Article

Genetic Structure of the Cave-Dwelling Catfish Pterocryptis

anomala (Siluriformes: Siluridae) in Southwest China
Renrong Huang, Jinmei Chen, Hongmei Li , Huan Cheng and Renyi Zhang *

School of Life Sciences, Guizhou Normal University, Guiyang 550025, China; [email protected] (R.H.);
[email protected] (J.C.); [email protected] (H.L.); [email protected] (H.C.)
* Correspondence: [email protected]

Simple Summary: This study elucidated the genetic structure and diversity of Pterocryptis
anomala populations in Southwest China based on two mitochondrial and two nuclear
genes and further investigated key driving factors influencing their current distribution
patterns. Through phylogenetic analysis of 255 individuals, we identified two distinct
evolutionary clades. The haplotype network also supported the division of all individuals
into two clades. The divergence time between the two clades was estimated at 13.73 million
years ago (Mya), which may have been influenced by the impact of the Qinghai–Tibet
Plateau (QTP) uplift on monsoonal systems. Our findings enhance the understanding of
the phylogeographic history of P. anomala in Southwest China and provide a theoretical
framework for targeted conservation strategies to safeguard freshwater biodiversity in this
ecologically critical region.

Abstract: The mountainous regions of Southwest China are biodiversity hotspots where
geographical isolation promotes genetic differentiation and species diversification. For
cave-dwelling species like the Pterocryptis anomala, how geographical isolation, historical
climate, and riverscapes have influenced their evolution remains largely unexplored. Based
on 255 samples from the Pearl River and the Yangtze River, this study integrated two
mitochondrial genes and two nuclear genes to analyze the genetic diversity and structure
of the P. anomala population. Phylogenetic trees based on mitochondrial DNA revealed
two distinct clades of P. anomala, while nuclear DNA loci showed no clear separation.
Academic Editor: Zissis Mamuris Spatial Analysis of Molecular Variance (SAMOVA) confirmed two groups: Clade I (the
Received: 8 March 2025
Yangtze, the Guijiang, and the Duliujiang Rivers) and Clade II (the Nanpanjiang, Hongshui,
Revised: 10 April 2025 Dahuanjiang, Youjiang, and Rongjiang Rivers). The divergence time between the two
Accepted: 18 April 2025 clades was estimated at 13.73 million years ago, which was potentially linked to the impact
Published: 23 April 2025 of the QTP uplift on monsoonal systems. The star-like network analysis and neutrality
Citation: Huang, R.; Chen, J.; Li, H.; test results indicated that the population of Clade I has maintained a stable state over a
Cheng, H.; Zhang, R. Genetic Structure long period, while the population of Clade II showed a trend of expansion. Additionally,
of the Cave-Dwelling Catfish geographical features such as the Nanling Mountains and the two major river systems may
Pterocryptis anomala (Siluriformes:
have obstructed gene flow, leading to genetic differentiation. These findings improved
Siluridae) in Southwest China. Animals
2025, 15, 1202. https://doi.org/
our understanding of this species’ evolutionary history and population structure, offering
10.3390/ani15091202 valuable insights for conservation efforts.

Copyright: © 2025 by the authors.

Keywords: Pterocryptis anomala; Southwest China karst; conservation; genetic differentiation;
Licensee MDPI, Basel, Switzerland.
This article is an open access article
population structure
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license
(https://creativecommons.org/
licenses/by/4.0/).

Animals 2025, 15, 1202 https://doi.org/10.3390/ani15091202

Animals 2025, 15, 1202 2 of 20

1. Introduction
Southwest China stands as a global biodiversity hotspot, shaped by its exceptional
geological history and climatic complexity [1,2]. This region, a Quaternary glacial refuge [3],
occupies a unique biogeographic position at the convergence of Pacific and Indian mon-
soon systems. Its dramatic elevational gradients—spanning the Qinghai–Tibet Plateau
(QTP), the Yunnan–Guizhou Plateau, and the Sichuan Basin—have created three distinct
topographic tiers that foster the ecological adaptive differentiation of species. Major moun-
tain systems, including the Himalayas, Kunlun Mountains, and Hengduan Mountains,
act as formidable biogeographic barriers, driving species diversification [4]. The region’s
hydrologic architecture further enhances its biological significance. Additionally, the region
boasts numerous rivers and lakes, with the central and northern parts dominated by rivers
in the Yangtze River basin. Meanwhile southern networks converge into the Pearl River
Basin—South China’s largest fluvial system comprising the East, West, and North Rivers,
along with tributaries like the Beipan and the Liujiang Rivers [5]. These river networks
have been profoundly shaped by Neogene uplift events, particularly the formation of the
Nanling Mountains, which established genetic barriers between the Yangtze and the Pearl
River biotas [6–8]. Mao’er Mountain, located in the northeastern part of Guangxi, is the
main peak of Yuecheng Ridge—one of the five major ridges of the Nanling Mountains. It
served as the source area for several important rivers, including the Lijiang, the Zijiang,
and the Xunjiang Rivers, and is also one of the sources of the Liujiang River. Its strategic
position bridges the Yangtze River and the Pearl River [9,10]. Furthermore, the uplift of
the QTP was closely related to the collision between the Indian Plate and the Asian Plate
in the early Cenozoic, which may have been the largest orogenic event in the history of
the Earth [11]. This collision not only triggered the significant uplift of the QTP but also
profoundly influenced the climate pattern of Asia [12]. The formation of the East Asian
monsoon, the South Asian monsoon, and the plateau monsoon, which play a controlling
role in the climate of Southwest China, has been closely related to the uplift of the QTP [13].
In addition, the uplift of the QTP also promoted the development of large-scale drainage
patterns and provided unique environmental conditions that facilitated speciation and
biodiversity of organisms on and around the plateau [11,14]. For example, the species
diversity of Sinocyclocheilus Fang, 1936 is closely related to the periodic uplift and erosion
of the QTP [15]. It can be seen that the formation of the Asian monsoon and its impact on
the climate evolution of Southwest China are important manifestations of the uplift of the
QTP in terms of climate and biodiversity. This orographic–climatic interplay maintains the
humidity gradients and thermal stability essential for preserving paleoendemic lineages in
karst ecosystems [16].
The vast karst landscapes in Southwest China, harboring the world’s most extensive
cave systems, are one of the typical karst development areas shaped by the uplift of the QTP
and the Asian monsoon climate [17,18], sustained unique subterranean ecosystems with
diverse troglobitic communities highly sensitive to environmental perturbations [19,20].
These hypogean habitats hosted remarkable endemic radiations, including cave-adapted
genera such as Triplophysa Rendahl, 1933 [21], Sinocyclocheilus [22], and Protocobitis Yang
& Chen, 1993 [23]. Cavefish (defined as species requiring groundwater environments
for critical life-history stages [24,25]) have been categorized into two ecological guilds:
typical cavefish exhibiting obligate subterranean adaptations (e.g., anophthalmia and
depigmentation) and atypical cavefish lacking such specialization [26]. The diversifica-
tion of typical cavefish has been influenced by both cave-specific selection pressures and
macro-evolutionary drivers such as the Neogene uplift of the QTP. The uplift of the QTP
fragmented the ancestral ranges, thus promoting allopatric speciation [27]. Notably, Sinocy-
clocheilus represents one of East Asia’s most spectacular freshwater fish radiations, with
Animals 2025, 15, x FOR PEER REVIEW 3 of 21

