Access to this full-text is provided by Wiley.
Content available from Ecology and Evolution
This content is subject to copyright. Terms and conditions apply.
14256
|
Ecology and Evolution. 2020;10:14256–14271.www.ecolevol.org
1 | INTRODUCTION
Extreme habitats often foster the evolution of adaptive features
(regressive evolution). Troglobites, a representative animal group
that has adapted to harsh environments, exhibit a set of sensory,
morphological, physiological, and behavioral trait s that have
arisen from long-term adaptation to the perpetual darkness of
the caves. The study on these traits will contribute not only to an
improved understanding of the genetic basis of this evolutionar y
process but also help to reveal the processes of speciation and the
function of individual genes. Loss of pigmentation, a character-
istic trait of cave-dwelling species, has received much attention
in recent years. However, the regressive evolution mechanisms
of reduction of skin pigmentation in cave animals need further
investigation (Jef fery, 2009). Some cave-dwelling creatures have
been studied to decipher the link between the pigmentation and
Received: 2 February 2020
|
Revised: 28 July 2020
|
Accepted: 28 October 2020
DOI: 10.1002/ece3.7024
ORIGINAL RESEARCH
Comparative transcriptomics reveals the molecular genetic
basis of pigmentation loss in Sinocyclocheilus cavefishes
Chunqing Li1 | Hongyu Chen1,2 | Yinchen Zhao2 | Shanyuan Chen1 | Heng Xiao1
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provide d the orig inal work is proper ly cited .
© 2020 The Authors. Ecology and Evolution published by John W iley & Sons Ltd
Chunqi ng Li and Hongy u Chen contribu ted equally to t his work .
1Yunnan Key Laboratory for Plateau
Mounta in Ecolog y and Res toration of
Degraded Enviro nment s, School of Ecolog y
and Environmental Sciences, Yunnan
University, Kunming, China
2School of Life Scie nces, Yunnan Univer sity,
Kunming, China
Correspondence
Shanyuan Chen and Heng Xiao, School
of Ecolog y and Environment al Sciences,
Yunnan University, Kunming 650500, China.
Emails: chensy@ynu.edu.cn (SC); xiaoheng@
ynu.edu.cn (HX)
Funding information
This stu dy was par tially suppor ted by the
Nationa l Natura l Science Foundation of
China (31560111) and the Top Young Talents
Program of Ten-Thousand Plan of Yunnan
Province (YNWR-QNBJ-2018-024).
Abstract
Cave-dwelling animals evolve distinct troglomorphic traits, such as loss of eyes,
skin pigmentation, and augmentation of senses following long-term adaptation to
perpetual darkness. However, the molecular genetic mechanisms underlying these
phenotypic variations remain unclear. In this study, we conducted comparative his-
tology and comparative transcriptomics study of the skin of eight Sinocyclocheilus
species (Cypriniformes: Cyprinidae) that included surface- and cave-dwelling species.
We analyzed four surface and four cavefish species by using next-generation se-
quencing, and a total of 802,798,907 clean reads were generated and assembled into
505,495,009 transcripts, which contributed to 1,037,334 unigenes. Bioinformatic
comparisons revealed 10,629 and 6,442 significantly differentially expressed uni-
genes between four different surface-cave fish groups. Further, tens of differentially
expressed genes (DEGs) potentially related to skin pigmentation were identified.
Most of these DEGs (including GNAQ, PK A, NRAS, and p38) are downregulated in
cavefish species. They are involved in key signaling pathways of pigment synthesis,
such as the melanogenesis, Wnt, and MAPK pathways. This trend of downregulation
was confirmed through qPCR experiments. This study will deepen our understanding
of the formation of troglomorphic traits in cavefishes.
KEYWORDS
cavefish, comparative transcriptomics, pigmentation, Sinocyclocheilus, troglomorphic trait
|
14257
LI et aL.
the underlying genetic mechanisms. For example, previous re-
search on cave tetra revealed that the loss-of-function mutation
of alleles of oca2 (Oculocutaneous albinism II) leads to albinism
in these fishes (Prot as et al., 20 06). Gross et al. discovered that
the “brown” phenotype of Astyanax mexicanus may be responsible
by the sequence mutations of the M C1R (melanocortin-1 recep-
tor) gene (Gross et al., 2009). Another study showed that albinism
in cave-dwelling planthoppers (Oliarus polyphemus) is caused by a
defect in the conversion of L-tyrosine to L-DOPA in the melanin
synthesis pathway (Bilandzija et al., 2012).
Color pat terns of fish skins are related to the distribution and
composition of pigment cells, including melanophores, erythro-
phores, xanthophores, iridophores, leucophores, and cyanophores,
which are derived from neural crest cells (Parichy, 2006). The for-
mation of fish skin pigmentation is orchestrated by multiple genes
and regulatory factors at different genetic levels. More than 125
pigment genes (e.g., TYR [tyrosinase], SILV [premelanosome protein
gene], SOX10 [SRY-box transcription factor 10], MITF [microphthal-
mia-associated transcription factor], MC1R ) have been identified
to perform critical functions in pigmentation in many fish species
(Parichy, 2006).
The wild f reshwater tel eost genus Sinocyclocheilus (Cypriniformes:
Cyprinidae) is a treasured endemic fish genus, distributed in the
karst regions of the east and northwest Yungui Plateau, Guangxi,
southwestern China. This genus is composed of more than 55 know
species (Meng et al., 2013). Owing to distinct differences in terms
of phenotype and habitat, both morphotypes (cave and sur face)
exist within one genus. High species diversity and skin phenotypic
variation make Sinocyclocheilus particularly suitable for studying the
molecular genetic mechanisms underlying the evolution of differen-
tial pigmentation. Previous studies on three Sinocyclocheilus species
(Sinocyclocheilus graham [surface fish], Sinocyclocheilus rhinocerous
[cavefish], and Sinocyclocheilus anshuiensis [cavefish]) have shown
that the expression of oca2, Ty r, Tyr p1 (tyrosinase-related protein1),
and Mpv17 (mitochondrial inner membrane protein 17) was lower
in the skin of Sinocyclocheilus cavefish than that in the surface fish
(Yang et al., 2016). However, considering the high species diversity
and habit at differences, it is necess ary to study mor e Sinocyclocheilus
fish species to understand additional evolutionary mechanisms un-
derlying pigment loss.
In this study, we conducted comparative histology and
comparative transcriptomics analysis to examine pheno-
typic data and transcript profiles from the skins of four sur face
fishes (Sinocyclocheilus maculatus, Sinocyclocheilus qiubeiensis,
Sinocyclocheilus jii, and Sinocyclocheilus oxycephalus) and four cave-
fish species (Sinocyclocheilus tianlinensis, Sinocyclocheilus brevibarba-
tus, Sinocyclocheilus broadihornes, and S. rhinocerous), all belonging to
genus Sinocyclocheilus. The main goals of this study were to (a) obt ain
phenotypic data of different body colors at the histological level,
and (b) obtain the skin transcriptome profiles of four Sinocyclocheilus
cave-dwelling and four Sinocyclocheilus surface fish species with
Illumina sequencing technology, (c) select pigmentation-related dif-
ferentially expressed genes (DEGs) by comparing four cave-surface
fish groups based on skin transcriptome profiles, and, finally, (d) to
identif y and validate key candidate genes linked to the dif ferential
skin pigmentation between cave and surface fish species through
pathway and function annotation of DEGs and qPCR. Our results are
expected to provide novel insight s into the transcriptional regulation
of pigment-related genes in adaptive evolution, leading to a better
understanding of the molecular regulation mechanisms of troglo-
morphic traits in cavefishes.
