a article accepted - genedenovo.com · systematic identification and analysis of...
TRANSCRIPT
Systematic identification and analysis of heat-stress-responsive lncRNAs, circRNAs and
miRNAs with associated co-expression and ceRNA networks in cucumber (Cucumis sativus L.)
Xueying Hea, Shirong Guo
a, Ying Wang
a, Liwei Wang
a, Sheng Shu
a and Jin Sun
a,b,*
a College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China
b Nanjing Agricultural University (Suqian) Academy of Protected Horticulture, Suqian 223800, China
Correspondence
*Corresponding author,
e-mail: [email protected]
Researchers have shown that long noncoding RNAs (lncRNAs) and circular RNAs (circRNAs) act as
competitive endogenous RNAs (ceRNAs) and are mutually regulated by competition for binding to
common microRNA response elements (MREs). However, a comprehensive identification and
analysis of lncRNAs and circRNAs as ceRNAs have not yet been completed in cucumber (Cucumis
sativus L.) exposed to high-temperature stress. In our study, 32 663 coding transcripts, 2085 lncRNAs,
2477 circRNAs and 348 differentially expressed miRNAs were identified using RNA sequencing. In
addition, six heat-stress-responsive miRNAs (five known and one novel miRNAs) and eight lncRNAs
were selected for qPCR to confirm their expression profiles. By analyzing the cis effects of lncRNAs,
we constructed a lncRNA-mRNA co-expression network. Based on the results, the corresponding
lncRNAs play a regulatory role in the stress response in cucumber plants. In our study, the PatMatch
software was used to predict the potential function of lncRNAs and circRNAs as ceRNAs. A total of
18 lncRNAs and seven circRNAs were predicted to bind to 114 differentially expressed miRNAs and
compete with 359 mRNAs for miRNA binding sites. These mRNAs are predicted to be involved in
various pathways, such as plant hormone signal transduction, plant-pathogen interaction and
glutathione metabolism. Among them, TCONS_00031790, TCONS_00014332, TCONS_00014717,
TCONS_00005674, novel_circ_001543 and novel_circ_000876 may interact with miR9748 by plant
hormone signal transduction pathways in response to high-temperature stress. Moreover,
indole-3-acetic acid (IAA) and 1-aminocyclopropane-l-carboxylic acid (ACC) levels decreased in the
high-temperature treatment group, indicating that IAA and ethylene signaling might be involved in
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/ppl.12997
2
response to high-temperature stress. In this study, we conducted a full transcriptomic analysis in
response to high-temperature stress in cucumber and, for the first time, integrated the potential ceRNA
functions of lncRNAs/circRNAs. The results provide a basis for studying the potential functions of
lncRNAs/circRNAs in response to high-temperature stress.
Introduction
High temperature is an important factor that limits plant growth and productivity. Heat stress (high
temperature) impedes intracellular homeostasis and can lead to leaf lesions, severe delays in growth
and development, risk of disease and even death (Bita and Gerats 2013, Liu et al. 2013). Studies have
shown that under high-temperature stress, the bioaccumulation of cucumber seedling leaves is
significantly inhibited, and the chlorophyll concentration is reduced (Zhou et al. 2016). As a result,
plants have developed specific adaptation mechanisms to address high-temperature stress.
A large number of studies have confirmed that miRNA and long noncoding RNA (lncRNA) play an
important role in plant stress, and their regulatory mechanisms have been revealed in plants (Xin et al.
2011, Zhu and Wang 2012). Under the induction of high temperature, miRNA affects and participates
in the process of plant growth by interacting with multiple genes. Genes respond to high-temperature
stress via changes in expression levels. Wang identified the first set of miRNAs associated with the
exogenous Spd-mediated improvement of high-temperature tolerance in cucumber seedlings (Wang et
al. 2018). For example, 34 specific lncRNAs were identified under heat stress and 192 target genes
were regulated by lncRNAs, most of which were thermos-responsive genes (Song et al. 2016a). Heat
stress does not induce lnc-173 expression, although its target gene SUCROSE SYNTHASE4, responds
to high temperatures (Di et al. 2014). In Populus simonii, the expression level of
PsiLncRNA00268512 is dynamic in response to heat stress (Song et al. 2016b). In addition to
participating in high-temperature stresses, lncRNAs are also involved in a series of physiological and
biochemical metabolic pathways during plant growth and development. In rice RNA-Seq data, most of
the 2063 lncRNAs were preferentially expressed during rice reproduction, and some lncRNAs also
induced rice reproductive defects (Zhang and Chen 2013). CSM10-lncRNA is differentially expressed
in different tissues, different developmental stages and different photoperiods of cucumber and may be
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
3
involved in the regulation of growth and development of these tissues (Cho et al. 2005). However, it is
unclear whether lncRNA plays a specific physiological role in plants. Although some work has been
done on the roles of lncRNAs in plants, little is known about the functions of heat-stress-responsive
lncRNAs in cucumbers.
At present, an increasing number of studies have shown that circular RNA (circRNA) is involved in
the regulation of plant response to various biotic and abiotic stresses. Wang et al. (2017a) identified 88
circRNAs in wheat, of which 62 were differentially expressed under water stress, indicating that these
circRNAs may play a role in response to water stress. In tomato, Zuo et al. (2016) identified 854
circRNAs, of which 163 circRNAs were differentially expressed in tomato fruits under control and
chilling treatments, and 102 circRNAs could be combined as molecular sponges with 24 miRNAs.
Plant circRNAs showed different expression patterns, and 27 rice exonic circRNAs were found to be
differentially expressed under phosphate-sufficient and -starvation conditions (Ye et al. 2015).
Overexpression of Vv-circATS1, a circRNA derived from glycerol-3-P acyltransferase (ATS1),
improved cold tolerance in Arabidopsis, while the linear RNA derived from the same sequence is not
able to do the same (Gao et al. 2019). Recently, transcriptome analyses of plant drought response and
transgenic studies in Arabidopsis thaliana have revealed a relationship between circRNA expression
and drought resistance, indicating that circRNAs play a key role in plant drought resistance responses
and can also be used as effective biomarkers for genetic improvement of crop drought resistance
(Zhang et al. 2019). Heat-induced circRNAs might participate in plant response to heat stress through
circRNA-mediated competitive endogenous RNA (ceRNA) networks (Pan et al. 2018). In addition to
participating in biotic and abiotic stresses, circRNAs are also involved in a series of physiological and
biochemical metabolic pathways during plant growth and development. Recently, Wang used
'Zhongcai No. 4' and LeERF1 transgenic tomato plants to verify the possible role of circRNA in
regulating ethylene metabolism-related pathways in tomato fruits. The results showed that 102 target
mRNAs of 39 circRNAs were involved in the pathways involved in ethylene synthesis and signal
transduction (Wang et al. 2017b). The research showed that circRNA is involved in the regulation of
pigment accumulation during tomato fruit ripening and overexpression of circRNA related to pigment
synthesis can significantly reduce the expression level of the parental gene and the color of the
response (Tan et al. 2017). The research has found that circular RNA might play an important role in
development of moso bamboo by regulating the splicing of several rapid-growth related genes (Wang
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
4
et al. 2019). It has been reported that 53 differentially expressed circRNAs were predicted as the
corresponding nine miRNAs sponges and play a role in sea buckthorn fruit ripening process (Zhang et
al. 2019). Similar research found that circRNAs play an important role in the regulation and control of
tomato fruits (Yin et al. 2018). CircRNA is widely distributed in plants and has been identified in
several plants such as Arabidopsis, rice, tomato and soybean using deep RNA-seq and bioinformatics
tools (Lu et al. 2015, Chen et al. 2017, Conn et al. 2017, Tan et al. 2017, Zhao et al. 2017, Zhou et al.
2018). However, at present, comprehensive studies of circRNA in plants are lacking, and the functions
of circRNA have not yet been clarified.
RNAs have been reported to regulate each other via competition in combination with common
microRNA response elements (MREs), which constitute a ceRNA (Salmena et al. 2011). The theory of
competing endogenous RNAs has been demonstrated and is now widely accepted (Salmena et al. 2011,
Ala et al.2013, Xu et al. 2016), and it includes protein-coding RNA and noncoding RNA, such as
pseudogene transcripts, lncRNA, and circRNA. The first example of ceRNA was found in Arabidopsis,
and the study showed that the noncoding RNA IPS1 could influence the expression level of PHO2 by
binding to miR399 (Franco-Zorrilla et al. 2007). Recent studies have also shown that circRNAs could
act as ceRNAs. For example, two circRNAs, ciRs-7/CDR1 and Sry, have been reported to be miRNA
sponges in humans (Hansen et al. 2013a, 2013b, Sebastian 2013). In addition, circRNAs, as ceRNAs,
are involved in the complex regulation of ethylene in tomato fruits (Wang et al. 2017b). Competition
of lncRNAs with other miRNA sponges has an important role in plants and animals (Franco-Zorrilla et
al. 2007, Sumazin et al. 2011, Wang et al. 2013, Wu et al. 2013, Zhang et al. 2014). Many studies have
demonstrated that lncRNAs play an important role as a competitive platform for miRNAs and mRNAs
in pathological and physiologically relevant processes (Tay et al. 2014). However, to the best of our
knowledge, the ceRNA network has not been used to study the function of lncRNAs/circRNAs in
cucumber.