Animals 2025, 15, 1202 3 of 20

these advances, the evolutionary mechanisms structuring atypical cavefish populations

remain
over 80poorly resolved.
described speciesPterocryptis anomala (Herre,
exhibiting cave-related 1933), aninnovations
morphological atypical cave-dwelling
[22]. Despite cat-
fish co-occurring with obligate subterranean lineages like Sinocyclocheilus,
these advances, the evolutionary mechanisms structuring atypical cavefish has presented
populations
an remain poorly to
ideal model resolved. Pterocryptis
disentangle anomala
isolation (Herre, 1933),
mechanisms inan atypical
karst cave-dwelling
ecosystems. Thiscatfish
leads us to
co-occurring with obligate subterranean lineages like Sinocyclocheilus,
pose the question: Does its genetic differentiation primarily reflect (1) isolation due has presented anto ge-
ideal model to disentangle isolation mechanisms in karst ecosystems. This leads us to pose
ographical barriers or (2) hydrological segregation across surface river networks? Resolv-
the question: Does its genetic differentiation primarily reflect (1) isolation due to geograph-
ing this dichotomy requires integrative phylogeographic approaches to partition isola-
ical barriers or (2) hydrological segregation across surface river networks? Resolving this
tion-by-cave versus isolation-by-river effects.
dichotomy requires integrative phylogeographic approaches to partition isolation-by-cave
Pterocryptis
versus anomala, effects.
isolation-by-river a benthic-dwelling catfish endemic to cave and montane stream
ecosystems in Southwest
Pterocryptis anomala, aChina, occupies catfish
benthic-dwelling a unique biogeographic
endemic niche spanning
to cave and montane stream the
Yangtze
ecosystems in Southwest China, occupies a unique biogeographic niche spanningprovinces
River and the Pearl River, including Hunan, Guizhou, and Guangxi the
(Figure
Yangtze1) River
[28]. As anda the
poor disperser
Pearl with strict
River, including habitat
Hunan, fidelityand
Guizhou, to subterranean and head-
Guangxi provinces
(Figure
water 1) [28]. As a [29,30],
environments poor disperser with strict
this species serves habitat
as anfidelity to subterranean
exceptional model forand head-
investigating
water environments [29,30], this species serves as an exceptional model
how historical climatic oscillations, riverine landscape dynamics, and geographic isola- for investigating
how
tion historical
shape climatic
freshwater oscillations,
population riverine landscape
structure. We employed dynamics, and geographic
a multilocus isola-
phylogeographic
tion shape freshwater population structure. We employed a multilocus phylogeographic
approach, sequencing two mitochondrial genes (the cytochrome c oxidase subunit I, COI,
approach, sequencing two mitochondrial genes (the cytochrome c oxidase subunit I, COI,
and cytochrome b gene, Cyt b) and two nuclear loci (recombination activating gene 1,
and cytochrome b gene, Cyt b) and two nuclear loci (recombination activating gene 1,
RAG1, and pleiomorphic adenoma gene-like 2, PLAGL2), to analyze genetic structure and
RAG1, and pleiomorphic adenoma gene-like 2, PLAGL2), to analyze genetic structure and
differentiation
differentiationacross
acrossall
all populations representing
populations representing the
the Yangtze
Yangtze River
River andand the Pearl
the Pearl River.River.
Specific
Specificobjectives
objectiveswere
were to 1) delineate
to (1) spatialgenetic
delineate spatial genetic architecture
architecture across
across major major
hydro-hydro-
graphic
graphic divides
dividesand and 2)
(2)identify drivers(e.g.,
identify drivers (e.g.,paleo-drainage
paleo-drainage rearrangements
rearrangements and karst
and karst
barriers) underlying
barriers) underlyingcontemporary
contemporary distribution patterns. These
distribution patterns. Theseanalyses
analyseswill
willelucidate
elucidate the
the phylogeographic history of P. anomala populations in Southwest
phylogeographic history of P. anomala populations in Southwest China and establish China and establish a a
theoreticalfoundation
theoretical foundation forfor conservation
conservation strategies.
strategies.

Figure 1. 1.
Figure Sampling regions
Sampling regionsofofthe
theP.P.anomala
anomala in
in Southwest China. Different
Southwest China. Differentsites
siteswere
were colored ac-
colored
according
cording to the
to the tributary.
tributary. Locationcodes
Location codeswere
were consistent
consistent with
withthose
thoseshown in Table
shown 1. 1.
in Table

Table 1. Basic information for sampling sites.

Population Code Drainage System Sampling Site Longitude (°E) Latitude (°N) Sample Size
N1 Nanpanjiang River Gaoka Village, Guizhou 104.8581 25.0310 2
N2 Nanpanjiang River Louxia Town, Guizhou 104.8766 25.3446 2
Animals 2025, 15, 1202 4 of 20

Table 1. Basic information for sampling sites.

Population
Drainage System Sampling Site Longitude (◦ E) Latitude (◦ N) Sample Size
Code
N1 Nanpanjiang River Gaoka Village, Guizhou 104.8581 25.0310 2
N2 Nanpanjiang River Louxia Town, Guizhou 104.8766 25.3446 2
N3 Nanpanjiang River Haiziba Village, Guangxi 105.1925 24.8566 3
N4 Nanpanjiang River Zuoshe Village, Guizhou 105.1196 25.2569 9
N5 Nanpanjiang River Kewang Village, Guizhou 105.2134 25.1208 5
H1 Hongshui River Aozizhai Village, Guizhou 105.6380 26.0297 22
H2 Hongshui River Dongtang Village, Guangxi 107.5182 25.1513 7
H3 Hongshui River Dalongguan Village, Guangxi 106.4319 24.5694 6
H4 Hongshui River Bali Village, Guangxi 106.4669 24.5272 6
H5 Hongshui River Dajing Village, Guizhou 106.8625 25.5571 16
Ganlongdong Village,
H6 Hongshui River 105.6773 25.2779 24
Guizhou
H7 Hongshui River Jiazuan Town, Guangxi 107.1531 24.1990 15
H8 Hongshui River Shanglin County, Guangxi 108.6050 23.4324 5
H9 Hongshui River Fengting Village, Guizhou 106.5623 25.4134 17
H10 Hongshui River Bina Village, Guizhou 106.7782 25.7632 2
DH Dahuanjiang River Jialiang Town, Guizhou 107.9958 25.2354 5
DL Duliujiang River Bakai Town, Guizhou 108.3740 25.8536 2
RJ Rongjiang River Rongan County, Guangxi 109.3975 25.2245 10
JX Youjiang River Longteng Village, Guangxi 106.3436 23.0566 16
GJ Guijiang River Dalingshan Village, Guangxi 110.7352 24.9354 15
Y1 Yangtze River Heboling Village, Hunan 111.9863 25.7043 40
Y2 Yangtze River Guping Village, Hunan 111.2851 26.7840 24

2. Materials and Methods

2.1. Sample Collection
A total of 22 sites of 255 specimens were collected from both the Yangtze River and
the Pearl River for molecular analysis. The geographical distribution of these populations
is depicted in Figure 1. Freshly collected specimens from the field were preserved in 75%
ethanol and subsequently stored in a −20 ◦ C freezer upon arrival at the laboratory, prior to
DNA extraction. All collected specimens are now archived at the College of Life Sciences,
Guizhou Normal University, in Guizhou Province, China. The name, location, and sample
size of each population are detailed in Table 1.

