MK , et al. Genetic structure and domestication history of the grape . Proc Natl Acad Sci U S A . 2011 ; 108 : 3530 - 5 .
9. Barker CL , Donald T , Pauquet J , Ratnaparkhe MB , Bouquet A , Adam-Blondon AF , et al. Genetic and physical mapping of the grapevine powdery mildew resistance gene, Run1, using a bacterial artificial chromosome library . Theor Appl Genet . 2005 ; 111 : 370 - 7 .
10. Hoffmann S , Di Gaspero G , Kovacs L , Howard S , Kiss E , Galbacs Z , et al. Resistance to Erysiphe necator in the grapevine 'Kishmish vatkana' is controlled by a single locus through restriction of hyphal growth . Theor Appl Genet . 2008 ; 116 : 427 - 38 .
11. Dalb MA , Ye GN , Weeden NF , Wilcox WF , Reisch BI . Marker-assisted selection for powdery mildew resistance in grapes . J Am Soc Hortic Sci . 2001 ; 126 : 83 - 9 .
12. Fischer BM , Salakhutdinov I , Akkurt M , Eibach R , Edwards KJ , Topfer R , et al. Quantitative trait locus analysis of fungal disease resistance factors on a molecular map of grapevine . Theor Appl Genet . 2004 ; 108 : 501 - 15 .
13. Merdinoglu D , Wiedeman-Merdinoglu S , Coste P , Dumas V , Haetty S , Butterlin G , et al. Genetic analysis of downy mildew resistance derived from Muscadinia rotundifolia . Acta Hort . 2003 ; 603 : 451 - 6 .
14. Peressotti E , Wiedemann-Merdinoglu S , Delmotte F , Bellin D , Di Gaspero G , Testolin R , et al. Breakdown of resistance to grapevine downy mildew upon limited deployment of a resistant variety . BMC Plant Biol . 2010 ; 10 : 147 .
15. Di Gaspero G , Copetti D , Coleman C , Castellarin SD , Eibach R , Kozma P , et al. Selective sweep at the Rpv3 locus during grapevine breeding for downy mildew resistance . Theor Appl Genet . 2012 ; 124 : 277 - 86 .
16. Wan YZ , Schwaninger H , Li D , Simon CJ , Wang YJ , Zhang CH . A review of taxonomic research on Chinese wild grapes . Vitis . 2008 ; 47 : 81 - 8 .
17. Carroll JE , Wilcox WF . Effects of humidity on the development of grapevine powdery mildew . Phytopathology . 2003 ; 93 : 1137 - 44 .
18. Wang Y , Liu Y , He P , Chen J , Lamikanra O , Lu J. Evaluation of foliar resistance to Uncinula necator in Chinese wild Vitis species . Vitis . 1995 ; 34 : 159 - 64 .
19. Wan YZ , Schwaninger H , He PC , Wang YJ . Comparison of resistance to powdery mildew and downy mildew in Chinese wild grapes . Vitis . 2007 ; 46 : 132 - 6 .
20. Wang Y , Liu Y , He P , Lamikanra O , Lu J. Resistance of Chinese Vitis species to Elsinoe ampelina (de Bary) Shear . Hortscience. 1998 ; 33 : 123 - 6 .
21. Li D , Wan Y , Wang Y , He P. Relatedness of resistance to anthracnose and to white rot in Chinese wild grapes . Vitis . 2008 ; 47 : 213 - 5 .
22. Zhu L , Wen XY , Li WW . Studies on photosynthetic characteristic of a Chinese wild species grapevine . Acta Hortic Sin . 1994 ; 21 : 31 - 4 .
23. Shi J , He M , Cao J , Wang H , Ding J , Jiao Y , et al. The comparative analysis of the potential relationship between resveratrol and stilbene synthase gene family in the development stages of grapes (Vitis quinquangularis and Vitis vinifera) . Plant Physiol Biochem . 2014 ; 74 : 24 - 32 .
24. Venturini L , Ferrarini A , Zenoni S , Tornielli GB , Fasoli M , Dal Santo S , et al. De novo transcriptome characterization of Vitis vinifera cv. Corvina unveils varietal diversity . BMC Genomics . 2013 ; 14 : 41 .
25. Da Silva C , Zamperin G , Ferrarini A , Minio A , Dal Molin A , Venturini L , et al. The high polyphenol content of grapevine cultivar tannat berries is conferred primarily by genes that are not shared with the reference genome . Plant Cell . 2013 ; 25 : 4777 - 88 .
26. Fekete C , Fung RW , Szabo Z , Qiu W , Chang L , Schachtman DP , et al. Up-regulated transcripts in a compatible powdery mildew-grapevine interaction . Plant Physiol Biochem . 2009 ; 47 : 732 - 8 .