Based on the hypothesis that lncRNAs/circRNAs compete with genes to play an important role in
cucumbers undergoing heat stress, we used the ceRNA network to study the functions of these
lncRNAs/circRNAs. Using high-throughput sequencing and bioinformatics analysis, we constructed
ceRNA networks of lncRNAs, circRNAs, miRNAs and mRNAs and analyzed their potential
regulatory roles with GO and KEGG. In this study, we conducted a transcriptome analysis of
cucumbers in response to high-temperature stress and integrated, for the first time, the potential
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
5
ceRNA functions of lncRNAs/circRNAs in the process of cucumber hyperthermia stress, which
provided a basis for studying the potential functions of lncRNAs/circRNAs in response to
high-temperature stress.
Materials and methods
Plant material and growth conditions
The Cucumis sativus cultivar ‘Improved Jinchun 2’ was used in this study. The surface-sterilized seeds
were grown in pots containing nursery substrates in a controlled-environment growth chamber
programmed for 12/12 h at 28/18°C for day/night. Seedlings at the three-leaf stage were transferred to
growth chambers set to 42/32°C as the high-temperature treatment and 28/18°C as the control for 7 d.
Each treated sample was obtained by homogeneously mixing the completely expanded third leaves of
eight seedlings for two biological replicates (Li et al. 2014). All leaf samples were collected from
control and treated plants, frozen in liquid nitrogen and stored at -80°C for RNA extraction.
Strand-specific library construction and sequencing
After total RNA was extracted, rRNAs were removed to retain mRNAs and ncRNAs. The enriched
mRNAs and ncRNAs were fragmented into short fragments using fragmentation buffer, and reverse
transcribed into cDNA with random primers. Second-strand cDNA was synthesized by DNA
polymerase I, RNase H, dNTP (dUTP instead of dTTP) and the appropriate buffer. Next, the cDNA
fragments were purified with a QiaQuick PCR extraction kit (Qiagen), and they were then end
repaired, supplemented with poly(A), and ligated to Illumina sequencing adapters. Then, UNG
(Uracil-N-Glycosylase) was used to digest the second-strand cDNA. The digested products were size
selected by agarose gel electrophoresis, PCR amplified, and sequenced using an Illumina HiSeqTM
2500 (Gene Denovo Biotechnology Co.). Raw reads were processed by deleting adapter reads and
low-quality labels, with all subsequent analyses performed using clean reads.
Small RNA libraries construction and sequencing
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
6
Total RNA was extracted from three replicates of each treatment with TRIzol reagent (Invitrogen).
High-purity (OD260/280 between 1.8 and 2.2) and high-integrity (RNA integrity number, RIN ≥ 7.5)
RNA samples from the CW (control treated with water) and HW (high temperature treated with water)
groups were selected to construct the sRNA libraries (CW and HW). Then, high-throughput
sequencing was performed on a HiSeq 2000 instrument (Illumina). The data were processed by the
following steps: (1) Total RNA was extracted from the samples via PAGE gel separation of RNA
segments of different size. A stripe between 18 and 30 nt was cut out and small RNAs were recovered.
(2) A 3' connection system was constructed, blending and centrifugation with 5000 rpm were
performed. Then 3' adaptors were ligated to the small RNAs based on a suitable temperature within a
specific period of time. (3) The same steps as in (2) were done for the 5' adaptors. (4) On a PCR
machine, the adaptor-ligated products were reverse-transcribed into double-stranded sequences, and
these double-stranded sequences were then PCR-amplified according to certain procedures, which
were: denaturation at 98 °C for 30s, followed by 6-16 cycles at 98°C for 10s, 65 °C for 30s, and 72 °C
for 30s, and a final extension at 72 °C for 5 min. (5) PAGE gel recycling and PCR product purification
were performed, and the product was placed in EB solution (10 mM Tris-HCL, pH of 8.0). (6) An
Agilent 2100 Bioanalyzer (Agilent Technologies) and ABI Step One Plus Real-Time PCR System
(Applied Biosystems) were used to determine the quality and yield of the RNA libraries.
Identification of differentially expressed lncRNA, circRNA and mRNA
Using the TopHat version 2.0.9, clean reads from both cDNA libraries were mapped to version 2 of the
cucumber genome sequence (http://www.icugi.org/cgi-bin/ICuGI/genome/home.cgi?ver=2&organism
= cucumber & cultivar = Chinese-long; Daehwan 2013). The known mRNAs were identified
according to the cucumber genomic sequence annotation. The protein coding potential of the new
transcripts was assessed using CNCI (version 2) and CPC (http://cpc.cbi.pku.edu.cn/; Kong et al. 2007,
Sun et al. 2013). The intersection of the two results was selected as the lncRNA. Cuffdiff was used to
calculate the fragments per kilobase of exon per million mapped reads (FPKM) scores for transcripts
in each library (Trapnell et al. 2012). Differentially expressed lncRNAs and mRNAs between the two
libraries were identified by edgeR (Robinson et al. 2010).
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
7
Then, 20-mers from both ends of the unmapped reads were extracted and aligned to the reference
genome to find unique anchor positions within the splice site. Anchor reads that aligned in the reverse
orientation (head to tail) indicated circRNA splicing and were subjected to find_circ to identify
circRNAs (Sebastian 2013). The anchor alignments were then extended such that the complete read
alignments and the breakpoints were flanked by GU/AG splice sites. A candidate circRNA was
identified if it was supported by at least two unique back-spliced reads at least in one sample.
To quantify circRNAs, back-spliced junction reads were scaled to RPM (reads per million mapped
reads), and the formula is shown as follows:
Eqn. 1
where C is the number of back-spliced junction reads that uniquely aligned to a circRNA, and N is the
total number of back-spliced junction reads. The RPM method is able to eliminate the influence of
different amounts of sequencing data on the calculation of circRNA expression. Therefore, the
calculated expression can be directly used to compare differential expression among samples.
We used FDR < 0.05 and | log2(FC) | > 1 as a threshold for assessing significantly differentially
expressed lncRNAs, circRNAs and mRNAs. Then, a bioinformatics analysis of candidate lncRNAs,
circRNAs and mRNAs was performed.
Screening of miRNAs responsive to high-temperature stress
The expression levels of known miRNAs and novel miRNAs in the two libraries were calculated and
normalized as transcripts per million according to the following formula: normalized expression =
actual miRNA count/total count of clean reads × 1 000 000. The expression of miRNAs with an
abundance of zero was modified to 0.01 for further analysis (Murakami et al. 2006). Then, the
normalized results were used to calculate the fold change and P-value. To avoid errors, miRNAs only
expressed in one library were removed and not involved in the differential expression analysis.
We analyzed the expression of the miRNAs in the two libraries (CW and HW). Then, these libraries
were compared pairwise to find the differentially expressed miRNAs. The differential expression of
miRNAs was calculated by the following formula: fold change = log2 (HW/CW). A miRNA was
considered to be differentially expressed between the two compared libraries in each comparison pair
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
8
when | fold change | = | (log2 (HW/CW) | > 1 and P-value < 0.05. The p-value was calculated as
previously described (Audic and Claverie 1997).
Target gene prediction and functional enrichment analysis
The cist-acting lncRNAs targeted neighboring genes (Ponjavic et al. 2009, Orom et al. 2010). We
searched for 10 kb of coding genes upstream and downstream of all identified lncRNAs and predicted
their function. The miRNA-mRNA, miRNA-lncRNA and miRNA-circRNA target genes were
predicted by patmatch_v1.2 software using small RNA sequencing and RNA-seq data (Yan et al.
2005).
All differentially expressed mRNAs were studied using GO annotation and KEGG pathway analyses
as described previously (Zhao et al. 2015, Chen et al. 2016). GO terms enrichment was determined
with Blast2GO by reference to the GO database (Conesa et al. 2005). At the same time, a KEGG
pathway analysis was performed with reference to the KEGG pathway database.
Network visualization
The ceRNA network was constructed based on ceRNA theory as follows: (1) The correlations in
expression between mRNA and miRNA or lncRNA and miRNA were evaluated using the Spearman
Rank correlation coefficient (SCC). Pairs with SCC < -0.7 were selected as negatively co-expressed
lncRNA–miRNA pairs or mRNA-miRNA pairs, where both mRNA and lncRNA were miRNA target
genes, and all RNAs were differentially expressed. (2) The correlation in expression between lncRNA
and mRNA was evaluated using the Pearson correlation coefficient (PCC). Pairs with PCC > 0.9 were
selected as co-expressed lncRNA–mRNA pairs, where both the mRNA and lncRNA in each pair were
targeted and co-expressed negatively with a common miRNA. As a result, only the gene pairs with a
P-value less than 0.05 were selected.
𝑝 − 𝑣𝑎𝑙𝑢𝑒 1 − 𝐹(𝑥/𝑈, ,𝑁) Eqn. 2
𝑝 − 𝑣𝑎𝑙𝑢𝑒 1 − ∑(𝑀𝑖 )(
𝑈−𝑀𝑁−𝑖 )
(𝑈𝑁)
𝑥− 𝑖= Eqn. 3
For a given gene pair (A,B), we denoted all their regulatory miRNAs as miRNA sets C (regulating
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
9
gene A) and D (regulating gene B). In the above equations, x stands for the number of common
miRNAs that regulate both genes, U is the total number of miRNAs in this work, M is the size of
miRNA set C, and N is the size of miRNA set D.
The lncRNA-mRNA co-expression and ceRNA regulatory network was constructed by assembling all
co-expression competing triplets, which were identified above, and visualized using Cytoscape 3.3.0
software (The Cytoscape Consortium, USA). Nodes in the ceRNA network include miRNAs, mRNAs,
lncRNAs, and circRNAs.