2.2. Laboratory Procedures and Sequence Analyses

Genomic DNA was extracted from P. anomala muscle tissues using the traditional
high-salt method [31]. We amplified two mitochondrial loci and two nuclear loci via
polymerase chain reaction (PCR). Primer sequences are provided in Table S1. Each 35 µL
PCR mixture contained 17.5 µL 2× Taq Plus MasterMix (CWBIO, Taizhou, China), 1 µL
of each forward/reverse primer (10 µM), 1 µL template DNA (~50 ng/µL), and 14.5 µL
ddH2 O. The thermal cycling conditions were as follows: initial denaturation at 95 ◦ C for
5 min, followed by 35 cycles, each cycle consisting of 95 ◦ C for 45 s, annealing at 55−57 ◦ C
for 30 s (annealing temperatures: 55 ◦ C for COI, Cyt b, and PLAGL2, 55 ◦ C for RAG1 in the
first stage, and 57 ◦ C for RAG1 in the second stage), and extension at 72 ◦ C for 1 min, with
a final extension at 72 ◦ C for 10 min. For PLAGL2 and RAG1, a nested PCR strategy was
implemented, with first-stage products serving as templates for second-stage amplification.
In the aforementioned analysis, the nuclear sequences involved are all exon regions. All
amplicons were verified by 1.0% agarose gel electrophoresis before bidirectional Sanger
sequencing at Sangon Biotech (Shanghai, China) using BigDye Terminator v3.1 chemistry
on an ABI 3730xl analyzer.
Animals 2025, 15, 1202 5 of 20

All sequences were proofread and assembled with the DNA analysis package DNAS-
TAR Lasergene’s SeqMan v7.0 (DNASTAR Inc., Madison, WI, USA). Initially, sequence
alignments were conducted utilizing Muscle within MEGA version 11.0 [32]. Subsequently,
each protein-coding sequence was translated to verify and designate codon positions.
Lastly, the counts of transversions, variable sites, and parsimony-informative sites were de-
termined using MEGA and checked manually. The mitochondrial fragments Cyt b and COI
were concatenated into one mitochondrial locus (mtDNA) for subsequent analyses. Based
on mtDNA and nuclear genes, haplotypes were defined using DnaSP v5.0 software [33].
The PHASE algorithm, integrated within the DnaSP, was employed to phase the double
chromatograph peaks of nuclear gene sequences, utilizing the default parameter settings.
Phasing results were validated based on a posterior probability threshold exceeding 85%,
and these validated results were subsequently utilized for further analysis [34]. Sequences
of all individuals were deposited in GenBank.

2.3. Molecular Phylogenetic Analysis

Utilizing all haplotypes from the concatenated mtDNA sequences (COI and Cyt b),
the phylogenetic relationships among populations of the cavefish P. anomala distributed in
Southwestern China were reconstructed. This was accomplished through the application
of Bayesian Inference (BI) with MrBayes V3.2.1 software [35] and Maximum Likelihood
(ML) analysis within the IQ-tree module of PhyloSuite v1.2.2 software [36]. Clarias fuscus
(Lacepède, 1803), Ictalurus punctatus (Rafinesque, 1818), Ompok bimaculatus (Bloch, 1794),
Heteropneustes fossilis (Bloch, 1794), and Pterocryptis cochinchinensis (Valenciennes, 1840)
were selected as the outgroups for phylogenetic studies. Based on the Akaike Information
Criterion (AIC), the optimal nucleotide substitution model, GTR + I + G, was selected for
BI analysis using MrModelTest v.2 [37]. The total number of generations for Markov Chain
Monte Carlo (MCMC) computation was 2,000,000, with the analysis starting from a random
tree and sampling once every 1000 generations. The average standard deviation of split
frequencies was less than 0.01, indicating that the sampling of the posterior distribution
was sufficient. The ML tree was constructed using IQ-tree module, employing the TVM + F
+ G model and incorporating the ultrafast bootstrap method. Node support was evaluated
through a bootstrap process comprising 5000 replicates. By conducting statistics on the
remaining trees, a consensus tree with posterior probability values greater than 50% was
generated. Then, the first 25% of the resulting phylogenetic trees was discarded as part
of the burn-in process. Finally, the phylogenetic trees were visualized and edited using
Interactive Tree of Life (iTOL) v4 [38].

2.4. Genetic Diversity and Genetic Differences

Genetic diversity was assessed using DnaSP to calculate haplotype diversity (h), the
number of haplotypes (nh), the average number of nucleotide differences (k), the number of
variable sites (s), and nucleotide diversity (π). Median-joining networks were constructed
using Popart v.1.7 [39] with default settings based on mtDNA and two nuclear genes data
to explore the relationships among their respective haplotypes. To assess the degree of
genetic differentiation among P. anomala from different clades, two statistical methods were
employed. Pairwise divergences were calculated both between and within the clades using
the Kimura 2-parameter (K2P) model [40] in MEGA. The pairwise fixation index (Fst) was
estimated using Arlequin v.3.5 [41]. To examine the correlation between genetic distance
and geographical distance, both matrices were analyzed using Mantel tests in R with the
vegan package v2.6-4 [42,43]. The geographical distance matrix was calculated based on
the sampling locations (Table S2).
Animals 2025, 15, 1202 6 of 20

2.5. Analysis of Population Structure and Historical Demographic Changes

Based on the aforementioned analyses, all individuals were successfully assigned to
corresponding haplotypes. To elucidate the population genetic structure of these haplo-
types, we performed Bayesian clustering analysis on the haplotype data using STRUCTURE
software v2.3.4 [44]. Initially, the burn-in process was set to 105 steps, followed by 106
Markov chain Monte Carlo (MCMC) iterations. The entire analysis was replicated 10 times
for a range of K values from 1 to 10. The optimal number of clusters (K) was determined
using STRUCTURE HARVESTER [45] and applying the Delta K method [46]. Subsequently,
to consolidate the results from each K replication, CLUMPP v1.1.2 [47] was utilized. Finally,
bar plots were generated using DISTRUCT v1.1 [48] to visualize the clustering outcomes.
To understand the genetic structure of P. anomala, AMOVA analysis was conducted using
Arlequin to calculate the distribution of genetic variation within and among populations,
as well as the genetic differentiation coefficient between populations. For the AMOVA
analysis, P. anomala was divided into two geographical groups based on the results of the
phylogenetic analysis and possible geographical barriers. The analysis was conducted with
1000 permutations for significance testing. To explore the spatial dynamics of population
history and test whether the sequences conform to the expectations of neutrality, Tajima’s
D [49] and Fu’s Fs [50] were calculated, and 1000 simulations were performed for each
statistic to create a 95% confidence interval. Mismatch distribution analysis was conducted
using DnaSP to explore whether past populations had undergone expansion.