2 | MATERIALS AND METHODS
2.1 | Sample collection and photograph with a
digital camera
The eight Sinocyclocheilus species were captured in Yunnan
Province and Guangxi Zhuang Autonomous Region, China, with
three biological replicates for each species (Table 1). Pictures were
taken of fresh fish immediately after capture. After general anes-
thesia with 30 mg/L of MS-222 anesthetic (3-aminobenzoic acid
ethyl ester methanesulfonate; Sigma-Aldrich), all lateral skin tis-
sues were surgically excised, divided into two sections, collected
into two sterile tubes for comparative histological and transcrip-
tomics (tissue immersed into RNAlater) analysis, and placed in
liquid nitrogen. After euthanizing the fish, other tissues (muscles)
and organs (heart, liver, brain, etc.) were collected for other stud-
ies. The samples were stored at −80°C in the laboratory. Since
TABLE 1 Sampling details of each Sinocyclocheilus species
Species Location Typ e
Sunlight intensit y
in Habitat Gender Age Number
MSinocyclocheilus maculatus N:23°4′; E:104°16′ Surface High All male Adult 3
QSinocyclocheilus qiubeiensis N:24°05′; E:104°13′ Surface High All male Adult 3
ji Sinocyclocheilus jii N:25°20′; E:110°9′ Surface High All male Adult 3
JSinocyclocheilus oxycephalus N:24°48′; E:103°18′ Surface High All male Adult 3
TSinocyclocheilus tianlinensis N:24°60′; E:106°30′ C ave No sunlight All male Adult 3
DSinocyclocheilus brevibarbatus N:24°10′; E:108°10′ Cave No sunlight All male Adult 3
KSinocyclocheilus broadihornes N:24°48′; E:103°18′ Cave No sunlight All male Adult 3
XSinocyclocheilus rhinocerous N:24°46′; E:104°17′ Cave We ak All male Adult 3
14258
|
LI et aL.
the cavefish die soon af ter being removed from the cave habitat,
to avoid RNA degradation, sterile instruments were prepared in
advance, and all tissue sampling was performed in the field im-
mediately after the fish were collected from the water. During the
sampling process, all animal experiments were conducted with the
approval of the local Ethical Committee at Yunnan University, in
accordance with China's local and global ethical policies (Grant No:
ynucae 20190056), and all procedures were approved and assisted
by the local government.
For surface fish, we took fish images immediately after fishing.
However, for cavefish, the sampling is done in deep caves (com-
pletely dark), and therefore, it cannot be photographed immediately
(flash photography would produce chromatic aberrations). Thus, we
temporarily placed the fish in a portable fishing case and suppor ted
with oxygen using a small aeration pump. Once outside the cave,
the cavefish was sheltered from light the whole time to prevent skin
pigment deterioration, except for taking photos under natural light.
2.2 | Paraffin sections
The lateral skin samples from collected fishes were cut into
0.8 cm × 1.0 cm × 0.2 cm pieces and fixed with 10% neutral for-
maldehyde for 10‒24 hr, washed, and dehydrated through a graded
series of ethanol (50%, 70%, 80%, 95%, and 100%). Paraffin em-
bedding was carried out using a paraffin embedding station (Leica
EG1140H; Leica Microsystems). The paraffin blocks were sliced into
6-μm-thick sections (Leica RM2016; Leica Microsystems), dried at
38°C for 7‒24 hr in the oven, and stained with haematoxylin–eosin.
Sections were imaged using an optical microscope (OLYMPUS BX
51; Olympus).
2.3 | RNA quantification and quality
RNA purity was checked using a NanoPhotometer® spectro-
photometer (IMPLEN), and RNA concentration was measured
using a Qubit® RNA Assay Kit in a Qubit
®2.0 Fluorometer (Life
Technologies). RNA samples were run on 1% agarose gels, and their
integrity was checked using an RNA Nano 6000 Assay Kit and the
Agilent Bioanalyzer 2100 system (Agilent Technologies).
2.4 | Library preparation for
transcriptome sequencing
A total of 3 μg of RNA sample from each fish was used as input mate-
rial. Sequencing libraries were generated using a NEBNext®Ult ra™
RNA Library Prep Kit for Illumina® (New England BioLabs) follow-
ing the manufacturer's recommendations, and index codes were
added to attribute sequences to each sample. Briefly, mRNA was
purified from total RNA using poly-T oligo-attached magnetic beads.
Fragmentation was carried out using divalent cations at elevated
temperature in NEB Next First Strand Synthesis Reaction Buffer
(5×). The first-strand synthesis of complementary DNA (cDNA) was
completed using random hexamer primer and M-MuLV Reverse
Transcriptase (RNase H-). Nex t, the synthesis of second-strand
cDNA was carried out using DNA Polymerase I and RNase H. Then,
the synthesized cDNA was subjected to end repair and remaining
overhangs were converted into blunt ends via polymerase treat-
ment. After adenylation of the 3′ ends of DNA fragments, NEBNext
Adaptor with a hairpin loop structure was ligated to prepare for
hybridization. The size of target fragments selected for cDNA li-
brary was 150–200 bp in length. The fragments were purified with
AMPure XP system (Beckman Coulter). Then, size-selected, adap-
tor-ligated cDNA was incubated at 37°C for 15 min, followed by
5 min at 95°C by using USER Enzyme (3 μl; New England BioLabs)
before PCR. Finally, the library quality was assessed on the Agilent
Bioanalyzer 2100 system after PCR using Phusion High-Fidelit y
DNA polymerase, universal PCR primers, and index (X) primer (New
England BioLabs) and product purification (AMPure XP system;
Beckman Coulter).
2.5 | Clustering and sequencing
Clone clusters were generated on Illumina cBot Cluster Generation
System, using TruSeq PE Cluster Kit v3-cBot-HS (Illumina), according
to the manufacturer's instructions. After cluster generation, high-
throughput sequencing of library preparations was performed on
Illumina Hiseq 2000, and paired-end reads were generated.
2.6 | Data analysis
The raw reads were cleaned by removing reads containing adapter,
reads containing poly-N, and low-quality reads through in-house
Perl scripts. Furthermore, to ensure the high quality of the data used
for downstream analyses, the quality (Q20, Q30, GC-content, and
sequence duplication level of the clean data) of the clean data was
determined. Clean and high-quality transcriptome from the eight
fish species were assembled using Trinity (min_kmer_cov: 2; other
parameters: default values) (Grabherr et al., 2011). The whole se-
quence data were submitted to the DRYAD database (https://datad
r y a d . o r g / s t a s h / s h a r e / t 5 c Z I X o V U g y h p z E P 6 z - G N 6 x j c 5 E U 3 T v P P E
wbdIo 7siI).
2.7 | Gene function annotation
Gene functions of unigenes were annotated based on the homology
searches of the following major public databases: NCBI nonredun-
dant protein (nr), protein families (Pfam), euKaryotic Orthologous
Groups (KOG)/Clusters of Orthologous Groups of proteins (COG)/
nonsupervised orthologous groups (eggNOG), Swiss-Prot (a manu-
ally annot ated and reviewed protein sequence database), Kyoto
|
14259
LI et aL.
Encyclopedia of Genes and Genomes (KEGG), and Gene Ontology
(GO).
2.8 | Differential expression analysis
According to different pairing modes, we investigated more than
30 combinations of pair-group among the eight species, and four
species pairs (M vs. K , Q vs. X, ji vs. T, J vs. D; Table 1) were ran-
domly selected for further description. The target genes were
identified according to the following three criteria: (a) genes with
expression log2
|Foldchange| ≥ 2 an d fa l s e di s c over y ra tes (FDR) ≤0 . 0 01
were considered to be dif ferentially expressed. (b) The functions
and pathways of these DEGs were involved in the formation of
fish skin pigmentation. (c) The regulation trends (upregulation or
downregulation) of these DEGs must be consistent across at least
three of the four groups. Differential expression analysis of all four
groups was performed using the DESeq R package (1.10.1) (Love
et al., 2014). DESeq analyses count data from high-throughput se-
quencing to determine differential expression using a model based
on a negative binomial distribution. The resulting P-values were
adjusted using the Benjamini–Hochberg procedure to factor in
FDR. DEGs were determined by comparing levels of gene expres-
sion within each group after using Bowtie and RSEM to calculate
fragments per kilobase of exon per million mapped reads (FPKM)
(Langmead et al., 2009; Li & Dewey, 2011). Genes with expression
log2
|Foldchange| ≥ 2 and FDR ≤ 0.001 were considered to be differ-
entially expressed.
2.9 | DEG functional annotation
GO enrichment analysis of DEGs was implemented by the topGO
R packages (Alexa & Rahnenfuhrer, 2006), using a Kolmogorov–
Smirnov test. KEGG (Kanehisa et al., 2004) is a database for un-
derstanding high-level functions and utilities of biological systems,
such as the cell, organism, and ecosystem, from molecular-level
information, especially large-scale molecular datasets generated
by genome sequencing and other high-throughput experimental
technologies (http://www.genome.jp/kegg/). We used KOBAS (Xie
et al., 2011) soft ware to test the statistical enrichment of DEGs in
KEGG pathways.