Quantitative real-time PCR (qRT-PCR)
We performed qRT-PCR to confirm the quality of the high-throughput sequencing and the expression
patterns of miRNA and lncRNA. The main steps for the verification of miRNAs were as follows.
Small RNAs were extracted from the leaves of plants in the CK and HT groups using the miRcute
miRNA isolation kit (Tiangen). A Mir-X miRNA First-Strand Synthesis Kit (TaKaRa) was used for
first-strand cDNA synthesis from miRNAs. U6 snRNA served as the internal control for the miRNA
expression analysis (Li et al. 2014). qRT-PCR was performed with a SYBR PrimeScriptTM
RT-PCR
Kit (TaKaRa) on a Step One™ Real-time PCR System (Applied Biosystems) based on the
manufacturer’s instructions.
The main steps for the validation of lncRNAs were performed as follows. Total RNAs were extracted
from the leaves of two treated samples using an RNA simple Total RNA kit (Tiangen). Then, the total
RNA was used for first-strand cDNA synthesis using a PrimeScript™ II First Strand cDNA Synthesis
Kit (TaKaRa). Finally, real-time PCR was performed on a Step One ™ Real-time PCR System
(Applied Biosystems) using a SYBR PrimeScriptTM
RT-PCR Kit (TaKaRa). The lncRNA primers were
designed by Beacon Designer 7.9. The cucumber actin gene was used as an internal reference to
normalize the qRT-PCR data.
The primers used are shown in Table S1. The relative expression level (fold change) was expressed as
2-ΔΔCt
. The data were statistically analyzed using SAS Version 9.0 software (SAS Institute). The
significance level was set to a P-value < 0.05. All qRT-PCRs were performed in triplicates.
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
10
Measurement of endogenous hormone contents
Approximately 500 mg of frozen leaf samples from the control and high temperature treatment were
used for endogenous phytohormone extraction. To link the RNA sequencing results to endogenous
hormone contents, the examined leaves were subjected to the same treatment used for sequencing. The
auxin indole-3-acetic acid (IAA) and the direct biosynthetic precursor of ethylene,
1-aminocyclopropane-l-carboxylic acid (ACC), contents were measured by previously described
methods (Dobrev and Vankova 2012). The hormones were isolated based on a previously published
protocol (Ağar et al. 2006). High-performance liquid chromatography-mass spectrometry (AB 5500)
was used to detect and quantify the hormones following a previously reported protocol (Pan et al.
2010). Standard IAA samples were purchased from Sigma-Aldrich, and standard ACC samples were
purchased from J&K Scientific. The results were analyzed using three replicates.
Results
Identification of lncRNAs and mRNAs responsive to high-temperature stress and their function
analysis
Identification of differentially expressed lncRNAs and mRNAs
To identify the lncRNAs involved in response to high-temperature stress in cucumber leaves, we
constructed four cDNA libraries (HT-1, HT-2, CK-1, and CK-2) and sequenced the libraries using the
Illumina HiSeq ™ 2500 platform (Fig. 2A). A total of 63 966 532 800 raw reads were generated for all
four libraries. After discarding the adaptor sequences and low-quality reads, we obtained
62 219 581 464 clean reads (Table 1). After removing the rRNA genes, the clean reads were mapped to
version 2 of the cucumber genome sequence. The percentage of clean reads in each library ranged
from 76.91-79.80% (Table 2). The mapping sequences in each library were assembled, and a total of
25 600 unique assembly transcripts were obtained. The mapping rate was low, which can be explained
by the following factors: (1) the V2 version of the cucumber reference genome sequence has only the
nuclear genome and does not contain the chloroplast genome; and (2) sampling cucumber leaves
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
11
yields a relatively high chloroplast content.
The expression levels of lncRNAs and mRNAs were estimated by the FPKM value using Cuffdiff. We
identified a total of 32 663 mRNAs and 2085 lncRNAs. Compared with the control group, 108
lncRNAs and 2130 mRNAs were differentially expressed in the leaves of cucumber treated with high
temperature. Fifty-six lncRNAs and 1533 mRNAs were upregulated, while 52 lncRNAs and 597
mRNAs were downregulated (Fig. 1A and Table S2). The apparent variations in lncRNAs and mRNAs
between the two groups are visually displayed with heatmaps (Fig. 1B, C). During the
high-temperature treatment, the number of upregulated mRNAs was greater than that of
downregulated mRNAs, while the upregulated and downregulated lncRNAs presented similar
numbers.
Next, the identified lncRNAs were compared to the genomic characteristics of the protein-coding
genes in cucumber. The average exon length of the mRNAs was longer than that of the lncRNAs (Fig.
2B). Most lncRNAs contained only one exon, whereas over 90% of the mRNAs had multiple exons
(Fig. 2C). Meanwhile, the lncRNAs in cucumbers had fewer and shorter exons than the mRNAs. The
GC content of the lncRNAs was also lower than that of the mRNAs (Fig. 2D).
GO/KEGG pathway analysis of differentially expressed mRNAs
Differentially expressed mRNAs were significantly enriched in 18 GO terms under biological process,
13 GO terms under cellular component and 10 GO terms under molecular function in the
high-temperature treatment of cucumber leaves (Fig. 3A). In our study, we found that upregulated and
downregulated mRNAs were most abundant in GO terms such as metabolic process, cellular process,
single-organism process and catalytic activity. In addition, the GO analysis indicated that these
differentially expressed mRNAs may be involved in many biological processes (including response to
organic substance, response to endogenous stimulus, response to stimulus, biological regulation,
regulation of biological process, response to stress, etc.). A bioinformatics analysis was performed to
predict the potential function of lncRNA by a GO and KEGG enrichment analysis of mRNA, and the
results showed that some lncRNAs may regulate the mRNA response to high-temperature stress (GO:
0050896 response to stimulus, GO: 0006950 response to stress, GO: 0009408 response to heat).
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
12
The differentially expressed mRNAs were enriched in 114 KEGG pathways, and the results showed
that three significant enrichment pathways were photosynthesis-antenna proteins, glutathione
metabolism and photosynthesis. Following these three groups, mRNAs were enriched in histidine
metabolism, brassinosteroid biosynthesis, phenylpropanoid biosynthesis, plant hormone signal
transduction and protein processing in the endoplasmic reticulum (Fig. 3B).
Co-expression analysis of lncRNAs and mRNAs and functional prediction
One of the functions of lncRNAs is cis-regulation of their neighboring genes on the same allele,
thereby regulating transcriptional or posttranscriptional gene expression. Because cis-acting lncRNAs
target neighboring genes (Ponjavic et al. 2009, Orom et al. 2010), we searched for 10 kb of coding
genes upstream and downstream of all identified lncRNAs and predicted their functions. By analyzing
the cis-regulation of lncRNAs, we constructed a co-expression network of lncRNAs and mRNAs.
Most mRNAs and lncRNAs were one-to-one matches. However, there were also one-to-many matches
between the lncRNAs and mRNAs (Table S3). Considering that graphics cannot display the enormous
amount of network information between lncRNAs and mRNAs, we selected more mRNAs for
co-expression with lncRNAs to make the network diagram (Fig. 4). As shown in Fig. 4, one lncRNA
interacted with nine mRNAs, two different lncRNAs interacted with seven mRNAs, and two different
lncRNAs interacted with six mRNAs.
LncRNAs located upstream of a protein-coding gene may overlap with a promoter region or
cis-regulate the element and may regulate the expression of genes in their vicinity at the transcriptional
or posttranscriptional level. LncRNAs located downstream of protein-coding genes can initiate
transcription from 3'UTRs or downstream regions and may be involved in intergenic regulatory
interactions. To predict the functions of lncRNAs, the upstream and downstream genes of lncRNAs
were analyzed. GO and KEGG enrichment analyses of lncRNAs and their upstream and downstream
differentially expressed genes were performed. The results showed that these genes were enriched in
the photosystem, response to endogenous stimulus, defense response, signal transduction and
regulation of biological process GO terms (Table S4). Interestingly, we found that protein-coding
genes, such as Csa4M314390.1 (ERF) and Csa5M613470.1 (MYB_related), were involved in
response to stimulus (GO:0050896), and these results showed the possible role of lncRNA in
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
13
transcriptional regulation of gene expression. The KEGG enrichment analysis showed that these genes
were enriched in photosynthesis-antenna proteins, glutathione metabolism, photosynthesis, histidine
metabolism and other pathways, indicating that high-temperature stress had a great impact on plant
photosynthesis (Table S4).
Identification of circRNAs responsive to high-temperature stress and functional analysis
To identify circRNAs and analyze their functions, high-throughput sequencing was performed in the
control and high-temperature treatment cucumber leaves using the Illumina HiSeq ™ 2500 platform.
After screening, we found a total of 2477 novel circRNAs (Table S5). Cuffdiff was used to assess their
expression level by FPKM, and the results showed that in the high-temperature treatment group, there
were five upregulated and one downregulated circRNA. To further understand the potential functions
of circRNAs, GO and KEGG analyses of the source genes of circRNAs were performed. The results
showed that the circRNAs were enriched in 44 GO terms (18 GO terms under biological process, 15
GO terms under cellular component and 11 GO terms under molecular function), suggesting that these
circRNAs may be involved in the regulation of many biological processes (including growth, response
to stimulus, biological regulation, and signaling). Thus, it was predicted that circRNAs may regulate
gene response to high-temperature stress (GO: 0050896 response stimulus; Table S6). The source
genes of circRNAs were involved in 112 KEGG pathways and significantly enriched in four KEGG
pathways, photosynthesis-antenna proteins, carbon metabolism, glyoxylate and dicarboxylate
metabolism and photosynthesis (Table S6).
