Content uploaded by Zhou Weiwei
Author content
All content in this area was uploaded by Zhou Weiwei on Apr 25, 2023
Content may be subject to copyright.
Submitted 17 October 2022
Accepted 24 March 2023
Published 24 April 2023
Corresponding author
Xiuguang Mao,
xgmao@sklec.ecnu.edu.cn
Academic editor
Kush Shrivastava
Additional Information and
Declarations can be found on
page 17
DOI 10.7717/peerj.15231
Copyright
2023 Chen et al.
Distributed under
Creative Commons CC-BY 4.0
OPEN ACCESS
Sex differences in gene expression and
alternative splicing in the Chinese
horseshoe bat
Wenli Chen, Weiwei Zhou, Qianqian Li and Xiuguang Mao
School of Ecological and Environmental Sciences, East China Normal University, Shanghai, China
ABSTRACT
Sexually dimorphic traits are common in sexually reproducing organisms and can
be encoded by differential gene regulation between males and females. Although
alternative splicing is common mechanism in generating transcriptional diversity, its
role in generating sex differences relative to differential gene expression is less clear.
Here, we investigate the relative roles of differential gene expression and alternative
splicing between male and female the horseshoe bat species, Rhinolophus sinicus.
Horseshoe bats are an excellent model to study acoustic differences between sexes.
Using RNA-seq analyses of two somatic tissues (brain and liver) from males and females
of the same population, we identified 3,471 and 2,208 differentially expressed genes
between the sexes (DEGs) in the brain and liver, respectively. DEGs were enriched with
functional categories associated with physiological difference of the sexes (e.g.,gamete
generation and energy production for reproduction in females). In addition, we also
detected many differentially spliced genes between the sexes (DSGs, 2,231 and 1,027
in the brain and liver, respectively) which were mainly involved in regulation of RNA
splicing and mRNA metabolic process. Interestingly, we found a significant enrichment
of DEGs on the X chromosome, but not for DSGs. As for the extent of overlap between
the two sets of genes, more than expected overlap of DEGs and DSGs was observed
in the brain but not in the liver. This suggests that more complex tissues, such as the
brain, may require the intricate and simultaneous interplay of both differential gene
expression and splicing of genes to govern sex-specific functions. Overall, our results
support that variation in gene expression and alternative splicing are important and
complementary mechanisms governing sex differences.
Subjects Evolutionary Studies, Genetics, Molecular Biology, Veterinary Medicine, Zoology
Keywords Differential gene expression, Alternative splicing, Sexual differences, Bats
INTRODUCTION
Sex differences in phenotypes (e.g., morphology, physiology and behavior) are quite
common across a wide range of sexually reproducing organisms. Most of sexually dimorphic
traits can be achieved by differential gene expression between the sexes, defined as sex-
biased gene expression (Ellegren & Parsch, 2007). In the last two decades, sex-biased gene
expression has been extensively studied in numerous species including humans, and
these studies have shown that sex-biased gene expression is present ubiquitously among
How to cite this article Chen W, Zhou W, Li Q, Mao X. 2023. Sex differences in gene expression and alternative splicing in the Chinese
horseshoe bat. PeerJ 11:e15231 http://doi.org/10.7717/peerj.15231
different tissues in these organisms (Rinn & Snyder, 2005;Ingleby, Flis & Morrow, 2015;
Mank, 2017), including human (Mayne et al., 2016;Oliva et al., 2020).
Alternative splicing (AS), as another important form of gene regulation, is a widespread
phenomenon among eukaryotes (Kim, Magen & Ast, 2007) and contributes greatly to
the complexity of organisms and adaptive evolution by creating multiple proteins from a
single gene (Nilsen & Graveley, 2010;Singh & Ahi, 2022). Because males and females largely
share an identical genome, sex-biased AS can act as an alternative mechanism, relative to
sex-biased gene expression, to produce sexually dimorphic traits, in particular when
pleiotropic constraints limit changes of gene expression level (Rogers, Palmer & Wright,
2021). Indeed, sex-specific AS has been documented in a number of animal species,
e.g.,Drosophila (Telonis-Scott et al., 2009;Gibilisco et al., 2016), primate (Blekhman et al.,
2010), fish (Naftaly, Pau & White, 2021), and human (Karlebach et al., 2020). However,
very few studies have attempted to investigate the relative roles of differential gene
expression and alternative splicing in sexual differences of animals (but see Rogers, Palmer
& Wright, 2021;Singh & Agrawal, 2021).
Bats belong to the order Chiroptera and comprise over 1400 species (Simmons &
Cirranello, 2020). Similar to other mammals, bats also exhibit many sexually dimorphic
traits (Camargo & De Oliveira, 2012;Grilliot, Burnett & Mendon¸
ca, 2014;Stevens & Platt,
2015;Wu et al., 2018). Most of the studies, in bats, focused on sex differences in
echolocation pulse frequency (reviewed in Siemers et al., 2005) due to its important role in
communication of bats (Jones & Siemers, 2011). Horseshoe bats are one of the most popular
groups to study acoustic differences between sexes because they emit constant frequency
(CF) in echolocation calls which can be assessed accurately by researchers (Siemers et al.,
2005).
In this study, using one horseshoe bat (Rhinolophus sinicus) as the system, we are the first
to explore sex differences of gene regulation (differential gene expression and alternative
splicing) in bats. Unlike most horseshoe bats showing overlap of call frequencies between
sexes, R. sinicus exhibits non-overlap of sex differences (Xie et al., 2017;Mao et al., 2013). In
addition, a high-quality chromosome-level genome has been generated for R. sinicus (Ren
et al., 2020). This genomic resource can help to quantify transcript expression accurately
and make it possible to perform alternative splicing analysis based on short-read RNA-seq
data. Specifically, we collected bat individuals in April when they arouse from hibernation
and start to feed extensively. For female bats, they also begin to prepare for reproduction.
We propose that if the sex differences are largely encoded by sex-biased gene expression
and/or alternative splicing, we expect to observe multiple differentially expressed or spliced
genes between the sexes which are associated with acoustic difference, feeding or female
reproduction. To test for our proposal, we obtained mRNA-seq data of brain and liver
from four individuals of each sex. Brain is responsible for regulation of almost all life
activities and was recently used to study acoustic differences between the sexes of frog
(Chen et al., 2022). Liver is the primary organ for metabolism and is related to feeding. In
addition, these two tissues have been commonly used to explore sex differences of gene
expression and/or alternative splicing in other animals (Naurin et al., 2011;Trabzuni et
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 2/23
Table 1 Detailed information of samples used in this study (modified from Chen et al., 2022).
Sample ID Sex Tissues Sampling locality Sampling date
180404 Male Brain and liver Jiangsu, China April 19, 2018
180406 Male Brain and liver Jiangsu, China April 19, 2018
180411 Male Brain and liver Jiangsu, China April 19, 2018
180401 Male Brain and liver Jiangsu, China April 19, 2018
180402 Female Brain and liver Jiangsu, China April 19, 2018
180403 Female Brain and liver Jiangsu, China April 19, 2018
180409 Female Brain and liver Jiangsu, China April 19, 2018
180410 Female Brain and liver Jiangsu, China April 19, 2018
al., 2013;Blekhman et al., 2010;Zheng et al., 2013; reviewed in Rinn & Snyder, 2005). Thus,
results from our current study may shed some light on sex-biased gene regulation in bats.
MATERIALS & METHODS
Sampling and mRNA-seq data collection
All samples used in this study were obtained from Chen & Mao (2022) and raw sequencing
reads were available from the NCBI Sequence Read Archive (SRA) under Bioproject
accession number PRJNA763734. Briefly, bats were captured using mist nets in Jiangsu,
China in April (Fig. 1A and Table 1) and only adult bats were sampled. Bats were euthanized
by cervical dislocation and tissues of brain and liver were collected for each bat. We chose
four males and four females in transcriptomics analysis (Fig. 1B). All 16 tissues were frozen
immediately in liquid nitrogen and stored in a −80 ◦C freezer. Sequencing libraries from
16 tissues were created with NEBNext®UltraTM RNA Library Prep Kit for Illumina®
(NEB, USA) and sequenced on an Illumina HiSeq X Ten platform (paired-end 150 bp).
Because R. sinicus is not in the list of state-protected and region-protected wildlife species
in the People’s Republic of China, no permission is required. Our sampling and tissue
collection procedures were approved by the National Animal Research Authority, East
China Normal University (approval ID Rh20200801).
RNA-Seq data trimming and mapping
Following Chen & Mao (2022), raw sequencing reads from each sample were processed
using TRIMMOMATIC version 0.38 (Bolger, Lohse & Usadel, 2014) with the parameters
of SLIDINGWINDOW:4:20. We further trimmed reads to 120 bp and removed those
with <120 bp in order to meet the requirement of rMATs (see below) that all input reads
should be of equal length. Then, filtered reads were mapped to a male R. sinicus reference
genome (a chromosome-level genome with scaffold N50 of >100 Mbp and annotation of
>20,000 genes, Ren et al., 2020) using HISAT2 version 2.2.0 (Kim, Langmead & Salzberg,
2015) with default settings. The resulting SAM files were converted to sorted BAM files
with SAMtools v1.11 (Li et al., 2009). The mRNA alignments in sorted BAM files were used
in both differential expression (DE) and alternative splicing (AS) analysis.