2.10 | Validation of DEGs by qPCR
RNA-seq results of a total of 9 DEGs were validated by qPCR; β-actin
(housekeeping gene) was used as an internal reference. qPCR analy-
sis using the same RNA samples as for the transcriptome profiling
was performed using qTOWER2.2 (Analytik Jena, Jena, Germany);
the primer sequences are listed in (Table S1). The amplification con-
ditions were as follows: initial denaturation at 95°C for 3 min, fol-
lowed by 39 cycles of denaturation at 95°C for 10 s and extension
at 58°C for 30 s. Finally, melting curves were generating from 60
to 95°C. All reactions were performed with three technical and
three biological replicates. Relative gene expression was calculated
with the 2−ΔΔCT method using qPCRsoft 3.2 software (The value of
threshold cycle Ct was used to calculate the DEG expression fold
change. ΔCt = Cttar get genes − Ctβ-actin, ΔΔCt = ΔCtcontrol − ΔCtIndicated
condition, log2
(FC) = 2−(ΔΔCt).
2.11 | Weighted gene coexpression network
analysis (WGCNA)
We used the R package WGCNA for weighted correlation network
analysis. By using gene expression data from the skin tissue of
cave and sur face fish, we constructed a coexpression network to
find additional important genes associated with each skin color
phenotype. First, we constructed a gene–gene similarity network
(Pearson's correlation) for all unigenes. In this analysis, we filtered
unigenes with expression quantity <1 to calculate the optimal
power value, and transformed these values into an adjacency ma-
trix under the soft power (beta = 7). Next, all unigenes were hier-
archically clustered, and the network was divided into modules.
Gene modules corresponding to the branches cutoff of the gene
tree were color-coded (networkType: signed; soft power: 7; min-
ModuleSize: 50; minKMEtoStay: 0.3; mergeCutHeight: 0.20). The
“gray” module contained unigenes that could not be associated
with any expression patterns. To find the core genes (hub genes)
that are significantly associated with the phenot ypic trait, we fo-
cused on the most highly correlated modules (R2>|.5|, p < .005)
and the top three most correlated genes were selected for further
analysis.
3 | RESULTS
3.1 | Color observations in Sinocyclocheilus skin
The eight Sinocyclocheilus species were photographed in the liv-
ing state with a digital camera (Figure 1), and skin color was noted.
Based on the body color, the four surface fish species were di-
vided into three different subtypes within the same type of habi-
tat. Sinocyclocheilus maculatus skin has a large number of black
flecks and appears golden yellow, whereas the skins of S. qiubeiensis
and S. oxycephalus exhibit yellow coloration but with fewer flecks.
Distinct from the other three surface fish species, S. jii has a char-
coal gray body color with little black spots. The four cave-dwell-
ing Sinocyclocheilus species were divided into two color subtypes:
S. broadihornes and S. rhinocerous belong to the gray translucent
body color type, and S. tianlinensis and S. brevibarbatus belong to the
pink translucent body color type.
The photographs taken of the same area on the lateral skin
of the eight species using a stereoscopic microscope showed the
presence of black dots composed of melanocytes in all surface
14260
|
LI et aL.
fishes. However, the melanocytes were not as evident in the cave-
fish. For the four surface fish species, the difference in the light
used for illumination at the time of taking the photograph may
explain the differences in melanocyte shape. The star-like shape
of melanocytes of S. qiubeiensis and S. jii may be due to the expan-
sion of intracellular melanosomes in melanocytes under low light
conditions. Interestingly, the melanocytes in cavefish differed
significantly in the distribution patterns, melanocytes abundance,
and intracellular arrangement of melanosomes. The melanocytes
in S. broadihornes and S. brevibarbatus skins were star-shaped,
and those in S. rhinocerous and S. tianlinensis had dendritic shape.
Furthermore, the fine-lined melanocytes in S. tianlinensis skin
were the faintest among the four cavefish species observed. The
melanocytes of the four sur face fish species were thicker and
darker than those of the cavefish species, and fewer melanin gran-
ules were observed in the cavefish skin (Figure S1). The detailed
distribution and composition patterns of melanocytes are shown
in Figure 2.
3.2 | Sequencing and de novo assembly
Sequencing of the skin transcriptomes of the eight Sinocyclocheilus
species using the Illumina Hiseq 2000 platform produced
802,798,907 clean reads and 237,946,156,516 bases from the
constructed sequencing libraries. The read and base numbers, GC-
content, and other parameters are presented in Table S2. High-
quality reads were de novo assembled using Trinit y, generating
505,495,009 transcripts. All transcripts contributed to 1,037,334
unigenes with an average length of 820.24 bp and N50 of 1,379 bp.
Information on sequence length distribution and other details of
transcripts and unigenes are shown in Table S3.
3.3 | Unigene functional annotation
A total of 131,766 unigenes from the eight species of Sinocyclocheilus
were annotated using several databases (including Nr, GO, COG,
FIGURE 1 Digital photographs of eight
Sinocyclocheilus species in living state
|
14261
LI et aL.
and KOG, Table S4). The BLASTx similarity analysis of the unigenes
against the NCBI Nr database found matches for most of the uni-
genes (126,654) in the Nr database. Most of the unigenes had the
highest homology with protein sequences of seven fish species
(Figure 3a): Danio rerio (75.04%), A. mexicanus (5.19%), Oncorhynchus
mykiss (2.37%), Cyprinus carpio (1.33%), Oreochromis niloticus (1.27%),
Esox lucius (1.22%), and Notothenia coriiceps (0.85%). To further an-
notate the functions of unigenes, three databases (COG, eggNOG,
and KOG) were aligned to unigenes. The results of the functional
classification of unigenes were as follows. In all three databases, the
cluster of general function prediction (R) had the largest proportion.
The proportion of signal transduction mechanisms (T) was second
in the NOG and KOG databases, whereas, in the COG database,
only a few unigenes were annotated in this functional class. In ad-
dition, the proportion of some clusters showed a consistent trend
across the three databases in cell motility (N), nuclear structures (Y)
and cell wall/membrane/envelope biogenesis (M), posttranslational
modification, protein turnover, and chaperones (O), and transcrip-
tion (K). However, the proportions of most categories in these three
databases were different, especially in S (function unknown) and L
(replication, recombination, and biogenesis; Figure S2). Further, GO
terms were assigned to 67,419 unigenes using GO database annota-
tion. The most abundant GO terms in the GO classes were binding
(GO: 0005488) and catalytic activity (GO: 0003824) in MF (molecu-
lar function); cell (GO: 0005623) and cell part (GO: 0044464) in CC
(cellular component); and cellular process (GO: 0009987) and meta-
bolic process (GO: 0008152) in BP (biological process; Figure 3b).
To further identify the biological pathways of these unigenes, they
were mapped to the KEGG database. A total of 60,928 unigenes
were mapped to 285 known KEGG pathways. The top five KEGG
pathways with the highest number of unigenes were the MAPK sign-
aling pathway (Ko04010), endocytosis (Ko04144), focal adhesion
(Ko04510), regulation of actin cytoskeleton (Ko04810), and calcium
signaling pathway (Ko04020). Furthermore, pigment-related path-
ways such as melanogenesis (Ko04916) and the Wnt signaling path-
way (Ko04310) were included in the top 30 pathways (Figure S3).
3.4 | DEG analysis
DEGs were identified in the four fish groups (M vs. K, Q vs. X , ji vs.
T, J vs. D).
FIGURE 2 The stereoscopic
microscopic photographs of skin paraffin
sections of eight Sinocyclocheilus species.
Melanophores are indicated by blue
arrows (magnification 40×). EL, epidermal
layer; DL, dermal layer. Melanocytes were
indicated with blue arrow
14262
|
LI et aL.
We obtained a total of 35,101 DEGs from the four fish groups.
M vs. K group had the highest number of DEGs (10,629 DEGs:
4,332 upregulated and 6,297 downregulated) followed by Q vs.
X (9,635 DEGs: 3,628 upregulated and 6,007 downregulated)
(Figure 4b). In every surface fish and cavefish group compared,
the proportion of downregulated genes was higher than that of
upregulated genes (Table S5). As shown in the Venn diagram, Q
vs. X and ji vs. T had the most shared DEGs (their habitats are not
more similar than other groups) The more details of unique and
shared DEGs between each surface-cave group are displayed in
Figure 4a.