Identification of miRNAs responsive to high-temperature stress and functional analysis
To explore the expression patterns of small RNAs (sRNAs) in response to high temperature, two RNA
libraries were constructed, i.e., CW and HW, and these two libraries generated 11 255 470 and
11 296 147 raw reads, respectively (Table 3). Of these raw reads, 10 865 804 and 10 854 344 were
retained after contaminants and low-mass sequences were removed. Among the HW/CW comparison
pairs, there were 19 428 531 consensus sequences that accounted for 89.45% of the total reads, and
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
14
these sequences were classified as 182 874 species, which accounting for 11.52%. In this comparison
pair, 546 579 sequences were HW-specific and 857 319 were CW-specific. To find miRNAs that
responded to high-temperature stress and identify their expression patterns, we compared the HW and
CW libraries. One hundred fifteen differentially expressed known miRNAs and 233 differentially
expressed novel miRNAs were obtained in the HW/CW comparison (Table S7).
To further understand the potential functions of miRNAs, we conducted GO and KEGG analyses of
the target genes of miRNAs. The results showed that all target genes were successfully assigned to the
corresponding 32 GO terms. The main subcategories were catalytic activity (GO: 0003824), binding
(GO: 0005488), metabolic process (GO: 0044710), and cellular process (GO: 0009987; Table S8). The
target genes of the miRNAs were involved in 30 KEGG pathways, which were significantly enriched
in the four KEGG pathways, including ribosome biogenesis in eukaryotes (ko03008), RNA
degradation (ko03018), RNA transport (ko03013), the mRNA surveillance pathway (ko03015) and
other metabolites. However, the p-values of the other pathways were greater than 0.05, indicating no
statistically significant difference (Table S8).
Construction of ceRNA network
At the transcriptome level, the mechanism by which ncRNAs regulate gene expression is revealed
through the ceRNA regulatory network. Based on the theory of ceRNA, lncRNAs, circRNAs and
mRNAs with the same miRNA binding sites were searched, and miRNAs were used as the core and
lncRNAs/circRNAs and mRNAs as target ceRNA regulatory networks. Thus, the ceRNA network was
constructed by integrating the expression profiles and regulatory relationships of mRNA, lncRNA,
circRNA and miRNA from RNA-seq and small RNA sequencing data. We found that 114 differentially
expressed miRNAs were associated with 359 mRNAs as well as eighteen lncRNAs and seven
circRNAs. Considering that graphics cannot display the enormous amount of network information
between miRNAs, lncRNAs, circRNAs and mRNAs, we selected miRNAs associated with more
mRNAs to make the ceRNA network diagram (Fig. 5). The ceRNA regulatory network contains 433
lncRNA-miRNAs, circRNA-miRNAs or mRNA-miRNAs pairs with 253 negative correlations and
180 positive correlations (Table S9).
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
15
To reveal their potential function, we performed GO and KEGG pathway analyses of 359 differentially
expressed mRNAs to predict the potential regulatory role of lncRNAs/circRNAs in the ceRNA
network (Fig. 6 and Table S10). As shown in Fig. 6A, all 359 differentially expressed mRNAs were
enriched in 17 GO terms under biological process, 11 GO terms under cellular component, and 10 GO
terms under molecular function. The results showed that these mRNAs were involved in the regulation
of many biological processes (including biological regulation, cellular process, response to stimulus,
and metabolic process), which predict lncRNAs/circRNAs targeted by miRNAs that may regulate
gene responses to high-temperature stress. It is noteworthy that in these GO terms, 40 mRNAs were
significantly enriched in response to stimulus (GO: 0050896; Fig. 6A and Table S10). Based on the
KEGG analysis, mRNAs were predicted to be involved in 48 pathways in which plant hormone signal
transduction, plant-pathogen interactions and glutathione metabolism were closely linked to
high-temperature stress (Fig. 6B and Table S10). TCONS_00002425, TCONS_00011544,
TCONS_00031257, Csa1M690240.1, Csa6M091930.1 and Csa7M405830.1 were involved in plant
hormone signal transduction pathways. Csa2M286450.1, Csa3M130890.2, Csa3M727960.1, and
Csa3M823060.1 were involved in plant-pathogen interaction pathways. Csa3M889840.1 was involved
in the glutathione metabolism pathway. Csa3M823060.1 was predicted to be the target gene for
miR6196, and the rest of the mRNAs were targeted by miR9748, while the lncRNAs
TCONS_00031790, TCONS_00014332, TCONS_00014717 and TCONS_00005674, circRNAs
novel_circ_001543 and novel_circ_000876 were predicted to bind to miR9748 (Fig. 5).
Based on the above results, we selected lncRNAs, circRNAs, miRNAs and mRNAs related to the
plant hormone signal transduction pathway (the level 1 classification of this pathway was
‘Environmental Information Processing’ and the level 2 classification was ‘Signal transduction’ in the
KEGG database) to further examine the ceRNA network (Fig. 7). TCONS_00031790,
TCONS_00014332, TCONS_00014717, TCONS_00005674, novel_circ_001543 and
novel_circ_000876 were predicted to bind to miR9748. Csa1M690240.1 (auxin-responsive protein
IAA16), Csa6M091930.1 (protein TIFY 9-like) and Csa7M405830.1 (ethylene response sensor 1)
were predicted as target genes of miR9748. These three mRNAs were the pivotal genes of the plant
hormone signal transduction pathway according to the KEGG analysis. This complex ceRNA network
indicated that TCONS_00031790, TCONS_00014332, TCONS_00014717, TCONS_00005674,
novel_circ_001543 and novel_circ_000876 may play regulatory roles in the plant hormone signal
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
16
transduction pathway through miR9748 and its target genes in response to high-temperature stress.
Validation of differentially expressed miRNAs and lncRNAs by qRT-PCR
To confirm the quality of the high-throughput sequencing and the expression patterns of miRNA and
lncRNA in the CK and HT groups, six heat-stress-responsive miRNAs (five known miRNAs and one
novel miRNA) and eight lncRNAs were randomly selected for qRT-PCR analysis. The expression
profiles are displayed in Fig. 8, and the primers are listed in Table S1. As shown in Fig. 8A, a similar
tendency was observed between the qRT-PCR and high-throughput sequencing results of the selected
miRNAs and lncRNAs expression, suggesting that the results of the high-throughput sequencing were
reliable. The expression of two miRNAs (miR172c, miR6196) was increased and reduced for four
miRNAs (miR827b, miR8597, miR9484 and novel_mir_261). The results suggested that the levels of
the tested miRNAs varied significantly during the process of high-temperature stress. Furthermore,
some miRNAs showed stage-specific expression, which was probably involved in the response to
high-temperature stress.
Additionally, we also validated the expression patterns of eight heat-stress-responsive lncRNAs. The
results showed that the differential expression of lncRNAs by qRT-PCR was consistent with the
high-throughput sequencing results. The expression profiles are displayed in Fig. 8B, and the primers
are listed in Table S1. For example, four lncRNAs (TCONS_00005674, TCONS_00014332,
TCONS_00000514 and TCONS_00017799) were significantly downregulated, and two lncRNAs
(TCONS_00031790 and TCONS_00011359) were significantly upregulated in the HT group,
consistent with the high-throughput sequencing results. In other words, the high-throughput
sequencing results were credible.
Endogenous hormone measurements
As mentioned above, the ceRNA network indicated that TCONS_00031790, TCONS_00014332,
TCONS_00014717, TCONS_00005674, novel_circ_001543 and novel_circ_000876 may play
regulatory roles in the plant hormone signal transduction pathway through miR9748 and its target
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
17
genes in response to high-temperature stress. To further investigate the roles of endogenous hormones
in response to high-temperature stress, the IAA and ACC contents were measured in cucumber.
Samples were collected from the leaves in the CK and HT groups. As shown in Fig. 9, both the IAA
and ACC levels decreased in the HT group, indicating that IAA and ethylene signaling might be
involved in response to high-temperature stress.
Discussion
Researchers have shown that lncRNAs and circRNAs act as ceRNAs and are mutually regulated by
competition for binding to common MREs. LncRNAs, circRNAs, miRNAs and mRNAs form
large-scale ceRNA cross-talk networks through MREs, which has exciting implications for gene
regulation at the posttranscriptional level during multiple physiological and pathophysiological
processes (Salmena et al. 2011, Ala et al. 2013). Although lncRNAs and circRNAs have been
identified and studied in plants, the function of most circRNAs remains unknown (Lin 2014, Wang et
al. 2014, Lu et al. 2015, Muthusamy et al. 2015). To investigate the function of lncRNAs and
circRNAs in response to high-temperature stress, we performed a full transcriptome analysis of
lncRNAs, circRNAs, miRNAs and mRNAs by high-throughput sequencing.
A total of 2085 lncRNAs and 2477 circRNAs were identified in cucumber leaves. Compared with
mRNAs, lncRNAs are shorter and present fewer exons and lower GC content. All the 2477 circRNAs
identified by high-throughput sequencing were novel circRNAs. Compared with the control group,
five of the circRNAs were significantly upregulated, and one of them was significantly downregulated.