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 3/23
(b)(a)
RNA-seq
VS
Brain
180409
180402
180403
180410
180401
180411
180404
180406
(c)
180402
180403
180409
180410
180401
180411
180404
180406
Liver
180401
180402
180403
180404
180406
180409
180410
180411
−6
−3
0
3
−5 05
Sex
Female
Male
180401
180402
180403
180404
180406
180409
180410
180411
−10
0
10
20
−10 0 10 20
PC1: 33% variance
PC2: 28% variance
Liver
Brain
(d)
Sample
Tissue
Sequencing
PC2: 18% variance
PC1: 52% variance
Sex
Female
Male
Sex
Female
Male
Sex
Female
Male
40°N
CHINA
VIETNAM
CAMBODIA
LAOS
THAILAND
25°N
35°N
30°N
20°N
15°N
10°N
100°E 105°E 110°E 115°E 120°E 125°E 130°E
Figure 1 Sampling, experimental design and clustering analysis. (A) Sampling locality in this study.
(B) Experimental design. Bats of females and males were collected and compared based on RNA-seq data
of two tissues (liver and brain). (C) Principal component analysis (PCA) based on normalized count ma-
trix of all genes in the brain and liver. (D) Hierarchical clustering and heatmap based on normalized count
matrix of all genes in the brain and liver.
Full-size DOI: 10.7717/peerj.15231/fig-1
Differential expression analysis
Mapped reads in each sample were quantified using featureCounts (Liao, Smyth & Shi,
2014) with default settings and normalized across samples using DESeq2 (Love, Huber
& Anders, 2014). To assess the similarity of expression patterns across samples in each
tissue, we conducted a principal component analysis (PCA) using PlotPCA function in
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 4/23
DESeq2 package (Love, Huber & Anders, 2014). In addition, we also performed hierarchical
clustering and heatmaps with the R package pvclust v2.2-0 (Suzuki & Shimodaira, 2006)
and pheatmap v1.0.12 (Kolde, 2012), respectively. These two analyses on all samples of
each tissue revealed one outlier (180401, Figs. 1C and 1D) which was excluded from the
downstream analyses. For each tissue, we filtered out the lowly expressed genes with an
average CPM (counts per million) <1 among individuals of each sex. Then we identified
sex-specific genes, including male-specific genes and female-specific genes, by comparing
the list of genes expressed in each sex. After that, shared genes in both sexes were used
to perform DE analysis with DESeq2 (Love, Huber & Anders, 2014) to identify sex-biased
genes (SBGs), including male-biased genes (MBGs) and female-biased genes (FBGs). We
determined SBGs with the Pvalue <0.05 after Benjamini and Hochberg adjustment for
multiple tests (padj <0.05, Benjamini & Hochberg, 1995). To investigate the grouping of
samples based on expression patterns across genes, we performed hierarchical clustering
and heatmaps based on Euclidean distances of rlog-transformed read counts of each SBG
using the R package pvclust v2.2-0 (Suzuki & Shimodaira, 2006) and pheatmap v1.0.12
(Kolde, 2012), respectively. The reliability of each node in clustering was determined using
bootstrap resampling (1,000 replicates).
Here, differentially expressed genes (DEGs) between males and females included
both sex-specific genes and sex-biased genes (DEGs-female: female-specific genes and
female-biased genes; DEGs-male: male-specific genes and male-biased genes).
Alternative splicing analysis
rMATs (v4.1.0) (Shen et al., 2014) was used to identify the AS events between the sexes in
each tissue. Five different types of AS events were detected by rMATs including skipped
exons (SE), mutually exclusive exons (MXE), alternative 50and 30splice site (A5’SS and
A3’SS), and retained intron (RI). rMATs assesses each splicing event by the PSI value
(percent spliced-in value) which indicates the proportion of an isoform in one group to the
other group at each splice site. Following Rogers, Palmer & Wright (2021), AS events were
determined using 0 <PSI <1 in at least half of the samples in each group to reduce the false
positives. To compare AS between groups, the inclusion difference (1PSI, average PSI of
one group minus average PSI of another group) was calculated for each AS event. Following
Grantham & Brisson (2018), significance of 1PSI between the two groups was determined
using the false discovery rate (FDR) <0.05 and 1PSI >0.1. Genes with significant 1PSI
were considered as differentially spliced genes (DSGs).
To characterize the transcriptional similarity of splicing across samples in each tissue, we
also performed hierarchical clustering and heatmaps based on Euclidean distances of the
PSI value of each DSG using the R package pvclust v2.2-0 and pheatmap v1.0.12. Following
Rogers, Palmer & Wright (2021), when a gene has multiple splice events the average PSI
value is used. Bootstrap resampling procedure was used to assess the reliability of each
node (1,000 replicates).
Chromosomal distribution of DEGs and DSGs
We test whether DEGs and DSGs were significantly enriched in X chromosome relative to
the autosomes. We compared the observed number of DEGs and DSGs to the corresponding
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 5/23
expected number. Non-random distribution was estimated using Fisher’s exact test and
significance was determined using a P-value <0.05.
Overlapping between DEGs and DSGs
We test for the overlap between DEGs and DSGs following Rogers, Palmer & Wright (2021).
Specifically, we first calculated the expected number of genes that are both DEGs and DSGs
as (total no. DEGs ×total no. DSG)/total no. expressed genes. Next, the representation
factor (RF) was calculated to compare the observed number of overlapped genes to the
expected number. RF >1 and RF <1 indicate more overlap than expected and less overlap
than expected, respectively. We used a hypergeometric test in R version 4.0.5 (R Core Team,
2021) to test for significance of comparisons between the observed and expected overlaps.
Significance was determined with a P-value <0.05.
Functional gene ontology analysis
Metascape (http://metascape.org) was used to perform functional enrichment analysis
on genes identified in DE and AS analyses with the Custom Analysis module (Zhou et al.,
2019). A total of 13,905 expressed genes identified in this study were used as the background
list. Significantly enriched gene ontology (GO) terms and KEGG pathways were determined
with corrected p-value using the Banjamini-Hochberg multiple test correction procedure
and q-value <0.05. Log (q-value) of −1.3 is equal to q-value of 0.05. Redundancy was
removed using the REVIGO clustering algorithm (http://revigo.irb.hr/) with the default
settings. We then used the R ggplot2 package to visualize the clustered GO terms.
RESULTS
Here, we obtained 16 tissue samples of RNA-seq data from Chen & Mao (2022) with an
average of 39,217,309 filtered pair reads per sample and an overall alignment rate of 98.11%
to the reference genome (Table S1). Based on these data, we identified and characterized
the differentially expressed genes and spliced genes between males and females. We also
compared these two sets of genes by exploring their distribution patterns in the genome
and the extent of their overlap to assess their relative roles in sex differences.
Identification and characterization of sex-specific genes
In the brain, we identified 232 female-specific and 133 male-specific genes among 13,456
expressed genes (Fig. 2A and Table 2). In contrast, we found more number of sex-specific
genes in the liver (458 and 230, female-specific and male-specific genes, respectively)
among 11,502 expressed genes (Fig. 2B and Table 2). Detailed sex-specific genes have been
described in Table S2.
To explore the functional categories of the sex-specific genes, we performed functional
enrichment analysis. In the brain, male-specific genes were enriched into 21 significant GO
terms and three KEGG pathways and most of them were related to digestion, fatty acid or
lipid transport, and histidine catabolic process (Fig. 2C and Table S3). For female-specific
genes, although not significant after accounting for multiple testing (q-value >0.05), they
were enriched into several interesting terms with uncorrected P<0.01, such as nuclear
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 6/23
13091 232
133
(a) Brain
10814 458230
(b) Liver
29
203 429
Brain
27
106 203
Liver
Female-specific genes
Male-specific genes
Brain Liver
(g)
(f)
GO on male-specific genes in brain
GO on male-specific genes in liver GO on female-specific genes in liver
immune receptor activity
cluster of actin−based cell projections
brush border
apical part of cell
external side of plasma membrane
monocarboxylic acid transport
chemokine−mediated signaling pathway
formate metabolic process
epidermal cell differentiation
formamide metabolic process
digestion
0.2 0.4 0.6
Rich Factor
3
9
15
−2.50
−2.25
−2.00
−1.75
−1.50
Count
Log10 (q-value)
Molecular function
Biological process
Cellular comonent
Molecular function
Biological process
Cellular comonent
Ndc80 complex
0.25 0.50 0.75 1.00
Rich Factor
Count
5
10
15
20
−1.375
−1.350
−1.325
−1.300
Log10 (q-value)
symporter activity
neurofibrillary tangle
postsynaptic intermediate filament cytoskeleton
exocytic vesicle
axon
vesicle−mediated transport in synapse
response to corticosterone
chemical synaptic transmission
regulation of neurotransmitter levels
0.2 0.4 0.6
Rich Factor
−5
−4
−3
−2
Count
10
20
30
Log10 (q-value)
Biological process
Cellular comonent
(c)
(d)
negative regulation
of nuclear division
adaptive immune
response
(e)
KDM5D
DDX3Y
EIF1AY
FOXL3
GTSF1
TMPRSS12
YBX2
♀♂
Sex-specific genes Sex-specific genes
♀♂
(h) GO on shared male-specific genes
digestion
0.025 0.050 0.075 0.100
Rich Factor
−2.4
Count
5
Log10 (q-value)
Biological
process
Figure 2 Identification and characterization of sex-specific genes. (A–B) Venn diagrams showing sex-
specific genes. (C–E) Significant Gene Ontology (GO) terms enriched on the sex-specific genes in the
brain (C) and liver (D and E). (F–G) Venn diagrams showing the number of shared sex-specific genes be-
tween brain and liver. In (F) and (G), four genes related to gamete generation and three Y-linked genes
were also shown, respectively. (H) Significant GO terms enriched on the shared male-specific genes. Rich
factor represents the proportion of sex-specific genes (male-specific and female-specific genes) or shared
sex-specific genes in a GO term to the total genes annotated in the same GO term. Significantly enriched
gene ontology (GO) terms were determined with corrected p-value using the Banjamini-Hochberg multi-
ple test correction procedure and q-value <0.05. Log (q-value) of −1.3 is equal to q-value of 0.05.