2.6. Estimations of Divergence Time

Based on the combined mitochondrial gene data, the divergence times among hap-
lotype lineages of all populations were calculated using BEAST v2.4.7 [51], employing
the GTR + I + G substitution model. Due to the lack of known fossil records for this
group, two calibration points were employed as suggested by Chen et al. [52] and Kappas
et al. [53], including (1) the first appearance of the African Clariidae being in the Lower
Eocene (34–56 Mya) and (2) the divergence of Ompok bimaculatus from P. cochinchinensis
(22.08 Mya). The outgroups include Clarisa fuscus, Ictalurus punctatus, Ompok bimaculatus,
Heteropneustes fossilis, and P. cochinchinensis. Three separate MCMC chains were run for
20 million generations each, sampling every 1000 generations. Analyses used a relaxed
lognormal clock model and a Yule process. Convergence was checked in TRACER v1.7 [54],
ensuring effective sample sizes exceeded 200. TreeAnnotator (BEAST v2.4.7 package) was
used to generate a maximum clade credibility (MCC) tree, discarding the first 25% as
burn-in. The resulting tree, including divergence time estimates, was visualized using
FigTree v1.4.3 [55].

3. Results
3.1. Sequence Information
The 253 sequences were obtained for the mtDNA sequences from all samples of P.
anomala. The aligned data of the sequences from P. anomala had lengths of 651 base pairs (bp)
for COI and 1138 bp for Cyt b. The concatenated mitochondrial genes comprised 1789 bp.
There were 31 haplotypes in concatenated mitochondrial gene sequences, among which
the frequency of Hap6 was the highest (37.15%), followed by Hap18 (13.83%) and Hap31
(9.49%); Hap6 were shared by two populations (Table S3). The number of polymorphic
sites (s) was 89, and the average number of nucleotide differences (k) was 22.284.
The 229 RAG1 sequences contained 26 polymorphic sites, with an average num-
ber of nucleotide differences (k) was 2.716 and the 255 PLAGL2 sequences contained
157 polymorphic sites and the average number of nucleotide differences (k) was 5.907. For
RAG1, 30 alleles were identified. Hap1 predominated at a frequency of 40.17%, followed
 Animals 2025, 15, 1202 7 of 20

by Hap6 (17.03%) and Hap3 (12.8%). Notably, Hap1 was shared across five populations
(Table S4). Similarly, PLAGL2 revealed 37 alleles, with Hap1 being dominant (53.73%),
followed by Hap7 (25.49%) and Hap15 (3.92%). Hap1 exhibited broader distribution,
occurring in seven populations (Table S5).

3.2. Phylogenetic Analysis and Haplotype Network

The phylogenetic trees based on concatenated mitochondrial gene sequences revealed
two main branches within the species P. anomala; the Clade I consisted of three minor
clades: I-1, the sample from DL, belonged to the Duliujiang River; I-2, the sample from
GJ, belonged to the Guijiang River; and I-3 and I-4, the samples from Y1 and Y2, which
belonged to the Yangtze River. Clade II consisted of two minor clades, II-1 and II-2,
which included the remaining P. anomala populations, all of which belonged to the Pearl
River (Figure 2). Haplotype network analyses revealed distinct genetic structuring, with all
studied individuals clustering into two major clades (Figure 3), a pattern that was consistent
with both BI and ML phylograms. However, due to the low intraspecific polymorphism
Animals 2025, 15, x FOR PEER REVIEW and with no significant population differentiation, we did not construct phylogenetic trees8 of 21
based on nuclear genes for the P. anomala populations.

Figure 2. 2.
Figure Phylogenetic
Phylogenetictrees
trees reconstructed basedonon
reconstructed based concatenated
concatenated mitochondrial
mitochondrial gene gene sequences;
sequences;
nodal
nodal values
values nearbranches
near branches indicate
indicate posterior
posteriorprobabilities
probabilitiesandand
bootstrap supports
bootstrap of BI/ML.
supports of BI/ML.
Animals 2025, 15, 1202 Figure 2. Phylogenetic trees reconstructed based on concatenated mitochondrial gene 8sequences;
of 20
nodal values near branches indicate posterior probabilities and bootstrap supports of BI/ML.

Figure 3. Network
Figure 3. Networkprofile
profilebased
based on mtDNAdata
on mtDNA dataofofthe
the populations.
populations. Each
Each haplotype
haplotype is represented
is represented
by abycircle.
a circle.
TheThe
sizesize of the
of the circle
circle is proportional
is proportional toto thathaplotype’s
that haplotype’sfrequency.
frequency. Different
Different colors
colors indi-
cateindicate different
different geographical
geographical locations.
locations. Locationcodes
Location codes are
are consistent
consistentwith
withthose shown
those in Table
shown 1,
in Table 1,
corresponding to Figure 1.
corresponding to Figure 1.
The median-joining network based on RAG1 and PLAGL2 analysis has a weblike
3.3.topology,
Genetic Diversity andwith
respectively, Population Differentiation
many singletons connected through multiple nodes, indicating
high
Thegenetic
geneticvariability
diversityfor P. anomala
of P. anomala(Figures S1 and
populations in S2). In the haplotype
Southwest networks,
China is shown in Table
some haplotypes were shared between clades, but none had a clear genealogical
2. Based on mtDNA data, the haplotype diversity (h) of the Duliujiang River site was the structure.
Three haplotypes based on RAG1 (Hap1, Hap3, and Hap6) were shared by the two clades
highest, with a value of 1.000 ± 0.500, while the Rongjiang River and Guijiang River sites
(Figure S1). Two haplotypes based on PLAGL2 (Hap1 and Hap9) were shared by the two
have exhibited the lowest haplotype diversity, both with a value of 0. All other popula-
clades (Figure S2). Overall, the absence of genetic substructure in nuclear markers was
tions had the h value exceeding 0.5, with the exception of the Youjiang River. Regarding
clearly demonstrated in the haplotype network. Specifically, no correlation was detected
between haplotype and geographic distribution (Figures S1 and S2).