27. Fung RW , Gonzalo M , Fekete C , Kovacs LG , He Y , Marsh E , et al. Powdery mildew induces defense-oriented reprogramming of the transcriptome in a susceptible but not in a resistant grapevine . Plant Physiol . 2008 ; 146 : 236 - 49 .
28. Li H , Xu Y , Xiao Y , Zhu Z , Xie X , Zhao H , et al. Expression and functional analysis of two genes encoding transcription factors, VpWRKY1 and VpWRKY2, isolated from Chinese wild Vitis pseudoreticulata . Planta . 2010 ; 232 : 1325 - 37 .
29. Wen Y , Wang X , Xiao S , Wang Y. Ectopic expression of VpALDH2B4, a novel aldehyde dehydrogenase gene from Chinese wild grapevine (Vitis pseudoreticulata), enhances resistance to mildew pathogens and salt stress in Arabidopsis . Planta. 2012 ; 236 : 525 - 39 .
30. Guo R , Xu X , Carole B , Li X , Gao M , Zheng Y , et al. Genome-wide identification, evolutionary and expression analysis of the aspartic protease gene superfamily in grape . BMC Genomics . 2013 ; 14 : 554 .
31. Consortium UP . Ongoing and future developments at the universal protein resource . Nucleic Acids Res . 2011 ; 39 : D214 - 219 .
32. AHRD (Automated Assignment of Human Readable Descriptions) . https://github.com/asishallab/AHRD-1
33. Karp PD , Paley S , Romero P. The Pathway Tools software . Bioinformatics . 2002 ;18 Suppl 1: S225 - 232 .
34. Kong L , Zhang Y , Ye ZQ , Liu XQ , Zhao SQ , Wei L , et al. CPC: assess the protein-coding potential of transcripts using sequence features and support vector machine . Nucleic Acids Res . 2007 ; 35 : W345 - 349 .
35. Dixon RA . Natural products and plant disease resistance . Nature . 2001 ; 411 : 843 - 7 .
36. Mithofer A , Boland W. Plant defense against herbivores: chemical aspects . Annu Rev Plant Biol . 2012 ; 63 : 431 - 50 .
37. Pieterse CM , Van der Does D , Zamioudis C , Leon-Reyes A , Van Wees SC . Hormonal modulation of plant immunity . Annu Rev Cell Dev Biol . 2012 ; 28 : 489 - 521 .
38. Santner A , Calderon-Villalobos LI , Estelle M. Plant hormones are versatile chemical regulators of plant growth . Nat Chem Biol . 2009 ; 5 : 301 - 7 .
39. DeYoung BJ , Innes RW . Plant NBS-LRR proteins in pathogen sensing and host defense . Nat Immunol . 2006 ; 7 : 1243 - 9 .
40. van der Biezen EA , Jones JD . The NB-ARC domain: a novel signalling motif shared by plant resistance gene products and regulators of cell death in animals . Curr Biol . 1998 ; 8 : R226 - 227 .
41. Ralph S , Park JY , Bohlmann J , Mansfield SD . Dirigent proteins in conifer defense: gene discovery, phylogeny, and differential wound- and insect-induced expression of a family of DIR and DIR-like genes in spruce (Picea spp.) . Plant Mol Biol . 2006 ; 60 : 21 - 40 .
42. Glazebrook J , Zook M , Mert F , Kagan I , Rogers EE , Crute IR , et al. Phytoalexin-deficient mutants of Arabidopsis reveal that PAD4 encodes a regulatory factor and that four PAD genes contribute to downy mildew resistance . Genetics . 1997 ; 146 : 381 - 92 .
43. Aarts N , Metz M , Holub E , Staskawicz BJ , Daniels MJ , Parker JE . Different requirements for EDS1 and NDR1 by disease resistance genes define at least two R gene-mediated signaling pathways in Arabidopsis . Proc Natl Acad Sci U S A . 1998 ; 95 : 10306 - 11 .
44. Xiao S , Ellwood S , Calis O , Patrick E , Li T , Coleman M , et al. Broad-spectrum mildew resistance in Arabidopsis thaliana mediated by RPW8 . Science . 2001 ; 291 : 118 - 20 .
45. Buschges R , Hollricher K , Panstruga R , Simons G , Wolter M , Frijters A , et al. The barley Mlo gene: a novel control element of plant pathogen resistance . Cell . 1997 ; 88 : 695 - 705 .
46. Licausi F , Ohme-Takagi M , Perata P. APETALA2/Ethylene Responsive Factor (AP2/ERF) transcription factors: mediators of stress responses and developmental programs . New Phytol . 2013 ; 199 : 639 - 49 .