3.5 | DEG functional and pathway analysis
in the databases
GO, KEGG, and COG annotations were performed to identify the
functions and biochemical pathways of DEGs to further filter the
candidate genes that may be responsible for differences in skin pig-
ment (Table S5). DEGs in all four groups were classified into three
functional categories (biological process, cellular component, and
molecular function) in the GO database. Most of the DEGs of the
four groups were annotated to seven GO subcategories, namely
“cellular process,” “single-organism process,” “metabolic process,”
FIGURE 3 Annotation of unigenes form the eight species of Sinocyclocheilus, according to Nr and GO. (a) Similarity analysis of
the unigenes against nonredundant Nr database; (b) Gene Ontology (GO) classification of unigenes for skin transcriptome of eight
Sinocyclocheilus
|
14263
LI et aL.
“cell,” “cell part,” “binding,” and “catalytic activity.” However, partial
subcategories were only annotated in some groups; for example,
the subcategory “channel regulator” was annotated only in groups
M vs. K and Q vs. X, and “extracellular matrix part” was found only
in groups ji vs. T and J vs. D. (Figure S4). GO enrichment analysis
of DEGs with Kolmogorov–Smirnov tests revealed that six GO sub-
categories, homophilic cell adhesion via plasma membrane adhe-
sion molecules, ionotropic glutamate receptor signaling pathway,
extracellular region, structural constituent of ribosome, ionotropic
glutamate receptor activity, and extracellular-glutamate-gated ion
channel activity were significantly enriched (p < .001) in at least
three of the four groups (Table S6).
To further study the function and BPs of DEGs, and identify tar-
get genes annotation of DEGs (four groups) was performed using
KEGG pathways (Figure S5). Under the category of the biochemi-
cal pathway in the KEGG classification, DEGs in all four compari-
son groups were associated with cellular processes, environmental
information processing, genetic information processing, human dis-
eases, metabolism, and organismal systems. Furthermore, the KEGG
subcategories with a high proportion of DEGs in the four groups
were endocytosis, protein processing in the endoplasmic reticulum,
ribosome, and MAPK signaling pathway. Interestingly, some path-
ways showed differences among the four groups, such as the car-
diac muscle contrac tion pathway, which was annotated in all groups
except for the Q vs. X group, and the RIG-l-like receptor signaling
pathway and the fatt y acid met abolism, which were only observed
in the ji vs. T and M vs. K groups, respectively. In this study, DEGs
in all four groups were partially annotated to some pigment-related
pathways such as the melanogenesis, MAPK signaling, and Wnt sig-
naling pathways.
3.6 | Target DEGs selection and expression
trend validation
After the functional analysis and pathway annotation of DEGs in
the four groups, we selected target genes according to the restric-
tions stated above. Five DEGs were found in the melanogenesis
FIGURE 4 (a) Venn diagram displaying
the number of differentially expressed
genes (DEGs) unique to or shared between
each surface-cave group Numbers in each
section of the figure show the number
of DEGs (log2|Foldchange| ≥ 2 and false
discovery rates (FDR) of ≤0.001). (b)
Number of differentially expressed genes
of four surface-cave fish groups. X-axis
shows the differences in expression given
as the log values, and the Y-axis shows
significant differences in expression as
negative log values in the Volcano plot;
upregulated genes and downregulated
genes are indicated by red dots and green
dots, respectively. Nondifferentially
expressed genes are indicated by black
dots. (a) M vs. K; (b) Q vs. X; (c) Ji vs. T; (d)
J vs. D
14264
|
LI et aL.
TABLE 2 Differentially expressed genes (DEGs) related to pigmentation in each surface-cave group
Pathway
Gene name FDR
MvsK QvsX ji vsT JvsD Mv sT MvsD MvsX Q vsT QvsD QvsK jivsD jivsK jivsX J vsT JvsK JvsX
Melanogenesis log2
|FC| log2
|FC|
GNAQ <0.001 −6.14 − 4. 76 −4. 59 −4.41 <|2.0 | −Inf −6.10 −2 .62 −8.02 <|2.0| −5.34 −3.66 −5.14 −5.86 −2.13 <|2.0|
AC <0.001 −5.04 −4.46 −4.26 −Inf −3.30 −Inf <|2.0| <|2.0| −Inf 1.37 −Inf −3.0 0 −3.31 <|2.0| −2 .55 <|2.0|
NRAS <0.001 −Inf −5. 53 −3.49 −4.80 −2.24 −3.90 −Inf −2.94 −Inf −Inf −5.39 −Inf −Inf −3.03 −Inf −2 .45
SFRP2 <0.001 −5.91 3.96 6 .17 8.51 <|2.0| 3.20 <|2.0| 5.61 4.56 5.13 <|2.0| <|2.0| 3.24 3.18 3.83 2.84
PKA <0.001 −5.73 −6.33 <|2.0| −3.68 <|2.0| −2.14 <|2.0| <|2.0| −5.24 −2 .16 −6.20 −4.55 −5.68 −3.39 <|2.0| −2 . 35
Pathway Gene name FDR
M versus K Q versus X ji versus T J versus D
log2
|FC|
Wnt signaling pathway DAAM1 <0.001 −5.38 −4.81 −6.10 −2 .58
MAPK signaling pathway TAO <0.001 −5.57 −8 .14 −7. 2 4 −5.39
P38 <0.001 −6.54 −5.00 −5.47 −3.63
Apoptosis BID <0.001 −5.27 −5.5 −4.58 −Inf
BAD <0.001 −4.76 −4.78 −4.45 −Inf
BAX <0.001 −3. 27 −Inf −4 .41 −4.30
Note: “Inf” and “−Inf” indicate that the gene is not expressed in the cavefish or surface fish. “<|2.0|” indicate that the log2
|FC|<2 (no significant difference)
|
14265
LI et aL.
pathway, and the expression of these genes in all sur face-cave pairs
was checked. Furthermore, seven DEGs involved in Wnt signaling,
MAPK signaling, and apoptosis pathways, which are putatively in-
volved in skin pigmentation, were selected from a total of 35,101
DEGs. Except for SFRP2, most of the DEGs were downregulated in
Sinocyclocheilus cavefishes (Table 2).
The expression of a total of 9 selected DEGs was verified by
qPCR to confirm the accuracy of the RNA-Seq data (Figure 5).
3.7 | Weighted gene coexpression network analysis
(WGCNA)
After filtering of genes with low levels of expression, the modules
were divided based on clustering genes with similar expression pat-
terns and calculating the correlation between the eigenvalue and
traits of each module. The coexpression network was used to link
gene expression to five skin color phenotypes quantitatively. As a
result, all analyzed genes resulted in a total of 35 color modules. We
found that the color subtypes of each cavefish and sur face fish have
different highly correlated modules (Figure 6). Here, we focused on
the five most correlated modules: “steel blue,” “dark green,” “grey60,”
“green-yellow,” and “black” (R2 > |.5|; p < .005) to determine hub
genes. The top three hub genes based on eigengene connectivity in
each module may play an important role in each trait. The details and
functions of these genes are presented in Table 3.
4 | DISCUSSION
For cave-dwelling fish species, the loss of evolutionarily undesir-
able traits is pivotal to control energy cost and may be one of the
reasons for the convergent evolution of regressive traits; however,
the regulatory mechanism underlying this process is complex (Protas
et al., 2007). Studies on skin pigments in A. mexicanus have repor ted
that different genes determine the degrees of skin pigment loss
(brown and albino) (Gross et al., 2009; Protas et al., 2006). Besides,
the mutations in these genes may be pleiotropic. Helena et al. found
that the albinism of A. mexicanus caused by loss-of-function mu-
tations in the oca2 gene is a by-product of catecholamine-related
behaviors (anesthesia resistance), which is related to the capacity
to be responsive to stimuli in the caves (reduced sleep) (Helena
et al., 2018), which indicates that the cause of reduced pigmentation
of cavefish may be more complicated than earlier believed, and there
may be additional, yet unidentified pigment-related genes.
The link between loss of pigmentation and the underlying ge-
netic mechanisms in cave-dwelling fishes c an be bet ter understood
by investigating the pattern of transcriptional regulation of genes
that may cause loss of skin pigmentation in Sinocyclocheilus cavefish.
In this study, we selected eight representative (skin color and habitat
type) Sinocyclocheilus species, which included four surface species
(with func tional eyes and pigmentation) and four cave species (with
reduced or absent eyes and pigmentation) to evaluate the pheno-
typic differences in skin pigmentation between the two ecological
FIGURE 5 Quantitative real-time PCR (qRT-PCR) validation of differentially expressed genes (DEGs) in each surface-cave group. The
expression profile of nine genes was analyzed by qRT-PCR using the same RNA as utilized for RNA-Seq. Each average RNA-Seq expression
value was plotted against the corresponding qRT-PCR value and fitted into a linear regression (conducted by Prism 8.4.0.)