3.3. Genetic Diversity and Population Differentiation

The genetic diversity of P. anomala populations in Southwest China is shown in Table 2.
Based on mtDNA data, the haplotype diversity (h) of the Duliujiang River site was the
highest, with a value of 1.000 ± 0.500, while the Rongjiang River and Guijiang River
sites have exhibited the lowest haplotype diversity, both with a value of 0. All other
populations had the h value exceeding 0.5, with the exception of the Youjiang River.
Regarding nucleotide diversity, the values at different sampling sites ranged from 0.0075
(the Yangtze River site) to 0 (the Rongjiang River and Guijiang River sites). Among them,
only two populations (the Yangtze River and Nanpanjiang River sites) had nucleotide
diversity (π) values higher than 0.005, while the π values for the remaining six sites fell
between 0 and 0.00137 (Table 2). The inter-clade genetic distances (Table 3) ranged from
3.1% (between minor clades I-1 and II-1) to 4% (between minor clades I-4 and II-2).
Based on RAG1 data, the haplotype diversity (h) of the Rongjiang River site was the
highest (0.845 ± 0.039), while the Duliujiang River and Dahuanjiang River sites exhibited
the lowest haplotype diversity, both with a value of 0. All other populations had the h
value exceeding 0.5, with the exception of the Duliujiang River, Dahuanjiang River, and
Hongshui River sites. Regarding nucleotide diversity, the values at different sampling sites
ranged from 0.00603 (the Rongjiang River site) to 0 (the Dahuanjiang and Duliujiang Rivers
sites). Among them, only the Rongjiang River site had nucleotide diversity (π) values
higher than 0.005, while the π values for the remaining seven sites fell between 0 and 0.0045
(Table 2). Based on PLAGL2 data, the haplotype diversity (h) of the Youjiang River site was
the highest (0.874 ± 0.064), while the Duliujiang and Dahuanjiang River sites exhibited
Animals 2025, 15, 1202 9 of 20

the lowest haplotype diversity, both with a value of 0. All other populations had the h
value lower than 0.5, with the exception of the Youjiang River and Rongjiang River sites.
Regarding nucleotide diversity, the values at different sampling sites ranged from 0.05209
(the Youjiang River site) to 0 (the Dahuanjiang River and Duliujiang River sites). Among
them, only the Youjiang River and Nanpanjiang River sites had nucleotide diversity (π)
values higher than 0.005, while the π values for the remaining seven sites fell between 0
and 0.00433 (Table 2). Compared to RAG1 and mtDNA data, the PLAGL2 gene exhibited a
lower level of genetic diversity in all populations (Table 2).

Table 2. Summary of genetic diversity based on mtDNA, RAG1, and PLAGL2 data in P. anomala in
Southwest China.

MtDNA RAG1 PLAGL2

River System N nh π h N nh π h N nh π h
0.00641 0.533 0.00451 0.665 0.03086 0.481
Nanpanjiang
21 4 ± ± 22 8 ± ± 22 7 ± ±
River
0.0011 0.111 0.0003 0.050 0.0158 0.131
0.00137 0.536 0.00060 0.361 0.00081 0.122
Hongshui River 120 15 ± ± 93 5 ± ± 112 8 ± ±
0.0002 0.052 0.0001 0.036 0.0003 0.042
0.00750 0.714 0.00103 0.558 0.00140 0.420
Yangtze River 64 8 ± ± 52 6 ± ± 67 13 ± ±
0.0005 0.032 0.0002 0.023 0.0004 0.077
0.00056 1.000
Duliujiang River 2 2 ± ± 1 1 0 0 2 1 0 0
0.0003 0.500
0.00056 0.700
Dahuanjiang
5 3 ± ± 4 1 0 0 2 1 0 0
River
0.0002 0.218
0.00007 0.125 0.00365 0.590 0.05209 0.874
Youjiang River 16 2 ± ± 26 9 ± ± 20 11 ± ±
0.0001 0.106 0.0004 0.063 0.0204 0.064
0.00603 0.845 0.00433 0.838
Rongjiang River 10 1 0 0 19 13 ± ± 15 8 ± ±
0.0005 0.039 0.0038 0.085
0.00020 0.083 0.00020 0.133
Guijiang River 15 1 0 0 12 2 ± ± 15 2 ± ±
0.0002 0.075 0.0002 0.112
The number of haplotypes (nh), nucleotide diversity (π), and haplotypes diversity (h).

Table 3. The Kimura 2-parameter (K2P) distances (%) of P. anomala populations inferred from
mtDNA data.

I-1 I-2 I-3 I-4 II-1 II-2

I-1
I-2 0.014
I-3 0.014 0.016
I-4 0.018 0.022 0.016
II-1 0.031 0.034 0.033 0.038
II-2 0.032 0.036 0.034 0.040 0.006

MtDNA-based pairwise genetic differentiation Fst between the 22 sites ranged from
−0.08 to 1.00 based on the mtDNA (Table S6). Pairwise estimates of Fst between populations
ranged from −0.09 to 0.98 based on the RAG1 and ranged from −0.33 to 0.98 based on the
PLAGL2 sequences (Tables S7 and S8). In addition, the Mantel test generated r values of
0.263 (p = 0.007), 0.397 (p = 0.006), and 0.308 (p = 0.028) for mtDNA, RAG1, and PLAGL2
 I-1
I-2 0.014
Animals 2025, 15, 1202
I-3 0.014 0.016 10 of 20

I-4 0.018 0.022 0.016

data,II-1
respectively,0.031
when evaluating0.034 0.033 and geographical
the genetic diversity 0.038 distance in the P.
II-2populations
anomala 0.032
(Figure 4). 0.036 0.034 0.040 0.006

Figure
Figure Scatter
4. 4. plotsplots
Scatter of genetic distancedistance
of genetic vs. geographical distance (km:distance
vs. geographical kilometer)(km:
for pairwise
kilometer) for
population comparisons inferred from mtDNA (a), RAG1 (b), and PLAGL2 (c).
population comparisons inferred from mtDNA (a), RAG1 (b), and PLAGL2 (c).
3.4. Population Genetic Structure and Historical Population Dynamics of P. anomala
3.4.SAMOVA
Population Genetic
results Structure
showed that theand Historical
highest Population
FCT value Dynamics
(FCT = 0.835) of P. anomala
was a grouping
arrangement
SAMOVA results showed that the highest FCT value (FCT = 0.835) with
of Clade I and the remaining population, which geographically aligned was a grou
Clade II (Table 4). AMOVA was used to find the best phylogeographic pattern for P. anomala
rangement of Clade I and the remaining population, which geographically align
from two grouping options identified by present drainage or results of genetic analyses
Clade5).IIHierarchical
(Table (Table 4). AMOVA
AMOVA was used
revealed to find according
that grouping the best to phylogeographic
the phylogenetic patter
anomala
tree from two
result resulted in thegrouping options identified
highest among-group by
variation (F CT present
= 0.793, p drainage
< 0.001) andorwas
results o
inferred
analyses to be the most
(Table 5). probable geographical
Hierarchical AMOVA subdivision (Table
revealed 5).grouping
that Finally, based on the to th
according
nuclear gene RAG1, genetic variation within populations was the main source of total
genetic tree result resulted in the highest among-group variation (FCT = 0.793, p
variation, with the proportion of variation within populations being 71.97%. Based on the
and was inferred to be the most probable geographical subdivision (Table 5).
nuclear gene PLAGL2, genetic variation within populations was the main source of total
based on
variation, the
with thenuclear
proportion gene RAG1,within
of variation genetic variation
populations within
being 88.02%.populations was t
 Animals 2025, 15, 1202 11 of 20

Table 4. SAMOVA results based on mtDNA (K = 2) of 22 P. anomala sites from Southwest China.