To date, the function of most lncRNAs is not fully understood. Constructing a co-expression network
of lncRNAs and mRNAs can facilitate lncRNA functional prediction (Cui et al. 2017). By analyzing
the cis effect of lncRNA, we constructed a lncRNA-mRNA co-expression network to further identify
the relationship between lncRNA and mRNA. The GO analysis showed that target genes were
enriched in the photosystem, response to endogenous stimulus, defense response, signal transduction,
and regulation of biological process. Interestingly, we found that the protein-coding genes
Csa4M314390.1 (ERF) and Csa5M613470.1 (MYB_related) were enriched in response to stimuli, and
the results showed the possible role of lncRNA in transcriptional regulation of gene expression, which
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
18
means that the corresponding lncRNAs play a regulatory role in stress response. Recent studies have
also shown that lncRNAs are involved in stress response (Aversano et al. 2015, Song et al. 2016a). A
KEGG pathway analysis of the target genes revealed that they are involved in photosynthesis-antenna
proteins, glutathione metabolism and photosynthesis, which indicated that high-temperature stress has
a large impact on plant photosynthesis. For example, Csa2M079660.1 (photosystem I reaction center
PsaE) and Csa3M060980.1 (photosystem I reaction center PsaG) are enriched in the photosynthetic
pathway. This suggested that they are involved in plant photosynthesis, which is also consistent with
previous reports (Zhao et al. 1993, O'Neill et al. 1994).
Reports have indicated that circRNAs play important roles in miRNA-mediated posttranscriptional
regulation of gene expression by acting as ceRNAs (Hansen et al. 2013a, 2013b, Sebastian 2013,
Wang et al. 2017b). Expression profiles of some plant circRNAs showed a positive correlation with
their parental genes (Ye et al. 2015, Pan et al. 2018). Parent genes of over 700 exonic circRNAs were
orthologues between rice and Arabidopsis, suggesting conservation of circRNAs in plants (Ye et al.
2015). Heat-induced circRNAs might participate in plant response to heat stress through
circRNA-mediated ceRNA networks (Pan et al. 2018). To date, the ceRNAs involved in the
high-temperature stress response of cucumbers have not been reported. Here, for the first time, we
constructed a ceRNA regulatory network of lncRNA/circRNA-miRNA-mRNA that responds to
high-temperature stress in cucumbers based on high-throughput sequencing data. Through the GO
analysis, 359 differentially expressed mRNAs were shown to be involved in the regulation of many
biological processes (including biological regulation, cellular process, response to stimulus, metabolic
process, etc.). In these GO terms, 40 mRNAs were significantly enriched in response to stimulus (GO:
0050896). These key response-stimulated genes, lncRNAs and circRNAs, form ceRNA networks by
targeting common miRNAs, and these networks may provide new evidence of the regulatory
mechanisms in response to high-temperature stress in cucumbers. The KEGG analysis predicted that
mRNAs of the ceRNA network were involved in plant hormone signal transduction, plant-pathogen
interactions and glutathione metabolism in response to high-temperature stress. Previous studies have
shown that plant-pathogen interactions were associated with plant hyperthermia (Chen et al. 2014). In
addition, Song et al. (2016a) studied the co-expression of lncRNAs with genes under different
temperature treatments in non-heading Chinese cabbage and found that plant hormone signal
transduction pathways were enriched according to KEGG analysis. Moreover, plant hormones have
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
19
been shown to induce abiotic stress tolerance through biosynthesis and signal transduction (Tran and
Pal 2014). Studies have shown that GSH synthesis activity in maize root cells increases during
high-temperature stress and increases GSH synthesis, which may be related to the ability of cells to
respond to high-temperature stress conditions (Nietosotelo and Ho 1986). Csa2M286450.1 (probable
calcium-binding protein CML32), Csa3M130890.2 (probable calcium-binding protein CML22),
Csa3M727960.1 (calcium-binding protein CML42) and Csa3M823060.1 (probable calcium-binding
protein CML35) were involved in the plant-pathogen interaction pathway, while Csa3M889840.1
(thylakoid lumenal 29 kDa protein) was involved in the glutathione metabolism pathway. It was
predicted that Csa3M823060.1 was a target gene of miR6196, while other mRNAs were targeted by
miR9748. Studies had shown that CML42 acts as a negative regulator of plant defenses by decreasing
COI1-mediated JA sensitivity and the expression of JA-responsive genes and is independent of
herbivore-induced JA biosynthesis; thus, CML42 might serve as a Ca2+
sensor that has multiple
functions in insect herbivory defense and abiotic stress responses (Vadassery et al. 2012). The CML37,
-38, and -39 transcripts are regulated by biotic and abiotic stress as well as hormone and chemical
treatment (Vanderbeld and Snedden 2007). CML9 and CML20 alters plant responses to ABA and
abiotic stress (Magnan et al. 2008, Wu et al. 2017). Mutum et al (2016) identified miR6196 from the
drought-tolerant rice variety Nagina 22. Yang et al (2017) found that miR6196 showed significant
differential expression under cold stress and was specifically differentially expressed in sugarcane
cultivars ROC22 (relatively cold-sensitive). We constructed the ceRNA network of TCONS_00031790,
TCONS_00014332, TCONS_00014717 and TCONS_00005674, novel_circ_001543,
novel_circ_000876, Csa1M690240.1, Csa6M091930.1, Csa7M405830.1 and miR9748. These mRNAs
were identified as important elements of the plant hormone signal transduction pathway based on the
KEGG analysis. Csa1M690240.1, Csa6M091930.1 and Csa7M405830.1 are the target genes of
miR9748. Studies have shown that chloroplast heat-shock protein 90 (HSP90), which plays a role in
protein processing in the endoplasmic reticulum, was affected by miR9722 and miR9748 (Cakir et al.
2016). The transcription factor MYC2 is involved in environmental information processing and plant
hormone signal transduction and is affected by miR9748 (Cakir et al. 2016). Csa1M690240.1 codes
for the cucumber auxin protein IAA16, and Csa7M405830.1 codes for the cucumber ethylene
response sensor 1 (ERS1) gene. Studies have shown that a high concentration of IAA can weaken the
cotton anther defense response to high-temperature stress (Min et al. 2014). High temperatures can
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
20
limit ethylene production (Kawakami et al. 2013), and ERF1 has been reported to play an active role
in regulating tolerance to salt stress, drought stress and heat stress by regulating specific genes that
respond to stress, thereby regulating the integration of JA, ET and abscisic acid signals (Cheng et al.
2013).
As mentioned above, the ceRNA network indicated that TCONS_00031790, TCONS_00014332,
TCONS_00014717, TCONS_00005674, novel_circ_001543 and novel_circ_000876 may play
regulatory roles in the plant hormone signal transduction pathway through miR9748 and its target
genes in response to high-temperature stress. Most of the target genes were involved in the ethylene-
and IAA-mediated signaling pathways. To further investigate the roles of endogenous hormones in
response to high-temperature stress, the IAA and ACC contents were measured in cucumber. The
results revealed that both the IAA and ACC levels were reduced in the HT group (Fig. 9), suggesting
that heat stress could reduce IAA and ethylene. We speculate that miR9748 regulated Csa1M690240.1
and Csa7M405830.1 alter plant IAA and ethylene responses. Ethylene synthesis by plants is very
sensitive to changes in temperature (Field 1985) and at high temperatures may be greatly reduced (Yu
et al. 1980, Robinson and Biddington 1990). Research has shown that high-temperature treatments
reduce ethylene production from filaments alone and from filaments with anthers attached (Biddington
and Robinson 1993). The ethylene precursor ACC inhibited filament growth in Fuchsia hybrida at
32°C (Jones and Koning 1986) and of Ipomoea nil at 30°C (Koning and Raab 1987). Compared with
the control, the high temperature treatment reduced the endogenous hormone content of IAA in
rapeseed plants (Zhou and Leul 1999). Wu found that high-temperature stress reduced IAA in rice (Wu
et al. 2016). Heat-induced reductions in IAA were reported in the anthers and developing grains of rice
(Wang et al. 2006, Tang et al. 2008). High temperature generally suppresses IAA biosynthesis (Sakata
et al. 2010). Moreover, Sakata et al. (2010) reported that IAA regulated pollen development and male
sterility under high temperature. Our findings showed that miR9748 mainly acted on interactions
between heat-stress-responsive and hormone pathways rather than directly acting on heat-stress
response pathways.
Through the GO and KEGG pathway analyses, competitive lncRNA/circRNA-miRNA-mRNA
regulatory networks were comprehensively integrated and predicted to respond to high-temperature
stress. In addition, TCONS_00031790, TCONS_00014332, TCONS_00014717 and
TCONS_00005674, novel_circ_001543 and novel_circ_000876 were predicted to interact with
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
21
miR9748 to regulate the heat shock response through the plant hormone signaling pathway. Our
research showed that specific lncRNAs and circRNAs might function as ceRNAs in response to
high-temperature stress. In this study, we conducted a full transcriptomic analysis in response to
high-temperature stress in cucumber and integrated, for the first time, the potential ceRNA function of
lncRNAs/circRNAs, laying a foundation for studying the potential functions and mechanisms of
lncRNAs/circRNAs in response to high-temperature stress.
Author contributions
X.H. performed the experiments, analyzed the data and wrote the manuscript. Y.W. and L.W.
contributed significantly to analyzing the data and preparing the manuscript. S.S. performed analyses
and contributed to constructive discussions. S.G. and J.S. conceived and designed the experiments. All
authors contributed to revising the manuscript. All authors read and approved the final manuscript.