Full-size DOI: 10.7717/peerj.15231/fig-2
Table 2 Summary of sex-specific and sex-biased genes identified between the sexes in the brain and
liver.
Tissue Brain Liver
Sex-specific Male-specific 133 (1.0%) 230 (2.0%)
Female-specific 232 (1.7%) 458 (4.0%)
Total 365 (2.7%) 688 (6.0%)
Sex-biased Male-biased 1567 (11.6%) 658 (5.7%)
Female-biased 1539 (11.4%) 862 (7.5%)
Total 3106 (23.0%) 1520 (13.2%)
DEGs Male 1700 (12.6%) 888 (7.7%)
Female 1771 (13.2%) 1320 (11.5%)
Total 3471 (25.8%) 2208 (19.2%)
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 7/23
division, meiotic cycle, gamete generation, and humoral immune response (Table S3).
In the liver, male-specific genes were enriched into 26 significant GO terms and one
KEGG pathway that were mainly involved in regulation of neurotransmitter levels, axon
development, and synaptic signaling (Fig. 2D and Table S3). It was notable that these
male-specific genes were also enriched into GO terms that were related to digestion and
feeding behavior (not significant, but uncorrected P<0.01, Table S3). For female-specific
genes, they were enriched into 16 significant GO terms and most were involved in adaptive
immune response and regulation of nuclear division (Fig. 2E and Table S3).
To investigate whether different tissues have functional similarities of sex difference,
we compared the lists of sex-specific genes identified in the brain and liver. We found 27
male-specific genes and 29 female-specific genes shared by brain and liver (Figs. 2F and
2G,Table S2). Functional enrichment analysis on 27 shared male-specific genes revealed
four significant GO terms and all of them were related to digestion (Fig. 2H and Table
S4). Interestingly, three of shared male-specific genes (KDM5D,DDX3Y and EIF1AY ) are
located on the Y chromosome and two of them (KDM5D and DDX3Y ) belong to ancestral
Y-linked genes (Couger et al., 2021). It was notable that the expression level of KDM5D in
the brain was over six-fold higher than in the liver, whereas the expression levels of other
two Y-linked genes were similar in these two tissues (Table S2). Functional enrichment
analysis on the 29 shared female-specific genes did not identify significant GO terms or
pathways. However, we found that four of them (FOXL3,GTSF1,TMPRSS12, and YBX2)
were associated with gamete generation, which was consistent with the enrichment analyses
on female-specific genes identified in the brain and liver, respectively (see above).
Identification and characterization of sex-biased genes
In the brain, a total of 3106 sex-biased genes (SBGs) were identified with similar numbers
of male-biased and female-biased genes, whereas in the liver, a total of 1520 SBGs were
found with more number of female-biased genes than male-biased genes (Figs. 3A–3D and
Table 2). Detailed sex-biased genes have been described in Table S2.
Functional enrichment analysis on female-biased genes in the brain identified 128
significant GO terms and 16 KEGG pathways and most of them were involved in
cytoplasmic translation, ATP synthesis coupled oxidative phosphorylation process,
ribosome biogenesis, and RNA splicing (Fig. 3E and Table S5). Male-biased genes identified
in the brain were enriched into 246 significant GO terms and 19 KEGG pathways and
most of them were associated with synaptic signaling, axonogenesis, regulation of cell
development and growth, actin cytoskeleton organization, learning and cognition, positive
regulation of cellular catabolic process, and circadian regulation of gene expression (Fig.
3F and Table S5). Similar to female-biased genes in the brain, functional enrichment
analysis on female-biased genes in the liver revealed 182 significant GO terms and 23
KEGG pathways and most of them were involved in cytoplasmic translation, ATP synthesis
coupled oxidative phosphorylation process, and ribosome biogenesis (Fig. 3G and Table
S5). In the liver, we found similar functional categories on sex-biased genes as in the brain
above. Specifically, male-biased genes in the liver were enriched into 301 significant GO
terms and 54 KEGG pathways and they were mostly associated with cellular catabolic
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 8/23
Female-biased genes
Male-biased genes
Sex
Male
Female
0
20
40
60
−6 −4 −2 0 2
Log2 Fold change
−Log10 (padj)
0
10
20
30
−2.5 0.0 2.5 5.0
Log2 Fold change
−Log10 (padj)
100
100 100
100
(a) Brain (b) Liver (c) Brain (d) Liver
1539
1567
Female-biased genes
862
Male-biased genes
658
180406
180410 180411
180404
180403180402180409 180404
180409 180411
180406
180403180402180410
(e) GO on female-biased genes in brain
(f) GO on male-biased genes in brain
(g) GO on female-biased genes in liver
(h) GO on male-biased genes in liver
0.2 0.4 0.6 0.8
RichFactor
Count
20
40
60
−10
−5
0.2 0.4 0.6 0.8
RichFactor
−60
−40
−20
Count
25
50
75
0.3 0.5 0.7
RichFactor
Count
30
60
90
120
−6
−5
−4
−3
−2
0.25 0.50 0.75 1.00
RichFactor
Count
10
20
30
40
50
−6
−5
−4
−3
−2
Log10 (q-value)
Log10 (q-value)
Log10 (q-value)
Log10 (q-value)
Brain Liver
4431096 419
Brain Liver
2791288 379
(l) GO on shared female-biased genes
(k) GO on shared male-biased genes
(i) Female-biased gene (j) Male-biased gene
0.1 0.2 0.3 0.4
RichFactor
Count
10
20
30
40
50
−40
−30
−20
−10
Log10 (q-value)
0.1 0.2 0.3 0.4
RichFactor
Count
10
20
−3.0
−2.5
−2.0
−1.5
Log10 (q-value)
Molecular function
Cellular comonent
Molecular function
Biological process
Cellular comonent
−2
−1
0
1
2
z-score
regulation of RNA splicing
mitochondrial gene expression
proton transmembrane transport
purine−containing compound metabolic process
ribose phosphate metabolic process
mitochondrion organization
generation of precursor metabolites and energy
energy derivation by oxidation of organic compounds
ATP metabolic process
Biological process
mRNA binding
ubiquitin−like modifier activating enzyme activity
protein tag
proteasome−activating activity
ligase activity
oxidoreductase activity, acting on NAD(P)H
proton transmembrane transporter activity
structural constituent of ribosome
cytochrome complex
respirasome
ribosome
structural constituent of cytoskeleton
transcription coregulator activity
PML body
site of polarized growth
cell leading edge
cerebellar mossy fiber
excitatory synapse
neuronal cell body
cell body
biological process involved in intraspecie
neurotransmitter transport
actin filament−based process
ocalization behavior
cell−cell adhesion
regulation of postsynaptic membrane neurot
cell growth
head development
chemical synaptic transmission
cell junction organization
carbohydrate derivative biosynthetic process
negative regulation of amyloid fibril formation
purine−containing compound metabolic process
mitochondrial gene expression
generation of precursor metabolites and energy
energy derivation by oxidation of organic compounds
mitochondrion organization
ATP metabolic process
Biological process
protein tag
ubiquitin−protein transferase regulator activityregulator activity
2 iron, 2 sulfur cluster binding
antioxidant activity
active transmembrane transporter activity
oxidoreductase activity, acting on a heme group of donors
xidoreductase activity
structural molecule activity
structural constituent of ribosome
prefoldin complex
respirasome
ribosome
Molecular function
Cellular comonent
nucleobase−containing compound transport
actin filament−based process
rhythmic process
cell activation
growth
developmental growth
regulation of miRNA transcription
regulation of fibroblast proliferation
regulation of growth
Biological process
armadillo repeat domain binding
protein tyrosine/threonine phosphatase activity
chromatin DNA binding
transcription coregulator activity
chromatin binding
myosin II filament
endocytic vesicle
basal part of cell
basolateral plasma membrane
focal adhesion
transcription regulator complex
Molecular function
Cellular comonent
SMAD binding
chromatin binding
transcription coregulator activity
transcription regulator complex
protein localization to cell junction
chromatin organization
vasculature development
regulation of transcription from RNA polymerase
II promoter in response to oxidative stress
regulation of cell adhesion
histone modification
Molecular function
Biological process
Cellular comonent
protein tag
ubiquitin−protein transferase inhibitor activity
active transmembrane transporter activity
rRNA binding
oxidoreductase activity, acting on a heme group of donors
oxidoreductase activity
structural molecule activity
structural constituent of ribosome
ough endoplasmic reticulum
cytochrome complex
respirasome
cellular detoxification
carbohydrate derivative biosynthetic process
proton transmembrane transport
purine−containing compound metabolic process
mitochondrial gene expression
generation of precursor metabolites and energy
energy derivation by oxidation of organic compounds
mitochondrion organization
ATP metabolic process
Molecular function
Biological process
Cellular comonent
growth
Figure 3 Identification and characterization of sex-biased genes. (A–B) Volcano plots showing sex-
biased gene expression in the brain (A) and liver (B). (C–D) Hierarchical clustering and heatmaps show-
ing expression patterns of sex-biased genes in the brain (C) and liver (D). Numbers on each node indicate
the bootstrap support values. (E–H) Significant Gene Ontology (GO) terms enriched on sex-biased genes
in brain (E, female-biased genes; F, male-biased genes) and liver (G, female-biased genes; H, male-biased
genes). (I–J) Venn diagrams showing the number of shared sex-biased genes between brain and liver. (K–
L) Significant GO terms enriched on the shared genes (K, male-biased genes; J, female-biased genes). Rich
factor represents the proportion of sex-biased genes (male-biased and female-biased genes) or shared sex-
biased genes in a GO term to the total genes annotated in the same GO term. Significantly enriched gene
ontology (GO) terms were determined with corrected p-value using the Banjamini-Hochberg multiple test
correction procedure and q-value <0.05. Log (q-value) of −1.3 is equal to q-value of 0.05.