-8 -6 -4 -2 0
-6
-4
-2
0
q-PCR(Log2)
M vs K
RNA-Seq(Log2)
R squared 0.5116
-10 -5 510
-10
-5
5
10
ji vs T
q-PCR(Log2)
RNA-Seq(Log2)
R squared 0.7083
-10-
55
-10
-5
5
Q vs X
RNA-Seq(Log2)
q-PCR(Log2) R squared 0.5284
-10-5510
-5
5
10
J vs D
q-PCR(Log2)
RNA-Seq(Log2)
R squared 0.8641
14266
|
LI et aL.
types of Sinocyclocheilus species. We aimed at identifying and con-
firming key candidate genes linked to the diverse skin pigmentation
between cave and sur face fish species through comparative tran-
scriptomics analysis.
4.1 | Relationship between skin color and habitat of
Sinocyclocheilus species
Skin color can be influenced by environmental interactions (Leclercq
et al., 2010). During the long-term sampling process and looking at
lots of other of species within this genus, we found that the skin
color of Sinocyclocheilus species varies widely among species of the
surface also cave-dwelling fish, but remains consistent within popu-
lations. The eight representative Sinocyclocheilus species selected in
this study covered almost all the color types we have observed in
the wild. Our comparative histological analysis of the two ecotypes
of Sinocyclocheilus species clearly indicated the differences in dis-
tribution patterns, quantity, and melanosome intracellular arrays of
melanocytes in the skin between cavefish species and surface fish
species, suggesting that different habitat s may be one of the crucial
reasons for this difference. Melanocytes synthesize melanin to pro-
tect skin by absorbing ultraviolet radiation (UV) (Kim et al., 2018).
Previous s tudies on fish , such as whitefis h larvae (Coregonus lavaretus)
(Winberg, 2000), red sea bream (Pagrosomus major) (Kumai, 2005),
and rainbow trout (O. mykiss; Lit tle, 1995) showed that the concen-
tration of skin melanin and eumelanin precursor increase in response
to UV exposure to adapt to the changing environment. Melanin can
absorb specific wavelengths of light to protect the skin against UV
radiation, resulting in the dark gray skin of zebrafish (Kelsh, 2004).
However, troglobites live in completely enclosed underground habi-
tats characterized by permanent darkness, and complete lack of
autochthonous photosynthesis, leading to limited food resources
(Bussotti et al., 2018). When adapting to these extreme environ-
ments, the accumulation of melanin in cavefish skin is a waste of lim-
ited resources (Culver & Pipan, 2009). Interestingly, among the four
cavefish species, digital photography showed that the skin color of
S. rhinocerous was the darkest but still transparent, and the stereo-
scopic microscopic photographs and paraffin section also confirmed
that the number of melanocytes in the S. rhinocerous was greater
than that in the other cavefish species. Li et al. (200 0) studied the
cave habitats at sampling sites of S. rhinocerous and found that this
FIGURE 6 Coexpression modules
conducted by WGCNA. (a) hierarchical
cluster tree constructed by WGCNA
shows coexpression modules (the gray
module represent s genes that are not
assigned to specific modules). Each
branch in the tree point connec ts to a
gene. Genes were assigned to network
modules by dynamic tree cut after the
dynamic tree cut algorithm has been used
to identify all modules. minModuleSize:30
(lowest number of genes in each module)
and Merged dynamics (the modules with
similar expression patterns (80%) were
then merged.); (b) Trait–module associated
heat map. Column: traits; row: modules.
The number in each grid represents
the correlations between the module
and corresponding traits (correlation
coefficients) and p-value (in parentheses).
Color of the grid indic ates the correlation
(the deeper the color, the stronger the
correlation)
|
14267
LI et aL.
species occupies funnel-shaped doline caves, which partially receive
solar radiation; in contrast, other cavefishes, such as Sinocyclocheilus
altishoulderus, live in completely enclosed underground rivers. This
half-open environment may partially explain its unique skin color.
Based on comparative histolog y, we divided the Sinocyclocheilus
cavefish and surface fish body colors into two and three subt ypes,
respectively. The coexpression network of WGCNA was used to link
gene expression with five skin color phenotypes quantitatively. The
“steel blue” module and the “dark green” module were positively cor-
related with the cavefish gray and pink color traits, respectively, indi-
cating that the two modules might play an important role in cavefish
pigmentation. We found hub genes among the top three hub genes,
C1q-li ke (complement C1q) and FN1 (fibronectin 1), in the “gray-steel
blue” module whose expression was related to solar ultraviolet radi-
ation (UVR). Mei et al. found that C1q-like is involved in UV-induced
apoptosis in zebrafish. After the UV exposure, the transcripts of
C1q-like were upregulated 3–4 fold (Mei et al., 2008). Fibronectin is
a globular glycoprotein ubiquitous in the dermal extracellular matrix
(ECM), and exposure to UV irradiation modulates FN1 expression,
thereby enhancing ECM degradation (Hibbert et al., 2017; Parkinson
et al., 2015). Fur thermore, in the “pink-dark green” module, PLEC
(plectin) is considered the top hub gene. Previous studies of human
skin color have shown that the attenuated expression of PLEC leads
to increased melanosome uptake by keratinocytes and skin hyper-
pigmentation after UVA exposure (Coelho et al., 2015). This finding
may reconfirm that the differences in body color bet ween cavefish
in this study are related to the UV intensity among different habitats.
Wild Sinocyclocheilus species establish complex interactions
with their habitats, which may require the development of differ-
ent mechanisms or regulatory genes for each skin color subtype of
cavefish such as in S. rhinocerous and S. altishoulderus. Furthermore,
some surface fish or cavefish may occasionally change their habi-
tat through the underground river for predation (Chen et al., 2019),
which may be part of the reason for the diversity of body colors
between the same type of fish. Complex wild habitats may have di-
verse influencing factors such as seasonal changes and diet, which
may affect the formation of host skin pigmentation. Nevertheless,
further research is needed to reveal the exact mechanisms underly-
ing these variations.
4.2 | DEGs related to loss of pigmentation of
Sinocyclocheilus fish
The distribution of the genus Sinocyclocheilus is very narrow, and
the habitat of each species is merely a single cave or one sur face
waterbody (Lunghi et al., 2019). Furthermore, a large number of
variables (age, geographical location, and species) may affect gene
transcription. As such, in the grouping, these surface-cave fish pairs
were combined only based on the “habitat type” to find the DEGs
of each group, before comparisons were made. We believe that the
candidate genes selected in this way could truly reflect the relation-
ship between skin pigmentation and the two habitat types (cave and
surface waterbody), rather than the effect of other variables.
TABLE 3 TOP 3 Hub genes identified by WGCNA in each skin color traits
Skin Color-module Unigene id Gene Gene function
gray-steel blue (Cavefish) BMK_Unigene_053214 C1q-like Inhibit the apoptosis et al
BMK_Unigene_030559 fndc5 Fulfill manifold functions in tissue development and regulation of
cellular metabolism.
BMK_Unigene_030561 FN1 Fibronectins are involved in cell adhe sion, cell motilit y,
opsonization, wound healing, and maintenance of cell shape.
pink-dark green (Cavefish) BMK_Unigene_033481 PLEC Cytoskeleton-associated protein which links the keratin
intermediate filaments to the transmembrane proteins of the
hemidesmosomes.
BMK_Unigene_276989 Tpbgl Trophoblast glycoprotein-like
BMK _Uni gene _160 469 CR LF1 In complex with CLCF1, forms a heterodimeric neurotropic
cytokine that plays a crucial role during neuronal development
Golden-grey60 (Surface fish) B MK _Un ige ne_176217 H1GD1A Subunit of cytochrome c oxidase
BMK_Unigene_181645 TMP3 Binds to ac tin filaments in muscle and nonmuscle cells
BMK_Unigene_213257 CALML6 Calmodulin-like protein 6 isoform X1
Yellow-black (Surface fish) BMK_Unigene_135409 FAM13B GTPase activator activity
BMK_Unigene_143803 USP2 Hydrolase that deubiquitinates polyubiquitinated target proteins
such as MDM2 , MDM4 and CCND1
BMK_Unigene_130832 ATP5PF Mitochondrial membrane ATP synthase
Grey-green-yellow (Surface fish) BMK_Unigene_029491 PUP2 Proteasome subunit alpha type-5
BMK_Unigene_084262 lmnb2 Lamin B2
BMK_Unigene_099265 SNRPG Small nuclear ribonucleoprotein G
14268
|
LI et aL.