Acknowledgements – We thank Gene Denovo Biotechnology Co. (Guangzhou, China) for their help
with the RNA-Seq and bioinformatics analyses. This work was supported by the National Key
Research and Development Program of China (2018YFD1000800) and National Natural Science
Foundation of China (31872152).
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon
reasonable request.
References
Ağar G, Türker M, Battal P, Erez ME (2006) Phytohormone levels in germinating seeds of Zea mays L.
exposed to selenium and aflatoxines. Ecotoxicology 15: 443-450
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
22
Ala U, Karreth FA, Bosia C, Pagnani A, Taulli R, Léopold V, Tay Y, Provero P, Zecchina R, Pandolfi
PP (2013) Integrated transcriptional and competitive endogenous RNA networks are
cross-regulated in permissive molecular environments. Proceedings of the National Academy of
Sciences 110: 7154-7159
Audic S, Claverie JM (1997) The Significance of Digital Gene Expression Profiles. Genome Res 7:
986-995
Aversano R, Contaldi F, Ercolano MR, Grosso V, Iorizzo M, Tatino F, Xumerle L, Dal Molin A,
Avanzato C, Ferrarini A, Delledonne M, Sanseverino W, Cigliano RA, Capella-Gutierrez S,
Gabaldon T, Frusciante L, Bradeen JM, Carputo D (2015) The Solanum commersonii Genome
Sequence Provides Insights into Adaptation to Stress Conditions and Genome Evolution of Wild
Potato Relatives. Plant Cell 27: 954-968
Biddington NL, Robinson HT (1990) Variations in response to high temperature treatments in anther
culture of Brussels sprouts. Plant Cell, Tissue and Organ Culture 22: 48-54
Biddington NL, Robinson HT (1993) High temperature enhances ethylene promotion of anther
filament growth in Brussels sprouts (Brassica oleracea var. gemmifera). Plant Growth Regul 12:
29-35
Bita CE, Gerats T (2013) Plant tolerance to high temperature in a changing environment scientific
fundamentals and production of heat stress-tolerant crops. Front Plant Sci 4: 273
Cakir O, Candar-Cakir B, Zhang B (2016) Small RNA and degradome sequencing reveals important
microRNA function in Astragalus chrysochlorus response to selenium stimuli. Plant Biotechnol J
14: 543-556
Chen J, Yin W, Xia X (2014) Transcriptome profiles of populus euphratica upon heat shock stress.
Current Genomics 15: 326-340
Chen R, Liu L, Xiao M, Wang F, Lin X (2016) Microarray expression profile analysis of long
noncoding RNAs in premature brain injury: A novel point of view. Neuroscience 319: 123-133
Chen G, Cui J, Wang L, Zhu Y, Lu Z, Jin B (2017) Genome-Wide Identification of Circular RNAs in
Arabidopsis thaliana. Frontiers in plant science 8: 1678
Cheng MC, Liao PM, Kuo WW, Lin TP (2013) The Arabidopsis ETHYLENE RESPONSE FACTOR1
regulates abiotic stress-responsive gene expression by binding to different cis-acting elements in
response to different stress signals. Plant Physiol 162: 1566-1582
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
23
Cho J, Koo DH, Nam YW, Han CT, Lim HT, Bang JW, Hur Y (2005) Isolation and characterization of
cDNA clones expressed under male sex expression conditions in a monoecious cucumber plant
(Cucumis sativus L. cv. Winter Long). Euphytica 146: 271-281
Conesa A, Gotz S, Garcia-Gomez JM, Terol J, Talon M, Robles M (2005) Blast2GO: a universal tool
for annotation, visualization and analysis in functional genomics research. Bioinformatics 21:
3674-3676
Conn VM, Hugouvieux V, Nayak A, Conos SA, Capovilla G, Cildir G, Jourdain A, Tergaonkar V,
Schmid M, Zubieta C, Conn SJ (2017) A circRNA from SEPALLATA3 regulates splicing of its
cognate mRNA through R-loop formation. Nature Plants 3: 17053.
Cui J, Luan Y, Jiang N, Bao H, Meng J (2017) Comparative transcriptome analysis between resistant
and susceptible tomato allows the identification of lncRNA16397 conferring resistance to
Phytophthora infestans by co-expressing glutaredoxin. Plant J 89: 577-589
Daehwan K, Geo P, Cole T, Harold P, Ryan K., Steven LS (2013) TopHat2_ accurate alignment of
transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biology 14:
R36
Di C, Yuan J, Wu Y, Li J, Lin H, Hu L, Zhang T, Qi Y, Gerstein MB, Guo Y, Lu ZJ (2014)
Characterization of stress-responsive lncRNAs in Arabidopsis thaliana by integrating expression,
epigenetic and structural features. Plant J 80: 848-861
Dobrev PI, Vankova R (2012) Quantification of abscisic Acid, cytokinin, and auxin content in
salt-stressed plant tissues. Methods Mol Biol 913: 251-261
Field RJ (1985) THE EFFECT OF TEMPERATURE ON ETHYLENE PRODUCTION BY PLANT
TISSUES. pp 47-69
Franco-Zorrilla JM, Valli A, Todesco M, Mateos I, Puga MI, Rubio-Somoza I, Leyva A, Weigel D,
Garcia JA, Paz-Ares J (2007) Target mimicry provides a new mechanism for regulation of
microRNA activity. Nat Genet 39: 1033-1037
Gao Z, Li J, Luo M, Li H, Chen Q, Wang L, Song S, Zhao L, Xu W, Zhang C, Wang S, Ma C (2019)
Characterization and cloning of grape circular RNAs identified the cold resistance-related
Vv-circATS1. Plant physiology pp-01331
Haiyan L, Yuanyuan D, Hailong Y, Nan W, Jing Y, Xiuming L, Yanfang W, Jinyu W, Xiaokun L (2011)
Characterization of the stress associated microRNAs in Glycine max by deep sequencing. BMC
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
24
Plant Biology 11: 170
Hansen TB, Jensen TI, Clausen BH, Bramsen JB, Finsen B, Damgaard CK, Kjems J (2013a) Natural
RNA circles function as efficient microRNA sponges. Nature 495: 384-388
Hansen TB, Kjems J, Damgaard CK (2013b) Circular RNA and miR-7 in cancer. Cancer Res 73:
5609-5612
Jones LS, Koning RE (1986) Role of Growth Substances in the Filament Growth Fuchsia hybrida cv
"Brilliant". American Journal of Botany 73: 1503-1508
Kawakami EM, Oosterhuis DM, Snider JL, FitzSimons TR (2013) High Temperature and the Ethylene
Antagonist 1-Methylcyclopropene Alter Ethylene Evolution Patterns, Antioxidant Responses, and
Boll Growth in Gossypium hirsutum. American Journal of Plant Sciences 4: 1400-1408
Kong L, Zhang Y, Ye ZQ, Liu XQ, Zhao SQ, Wei L, Gao G (2007) CPC: assess the protein-coding
potential of transcripts using sequence features and support vector machine. Nucleic Acids Res 35:
W345-349
Koning RE, Raab MM (1987) Parameters of filament elongation in Ipomoea nil (convolvulaceae).
American Journal of Botany 74: 510-516
Li C, Li Y, Bai L, Zhang T, He C, Yan Y, Yu X (2014) Grafting-responsive miRNAs in cucumber and
pumpkin seedlings identified by high-throughput sequencing at whole genome level. Physiol
Plant 151: 406-422
Lin L, Steven RE, Rena S, Katherine P, Cheng-Ting Y, Wei W, Antony MC, Scott AG, Rex AC, John
EF, Matthew MSE, Michael JS, Jianming Y, Patrick SS, Marja CPT, Nathan MS, Gary JM (2014)
Genome-wide discovery and characterization of maize long non-coding RNAs. Genome Biology
15: R40
Liu F, Wang W, Sun X, Liang Z, Wang F (2013) RNA-Seq revealed complex response to heat stress on
transcriptomic level in Saccharina japonica (Laminariales, Phaeophyta). Journal of Applied
Phycology 26: 1585-1596
Lu T, Cui L, Zhou Y, Zhu C, Fan D, Gong H, Zhao Q, Zhou C, Zhao Y, Lu D, Luo J, Wang Y, Tian Q,
Feng Q, Huang T, Han B (2015) Transcriptome-wide investigation of circular RNAs in rice. RNA
21: 2076-2087
Magnan F, Ranty B, Charpenteau M, Sotta B, Galaud JP, Aldon D (2008) Mutations in AtCML9, a
calmodulin-like protein from Arabidopsis thaliana, alter plant responses to abiotic stress and
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
25
abscisic acid. Plant J 56: 575-589
Min L, Li Y, Hu Q, Zhu L, Gao W, Wu Y, Ding Y, Liu S, Yang X, Zhang X (2014) Sugar and auxin
signaling pathways respond to high-temperature stress during anther development as revealed by
transcript profiling analysis in cotton. Plant Physiol 164: 1293-1308
Murakami Y, Yasuda T, Saigo K, Urashima T, Toyoda H, Okanoue T, Shimotohno K (2006)
Comprehensive analysis of microRNA expression patterns in hepatocellular carcinoma and
non-tumorous tissues. Oncogene 25: 2537-2545
Muthusamy M, Uma S, Backiyarani S, Saraswathi MS (2015) Genome-wide screening for novel,
drought stress-responsive long non-coding RNAs in drought-stressed leaf transcriptome of
drought-tolerant and -susceptible banana (Musa spp) cultivars using Illumina high-throughput
sequencing. Plant Biotechnology Reports 9: 279-286
Mutum RD, Kumar S, Balyan S, Kansal S, Mathur S, Raghuvanshi S (2016) Identification of novel
miRNAs from drought tolerant rice variety Nagina 22. Sci Rep 6: 30786
Nietosotelo J, Ho TH (1986) Effect of heat shock on the metabolism of glutathione in maize roots.