Full-size DOI: 10.7717/peerj.15231/fig-3
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 9/23
process, response to hormone and nutrient levels, regulation of growth and fibroblast
proliferation, circadian rhythm, and immune function (Fig. 3H and Table S5).
Similar to the analysis on sex-specific genes above, we also compared the lists of
sex-biased genes identified in brain and liver and found 722 shared SBGs, including 279
male-biased genes and 443 female-biased genes (Figs. 3I and 3J). Interestingly, we also found
12 SBGs which have opposite expression patterns between the two tissues. Specifically,
seven of them were female-biased in the brain but male-biased in the liver; five of them were
male-biased in the brain but female-biased in the liver (Table S2). Functional enrichment
analysis on 279 shared male-biased genes identified 57 significant GO terms and 7 KEGG
pathways and most of them were related to regulation of mRNA catabolic process and
stability, hemopoiesis, immune system development, and chromatin organization (Fig. 3K
and Table S6). For 443 shared female-biased genes, they were enriched into 144 significant
GO terms and 18 KEGG pathways which were mostly associated with energy production
via oxidative phosphorylation in the mitochondria and ribosome biogenesis (Fig. 3L and
Table S6). This was consistent with the separate enrichment analyses on female-biased
genes in the brain and liver, respectively (see above).
Alternative splicing analysis
Using rMATs, we found lots of alternative splicing events between sexes in two somatic
tissues. Similar to previous studies (e.g.,Rogers, Palmer & Wright, 2021), MXE and SE
are more common than other three types of splicing in both brain and liver (Table 3).
Hierarchical clustering analysis classified males and females into different clusters in both
tissues (Figs. 4A and 4B). As for differentially spliced genes (DSGs) between sexes, we
found over twice number of DSGs in the brain than in the liver (2231 and 1027 in the brain
and liver, respectively, Table 3 and Table S7). Functional enrichment analysis on DSGs in
the brain revealed 84 significant GO terms and four KEGG pathways which were mostly
related to synaptic signaling, cognition or learning, regulation of RNA splicing and mRNA
processing (Fig. 4C and Table S8). In the liver, DSGs were enriched into 180 significant GO
terms and 20 KEGG pathways and most of them were involved in catabolic and metabolic
processes, regulation of RNA splicing and mRNA processing, humoral immune response,
and regulation of coagulation (Fig. 4D and Table S8). By comparing the lists of DSGs in
the brain and liver, we found 387 DSGs shared by these two tissues (Fig. 4E) which were
enriched into 13 significant GO terms mostly associated with mRNA metabolic process
and regulation of RNA splicing (Fig. 4F and Table S9).
Comparisons of gene differential expression and alternative splicing
To compare the two forms of gene expression regulation, we first explored the difference of
chromosomal distribution patterns for DEGs and DSGs. We found that DEGs in females
were significantly enriched on the X chromosome in both brain and liver, whereas DEGs
in males were less enriched in either brain or liver (Table 4 and Figs. 5A and 5B). For all
DEGs, significant enrichment on the X chromosome was observed in the brain but not in
the liver. Contrast to the case in DEGs, we did not observe significant enrichment of DSGs
on the X chromosome in either brain or liver (Table 4 and Figs. 5A and 5B).
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 10/23
Table 3 Summary of alternative splicing (AS) events and differentially spliced genes (DSGs) identified
between the sexes in the brain and liver.
Tissue Brain Liver
A3SS 336 189
A5SS 341 136
MXE 1766 912
RI 391 192
SE 1113 432
Splicing
events
Total 3940 1861
A3SS 273 145
A5SS 288 114
MXE 1202 548
RI 336 154
SE 787 292
DSGs
Total 2231 (16.6%) 1027 (8.9%)
Sex
Male
Female
(a) Brain (b) Liver
Brain
3871844 640
Liver
100 100 100
100
(f) GO on shared DSGs(e) Shared DSGs
180404 180402
180411
180406 180409180403180410
180411 180410
180406
180404 180403180402180409
mRNA binding
nuclear speck
mRNA metabolic process
regulation of RNA splicing
0.07 0.09 0.11 0.13 0.15
RichFactor
−6
−5
−4
−3
−2
Count
24
32
40
Log10 (q-value)
(c) GO on DSGs in brain
(d) GO on DSGs in liver
mRNA binding
tau protein binding
glutamate receptor activity
nuclear speck
site of polarized growth
cell body
GO:0005938 cell cortex
neuronal cell body
main axon
glutamatergic synapse
regulation of nervous system process
regulation of endocytosis
establishment or maintenance of epithelial cell apical/basal polarity
neurotransmitter transport
regulation of membrane potential
cell junction organization
cognition
0.3 0.4 0.5
RichFactor
Count
30
60
90
120
−6
−5
−4
−3
−2
Log10 (q-value)
endopeptidase inhibitor activity
lipid transporter activity
flavin adenine dinucleotide binding
oxidoreductase activity
mRNA binding
peroxisome
blood microparticle
sulfur compound metabolic process
protein−containing complex remodeling
regulation of coagulation
organic hydroxy compound metabolic process
regulation of plasma lipoprotein particle levels
regulation of small molecule metabolic process
amine metabolic process
nitrogen cycle metabolic process
lipid homeostasis
plasma lipoprotein particle clearance
acute−phase response
organophosphate ester transport
mRNA metabolic process
0.2 0.3 0.4 0.5 0.6
RichFactor
−10
−8
−6
−4
−2
Count
20
40
60
80
Log10 (q-value)
Molecular function
Biological process
Cellular comonent
Molecular function
Biological process
Cellular comonent
Molecular function
Biological process
Cellular comonent
−2
−1
0
1
2
z-score
Figure 4 Characterization of differentially spliced events and differentially spliced genes (DSGs).
(A–B) Hierarchical clustering and heatmaps showing alternative splicing level in the brain (A) and liver
(B). This analysis was based on Euclidean distances of the PSI value of each DSG. The PSI value (percent
spliced-in value) represents the proportion of an isoform in one group to the other group at each splice
site, ranging from 0 to 1. Numbers on each node indicate the bootstrap support values. (C–D) Significant
Gene Ontology (GO) terms enriched on DSGs in brain (C) and liver (D). (E) Venn diagrams showing the
number of shared DSGs between brain and liver. (F) Significant GO terms enriched on the shared DSGs.
Rich factor represents the proportion of DSGs in a GO term to the total genes annotated in the same GO
term. Significantly enriched gene ontology (GO) terms were determined with corrected p-value using
the Banjamini-Hochberg multiple test correction procedure and q-value <0.05. Log (q-value) of −1.3 is
equal to q-value of 0.05.
Full-size DOI: 10.7717/peerj.15231/fig-4
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 11/23
Table 4 Tests for enrichments of DEGs and DSGs on the X chromosome using Fisher’s exact test.