Through comparative histology studies, we found that surface
fish have a large number of melanocytes on the skin compared to
cavefish. Previous researches have shown that two types of mel-
anin (eumelanin and pheomelanin) are produced (Jiang et al., 2014)
and stockpiled by the melanosomes in melanophores (Slominski
et al., 2004). Previous studies have shown that the expression of
melanin is regulated by multiple genes in the melanogenesis path-
way, such as Mc1r, MITF, GNAQ (α melanocyte-stimulating hormone).
Accordingly, in this study, we first focused on the melanogenesis
pathway for candidate DEGs screening and found five genes: GNAQ,
AC (adenylate cyclase), NRAS (NRAS proto-oncogene, GTPase),
SFRP2 (secreted frizzled-related protein 2), and PKA (protein kinase)
with significant differential expression, and further verified the ex-
pression trends of these five genes in all sur face-cave pairs.
In the melanogenesis pathway, Mc1r on the cell membrane
is activated by α-MSH, which results in the stimulation of Gnαq,
thereby activating AC to increase the production of cAMP and
PKA activation. PKA upregulates the expression of MITF, leading to
a rise in melanin synthesis (Buscà & Ballot ti, 2000; García-Borrón
et al., 2005; Oscar & Van, 2016; Van Raamsdonk et al., 2009), even-
tually leading to darker skin coloration through the intracellular dis-
persal of membrane-bound pigment granules (melanosomes) within
the melanophore (García-Borrón et al., 2005; Hoekstra, 2006; Lin
& Fisher, 2007; Yamaguchi et al., 2007). Notably, we found three
genes, GNAQ, AC, and PKA, that showed a significant downregula-
tion trend in most cavefish skins.
GNAQ encodes Gnαq, a subunit of a trimeric G protein complex
that binds to the endothelin B receptor in melanocytes (Raamsdonk
et al., 2004), and plays an impor tant role in the regulation of pigment
formation (Van Raamsdonk et al., 2009). We found that GNAQ was
significantly downregulated in more than 75% of surface-cave pairs
(12/16 pairs). Recent research found that the expression of GNAQ
in black mouse skin was significantly higher than that in the white
mouse skin (Yin et al., 2015). In fish species, GNAQ showed a signif-
icant association with skin pigmentation in three spine sticklebacks
(Gasterosteus aculeatus; Greenwood et al., 2011). Furthermore, Gessi
et al. (2013) found that a point mutation in GNAQ may be related
to the development of primary melanocytic tumors in humans.
They identified six cases harboring mutations in codon 209 of the
GNAQ gene. As members of a cascade of regulatory genes in the
melanogenesis pathway, GNAQ, AC, and PKA control the process
of melanin production by regulating the expression of upstream or
downstream genes (Bennett & Lamoreux, 2003). This suggests that
the above-mentioned genes, which are downregulated in cavefish
species, are likely to be closely related to the loss of pigmentation in
Sinocyclocheilus cavefish.
In addition to the melanogenesis pathway, genes (Tyr, TYRP1,
P38, etc.) in Wnt signaling pathway (Fujimura et al., 2009), MAPK
signaling pathway and apoptosis pathway (Squarzoni et al., 2011),
have been indicated to be involved in the regulation of pigmen-
tation. Tyrosinase, encoded by the Ty r gene, is the rate-limiting
enzyme in melanogenesis and is an impor tant regulatory factor
in the synthesis of melanin. Previous studies reported that loss
or downregulation of Tyr gene leads to an albino phenotype in ze-
brafish (Page-McC aw et al., 20 04), duck (Li et al., 2012), mouse
(Ito & Wakamatsu, 2011) and rabbit (Song et al., 2017). Yang
et al. (2016) carried out comparative transcriptomic analyses of
three Sinocyclocheilus fish species, S. graham (surface fish), S. rhi-
nocerous (cavefish), and S. anshuiensis (cavefish) and suggested
that Tyr, Tyrp1, and Mpv 17 (mitochondrial inner membrane protein)
lead to the loss of iridophores in Sinocyclocheilus cavefish ( Yang
et al., 2016). In th e prese nt stud y, we did not find significa nt differ-
ences in the expression of Tyr or Ty rp1. However, we observed that
p38 (mitogen-activated protein kinase 14) was significantly down-
regulated in cavefish species compared to the surface fish. In the
MAPK signaling pathway, p38 signaling is activated by α-MSH and
UV radiation, thereby promoting the phosphorylation of the USF-1
(MITF-like transcription factor) to activate the Ty r promoter (Bellei
et al., 2010; Galibert et al., 2001). Furthermore, in this study, some
key pigment-related genes identified in other Sinocyclocheilus
fish species, such as Tyr, Tyr, and Oca2 ( Yang et al., 2016), did not
show significant dif ferential expression in all cavefish groups.
Interestingly, some genes associated with the Wnt, MAPK, and
apoptosis pathway, such as DA AM1 (disheveled associated acti-
vator of morphogenesis) and BID (BH3 interacting domain death
agonist) were significantly downregulated in the skin of most
Sinocyclocheilus cavefish species studied. This may be partially ex-
plained by the spatiotemporal specificity of the transcriptome, or
by the fact that these genes show pleiotropy while par ticipating in
skin pigment formation. Furthermore, a number of factors could
affect the expression of genes in the skin, such as season and in-
dividual developmental status. In summary, 11 of DEGs, including
GNAQ, PKA, NRAS, and p38 etc, which are involved in key pig-
ment regulation pathways, were identified through comparative
transcriptomics. We infer that these genes may participate in the
partial regulation of skin pigmentation, and the downregulation of
these genes may lead to the pigment ation loss in Sinocyclocheilus
cavefish species. Furthermore, we found that the trend of gene
expression was different in each group (Table 2), and some genes
even no significant difference in some group may due to dif fer-
ent mechanisms of pigment loss, which could be responsible for
these different species. Therefore, future studies should include
gene knockout or overexpression studies to validate the function
of these genes.
In this study, we found that the results of qPCR of some pairs
may not perfectly support RNA-seq but they used the same RNA. In
order to minimize RNA degradation, in this study, we sent the skin
tissue to the sequencing company directly to extract the RNA, and
then sent the RNA back to our laboratory to conduct qPCR, we think
this may cause the degradation of a small amount of RNA during
the sample delivery. Furthermore, since the two experiments are
not conduc t at the same laboratory, the final result s may have some
differences.
In conclusion, to further elucidate the underlying genetic mech-
anism of skin pigment loss in Sinocyclocheilus cavefish species, we
conducted a comparative histological, WGCNA, and comparative
|
14269
LI et aL.
transcriptomics analyses of the skin of Sinocyclocheilus fishes dwell-
ing on surface and in c aves, and found differences in the distribution,
quantity, and morphology of skin melanocytes. A total of 35,101
DEGs were found in four surface-cave groups. Through GO, COG,
and KEGG pathway annotations, we identified 11 candidate genes
which showed significant differential expression in Sinocyclocheilus
cave and sur face fish species analyzed in this study, that may par-
ticipate in the regulation of skin pigmentation. Most of the DEGs
were downregulated in Sinocyclocheilus cavefishes as validated by
qPCR. However, these genes require further functional validation in
Sinocyclocheilus. This study provides a strong foundation for better
understanding of the molecular genetic mechanisms underlying tro-
glomorphic traits in cavefish species.
CONFLICT OF INTEREST
The authors declare that they have no competing interests.
AUTHOR CONTRIBUTION
Chunqing Li: Formal analysis (lead); Investigation (lead); Methodology
(lead). Hongyu Chen: Formal analysis (lead); Investigation (lead);
Methodology (lead); Writing-original draft (lead); Writing-review &
editing (lead). Yinchen Zhao: Methodology (supporting); Resources
(supporting). Shanyuan Chen: Conceptualization (lead); Funding
acquisition (lead); Project administration (lead); Resources (equal);
Supervision (equal); Writing-original draft (equal); Writing-review
& editing (equal). Heng Xiao: Conceptualization (lead); Funding ac-
quisition (lead); Project administration (equal); Resources (equal);
Supervision (equal).