Plant Physiology 82: 1031-1035
O'Neill SD, Zhang XS, Zheng CC (1994) Dark and circadian regulation of mRNA accumulation in the
short-day plant Pharbitis nil. Plant Physiology 104: 569-580
Orom UA, Derrien T, Beringer M, Gumireddy K, Gardini A, Bussotti G, Lai F, Zytnicki M,
Notredame C, Huang Q, Guigo R, Shiekhattar R (2010) Long noncoding RNAs with
enhancer-like function in human cells. Cell 143: 46-58
Pan T, Sun X, Liu Y, Li H, Deng G, Lin H, Wang S (2018) Heat stress alters genome-wide profiles of
circular RNAs in Arabidopsis. Plant Molecular Biology 96: 217-229
Pan X, Welti R, Wang X (2010) Quantitative analysis of major plant hormones in crude plant extracts
by high-performance liquid chromatography–mass spectrometry. Nature Protocols 5: 986-992
Ponjavic J, Oliver PL, Lunter G, Ponting CP (2009) Genomic and transcriptional co-localization of
protein-coding and long non-coding RNA pairs in the developing brain. PLoS Genet 5: e1000617
Robinson MD, McCarthy DJ, Smyth GK (2010) edgeR: A Bioconductor package for differential
expression analysis of digital gene expression data. Bioinformatics 26: 139-140
Sakata T, Oshino T, Miura S, Tomabechi M, Tsunaga Y, Higashitani N, Miyazawa Y, Takahashi H,
Watanabe M, Higashitani A (2010) Auxins reverse plant male sterility caused by high
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
26
temperatures. Plant Signaling and Behavior 11: 8569-8574
Salmena L, Poliseno L, Tay Y, Kats L, Pandolfi PP (2011) A ceRNA Hypothesis: The Rosetta Stone of
a Hidden RNA Language? Cell 146: 353-358
Sebastian M, Marvin J, Antigoni E, Francesca T, Janna K, Agnieszka R, Luisa M, Sebastian DM, Lea
HG, Mathias M, Alexander L, Ulrike Z, Markus L, Christine K, Ferdinand le N, Nikolaus R
(2013) Circular RNAs are a large class of animal RNAs with regulatory. potency. Nature 495:
333-338
Song X, Liu G, Huang Z, Duan W, Tan H, Li Y, Hou X (2016a) Temperature expression patterns of
genes and their coexpression with LncRNAs revealed by RNA-Seq in non-heading Chinese.
cabbage. BMC Genomics 17: 297
Song Y, Ci D, Tian M, Zhang D (2016b) Stable methylation of a non-coding RNA gene regulates gene
expression in response to abiotic stress in Populus simonii. Journal of Experimental Botany 67:
1477-1492
Sumazin P, Yang X, Chiu HS, Chung WJ, Iyer A, Llobet-Navas D, Rajbhandari P, Bansal M, Guarnieri
P, Silva J, Califano A (2011) An extensive microRNA-mediated network of RNA-RNA
interactions regulates established oncogenic pathways in glioblastoma. Cell 147: 370-381
Sun L, Luo H, Bu D, Zhao G, Yu K, Zhang C, Liu Y, Chen R, Zhao Y (2013) Utilizing sequence
intrinsic composition to classify protein-coding and long non-coding transcripts. Nucleic Acids
Res 41: e166
Tan J, Zhou Z, Niu Y, Sun X, Deng Z (2017) Identification and functional characterization of tomato
circrnas derived from genes involved in fruit pigment accumulation. Scientific Reports 7: 8594
Tang R, Zheng J, Jin Z, Zhang D, Huang Y, Chen L (2007) Possible correlation between high
temperature-induced floret sterility and endogenous levels of IAA, GAs and ABA in rice (Oryza
sativa L.). Plant Growth Regulation 54: 37-43
Tay Y, Rinn J, Pandolfi PP (2014) The multilayered complexity of ceRNA crosstalk and competition.
Nature 505: 344-352
Tran LSP, Pal S (2014) Phytohormones: A Window to Metabolism, Signaling and Biotechnological
Applications. Springer, New York
Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL,
Pachter L (2012) Differential gene and transcript expression analysis of RNA-seq experiments
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
27
with TopHat and Cufflinks. Nat Protoc 7: 562-578
Vadassery J, Reichelt M, Hause B, Gershenzon J, Boland W, Mithofer A (2012) CML42-mediated
calcium signaling coordinates responses to Spodoptera herbivory and abiotic stresses in
Arabidopsis. Plant Physiol 159: 1159-1175
Vanderbeld B, Snedden WA (2007) Developmental and stimulus-induced expression patterns of
Arabidopsis calmodulin-like genes CML37, CML38 and CML39. Plant Mol Biol 64: 683-697
Wang F, Cheng F, Liu Y, Zhong L, Zhang G (2006) Dynamic changes of plant hormones in developing
grains at rice filling stage under different temperatures. Acta Agronomica Sinica 32: 25-29
Wang PL, Bao Y, Yee MC, Barrett SP, Hogan GJ, Olsen MN, Dinneny JR, Brown PO, Salzman J
(2014) Circular RNA is expressed across the eukaryotic tree of life. Plos One 9: e90859
Wang Y, Guo S, Wang L, Wang L, He X, Shu S, Sun J, Lu N (2018) Identification of microRNAs
associated with the exogenous spermidine-mediated improvement of high-temperature tolerance
in cucumber seedlings (Cucumis sativus L.). BMC genomics 19: 218-285
Wang Y, Yang M, Wei S, Qin F, Zhao H, Suo B (2017a) Identification of Circular RNAs and Their
Targets in Leaves of Triticum aestivum L. under Dehydration Stress. Frontiers in Plant Science 7
Wang Y, Wang Q, Gao L, Zhu B, Luo Y, Deng Z, Zuo J (2017b) Integrative analysis of circRNAs
acting as ceRNAs involved in ethylene pathway in tomato. Physiol Plant 161: 311-321
Wang Y, Gao Y, Zhang H, Wang H, Liu X, Xu X, Zhang Z, Markus VK, Kaiqiang H, Wang H, Xi F,
Zhao L, Lin C, Gu L (2019) Genome-wide profiling of circular RNAs in the rapidly growing
shoots of moso bamboo (Phyllostachys edulis). Plant Cell Physiol 0: 1-20
Wang Y, Xu Z, Jiang J, Xu C, Kang J, Xiao L, Wu M, Xiong J, Guo X, Liu H (2013) Endogenous
miRNA sponge lincRNA-RoR regulates Oct4, Nanog, and Sox2 in human embryonic stem cell
self-renewal. Dev Cell 25: 69-80
Wu C, Cui K, Wang W, Li Q, Fahad S, Hu Q, Huang J, Nie L, Peng S (2016) Heat-induced
phytohormone changes are associated with disrupted early reproductive development and reduced
yield in rice. Scientific Reports 6: 34978
Wu HJ, Wang ZM, Wang M, Wang XJ (2013) Widespread long noncoding RNAs as endogenous target
mimics for microRNAs in plants. Plant Physiol 161: 1875-1884
Wu X, Qiao Z, Liu H, Acharya BR, Li C, Zhang W (2017) CML20, an Arabidopsis Calmodulin-like
Protein, Negatively Regulates Guard Cell ABA Signaling and Drought Stress Tolerance. Front
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
28
Plant Sci 8: 824
Xin M, Wang Y, Yao Y, Song N, Hu Z, Qin D, Xie C, Peng H, Ni Z, Sun Q (2011) Identification and
characterization of wheat long non-protein coding RNAs responsive to powdery mildew infection
and heat stress by using microarray analysis and SBS sequencing. BMC Plant Biology 11: 61
Xu XW, Zhou XH, Wang RR, Peng WL, An Y, Chen LL (2016) Functional analysis of long intergenic
non-coding RNAs in phosphate-starved rice using competing endogenous RNA network.