Tissue Observed Expected
Brain DEGs-female Autosomal 1675 1705.46
X-linked 96 65.54
pvalue of Fisher’s exact test 0.000
DEGs-male Autosomal 1643 1637.08
X-linked 57 62.92
pvalue of Fisher’s exact test 0.450
DEGs Autosomal 3318 3342.54
X-linked 153 128.46
pvalue of Fisher’s exact test 0.012
DSGs Autosomal 2163 2148.43
X-linked 68 82.57
pvalue of Fisher’s exact test 0.075
Liver DEGs-female Autosomal 1273 1275.36
X-linked 47 44.64
pvalue of Fisher’s exact test 0.000
DEGs-male Autosomal 864 857.97
X-linked 24 30.03
pvalue of Fisher’s exact test 0.329
DEGs Autosomal 2137 2133.32
X-linked 71 74.68
pvalue of Fisher’s exact test 0.694
DSGs Autosomal 1000 992.27
X-linked 27 34.73
pvalue of Fisher’s exact test 0.175
Notes.
Abbreviations: DSGs, differentially spliced genes; DEGs, differentially expressed genes, included both sex-specific genes and
sex-biased genes; DEGs-female, female-specific genes and female-biased genes; DEGs-male, male-specific genes and male-
biased genes.
Second, we test whether there is more overlap than expected between DEGs and DSGs.
We observed significant overlap between these two categories of genes in the brain (RF =
1.21, P<0.05) but not in the liver (RF =0.92, P>0.05, Figs. 5C and 5D). To explore
the functional differences between overlapped and non-overlapped DEGs and DSGs in
each tissue, we also performed enrichment analyses on each set of genes (Table S10).
Specifically, in the brain, we found that the overlapped DEGs and DSGs were mostly
involved in the regulation of RNA splicing and synaptic signaling, whereas the only DEGs
were in the processes of cytoplasmic translation, oxidative phosphorylation, ATP synthesis,
and ribosome biogenesis, and the only DSGs were in synaptic signaling (Table S11). In the
liver, we found that overlapped DEGs and DSGs were mostly associated with metabolic
and biosynthetic processes, regulation of RNA splicing, cytoplasmic translation, whereas
only DEGs were enriched into similar GO terms with only DEGs in brain, and only DSGs
were involved in the processes of metabolism and biosynthesis (Table S11).
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 12/23
(c) Brain (d) Liver
DEGs
6962775 1535
DSGs DEGs
1812027 846
DSGs
RF = 1.21
(P = 0.00)
RF = 0.92
(P = 0.10)
(a) Brain
*
***
sGSD ro sGED fo egatnecreP emosomorhc-X eht no
(575.49) (197.15)
0%
1%
2%
3%
4%
5%
6%
DEGs-
female
DEGs-
male
All DEGs DSGs
(b) Liver
sGSD ro sGED fo egatnecreP emosomorhc-X eht no
0%
1%
2%
3%
4%
DEGs-
female
DEGs-
male
All DEGs DSGs
***
Figure 5 (A–B) Enrichment of differentially expressed genes (DEGs) and differentially spliced genes
(DSGs) between the sexes on the X chromosome in the brain (A) and liver (B). (C–D) Venn diagrams
showing the overlap of DEGs and DSGs in the brain (C) and liver (D). Numbers in brackets are the ex-
pected number of overlapped DEGs and DSGs. DEGs-female: female-specific and female-biased genes;
DEGs-male: male-specific and male-biased genes. *P<0.05, ***P<0.001.
Full-size DOI: 10.7717/peerj.15231/fig-5
DISCUSSION
In this study, we used RNA-seq data of brain and liver, for the first time, to investigate sex
differences of gene expression and splicing in bats, a group of mammals exhibiting diverse
sexually dimorphic traits (see also in Introduction). Below, we first discussed the results
of differential expression analysis and alternative splicing analysis, respectively. Then, we
assessed the relative role of these two forms of gene regulation in sex differences.
Sex differences in differential gene expression
In April, bats arouse from hibernation for feeding and nutrition. Additionally, female bats
need to prepare for reproduction, including gamete generation, fertilization and gestation
(Oxberry, 1979). Consistent with the physiological differences between sexes, we found
that female-specific genes in both tissues were mostly involved in nuclear division and
gamete generation although the later functional category was not significantly enriched
(uncorrected p<0.01). Among them, four (FOXL3,GTSF1,TMPRSS12, and YBX2) should
be notable here. FOXL3 is a germ cell-intrinsic factor and it has been shown to be involved
in spermatogenesis and the initiation of oogenesis in female gonad of fishes (Nishimura
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 13/23
et al., 2015;Kikuchi et al., 2020). GTSF1, encoding gametocyte specific factor 1, has been
suggested to play important roles in postnatal oocyte maturation and prespermatogonia
in mammals (Krotz et al., 2009;Liperis, 2013;Yoshimura et al., 2018). In mice, TMPRSS12,
encoding transmembrane serine protease 12, has been found to be required for male
fertility (Zhang et al., 2022) and sperm motility and migration to the oviduct (Larasati et
al., 2020). Last, YBX2, encoding Y-box binding protein 2, has been proved to be important
in spermatogenesis in mice (He et al., 2019) and also in human (Hammoud et al., 2009). In
addition, a majority of female-biased genes in both tissues were associated with cytoplasmic
translation and ATP synthesis coupled oxidative phosphorylation process, which provides
energy demand for reproduction. Overall, our current study identified thousands of
differentially expressed genes between sexes (sex-specific and sex-biased genes) in two
somatic tissues which largely contribute to sex differences in physiology (e.g., female
reproduction). Thus, our results in bats support the well-known proposal that most sex
differences are caused by sex-biased gene expression (Ellegren & Parsch, 2007;Mank, 2017).
Also we found three notable Y-linked genes (KDM5D,DDX3Y and EIF1AY ) among the
list of male-specific genes in both tissues. KDM5D encodes a histone demethylase for H3K4
demethylation. This gene has also been identified as a male-specific gene and is required for
other sexually dimorphic genes in mouse embryonic fibroblasts (Mizukami et al., 2019).
A recent study indicated that the X chromosome paralog of KDM5D,KDM5C, could be
considered as a determinant of sex difference in adiposity due to its dosage difference
between sexes (Link et al., 2020). Here, KDM5C was also identified as a female-biased
gene in the brain, suggesting that this gene might also contribute to the sex difference
in the brain in bats. DDX3Y (also known as DBY ) encodes an ATP-dependent RNA
helicase and its main function is related to RNA metabolism. This gene has been shown
to be expressed widely across human tissues (Uhlén et al., 2015) and has been suggested
to play an important role in dimorphic neural development (Vakilian et al., 2015). These
combined results provide further evidences on the contribution of Y chromosome genes
beyond sex determination and support their important roles in sexual dimorphic traits of
adult nonreproductive tissues (see also Meyfour et al., 2019;Godfrey et al., 2020).
Sex differences in alternative splicing
Similar to previous studies in other animals (e.g.,Drosophila,Gibilisco et al., 2016; birds,
Rogers, Palmer & Wright, 2021; human, Trabzuni et al., 2013;Karlebach et al., 2020), we
also detected a large number of sex-biased spliced genes in bats (16.6% and 8.9% of
expressed genes in the brain and liver, respectively). These combined evidences from
different animals and tissues suggest that similar to sex-biased gene expression, sex-biased
alternative splicing might be also an important form of gene regulation in encoding sex
differences (Karlebach et al., 2020;Singh & Agrawal, 2021).
Although somatic tissues were used in this study, we still observed strong tissue effects
on alternative splicing between sexes with over twice number of DSGs identified in the
brain than in the liver. This tissue effects of sex-biased splicing has also been reported in
previous studies in birds (Rogers, Palmer & Wright, 2021) and Drosophila (Gibilisco et al.,
2016). However, in both previous studies, gonad and somatic tissues were used and they
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 14/23
found little sex-biased splicing in somatic tissues comparing to gonad tissues (Gibilisco et
al., 2016;Rogers, Palmer & Wright, 2021). Further evidences of tissue differences between
somatic and gonad tissues was from the hierarchical clustering analysis based on alternative
splicing level in Rogers, Palmer & Wright (2021), where males and females were mixed in the
somatic tissue but they clustered separately in the gonad tissues. However, our hierarchical
clustering analysis revealed that both somatic tissues showed clustering between males and
females. The difference between these two studies might be resulted from tissue effect on
different somatic tissues. Indeed, a recent study on 39 different tissues in human revealed
that a majority of alternative splicing events (97.6%) were specific to one tissue (Karlebach
et al., 2020).
Comparisons of the two forms of gene expression regulation
Our results showed that in both somatic tissues (brain and liver), DEGs in females (female-
specific and female-biased genes) were found to be more enriched than expected in X
chromosome, which is similar to previous studies in other organisms (e.g., fish, Leder et al.,
2010;Sharma et al., 2014; water strider, Toubiana et al., 2021; mouse, Khil et al., 2004;Yang
et al., 2006; human, Oliva et al., 2020). Enrichment of sex-biased genes in X chromosome
has been proposed to resolve sexual conflict or sexual dimorphism (Rice, 1984;Rice, 1987;
Rowe, Chenoweth & Agrawal, 2018) although this proposal has been recently questioned
(Ruzicka & Connallon, 2020).