ETHICS APPROVAL AND CONSENT TO PARTICIPATE
The Sinocyclocheilus fish species used in this study were caught from
wild water bodies, and no specific permissions were required. All ex-
periments were conducted after review and approval from the local
Ethical Committee at Yunnan University in accordance with China's
local and global ethical policies (Grant No: ynucae 20190056), and all
the procedures were approved and assisted by the local government .
CONSENT FOR PUBLICATION
Not applicable.
DATA AVAIL ABI LIT Y S TATEM ENT
All data generated or analyzed during this study are included in this
published article and its supplementary information files. The raw
reads produced in this study were deposited in the DRYAD database
( h t t p s : / / d a t a d r y a d . o r g / s t a s h / s h a r e / t 5 c Z I X o V U g y h p z E P 6 z - G N 6 x j
c5EU3 TvPPE wbdIo 7siI).
ORCID
Shanyuan Chen https://orcid.org/0000-0002-9524-9428
REFERENCES
Alexa, A., & Rahnenfuhrer, J. (2006). topGO: Enrichment analysis for Gene
Ontology. R Package Version.
Bellei, B., Maresca, V., Flori, E., Pitisci, A., Larue, L., & Picardo, M. (2010).
p38 regulates pigmentation via proteasomal degradation of tyrosi-
nase. Journal of Biological Chemistry, 285(10), 7288–7299. https://doi.
org/10.1074/jb c.M109.070573
Bennett, D. C., & Lamoreux, M. L. (2003). The color loci of mice – A
genetic century. Pigment Cell Research, 16(4), 333–344. https://doi.
org/10.1034/j.1600-0749.2003.00067.x
Bilandzija, H., Ćetković, H., & Jeffery, W. (2012). Evolution of albinism in
cave plant hoppers by a convergent defec t in the first step of melanin
biosynthesis. Evolution & Development, 14(2), 196–203. https://doi.
org/10.1111/j.1525-142X.2012.00535.x
Buscà, R., & Ballotti, R. (2000). Cyclic AMP a key messenger in the regu-
lation of skin pigmentation. Pigment Cell & Melanoma Research, 13(2),
60–69. https://doi.org /10.1034/j.1600- 0749.20 00.130203.x
Bussotti, S., Di Franco, A., Bianchi, C. N., Chevaldonné, P., Egea, L.,
Fanelli, E., Lejeusne, C., Musco, L., Navarro-Barranco, C., Pey, A .,
Planes, S., Vieux-Ingrassia, J. V., & Guidetti, P. (2018). Fish mitigate
trophic depletion in marine cave ecosystems. Scientific Reports, 8(1),
11–13. https://doi.org/10.1038/s41598-018-27491-1
Chen, H., Li, C., Liu, T., Chen, S., & Xiao, H. (2019). A metagenomic study
of intestinal microbial diversity in relation to feeding habits of sur-
face and c ave-dwelling Sinocyclocheilus species. Microbial Ecology,
79(2), 1–13. https://doi.org/10.1007/s00248-019-01409-4
Coelho, S. G., Valencia, J. C., Yin, L., Smuda, C., Mahns, A., Kolbe, L.,
Miller, S. A., Beer, J. Z., Zhang, G., Tuma, P. L., & Hearing, V. J. (2015).
UV exposure modulates hemidesmosome plasticity, contributing to
long-term pigmentation in human skin. Journal of Pathology, 236(1),
17–29. https://doi.org/10.1002/path.4497
Culver, D. C., & Pipan, T. (2009). Adaptations to subterranean life. The biol-
ogy of caves and other subterranean habitats. Oxford University Press.
Fujimura, N., Taketo, M. M., Mori, M., Korinek, V., & Kozmik, Z. (2009).
Spatial and temporal regulation of Wnt/beta-catenin signaling
is essential for development of the retinal pigment epithelium.
Developmental Biology, 334(1), 31–45.
Galibert, M. D., Carreira, S., & Goding, C. R. (2001). The Usf-1 transcrip-
tion factor is a novel target for the stress-responsive p38 kinase and
mediates UV-induced Tyrosinase expression. Embo Journal, 20(17),
5022–5031. https://doi.org/10.1093/emboj/ 20.17.5022
García-Borrón, J. C., Sánchez-Laorden, B. L., & Jiménez-Cervantes,
C. (2005). Melanocortin-1 receptor structure and functional
regulation. Pigment Cell Research, 18(6), 393–401. https://doi.
org /10.1111/j.160 0-0749.2005.00278.x
Gessi, M ., Hammes, J., Lauriola, L., Dörner, E., Kir fel, J., Kristiansen, G.,
… Pietsch, T. (2013). GNA11 and N-RAS mutations: alternatives for
MAPK pathway activating GNAQ mutations in primary melanocytic
tumours of the central ner vous system. Neuropathology and Applied
Neurobiology, 39(4), 417–425.
Grabherr, M. G., Haas, B. J., Yassour, M., Levin, J. Z., Thompson, D. A.,
Amit, I., Adiconis, X., Fan, L., Raychowdhury, R., Zeng, Q., Chen, Z.,
Mauceli, E., Hacohen, N., Gnirke, A ., Rhind, N., di Palma, F., Birren,
B. W., Nusbaum, C., Lindblad-Toh, K., … Regev, A. (2011). Full-
length transcriptome assembly from RNA-Seq data without a ref-
erence genome. Nature Biotechnology, 29(7), 644–652. https://doi.
org/10.1038/nbt .1883
Greenwood, A. K., Jones, F. C., Chan, Y. F., Brady, S. D., Absher, D. M.,
Grimwood, J., Schmutz, J., Myers, R . M., Kingsley, D. M., & Peichel,
C. L. (2011). The genetic basis of divergent pigment patterns in juve-
nile threespine sticklebacks. Heredity, 107(2), 155–166. https://doi.
org/10.1038/hdy.2011.1
Gross, J. B., Borowsky, R., & Tabin, C. J. (2009). A novel role for Mc1r in
the parallel evolution of depigmentation in independent populations
of the cavef ish Astyanax mexicanus. PLoS Genetics, 5(1), e1000326.
https://doi.org/10.1371/journ al.pgen.1000326
Helena, B., Lindsey, A., Li, M., Kenneth, J. R., & William, R. J.(2018).
Behavioural changes controlled by catecholaminergic systems
14270
|
LI et aL.
explain recurrent loss of pigmentation in cavefish. Proceedings of the
Royal Societ y B, 285(1878), 1–10.
Hibbert, S. A., Costello, P., O’Connor, C., Bell, M., Griffiths, C. E. M.,
Watson, R. E. B., & Sherratt, M. J. (2017). A new in vitro assay
to test UVR protection of dermal extracellular matrix compo-
nents by a flat spectrum sunscreen. Journal of Photochemistry and
Photobiology B: Biology, 175, 58–64. https://doi.org/10.1016/j.jphot
obiol.2017.08.020
Hoekstra, H. E. (2006). Genetics, development and evolution of adaptive
pigmentation in vertebrates. Heredity, 97(3), 222–234. https://doi.
org/10.1038/sj.hdy.6800861
Ito, S., & Wakamatsu, K. (2011). Human hair melanins: What we have
learned and have not learned from mouse coat color pigmenta-
tion. Pigment Cell & Melanoma Research, 24(1), 63 –74. https://doi.
org/10.1111/j.1755-148X. 2010.00755.x
Jeffery, W. R. (2009). Regressive evolution in astyanax cavefish. Annual
Review of Genetics, 43(1), 25– 47.
Jiang, Y., Zhang, S., Xu, J., Feng, J., Mahboob, S., Al-Ghanim, K. A ., Sun,
X., & Xu, P. (2014). Comparative transcriptome analysis reveals the
genetic basis of skin color variation in common carp. PLoS One, 9(9),
1–10. https://doi.org/10.1371/journ al.pone.0108200
Kanehisa, M., Goto, S., Kawashima, S., Okuno, Y., & Hattori, M. (20 04).
The KEGG resource for deciphering the genome. Nucleic Acids
Research, 32(Database issue), 277–279. https://doi.org/10.1093/nar/
gkh063
Kelsh, R. N. (2004). Genetics and evolution of pigment patterns
in fish. Pigment Cell Research, 17(4), 326–336. https://doi.
org/10.1111/j.1600-0749.2004.00174.x
Kim, M., Sh ibata, T., Kwon , S. , Park, T. J., & Kang, H. Y. (2018). Ultraviolet-
irradiated endothelial cells secrete stem cell factor and induce
epidermal pigmentation. Scientific Reports, 8(1), 4235. https://doi.
org/10.1038/s41598-018-22608-y
Kumai, H. (2005). The histological analysis, colorimetric evaluation,
and chemical quantification of melanin content in ‘suntanned’ fish.