Sum of Variance Percentage of

K Source of Variation d.f. p Value
Squares Components Variation
Among clades 1 2070.317 18.03249 83.53 0.000
Among populations within regions 20 711.765 3.39220 15.71 0.000
2
Within populations 231 37.495 0.16231 0.75 0.000
Total 252 2819.577 21.58701

Table 5. Analysis of molecular variance (AMOVA) of P. anomala based on mtDNA data. Significance
test: 1000 permutations.

Grouping Variation (%)

F CT F SC F ST
Options Among Clades Among Populations Within Clades Within Populations
1. The Pearl River and the Yangtze River
62.11 29.50 8.38 0.621 * 0.779 *** 0.916 ***
2. Phylogenetic tree groupings
79.27 12.50 8.23 0.793 * 0.603 *** 0.918 ***
Note: *: p < 0.05; ***: p < 0.001.

The STRUCTURE analysis based on mtDNA data demonstrated that the most likely
number of genetic clusters for the P. anomala populations was two (mean lnP(K) = –268.970;
mean Delta K = 888.563) (Figure 5a). Cluster 1 included populations I-1 (Guijiang River),
I-2 (Duliujiang River), and I-3 and I-4 (Yangtze River). Cluster 2 comprised populations
Animals 2025, 15, x FOR PEER REVIEW 12 of 21
II-1 (Nanpanjiang River) and II-2 (Hongshui River, Rongjiang River, Youjiang River, and
Dahuanjiang River) (Figure 5b).

Figure
Figure TheThe
5. 5. results
results of of structureanalysis:
structure analysis: (a)
(a)plot
plot of
of delta distribution ofofP.P.anomala
delta K distribution anomalasites;
sites; (b)
(b) histogram
histogram of theofstructure
the structure analysis
analysis forfor
thethe modelwith
model withKK==2;2;the x-axis shows
the x-axis showsclades,
clades,the propor-
the proportion
tion of each color represents the likelihood of each individual being assigned to the corresponding
of each color represents the likelihood of each individual being assigned to the corresponding clus-
cluster.
ter.
The historical population dynamics of P. anomala were analyzed using the two genetic
clusters
Table (CladeofI neutrality
6. Statistics and CladetestII)and
identified
mismatch from the AMOVA
distribution analysis.
analysis resultsThese
for twoanalyses
lineages of P.
provided
anomala. crucial insights into the evolutionary history of P. anomala (Table 6; Figure 6). Clade
I exhibited a multimodal distribution with steep curves, characterized by relatively high
Grouping (Populations)
levels of haplotype π diversity.
and nucleotide h TheTajima’s D (p
positive but Value)
nonsignificant Fu’s Fs (p
Tajima’s Value)
D and
Clade for
Fu’s Fs values I Clade I suggested
0.01030 that0.790 1.37841
a population (0.936)
expansion model could 15.00847 (0.995)
be rejected.
Clade
In contrast, II II showed
Clade 0.00262 0.681
a unimodal −1.49888
distribution with (0.040)
smooth curves,−5.68864
indicative(0.052)
of
Animals 2025, 15, 1202 12 of 20

population expansion. The negative and significant Tajima’s D and Fu’s Fs values for
Clade II further supported the inference of a historical population expansion. Additionally,
Figure 5. The results of structure analysis: (a) plot of delta K distribution of P. anomala sites; (b)
the haplotype network of Clade II, centered around haplotype Hap6, displayed a star-
histogram of the structure analysis for the model with K = 2; the x-axis shows clades, the proportion
like phylogenetic structure (Figure 3). This pattern was consistent with rapid population
of each color represents the likelihood of each individual being assigned to the corresponding clus-
expansion and further corroborated the demographic inference for Clade II.
ter.
Table 6. Statistics of neutrality test and mismatch distribution analysis results for two lineages of
P.Table 6. Statistics of neutrality test and mismatch distribution analysis results for two lineages of P.
anomala.
anomala.
Grouping
π h Tajima’s D (p Value) Fu’s Fs (p Value)
Grouping (Populations)
(Populations) π h Tajima’s D (p Value) Fu’s Fs (p Value)
Clade
Clade I I 0.01030 0.790
0.01030 0.790 1.37841
1.37841(0.936)
(0.936) 15.00847(0.995)
15.00847 (0.995)
Clade
Clade II II 0.00262 0.681
0.00262 0.681 −1.49888(0.040)
−1.49888 (0.040) −−5.68864 (0.052)
5.68864 (0.052)

Figure 6. Mismatch distributions of two clades of P. anolama inferred from mtDNA sequences.

3.5. Estimation of Divergence Time

Estimated divergence times based on the mtDNA data among the clades of P. anomala
are shown in Figure 7. The divergence age estimates indicated that the split between P.
anomala and P. cochinchinensis occurred around 17.69 million years ago (Mya; 95% HPD,
12.77–21.72 Mya; Figure 7). The divergence time analysis further revealed that the two
main clades diverged approximately 13.73 Mya (95% HPD = 8.50–18.81 Mya; Figure 7).
 Estimated divergence times based on the mtDNA data among the clades of P. anomala
are shown in Figure 7. The divergence age estimates indicated that the split between P.
anomala and P. cochinchinensis occurred around 17.69 million years ago (Mya; 95% HPD,
Animals 2025, 15, 1202 13 of 20
12.77–21.72 Mya; Figure 7). The divergence time analysis further revealed that the two
main clades diverged approximately 13.73 Mya (95% HPD = 8.50–18.81 Mya; Figure 7).

Figure7.7.Time
Figure Timetree
treeofofthe speciesP.P.anomala
thespecies anomalabased
basedon
onthe
themtDNA
mtDNAdata.
data. Black dots mark the
the calibra-
calibration
nodes. Numbers
tion nodes. at nodes
Numbers indicate
at nodes the main
indicate divergence
the main times
divergence of the
times haplotypes
of the haplotypesofof
the species.
the species.