Contrast to the case of DEGs, we did not observe a significant enrichment of DSGs in
X chromosome. Up to now, less studies have been performed to investigate the genomic
distributions of sex-biased DSGs. In addition, those few published studies revealed different
results. A recent study based on combined results of 39 tissues found that sex-biased DSGs
were significantly enriched in X chromosome (Karlebach et al., 2020). However, another
recent study on different tissues of Drosophila found that sex-biased DSGs identified in
the whole body were enriched in X chromosome while ones in the head were not enriched
(Singh & Agrawal, 2021). We proposed that the inconsistency between different studies
might be largely caused by different tissues used because there was a strong tissue effect on
sex-biased alternative splicing (Karlebach et al., 2020).
We observed more than expected overlap of DEGs and DSGs identified between the
sexes in the brain but less than expected overlap in the liver. This contrast result might be
caused by the difference of the extent of complexity between the two tissues. Compared to
liver, the brain is more complex and more involved in sex differences. Indeed, we observed
more number of DEGs and DSGs in the brain than the liver (brain: 3471 DEGs and 2231
DSGs; liver: 2208 DEGs and 1027 DSGs). Again, the previous studies on the extent of
overlap between the two sets of genes revealed different results. In Rogers, Palmer & Wright
(2021), less than expected overlap of DEGs and DSGs was observed in the gonad. However,
in Karlebach et al. (2020), the authors observed more than expected overlap between these
two sets of genes. This inconsistency between different studies might also result from
tissue specificity in sex-biased gene expression or alternative splicing possibly due to the
difference of the extent of complexity across tissues.
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 15/23
Overall, our current results, combined previous studies, suggested that the relative roles
of differential gene expression and alternative splicing in sex differences may have tissue
specificity. In addition, we found that the only DEGs and only DSGs in each tissue were
enriched into different functional categories. Thus, our study further supports that the
two forms of gene regulation might play complementary roles in encoding sex differences
(Rogers, Palmer & Wright, 2021;Singh & Agrawal, 2021;Karlebach et al., 2020).
Limitations of the study
In this study, we identified far more DSGs between males and females than DEGs in both
brain and liver, whereas a recent study detected far fewer DSGs between sexes than DEGs
in birds (Rogers, Palmer & Wright, 2021). This contrast may be resulted from different
kinds of tissues used between studies (reproductive tissue in (Rogers, Palmer & Wright,
2021) while somatic tissues in this study). In the future reproductive tissues of our study
system will be used to test whether there were different effects of differential expression and
splicing on sex-related regulatory networks between reproductive and nonreproductive
tissues.
To make comparable analysis on gene expression patterns, individuals of this study were
collected in the same population and at the same time. However, we still cannot confidently
determine whether the sampled individuals were at the same age. To reduce the effect of
age on gene expression, we only used adult bats in this study (Chen & Mao, 2022). In the
future, we can first determine the age of bats using DNA methylation profiles which use
noninvasive sampling (Wilkinson et al., 2021). Then, bats with the same age were used to
assess the sex differences of gene expression and splicing.
Similar to the majority of current studies on gene expression and splicing, here we used
bulk RNA-seq which may mask difference of gene expression and splicing between the sexes
because this sequencing strategy assess the difference of expression using the average level of
multiple cell types in the tissue. In the future, single-cell transcriptome analyses (Kulkarni
et al., 2019) will be promising to explore the difference of sex-biased gene expression and
splicing in different cell types (Kasimatis, Sánchez-Ramírez & Stevenson, 2021). In addition,
it will be interesting to examine specific regions of the brain to determine differentially
expressed and spliced genes in males and females in the future. Finally, it is difficult to
reconstruct isoforms with short-read RNA-seq. In the future, we can identify sex-specific
transcripts accurately using long-read RNA-seq (e.g., PacBio Iso-Seq) which can skip
the bioinformatics steps of reconstructing isoforms (e.g., in fishes, Naftaly, Pau & White,
2021).
CONCLUSIONS
In two somatic tissues of bats, we found many differentially expressed genes between the
sexes which largely contributed to their physiological differences. In addition, our results in
bats also support an important role of sex-biased alternative splicing in sex differences. As
for the relative roles of these two gene regulation forms, it may depend on specific tissues
used in the study.
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 16/23
ACKNOWLEDGEMENTS
We thank Sun Haijian, Wang JY, and Ding YT for assistance with sample collection. We also
thank Kush Shrivastava, Fernando Diaz and four anonymous reviewers for constructive
comments that improved the manuscript.
ADDITIONAL INFORMATION AND DECLARATIONS
Funding
This work was supported by The Scientific and Technological Innovation Plan of Shanghai
Science and Technology Committee (20ZR1417000). The funders had no role in study
design, data collection and analysis, decision to publish, or preparation of the manuscript.
Grant Disclosures
The following grant information was disclosed by the authors:
The Scientific and Technological Innovation Plan of Shanghai Science and Technology
Committee: 20ZR1417000.
Competing Interests
The authors declare there are no competing interests.
Author Contributions
•Wenli Chen conceived and designed the experiments, performed the experiments,
analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the
article, and approved the final draft.
•Weiwei Zhou and Qianqian Li analyzed the data, prepared figures and/or tables, and
approved the final draft.
•Xiuguang Mao conceived and designed the experiments, authored or reviewed drafts of
the article, and approved the final draft.
Animal Ethics
The following information was supplied relating to ethical approvals (i.e., approving body
and any reference numbers):
National Animal Research Authority, East China Normal University
Data Availability
The following information was supplied regarding data availability:
All data are available in the Supplemental Files and the sequencing reads are available at
NCBI Sequence Read Archive (SRA): PRJNA763734.
Supplemental Information
Supplemental information for this article can be found online at http://dx.doi.org/10.7717/
peerj.15231#supplemental-information.
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 17/23
REFERENCES
Benjamini Y, Hochberg Y. 1995. Controlling the false discovery rate—a practical and
powerful approach to multiple testing. Journal of the Royal Statistical Society. Series B
57(1):289–300 DOI 10.1111/j.2517-6161.1995.tb02031.x.
Blekhman R, Marioni JC, Zumbo P, Stephens M, Gilad Y. 2010. Sex-specific and
lineage-specific alternative splicing in primates. Genome Research 20(2):180–189
DOI 10.1101/gr.099226.109.
Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for Illumina
sequence data. Bioinformatics 30(15):2114–2120
DOI 10.1093/bioinformatics/btu170.
Camargo NFD, De Oliveira HF. 2012. Sexual dimorphism in Sturnira lilium (Chiroptera,
Phyllostomidae): can pregnancy and pup carrying be responsible for differences in
wing shape? PLOS ONE 7(11):e49734 DOI 10.1371/journal.pone.0049734.
Chen Z, Liu Y, Liang R, Cui C, Zhu Y, Zhang F, Zhang J, Chen X. 2022. Comparative
transcriptome analysis provides insights into the molecular mechanisms of high-
frequency hearing differences between the sexes of Odorrana tormota.BMC Genomics
23(1):1–13 DOI 10.1186/s12864-022-08536-2.
Chen W, Mao X. 2022. Impacts of seasonality on gene expression in the Chinese
horseshoe bat. Ecology and Evolution 12(5):e8923 DOI 10.1002/ece3.8923.
Couger MB, Roy SW, Anderson N, Gozashti L, Pirro S, Millward LS, Kim M, Kilburn
D, Liu KJ, Wilson TM, Epps CW, Dizney L, Ruedas LA, Campbell P. 2021. Sex
chromosome transformation and the origin of a male-specific X chromosome in the
creeping vole. Science 372(6542):592–600 DOI 10.1126/science.abg7019.
Ellegren H, Parsch J. 2007. The evolution of sex-biased genes and sex-biased gene
expression. Nature Reviews Genetics 8(9):689–698 DOI 10.1038/nrg2167.
Gibilisco L, Zhou Q, Mahajan S, Bachtrog D. 2016. Alternative splicing within and
between Drosophila species, sexes, tissues, and developmental stages. PLOS Genetics
12(12):e1006464 DOI 10.1371/journal.pgen.1006464.
Godfrey AK, Naqvi S, Chmátal L, Chick JM, Mitchell RN, Gygi SP, Skaletsky H, Page
DC. 2020. Quantitative analysis of Y-Chromosome gene expression across 36 human
tissues. Genome Research 30(6):860–873 DOI 10.1101/gr.261248.120.
Grantham ME, Brisson JA. 2018. Extensive differential splicing underlies phenotyp-
ically plastic aphid morphs. Molecular Biology and Evolution 35(8):1934–1946
DOI 10.1093/molbev/msy095.