Pigment Cell & Melanoma Research, 18(6), 465–468.
Langmead, B., Trapnell, C., Pop, M., & Salzberg , S. L. (2009). Ultrafast and
memory-efficient alignment of short DNA sequences to the human
genome. Genome Biology, 10 (3), R25. https://doi.org/10.1186/
gb-2009-10-3-r25
Leclercq, E., Taylor, J. F., & Migaud, H . (2010). Morphological skin co-
lour changes in teleosts. Fish & Fisheries, 11(2), 159–193. https://doi.
org /10.1111/j.1467-2979.20 09.0 0346.x
Li, B., & Dewey, C. N. (2011). RSEM: Accurate transcript quantification from
RNA-Seq data with or without a reference genome. BMC Bioinformatics,
12(1), 323. https://doi.org/10.1186/1471-2105-12-323
Li, W., Wu, D., Chen, A., & Tao, J. (200 0). The preliminary investigation
of geogrophical distribution and elcological adaptation for cave envi-
ronment of Sinocyclocheilus rhinocerous—a cyprinid fish. Journal of
Yunnan Agricultural Universit y, 15(1) , 1–4 .
Li, S., Wang, C., Yu, W., Zhao, S., & Gong, Y. (2012). Identification of genes
related to white and black plumage formation by RNA-seq from
white and black feather bulbs in ducks. PLoS One, 7(5), 1–10. https://
doi.org/10.1371/journ al.pone.0036592
Lin, J. Y., & Fisher, D. E. (2007). Melanocyte biology and skin pigmenta-
tion. Nature, 445(7130), 843–850.
Little, E. E. (1995). Skin component may protect fishes from ultraviolet-B
radiation. Environmental Science & Pollution Research, 2(1), 30–32.
https://doi.org /10.1007/BF029 87508
Love, M. I., Huber, W., & Anders, S. (2014). Moderated estimation of
fold change and dispersion for RNA-seq data with DESeq2. Genome
Biology, 15(12), 550. https://doi.org/10.1186/s13059-014-0550-8
Lunghi, E., Zhao, Y., Sun, X., & Zhao, Y. (2019). Morphometrics of eight
Chinese cavefish species. Scientific Data, 6(1), 233. https://doi.
org/10.1038/s41597-019-0257-5
Mei, J., Zhang, Q.-Y., Li, Z., Lin, S., & Gui, J.-F. (2008). C1q-like inhibits
p53-mediated apoptosis and controls normal hematopoiesis during
zebrafish embr yogenesis. Developmental Biology, 319(2), 273–284.
https://doi.org/10.1016/j.ydbio.2008.04.022
Meng, F., Ingo, B., Phillips, J. B., Lin, X., Tom, T., Zhang, C., &
Postlethwait, J. H. (2013). Evolution of the eye transcriptome
under constant darkness in Sinocyclocheilus cavefish. Molecular
Biology & Evolution, 30(7), 1–7. https://doi.org/10.1093/molbe v/
mst079
Oscar, U., & Van, R. C. D. (2016). Gnaq and Gna11 in the endothelin sig-
naling pathway and melanoma. Frontiers in Genetics, 7(59), 1–10.
Page-McCaw, P. S., Chung, S. C., Muto, A., Roeser, T., Staub, W., Finger-
Bai er, K . C., Korenbrot, J. I., & Baier, H. (200 4). Retinal network adap-
tation to bright light requires tyrosinase. Nature Neuroscience, 7(1 2),
1329–1336. https://doi.org/10.1038/nn1344
Parichy, D. M. (20 06). Evolution of danio pigment pattern development.
Heredity, 97(3), 200–207. https://doi.org/10.1038/sj.hdy.6800867
Parkinson, L. G., Toro, A., Zhao, H., Brown, K., Tebbutt, S. J., & Granville,
D. J. (2015). Granzyme B mediates both direct and indirect cleav-
age of extracellular matrix in skin after chronic low-dose ultraviolet
light irradiation. Aging Cell, 14(1), 67–77. https://doi.org/10.1111/
acel.12298
Protas, M., Conrad, M., Gross, J. B., Tabin, C., & Borowsky, R. (2007).
Regressive Evolution in the Mexican Cave Tetra, Astyanax mexi-
canus. Current Biology, 17 (5), 452–454. https://doi.org/10.1016/j.
cub.2007.01.051 https://doi.org/10.1016/j.cub.2007.01.051
Protas, M. E., Hersey, C., Kochanek, D., Zhou, Y. I., Wilkens, H., Jeffery,
W. R., Zon, L. I., Borowsky, R., & Tabin, C. J. (2006). Genetic anal-
ysis of cavefish reveals molecular convergence in the evolution of
albinism. Nature Genetics, 38(1), 107–111. https://doi.org/10.1038/
ng170 0
Raamsdonk, C. D. V., Fitch, K. R., Fuchs, H., Angelis, M. H. D., & Barsh,
G. S. (20 04). Effects of G-protein mutations on skin color. Nature
Genetics, 36 (9), 961–968. https://doi.org/10.1038/ng1412
Slominski, A., Tobin, D. J., Shibahara, S., & Wortsman, J. (200 4). Melanin
pigment ation in mammalian skin and it s hormonal regulation.
Physiological Reviews, 84(4), 1155–1228. https://doi.org/10.1152/
physr ev.00044.2003
Song, Y., Xu, Y., Deng, J., Chen, M., Lu, Y. I., Wang, Y., Yao, H., Zhou, L.,
Liu, Z., Lai, L., & Li, Z. (2017). CRISPR/Cas9-mediated mut ation of ty-
rosinase (Tyr) 3' UTR induce graying in rabbit. Scientific Reports, 7(1),
1–8. https://doi.org/10.1038/s41598-017-01727-y
Squarzoni, P., Parveen, F., Zanetti, L., Ristoratore, F., & Spagnuolo, A.
(2011). FGF/MAPK/Ets signaling renders pigment cell precursors
competent to respond to Wnt signal by directly controlling Ci-Tcf
transcription. Development, 138(7), 1421–1432.
Van Raamsdonk, C. D., Barsh, G. S., Wakamatsu, K., & Ito, S. (2009).
Independent regulation of hair and skin color by two G protein-cou-
pled pathways. Pigment Cell & Melanoma Research, 22(6), 819–826.
https://doi.org/10.1111/j.1755-148X. 20 09.00609.x
Winberg, S. (20 00). Skin darkening, a potential social signal in subor-
dinate Arctic charr (Salvelinus alpinus): The regulatory role of brain
monoamines and pro-opiomelanocortin-derived peptides. J.exp.biol,
203(11), 1711–1721 .
Xie, C., Mao, X., Huang, J., Ding, Y., Wu, J., Dong, S., Kong, L., Gao, G.,
Li, C. Y., & Wei, L. (2011). KOBAS 2.0: A web server for annotation
and identification of enriched pathways and diseases. Nucleic Acids
Research, 39(Web Server, issue), 316–322.
Yamaguchi, Y., Brenner, M., & Hearing, V. J. (2007). The regulation of skin
pigmentation. Journal of Biological Chemistry, 282(30), 27557–27561.
https://doi.org/10.1074/jbc.R7000 26200
Yang, J., Chen, X., Bai, J., Fang, D., Qiu, Y., Jiang, W., Yuan, H., Bian, C ., Lu,
J., He, S., Pan, X., Zhang, Y., Wang, X ., You, X., Wang, Y., Sun, Y., Mao,
D., Liu, Y., Fan, G., … Shi, Q. (2016). The Sinocyclocheilus cavefish
|
14271
LI et aL.
genome provides insights into cave adaptation. BMC Biology, 14 (1),
1–13. https://doi.org/10.1186/s12915-015-0223-4
Yin, Z. X ., Yu, X., Xin, Z., Zhun, W., Jieli, L ., & Quanhai, P. (2015).
Differential expression of Gnαq gene in white and black skin tissues
of mice. Journal of Zhejiang University (A griculture & Life Sciences),
41(4), 414–42 0.
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section.
How to cite this article: Li C, Chen H, Zhao Y, Chen S, Xiao H.
Comparative transcriptomics reveals the molecular genetic
basis of pigmentation loss in Sinocyclocheilus cavefishes. Ecol
Evol. 2020;10:14256–14271. https://doi.org/10.1002/
ece3.7024
Available via license: CC BY 4.0
Content may be subject to copyright.