4.4.Discussion
Discussion
4.1.
4.1.Genetic
GeneticDiversity
Diversity of of P.
P. anomala
anomala
Genetic
Genetic diversity is afundamental
diversity is a fundamental condition
condition for for species
species toto maintain
maintain their
theirevolution-
evolutionary
potential and is crucial for the formation, development, and sustainability
ary potential and is crucial for the formation, development, and sustainability of biodiver- of biodiver-
sity.
sity.ItItenables
enablesspecies
species toto adapt
adapt to to complex
complexand andchanging
changingenvironments,
environments, thereby
thereby ensuring
ensuring
their survival
their survival [56–58]. Nevo [59] proposed that genetic diversity is correlated with with
Nevo [59] proposed that genetic diversity is correlated envi- en-
vironmental heterogeneity
ronmental heterogeneity across
across different
different spatial
spatial scales,
scales, highlighting
highlighting the importance
the importance of di- of
diverse habitats
verse habitats inin maintaining
maintaining genetic
genetic variation.
variation. Based Based on mtDNA,
on mtDNA, this study
this study analyzedanalyzed
the
genetic
the geneticdiversity of populations
diversity of populations of P. anomala in CladeinI and
of P. anomala Clade Clade II. The
I and Clade results showed
II. The results
that thethat
showed genetic diversity
the genetic in CladeinI (h
diversity = 0.790,
Clade I (h π= =0.790,
0.01030)
π = was higher
0.01030) wasthan that than
higher in Clade
that in
II (h = 0.681, π = 0.00262). This difference may be closely related to the
Clade II (h = 0.681, π = 0.00262). This difference may be closely related to the geographical geographical envi-
ronment in which
environment CladeClade
in which II is located. FurtherFurther
II is located. analysisanalysis
indicated significant
indicated genetic differ-
significant genetic
entiation between the population in the Hongshui River and the
differentiation between the population in the Hongshui River and the populations in the populations in the Gui-
jiang andand
Guijiang Rongjiang
Rongjiang Rivers. This This
Rivers. differentiation may stem
differentiation mayfrom
stemthe limited
from dispersal
the limited abil-
dispersal
ity of P.
ability of anomala
P. anomalaitself, as well
itself, as theascombined
as well the combinedeffectseffects
of geographical isolationisolation
of geographical and envi-and
ronmental differences. Compared with other fish species, the
environmental differences. Compared with other fish species, the genetic diversity genetic diversity of Clade II of
is lower than that of the Pelteobagrus fulvidraco (Richardson, 1846) (h = 0.848, π = 0.0480)
Clade II is lower than that of the Pelteobagrus fulvidraco (Richardson, 1846) (h = 0.848, π
[60] and the Ptychidio jordani Myers, 1930 (h = 0.768, π = 0.0023) [61], which are also dis-
= 0.0480) [60] and the Ptychidio jordani Myers, 1930 (h = 0.768, π = 0.0023) [61], which are
tributed in the Pearl River. This difference may be closely related to the ecological habits
also distributed in the Pearl River. This difference may be closely related to the ecological
and habitat characteristics of P. anomala. Once a species adapts to cave-dwelling life, re-
habits and habitat characteristics of P. anomala. Once a species adapts to cave-dwelling
duced selective pressures and smaller population sizes can exacerbate genetic drift,
life, reduced selective pressures and smaller population sizes can exacerbate genetic drift,
thereby reducing genetic diversity [62]. Moreover, genetic differentiation coefficients (Fst)
based on mitochondrial DNA, RAG1, and PLAGL2 sequences, as well as Mantel tests,
showed significant genetic differentiation between populations of P. anomala in Clade I and
Clade II, revealing distinct geographical distribution patterns. These findings were also
supported by phylogenetic tree and haplotype network analyses. The observed genetic
differentiation may be due to the marked contrast between the tropical maritime climate
of the Pearl River and the temperate monsoon climate of the Yangtze River basin. This
significant north–south climatic difference may be one of the important factors leading to
genetic differentiation. In addition, there were also considerable differences between the
Pearl River and Yangtze River ecosystems in terms of water temperature, flow velocity,
Animals 2025, 15, 1202 14 of 20

water quality, and other aspects. These factors, acting together, further exacerbate the
genetic differentiation between the two populations [63].

4.2. Genetic Structure of P. anomala

In this study, by integrating mtDNA and nuclear DNA molecular markers of P. anomala
populations, we systematically analyzed the population genetic structure characteristics
of this species. Notably, we observed a significant incongruence between nuclear DNA
haplotypes and mitochondrial gene haplotype networks. Comparative analysis indicated
that the phylogeographical structure reconstructed based on mitochondrial gene markers
exhibited a more pronounced phylogeographical differentiation pattern than nuclear gene
data, while the phylogenetic topology constructed by nuclear genes failed to resolve distinct
evolutionary clades. This phenomenon of incongruence between mitochondrial and nuclear
DNA is widespread in many organisms [64,65]. Such incongruence typically implies
ancient hybridization events and incomplete lineage sorting (ILS) [66]. Mitochondrial
genes, due to their faster substitution rates and smaller effective population sizes, are more
susceptible to hybridization in phylogenetic analyses. In contrast, nuclear genes, with their
slower substitution rates and larger effective population sizes, are more prone to ILS in
phylogenetic analyses [66,67]. Additionally, the relatively slow mutation rate of nuclear
genes somewhat limits their ability to resolve population structure, a phenomenon that
has been verified in studies of the Baryancistrus xanthellus Rapp Py-Daniel, Zuanon & de
Oliveira, 2011 [68].
The phylogenetic trees and haplotype network based on mtDNA data revealed the
genetic differentiation characteristics of P. anomala populations. The research findings
indicated that P. anomala populations could be divided into two main clades, with their
distribution closely related to geographical regions (Figure 2). These results suggested
that the Yangtze River population shared a relatively close genetic relationship with the
populations of the Guijiang and Duliujiang Rivers. Similar patterns had also been reported
in Hemibarbus medius Yue, 1995 and Hemibarbus labeo (Pallas, 1776) [69], as well as Hypoph-
thalmichthys nobilis (Richardson, 1845) [70]. The close genetic relationships among subclades
within Clade I and the current distribution patterns of Clade I and Clade II can be attributed
to the following factors.
First, the monsoon climate impacts brought about by the uplift of the QTP, as well
as the formation of the Nanling Mountains, have played significant roles. Phylogenetic
analysis using BEAST estimated that these lineages diverged approximately 8.5–18.81 mya
(Figure 7), a period that was potentially influenced by the southwest monsoon [71]. During
the Miocene, as the QTP approached its present-day elevation, its uplift significantly
intensified the monsoon systems and altered marine biogeochemical processes. This
process not only drove the South Asian and East Asian monsoons into an enhanced phase
but also fundamentally changed the evolution of the regional topography [72–75]. The
mid-Miocene climatic optimum (approximately 17–14 mya) saw intensified monsoon
precipitation that dramatically accelerated karst cave system development [76]. Studies
have shown that the ancestral populations of the cave-dwelling and non-cave-dwelling
groups of the genus Triplophysa initially diverged around 15.3–13 million years ago, which
is inferred to be closely related to the continuous uplift of the plateau [72]. We speculate that
the differentiation of populations of the P. anomala inhabiting the same karst environment
is also likely to be significantly associated with the ongoing uplift of the QTP. Concurrently,
the formation of the Nanling Mountains as a geographical barrier played a significant
role in the genetic differentiation of these fish populations. During its early uplift, the
Nanling Mountains had a relatively low elevation, which did not effectively block gene
flow between freshwater fish populations on either side [69]. However, during the uplift
Animals 2025, 15, 1202 15 of 20