47. Jiao WB , Huang D , Xing F , Hu Y , Deng XX , Xu Q , et al. Genome-wide characterization and expression analysis of genetic variants in sweet orange . Plant J . 2013 ; 75 : 954 - 64 .
48. Lai J , Li R , Xu X , Jin W , Xu M , Zhao H , et al. Genome-wide patterns of genetic variation among elite maize inbred lines . Nat Genet . 2010 ; 42 : 1027 - 30 .
49. Wang XJ , Gaasterland T , Chua NH . Genome-wide prediction and identification of cis-natural antisense transcripts in Arabidopsis thaliana . Genome Biol . 2005 ; 6 : R30 .
50. Borsani O , Zhu J , Verslues PE , Sunkar R , Zhu JK . Endogenous siRNAs derived from a pair of natural cis-antisense transcripts regulate salt tolerance in Arabidopsis . Cell. 2005 ; 123 : 1279 - 91 .
51. Katiyar-Agarwal S , Morgan R , Dahlbeck D , Borsani O , Villegas Jr A , Zhu JK , et al. A pathogen-inducible endogenous siRNA in plant immunity . Proc Natl Acad Sci U S A . 2006 ; 103 : 18002 - 7 .
52. Chen S , Krinsky BH , Long M. New genes as drivers of phenotypic evolution . Nat Rev Genet . 2013 ; 14 : 645 - 60 .
53. Ellis J , Dodds P , Pryor T. The generation of plant disease resistance gene specificities . Trends Plant Sci . 2000 ; 5 : 373 - 9 .
54. Noel L , Moores TL , van Der Biezen EA , Parniske M , Daniels MJ , Parker JE , et al. Pronounced intraspecific haplotype divergence at the RPP5 complex disease resistance locus of Arabidopsis . Plant Cell . 1999 ; 11 : 2099 - 112 .
55. Parniske M , Hammond-Kosack KE , Golstein C , Thomas CM , Jones DA , Harrison K , et al. Novel disease resistance specificities result from sequence exchange between tandemly repeated genes at the Cf-4/9 locus of tomato . Cell . 1997 ; 91 : 821 - 32 .
56. Hu Z , Jiang QY , Ni ZY , Zhang H. Prediction and identification of natural antisense transcripts and their small RNAs in soybean (Glycine max) . BMC Genomics . 2013 ; 14 : 230 .
57. Swiezewski S , Liu F , Magusin A , Dean C. Cold-induced silencing by long antisense transcripts of an Arabidopsis Polycomb target . Nature . 2009 ; 462 : 799 - 802 .
58. Zipfel C. Plant pattern-recognition receptors . Trends Immunol . 2014 ; 35 : 345 - 51 .
59. Hayafune M , Berisio R , Marchetti R , Silipo A , Kayama M , Desaki Y , et al. Chitin-induced activation of immune signaling by the rice receptor CEBiP relies on a unique sandwich-type dimerization . Proc Natl Acad Sci U S A . 2014 ; 111 : E404 - 413 .
60. Yamaguchi Y , Huffaker A , Bryan AC , Tax FE , Ryan CA. PEPR2 is a second receptor for the Pep1 and Pep2 peptides and contributes to defense responses in Arabidopsis . Plant Cell . 2010 ; 22 : 508 - 22 .
61. Postel S , Kufner I , Beuter C , Mazzotta S , Schwedt A , Borlotti A , et al. The multifunctional leucine-rich repeat receptor kinase BAK1 is implicated in Arabidopsis development and immunity . Eur J Cell Biol . 2010 ; 89 : 169 - 74 .
62. Verica JA , He ZH . The cell wall-associated kinase (WAK) and WAK-like kinase gene family . Plant Physiol . 2002 ; 129 : 455 - 9 .
63. Shiu SH , Karlowski WM , Pan R , Tzeng YH , Mayer KF , Li WH . Comparative analysis of the receptor-like kinase family in Arabidopsis and rice . Plant Cell . 2004 ; 16 : 1220 - 34 .
64. Meyers BC , Kaushik S , Nandety RS. Evolving disease resistance genes . Curr Opin Plant Biol . 2005 ; 8 : 129 - 34 .
65. Ellis J , Dodds P , Pryor T. Structure , function and evolution of plant disease resistance genes . Curr Opin Plant Biol . 2000 ; 3 : 278 - 84 .
66. Dangl JL , Jones JD . Plant pathogens and integrated defence responses to infection . Nature . 2001 ; 411 : 826 - 33 .
67. Song J , Bradeen JM , Naess SK , Raasch JA , Wielgus SM , Haberlach GT , et al. Gene RB cloned from Solanum bulbocastanum confers broad spectrum resistance to potato late blight . Proc Natl Acad Sci U S A . 2003 ; 100 : 9128 - 33 .