Scientific reports 6: 20715
Yan T, Yoo D, Berardini TZ, Mueller LA, Weems DC, Weng S, Cherry JM, Rhee SY (2005) PatMatch:
a program for finding patterns in peptide and nucleotide sequences. Nucleic Acids Res 33:
W262-266
Yang Y, Zhang X, Su Y, Zou J, Wang Z, Xu L, Que Y (2017) miRNA alteration is an important
mechanism in sugarcane response to low-temperature environment. BMC Genomics 18: 833
Ye CY, Chen L, Liu C, Zhu QH, Fan L (2015) Widespread noncoding circular RNAs in plants. New
Phytologist 208: 88-95
Yin J, Liu M, Ma D, Wu J, Li S, Zhu Y, Han B (2018) Identification of circular RNAs and their targets
during tomato fruit ripening. Postharvest Biology and Technology 136: 90-98
Yu YB, Adams DO, Yang SF (1980) Inhibition of ethylene production by 2,4-dinitrophenol and high
temperature. Plant Physiol 66: 286-290
Zhang YC, Chen YQ (2013) Long noncoding RNA: new regulators in plant development. Biochemical
and Biophysical Research Communications 436: 111-114
Zhang YC, Liao JY, Li ZY, Yu Y, Zhang JP, Li QF, Qu LH, Shu WS, Chen YQ (2014) Genome-wide
screening and functional analysis identify a large number of long noncoding RNAs involved in
the sexual reproduction of rice. Genome Biology 15: 512
Zhang G, Diao S, Zhang T, Chen D, He C, Zhang J (2019) Identification and characterization of
circular RNAs during the sea buckthorn fruit development. RNA biology 16: 354-361
Zhang P, Fan Y, Sun X, Chen L, Terzaghi W, Bucher E, Li L, Dai M (2019) A large‐scale circular RNA
profiling reveals universal molecular mechanisms responsive to drought stress in maize and
Arabidopsis. The Plant Journal 1-17
Zhao J, Mühlenhoff U, Bryant DA, Golbeck JH (1993) Psae is required for in vivo cyclic electron
flow around photosystem І in the cyanobacterium Synechococcus sp. pcc 7002. Plant Physiology
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
29
103: 171-180
Zhao Z, Bai J, Wu A, Wang Y, Zhang J, Wang Z, Li Y, Xu J, Li X (2015) Co-LncRNA: investigating
the lncRNA combinatorial effects in GO annotations and KEGG pathways based on human
RNA-Seq data. Database (Oxford) 2015: bav082
Zhao W, Cheng Y, Zhang C, You Q, Shen X, Guo W, Jiao Y (2017) Genome-wide identification and
characterization of circular RNAs by high throughput sequencing in soybean. Scientific Reports 7:
5636.
Zhou W, Leul M (1999) Uniconazole-induced tolerance of rape plants to heat stress in relation to
changes in hormonal levels, enzyme activities and lipid peroxidation. Plant Growth Regulation 27:
99-104
Zhou H, Guo S, An Y, Shan X, Wang Y, Shu S, Sun J (2016) Exogenous spermidine delays chlorophyll
metabolism in cucumber leaves (Cucumis sativus L.) under high temperature stress. Acta
Physiologiae Plantarum 38: 224
Zhou R, Xu L, Zhao L, Wang Y, Zhao T (2018) Genome-wide identification of circRNAs involved in
tomato fruit coloration. Biochemical and biophysical research communications 499: 466-469
Zhu QH, Wang MB (2012) Molecular Functions of Long Non-Coding RNAs in Plants. Genes (Basel)
3: 176-190
Zuo J, Wang Q, Zhu B, Luo Y, Gao L (2016) Deciphering the roles of circRNAs on chilling injury in
tomato. Biochemical and Biophysical Research Communications 479: 132-138
Supporting Information
Table S1. Primers used for qRT-PCR of miRNAs and lncRNAs.
Table S2. List of differentially expressed lncRNAs and mRNAs from the two treatment groups.
Table S3. Differentially expressed protein-coding genes detected 10-kb upstream and downstream of
the lncRNAs.
Table S4. GO and KEGG enrichment analysis of protein-coding genes targeted by cis-acting
lncRNAs.
Table S5. List of circRNAs identified in the cucumber leaf libraries.
Table S6. GO and KEGG enrichment analysis of circRNA source genes.
Table S7. List of known and novel miRNAs identified in the cucumber leaf libraries.
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
30
Table S8. GO and KEGG enrichment analysis of protein-coding genes targeted by miRNAs.
Table S9. LncRNAs, mRNAs and circRNAs predicted to bind to miRNAs.
Table S10. GO terms and pathways enriched by target mRNAs through GO and KEGG analyses of
the ceRNA network.
Figure legends
Fig. 1. Differentially expressed lncRNAs and mRNAs in cucumber. (A) Number of upregulated and
downregulated lncRNAs and mRNAs. (B) and (C) Heatmap of differentially expressed mRNAs and
lncRNAs from four libraries (HT-1, HT-2, CK-1, and CK-2).
Fig. 2. Basic characteristics of lncRNAs in cucumber. (A) Flow chart of the method used to identify
the lncRNAs. (B) Distribution of exon lengths in lncRNAs and mRNAs. (C) Proportions of exon
numbers per transcript for lncRNAs and mRNAs. (D) GC contents of the lncRNAs and mRNAs.
Fig. 3. GO and KEGG enrichment analysis of differentially expressed mRNAs. (A) mRNAs were
significantly enriched in 41 GO terms (P-value < 0.05). Red indicate an increased level of expression,
while blue indicate a decreased level of expression. (B) KEGG pathways involving the top 20 terms.
QValue ranges from 0 to 1. The closer to zero the QValue is, the more significant the enrichment is. A
larger RichFactor value indicates a higher degree of enrichment.
Fig. 4. Multiple mRNAs interacted with one lncRNA. The lines indicate interaction.
Fig. 5. CeRNA regulatory network in cucumber. The ceRNA network is based on lncRNA/miRNA,
circRNA/miRNA, and miRNA/mRNA interactions. The lines represent sequence matching, and
lncRNAs or circRNAs connect expression correlated mRNAs via miRNAs.
Fig. 6. GO annotations and KEGG pathway analyses of 359 differentially expressed mRNAs. (A)
mRNAs were significantly enriched in 38 GO terms (P-value < 0.05). (B) KEGG pathways involving
the top 20 terms. QValue ranges from 0 to 1. The closer to zero the QValue is, the more significant the
enrichment is. A larger RichFactor value indicates a higher degree of enrichment.
Fig. 7. CeRNA network of lncRNAs/circRNAs-miRNAs-mRNA involved in KEGG pathway of Plant
hormone signal transduction. The lncRNAs coded as TCONS_00031790, TCONS_00014332,
TCONS_00014717 and TCONS_00005674, as well as novel_circ_001543 and novel_circ_000876,
were predicted to interact with miR9748. Csa1M690240.1, Csa6M091930.1, Csa7M405830.1 were
predicted to be target mRNAs of miR9748. These three mRNAs are pivotal genes in the Plant
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
31
hormone signal transduction pathway according to KEGG analysis.
Fig. 8. qRT-PCR analysis of several miRNAs (A) and lncRNAs (B). * P-value < 0.05, ** P-value <
0.01.
Fig. 9. Direct biosynthetic precursor of ethylene ACC (A) and auxin IAA (B) contents in leaves from
the CK and HT groups in cucumber.
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
Table 1. Base statistics before and after filtering form HT-1, HT-2, CK-1 and CK-2 lncRNA libraries.
Before Filter After FilterSample
total reads Q20 (%) Q30 (%) N (%) GC (%) clean reads Q20 (%) Q30 (%) N (%) GC (%)
HT-1
HT-2
CK-1
CK-2
16 640 895 000
15 811 860 600
15 123 343 500
16 390 433 700
15 652 223 999
(94.06%)
14 898 898 006
(94.23%)
14 259 408 356
(94.29%)
15 414 566 471
(94.05%)
14 559 951 092
(87.50%)
13 879 986 382
(87.78%)
13 292 965 086
(87.90%)
14 334 505 931
(87.46%)
8 144 531
(0.05%)
7 745 578
(0.05%)
7 347 134
(0.05%)
8 000 888
(0.05%)
6 974 562 474
(41.91%)
6 664 377 487
(42.15%)
6 359 434 947
(42.05%)
6 917 448 309
(42.20%)
16 168 783 366
15 369 275 680
14 747 065 130
15 934 457 288
15 333 665 497
(94.83%)
14 594 514 883
(94.96%)
14 008 253 421
(94.99%)
15 109 145 507
(94.82%)
14 323 684 725
(88.59%)
13 650 487 530
(88.82%)
13 107 755 080
(88.88%)
14 109 446 548
(88.55%)
7 638 541
(0.05%)
7 287 130
(0.05%)
6 931 190
(0.05%)
7 506 657
(0.05%)
6 753 978 148
(41.77%)
6 456 008 376
(42.01%)
6 183 504 648
(41.93%)
6 702 578 867
(42.06%)
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
Table 2. Summary of RNA-seq data and reads mapped to the Cucumis sativus reference genome.
Sample Total Reads Unmapped Reads Unique Mapped Reads Multiple Mapped reads Mapping Ratio
HT-1
HT-2
CK-1
CK-2
108 425 230
103 190 136
98 789 534
106 824 530
25 030 450 (23.09%)
23 280 734 (22.56%)
19 950 633 (20.20%)
21 676 892 (20.29%)
82 062 106 (75.69%)
78 560 258 (76.13%)
77 720 689 (78.67%)
83 946 280 (78.58%)
1 332 674 (1.23%)
1 349 144 (1.31%)
1 118 212 (1.13%)
1 201 358 (1.12%)
76.91%
77.44%
79.80%
79.71%
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
Table 3. Summary of cleaning data from CW and HW sRNA libraries. CW means control treated
with water and HW means high temperature treated with water.
CW HWType
Count Percent (%) Count Percent (%)
Total reads
High quality
3'adapter null
Insert null
5'adapter contaminants
Smaller than 18nt
Poly(A)
Clean reads
11 255 470
11 245 483
36 410
863
9749
332 565
92
10 865 804
100
0.32
0.01
0.09
2.96
0.00
96.62
11 296 147
11 284 130
48 705
2094
14 824
364 094
69
10 854 344
100
0.43
0.02
0.13
3.23
0.00
96.19
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.