Grilliot ME, Burnett SC, Mendon¸
ca MT. 2014. Sex and season differences in the
echolocation pulses of big brown bats (Eptesicus fuscus) and their relation to mating
activity. Acta Chiropterologica 16(2):379–386 DOI 10.3161/150811014X687332.
Hammoud S, Emery BR, Dunn D, Weiss RB, Carrell DT. 2009. Sequence alterations
in the YBX2 gene are associated with male factor infertility. Fertility and Sterility
91(4):1090–1095 DOI 10.1016/j.fertnstert.2008.01.009.
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 18/23
He Y, Lin Y, Zhu Y, Ping P, Wang G, Sun F. 2019. Murine PAIP1 stimulates translation
of spermiogenic mRNAs stored by YBX2 via its interaction with YBX2. Biology of
Reproduction 100(2):561–572 DOI 10.1093/biolre/ioy213.
Ingleby FC, Flis I, Morrow EH. 2015. Sex-biased gene expression and sexual conflict
throughout development. Cold Spring Harbor Perspectives in Biology 7(1):a017632
DOI 10.1101/cshperspect.a017632.
Jones G, Siemers BM. 2011. The communicative potential of bat echolocation pulses.
Journal of Comparative Physiology A 197:447–457
DOI 10.1007/s00359-010-0565-x.
Karlebach G, Veiga DF, Mays AD, Chatzipantsiou C, Barja PP, Chatzou M, Kesarwani
AK, Danis D, Kararigas G, Zhang XA, George J, Steinhaus R, Hansen P, Seelow
D, McMurry JA, Haendel MA, Yang J, Oprea T, Anczukow O, Banchereau J,
Robinson PN. 2020. The impact of biological sex on alternative splicing. BioRxiv
DOI 10.1101/490904.
Kasimatis KR, Sánchez-Ramírez S, Stevenson ZC. 2021. Sexual dimorphism through
the lens of genome manipulation, forward genetics, and spatiotemporal sequencing.
Genome Biology and Evolution 13(2):evaa243 DOI 10.1093/gbe/evaa243.
Khil PP, Smirnova NA, Romanienko PJ, Camerini-Otero RD. 2004. The mouse X
chromosome is enriched for sex-biased genes not subject to selection by meiotic sex
chromosome inactivation. Nature Genetics 36(6):642–646
DOI 10.1038/ng1368.
Kikuchi M, Nishimura T, Ishishita S, Matsuda Y, Tanaka M. 2020. Foxl3, a sexual
switch in germ cells, initiates two independent molecular pathways for commitment
to oogenesis in medaka. Proceedings of the National Academy of Sciences of the United
States of America 117(22):12174–12181 DOI 10.1073/pnas.1918556117.
Kim D, Langmead B, Salzberg SL. 2015. HISAT: a fast spliced aligner with low memory
requirements. Nature Methods 12(4):357–360 DOI 10.1038/nmeth.3317.
Kim E, Magen A, Ast G. 2007. Different levels of alternative splicing among eukaryotes.
Nucleic Acids Research 35(1):125–131 DOI 10.1093/nar/gkl924.
Kolde R. 2012. Pheatmap: pretty heatmaps. R package version 1.0.12. Available at https:
//CRAN.R-project.org/package=pheatmap.
Krotz SP, Ballow DJ, Choi Y, Rajkovic A. 2009. Expression and localization of the
novel and highly conserved gametocyte-specific factor 1 during oogenesis and
spermatogenesis. Fertility and Sterility 91(5):2020–2024
DOI 10.1016/j.fertnstert.2008.05.042.
Kulkarni A, Anderson AG, Merullo DP, Konopka G. 2019. Beyond bulk: a review of
single cell transcriptomics methodologies and applications. Current Opinion in
Biotechnology 58:129–136 DOI 10.1016/j.copbio.2019.03.001.
Larasati T, Noda T, Fujihara Y, Shimada K, Tobita T, Yu Z, Matzuk MM, Ikawa M.
2020. Tmprss12 is required for sperm motility and uterotubal junction migration
in mice. Biology of Reproduction 103(2):254–263 DOI 10.1093/biolre/ioaa060.
Leder EH, Cano JM, Leinonen T, O’Hara RB, Nikinmaa M, Primmer CR, Merilä
J. 2010. Female-biased expression on the X chromosome as a key step in sex
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 19/23
chromosome evolution in threespine sticklebacks. Molecular Biology and Evolution
27(7):1495–1503 DOI 10.1093/molbev/msq031.
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis
G, Durbin R. 1000 Genome Project Data Processing Subgroup. 2009. The se-
quence alignment/map format and SAMtools. Bioinformatics 25(16):2078–2079
DOI 10.1093/bioinformatics/btp352.
Liao Y, Smyth GK, Shi W. 2014. FeatureCounts: an efficient general purpose program
for assigning sequence reads to genomic features. Bioinformatics 30(7):923–930
DOI 10.1093/bioinformatics/btt656.
Link JC, Wiese CB, Chen X, Avetisyan R, Ronquillo E, Ma F, Guo X, Yao J, Allison M,
Chen Y-DI, Rotter JI, Moustafa JSE-S, Small KS, Iwase S, Pellegrini M, Vergnes
L, Arnold AP, Reue K. 2020. X chromosome dosage of histone demethylase
KDM5C determines sex differences in adiposity. The Journal of Clinical Investigation
130(11):5688–5702 DOI 10.1172/JCI140223.
Liperis G. 2013. The function of gametocyte specific factor 1 (GTSF1) in mammalian
oocyte and ovarian follicle development. PhD thesis, University of Leeds.
Love MI, Huber W, Anders S. 2014. Moderated estimation of fold change and dispersion
for RNA-seq data with DESeq2. Genome Biology 15(12):550
DOI 10.1186/s13059-014-0550-8.
Mank JE. 2017. The transcriptional architecture of phenotypic dimorphism. Nature
Ecology and Evolution 1(1):0006 DOI 10.1038/s41559-016-0006.
Mao X, He G, Zhang J, Rossiter SJ, Zhang S. 2013. Lineage divergence and histor-
ical gene flow in the Chinese horseshoe bat (Rhinolophus sinicus). PLOS ONE
8(2):e56786 DOI 10.1371/journal.pone.0056786.
Mayne BT, Bianco-Miotto T, Buckberry S, Breen J, Clifton V, Shoubridge C,
Roberts CT. 2016. Large scale gene expression meta-analysis reveals tissue-
specific, sex-biased gene expression in humans. Frontiers in Genetics 7:183
DOI 10.3389/fgene.2016.00183.
Meyfour A, Pahlavan S, Ansari H, Baharvand H, Salekdeh GH. 2019. Down-regulation
of a male-specific H3K4 demethylase, KDM5D, impairs cardiomyocyte differentia-
tion. Journal of Proteome Research 18(12):4277–4282
DOI 10.1021/acs.jproteome.9b00395.
Mizukami H, Kim JD, Tabara S, Lu W, Kwon C, Nakashima M, Fukamizu A. 2019.
KDM5D-mediated H3K4 demethylation is required for sexually dimorphic
gene expression in mouse embryonic fibroblasts. The Journal of Biochemistry
165(4):335–342 DOI 10.1093/jb/mvy106.
Naftaly AS, Pau S, White MA. 2021. Long-read RNA sequencing reveals widespread
sex-specific alternative splicing in threespine stickleback fish. Genome Research
31(8):1486–1497 DOI 10.1101/gr.274282.120.
Naurin S, Hansson B, Hasselquist D, Kim YH, Bensch S. 2011. The sex-biased brain:
sexual dimorphism in gene expression in two species of songbirds. BMC Genomics
12(1):1–11 Available at https://www.biomedcentral.com/1471-2164/12/37.
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 20/23
Nilsen TW, Graveley BR. 2010. Expansion of the eukaryotic proteome by alternative
splicing. Nature 463(7280):457–463 DOI 10.1038/nature08909.
Nishimura T, Sato T, Yamamoto Y, Watakabe I, Ohkawa Y, Suyama M, Kobayashi
S, Tanaka M. 2015. foxl3 is a germ cell–intrinsic factor involved in sperm-egg fate
decision in medaka. Science 349(6245):328–331
DOI 10.1073/pnas.1918556117.
Oliva M, Muñoz Aguirre M, Kim-Hellmuth S, Wucher V, Gewirtz AD, Cotter DJ,
Parsana P, Kasela S, Balliu B, Vinuela A, Castel SE, Mohammadi P, Aguet F, Zou
Y, Khramtsova EA, Skol AD, Garrido-Martin D, Reverter F, Brown A, Evans P,
Gamazon ER, Payne A, Bonazzola R, Barbeira AN, Hamel AR, Martinez-Perez
A, Soria JM, Pierce BL, Stephens M, Eskin E, Dermitzakis ET, Segre AV, Kyung
H, Engelhardt BE, Ardlie KG, Montgomery SB, Battle AJ, Lappalainen T, Guigo
R, Stranger BE. 2020. The impact of sex on gene expression across human tissues.
Science 369(6509):eaba3066 DOI 10.1126/science.aba3066.
Oxberry BA. 1979. Female reproductive patterns in hibernating bats. Reproduction
56(1):359–367 DOI 10.1530/jrf.0.0560359.