68. Weng K , Li Z-Q , Liu R-Q , Wang L , Wang Y-J , Xu Y. Transcriptome of Erysiphe necator-infected Vitis pseudoreticulata leaves provides insight into grapevine resistance to powdery mildew . Horti Res . 2014 ; 1 : 14049 .
69. Davis KR , Hahlbrock K. Induction of defense responses in cultured parsley cells by plant cell wall fragments . Plant Physiol . 1987 ; 84 : 1286 - 90 .
70. Gao F , Dai R , Pike SM , Qiu W , Gassmann W. Functions of EDS1-like and PAD4 genes in grapevine defenses against powdery mildew . Plant Mol Biol . 2014 ; 86 : 381 - 93 .
71. Singh K , Foley RC , Onate-Sanchez L. Transcription factors in plant defense and stress responses . Curr Opin Plant Biol . 2002 ; 5 : 430 - 6 .
72. Gu YQ , Yang C , Thara VK , Zhou J , Martin GB . Pti4 is induced by ethylene and salicylic acid, and its product is phosphorylated by the Pto kinase . Plant Cell . 2000 ; 12 : 771 - 85 .
73. McGrath KC , Dombrecht B , Manners JM , Schenk PM , Edgar CI , Maclean DJ , et al. Repressor- and activator-type ethylene response factors functioning in jasmonate signaling and disease resistance identified via a genome-wide screen of Arabidopsis transcription factor gene expression . Plant Physiol . 2005 ; 139 : 949 - 59 .
74. Zhong S , Joung JG , Zheng Y , Chen YR , Liu B , Shao Y , et al. High-throughput Illumina strand-specific RNA sequencing library preparation . Cold Spring Harb Protoc . 2011 ; 2011 : 940 - 9 .
75. Bolger AM , Lohse M , Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data . Bioinformatics . 2014 ; 30 : 2114 - 20 .
76. Quast C , Pruesse E , Yilmaz P , Gerken J , Schweer T , Yarza P , et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools . Nucleic Acids Res . 2013 ; 41 : D590 - 596 .
77. Langmead B , Trapnell C , Pop M , Salzberg SL . Ultrafast and memory-efficient alignment of short DNA sequences to the human genome . Genome Biol . 2009 ; 10 :R25.
78. Grabherr MG , Haas BJ , Yassour M , Levin JZ , Thompson DA , Amit I , et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome . Nat Biotechnol . 2011 ; 29 : 644 - 52 .
79. Benson DA , Cavanaugh M , Clark K , Karsch-Mizrachi I , Lipman DJ , Ostell J , et al. GenBank. Nucleic Acids Res . 2013 ; 41 : D36 - 42 .
80. Zheng Y , Zhao L , Gao J , Fei Z. iAssembler: a package for de novo assembly of Roche-454/Sanger transcriptome sequences . BMC Bioinformatics . 2011 ; 12 : 453 .
81. Seqclean program . http://sourceforge.net/projects/seqclean/files/
82. Jaillon O , Aury JM , Noel B , Policriti A , Clepet C , Casagrande A , et al. The grapevine genome sequence suggests ancestral hexaploidization in major angiosperm phyla . Nature . 2007 ; 449 : 463 - 7 .
83. Kent WJ . BLAT - The BLAST-like alignment tool . Genome Res . 2002 ; 12 : 656 - 64 .
84. Barrell D , Dimmer E , Huntley RP , Binns D , O'Donovan C , Apweiler R. The GOA database in 2009-an integrated Gene Ontology Annotation resource . Nucleic Acids Res . 2009 ; 37 : D396 - 403 .
85. Boyle EI , Weng S , Gollub J , Jin H , Botstein D , Cherry JM , et al. GO:: TermFinder-open source software for accessing Gene Ontology information and finding significantly enriched Gene Ontology terms associated with a list of genes . Bioinformatics . 2004 ; 20 : 3710 - 5 .
86. Finn RD , Bateman A , Clements J , Coggill P , Eberhardt RY , Eddy SR , et al. Pfam: the protein families database . Nucleic Acids Res . 2014 ; 42 : D222 - 230 .
87. Eddy SR . Accelerated profile HMM searches . PLoS Comput Biol . 2011 ; 7 : e1002195 .
88. Li H , Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform . Bioinformatics . 2009 ; 25 : 1754 - 60 .
89. Li H , Handsaker B , Wysoker A , Fennell T , Ruan J , Homer N , et al. Genome Project Data Processing S: The sequence alignment/map format and SAMtools . Bioinformatics . 2009 ; 25 : 2078 - 9 .