R Core Team. 2021. R: a language and environment for statistical computing. Version
4.0.5. Vienna: R Foundation for Statistical Computing. Available at https://www.r-
project.org .
Ren L, Wu C, Guo L, Yao J, Wang C, Xiao Y, Pisco AO, Wu Z, Lei X, Liu Y, Shi L, Han
L, Zhang H, Xiao X, Zhong J, Wu H, Li M, Quake SR, Huang Y, Wang J, Wang J.
2020. Single-cell transcriptional atlas of the Chinese horseshoe bat (Rhinolophus
sinicus) provides insight into the cellular mechanisms which enable bats to be viral
reservoirs. BioRxiv DOI 10.1101/2020.06.30.175778.
Rice WR. 1984. Sex chromosomes and the evolution of sexual dimorphism. Evolution
38(4):735–742 DOI 10.1111/j.1558-5646.1984.tb00346.x.
Rice WR. 1987. The accumulation of sexually antagonistic genes as a selective agent
promoting the evolution of reduced recombination between primitive sex chromo-
somes. Evolution 41(4):911–914
DOI 10.1111/j.1558-5646.1987.tb05864.x.
Rinn JL, Snyder M. 2005. Sexual dimorphism in mammalian gene expression. Trends in
Genetics 21(5):298–305 DOI 10.1016/j.tig.2005.03.005.
Rogers TF, Palmer DH, Wright AE. 2021. Sex-specific selection drives the evolution
of alternative splicing in birds. Molecular Biology and Evolution 38(2):519–530
DOI 10.1093/molbev/msaa242.
Rowe L, Chenoweth SF, Agrawal AF. 2018. The genomics of sexual conflict. The
American Naturalist 192(2):274–286 DOI 10.1086/698198.
Ruzicka F, Connallon T. 2020. Is the X chromosome a hot spot for sexually antagonistic
polymorphisms? Biases in current empirical tests of classical theory. Proceedings of
the Royal Society B 287(1937):20201869 DOI 10.1098/rspb.2020.1869.
Sharma E, Künstner A, Fraser BA, Zipprich G, Kottler VA, Henz SR, Weigel D, Dreyer
C. 2014. Transcriptome assemblies for studying sex-biased gene expression in the
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 21/23
guppy, Poecilia reticulata.BMC Genomics 15(1):1–21
DOI 10.1186/1471-2164-15-400.
Shen S, Park JW, Lu ZX, Lin L, Henry MD, Wu YN, Zhou Q, Xing Y. 2014. rMATS:
robust and flexible detection of differential alternative splicing from replicate RNA-
Seq data. Proceedings of the National Academy of Sciences of the United States of
America 111(51):E5593–E5601 DOI 10.1073/pnas.1419161111.
Siemers BM, Beedholm K, Dietz C, Dietz I, Ivanova T. 2005. Is species identity, sex, age
or individual quality conveyed by echolocation call frequency in European horseshoe
bats? Acta Chiropterologica 7(2):259–274 DOI 10.3161/150811005775162579.
Simmons NB, Cirranello AL. 2020. Bat species of the world: a taxonomic and geographic
database. Available at https://batnames.org.
Singh A, Agrawal AF. 2021. Two forms of sexual dimorphism in gene expression in
Drosophila melanogaster: their coincidence and evolutionary genetics. BioRxiv
DOI 10.1101/2021.02.08.429268.
Singh P, Ahi EP. 2022. The importance of alternative splicing in adaptive evolution.
Molecular Ecology 31(7):1928–1938 DOI 10.1111/mec.16377.
Stevens RD, Platt RN. 2015. Patterns of secondary sexual size dimorphism in New
World Myotis and a test of Rensch’s rule. Journal of Mammalogy 96(6):1128–1134
DOI 10.1093/jmammal/gyv120.
Suzuki R, Shimodaira H. 2006. Pvclust: an R package for assessing the uncertainty in
hierarchical clustering. Bioinformatics 22(12):1540–1542
DOI 10.1093/bioinformatics/btl117.
Telonis-Scott M, Kopp A, Wayne ML, Nuzhdin SV, McIntyre LM. 2009. Sex-specific
splicing in Drosophila: widespread occurrence, tissue specificity and evolutionary
conservation. Genetics 181(2):421–434 DOI 10.1534/genetics.108.096743.
Toubiana W, Armisén D, Dechaud C, Arbore R, Khila A. 2021. Impact of male trait
exaggeration on sex-biased gene expression and genome architecture in a water
strider. BMC Biology 19(1):1–17 DOI 10.1186/s12915-021-01021-4.
Trabzuni D, Ramasamy A, Imran S, Walker R, Smith C, Weale ME, Hardy J, Ryten M.
2013. Widespread sex differences in gene expression and splicing in the adult human
brain. Nature Communications 4(1):1–7 DOI 10.1038/ncomms3771.
Uhlén M, Fagerberg L, Hallström BM, Lindskog C, Oksvold P, Mardinoglu A, Siverts-
son A, Kampf C, Sjostedt E, Asplund A, Olsson I, Edlund K, Lundberg E, Navani
S, Szigyarto CA, Odeberg J, Djureinovic D, Takanen JO, Hober S, Alm T, Edqvist
P-H, Berling H, Tegel H, Mulder J, Rockberg J, Nilsson P, Schwenk JM, Hamsten
M, Feilitzen KV, Forsberg M, Persson L, Johansson F, Zwahlen M, Heijne GV,
Nielsen J, Pontén F. 2015. Tissue-based map of the human proteome. Science
347(6220):1260419 DOI 10.1126/science.1260419.
Vakilian H, Mirzaei M, Sharifi Tabar M, Pooyan P, Habibi Rezaee L, Parker L, Haynes
PA, Gourabi H, Baharvand H, Salekdeh GH. 2015. DDX3Y, a male-specific region
of Y chromosome gene, may modulate neuronal differentiation. Journal of Proteome
Research 14(9):3474–3483 DOI 10.1021/acs.jproteome.5b00512.
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 22/23
Wilkinson GS, Adams DM, Haghani A, Lu AT, Zoller J, Breeze CE, Arnold BD, Ball
HC, Carter GG, Cooper LN, Dechmann DKN, Devanna P, Fasel NJ, Galazyuk AV,
Gunther L, Hurme E, Jones G, Knornschild M, Lattenkamp EZ, Li CZ, Mayer F,
Reinhardt JA, Medellin RA, Nagy M, Pope B, Power ML, Ransome RD, Teeling
EC, Vernes SC, Zamora-Mejias D, Zhang J, Faure PA, Greville LJ, Herrera MLG,
Flores-Martinez JJ, Horvath S. 2021. DNA methylation predicts age and provides
insight into exceptional longevity of bats. Nature Communications 12(1):1–13
DOI 10.1038/s41467-021-21900-2.
Wu H, Jiang T, Huang X, Feng J. 2018. Patterns of sexual size dimorphism in horse-
shoe bats: testing Rensch’s rule and potential causes. Scientific Reports 8(1):1–13
DOI 10.1038/s41598-018-21077-7.
Xie L, Sun K, Jiang T, Liu S, Lu G, Jin L, Feng J. 2017. The effects of cultural drift on
geographic variation in echolocation calls of the Chinese rufous horseshoe bat
(Rhinolophus sinicus). Ethology 123(8):532–541 DOI 10.1111/eth.12627.
Yang X, Schadt EE, Wang S, Wang H, Arnold AP, Ingram-Drake L, Drake TA, Lusis AJ.
2006. Tissue-specific expression and regulation of sexually dimorphic genes in mice.
Genome Research 16(8):995–1004 DOI 10.1101/gr.5217506.
Yoshimura T, Watanabe T, Kuramochi-Miyagawa S, Takemoto N, Shiromoto Y, Kudo
A, Kanai-Azuma M, Tashiro F, Miyazaki S, Katanaya A, Chuma S, Miyazaki JI.
2018. Mouse GTSF 1 is an essential factor for secondary pi RNA biogenesis. EMBO
Reports 19(4):e42054 DOI 10.15252/embr.201642054.
Zhang J, Zhou X, Wan D, Yu L, Chen X, Yan T, Wu Z, Zheng M, Zhu F, Zhu H.
2022. TMPRSS12 functions in meiosis and spermiogenesis and is required for
male fertility in mice. Frontiers in Cell and Developmental Biology 10:757042
DOI 10.3389/fcell.2022.757042.
Zheng W, Xu H, Lam SH, Luo H, Karuturi RKM, Gong Z. 2013. Transcriptomic analyses
of sexual dimorphism of the zebrafish liver and the effect of sex hormones. PLOS
ONE 8(1):e53562 DOI 10.1371/journal.pone.0053562.
Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, Ben-
ner C, Chanda SK. 2019. Metascape provides a biologist-oriented resource
for the analysis of systems-level datasets. Nature Communications 10(1):1–10
DOI 10.1038/s41467-019-09234-6.
Chen et al. (2023), PeerJ, DOI 10.7717/peerj.15231 23/23