of the QTP, the Nanling Mountains experienced accelerated uplift [77]. This rapid uplift
transformed the Nanling Mountains into a geographical barrier between the northern
and southern river systems [69]. The continued uplift ultimately established a divide
between the Pearl River and Yangtze River systems. This divide prevented warm air
masses south of the mountains from moving northward and blocked cold air masses north
of the mountains from penetrating southward. Consequently, a climatic boundary formed
between the south subtropical and mid-subtropical zones, restricting gene flow and leading
to genetic differentiation [78]. In summary, the uplift of QTP and the formation of the
Nanling Mountains as a geographical barrier both contributed to the genetic differentiation
observed in fish populations in the region.
Second, river capture events played a role. Tectonic movements and river capture
events led to connections or diversions between the Yangtze River and Pearl River sys-
tems. Early geological and ichthyological studies showed that some rivers in the Pearl
River and Yangtze River were geographically very close, with a few even being directly
connected [79,80]. For example, the Lingqu Canal, which connected the Xiangjiang (a tribu-
tary of the Yangtze River) and Guijiang Rivers [5,80], may also have promoted gene flow
between populations of the Yangtze River and Pearl River systems.
The results of the haplotype network analysis showed that there is haplotype sharing
between the Pearl River and the Yangtze River (Figures S1 and S2). It is inferred that this
phenomenon is more likely the result of the combined effects of the dispersal of the Yangtze
River population and shared ancestral polymorphisms. According to the studies of Yang
et al., (2022) and Li et al., (2020), if ancestral polymorphisms are shared between the Pearl
River and Yangtze River, each basin should retain a subset of unique haplotypes [70,81].
However, the haplotype network reveals that, despite shared haplotypes, both basins
exhibit distinct unique haplotypes. This indicates that genetic differentiation between
populations is not solely attributable to ancestral polymorphism retention, and dispersal
of the Yangtze River population likely contributed to the observed pattern. Secondly, the
analysis of average genetic distances between populations (Table 3) showed that the genetic
distance between populations of Clade II and the four subclades of Clade I increased
gradually from north to south. This suggested that some ancestors of the P. anomala
populations in the Pearl River likely originated from the Yangtze River. It was worth noting
that, in the Pearl River and Yangtze River, genetic differentiation among different fish
populations exhibited diverse characteristics. For example, the genetic distance among
individuals of P. anomala ranged from 0% to 4%, a value significantly lower than that of
the sympatric species Squaliobarbus curriculus (Richardson, 1846) (with genetic distances
ranging from 0% to 7.43%) [78] but slightly higher than the genetic distance between
populations of Hypophthalmichthys molitrix (Valenciennes, 1844) across basins (at 2.3%) [70].
Although P. anomala (0–4%), H. molitrix (2.3%), and S. curriculus (0–7.43%) all showed clear
genetic differentiation between populations in the Pearl River and Yangtze River, this
degree of differentiation had not reached the level of speciation. This demonstrated that
there is no correspondence between genetic distance and species boundaries, and genetic
differentiation between populations, even up to 7.43%, might still fall within the range
of intraspecific variation. Therefore, when delimiting species, it is not sufficient to rely
solely on genetic distance as a single indicator. Instead, multidimensional evidence, such
as morphological differences, reproductive isolation, and ecological niche differentiation,
should be taken into account [82].

4.3. Implications for Conservation

Deciphering the genetic structure of specific species can provide robust support for
species management and conservation [83]. Given that P. anomala’s wild resources are
Animals 2025, 15, 1202 16 of 20

scarce and artificial breeding has not yet been successful, refining conservation units and
implementing in situ conservation is particularly crucial. Therefore, analyzing the genetic
structure of P. anomala across different river systems through phylogeographic analysis
is of great significance for accurately delineating its conservation units. In our study, we
observed two clades in P. anomala populations, which we proposed should be recognized
as two molecular operational taxonomic units (MOTUs). Although P. anomala is currently
listed as a species of Least Concern on the IUCN Red List due to its wide distribution
and seemingly stable population [84], our mitochondrial DNA (mtDNA) analysis revealed
relatively low haplotype and nucleotide diversity in the Youjiang, Rongjiang, and Guijiang
River basins. The genetic diversity of these populations was significantly lower than
that of Pelteobagrus fulvidraco and Hemibarbus medius in the same river basins. Moreover,
the analysis of genetic diversity using nuclear gene markers also showed relatively low
haplotype and nucleotide diversity in the Nanpanjiang, Duliujiang, Dahuanjiang, and
Hongshui Rivers. The genetic diversity in Duliujiang and Dahuanjiang Rivers might have
been limited by the small sample size. Cave-dwelling fish, due to their unique habitats, face
multiple threats including habitat degradation, environmental pollution, overexploitation
of resources, and invasion of non-native species. In China, most cave-dwelling fish have
limited distribution ranges and small population sizes, making them highly vulnerable
to even minor threats [85]. To protect these species, effective conservation measures must
be implemented, including strengthening habitat protection, reducing pollution sources,
managing resource exploitation sustainably, and preventing the invasion of non-native
species, to ensure their survival and reproduction.

5. Conclusions
This study has conducted a focused and comprehensive investigation into the phylo-
geography and diversity of P. anomala in Southwest China. The identification of two distinct
genetic lineages highlights the pivotal role of geological events and river systems in shaping
the evolutionary trajectory of the species. Influenced by geographical and environmental
factors, the relatively low genetic diversity and high genetic differentiation among evolu-
tionary branches underscore the necessity of considering fine-scale spatial heterogeneity
when devising conservation strategies. In summary, this study emphasizes the critical
impact of geological, climatic, and riverine factors on the distribution and evolutionary
history of P. anomala in the Yangtze and Pearl River and highlights the urgency of further
research to protect this species. We plan to expand the scope of our study and increase the
sample size in future research and combine more types of markers (such as SNP, SSR, etc.)
to comprehensively assess the population genetic structure of P. anomala, thereby providing
a more scientific and systematic basis for the formulation of its conservation strategies.

Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/ani15091202/s1, Figure S1: The distribution of the haplotypes
based on RAG1 in the P. anomala populations; each haplotype is represented by a circle. The size of the
circle is proportional to that haplotype’s frequency. Different colors indicate different clades; Figure
S2: The distribution of the haplotypes based on PLAGL2 in the P. anomala populations; each haplotype
is represented by a circle. The size of the circle is proportional to that haplotype’s frequency. Different
colors indicate different clades; Table S1: Primers information; Table S2: Geographic distance (km)
among populations of P. anomala; Table S3: The distribution of the haplotypes based on mtDNA
in the P. anomala populations; Table S4: The distribution of the haplotypes based on RAG1 in the
P. anomala populations; Table S5: The distribution of the haplotypes based on PLAGL2 in the P.
anomala populations; Table S6: Pairwise genetic differentiation of P. anomala populations inferred
from mtDNA; Table S7: Pairwise genetic differentiation of P. anomala populations inferred from
RAG1; Table S8: Pairwise genetic differentiation of P. anomala populations inferred from PLAGL2;
Animals 2025, 15, 1202 17 of 20

Table S9: Analysis of molecular variance (AMOVA) of P. anomala based on RAG1 and PLAGL2 data.
Significance test: 1000 permutations.

Author Contributions: Conceptualization, R.Z. and R.H.; methodology, H.L., J.C. and H.C.; software,
R.H.; formal analysis, R.H. and J.C.; investigation, writing—original draft preparation, H.L. and H.C.;
writing—review and editing, R.H. and R.Z.; supervision, R.Z. All authors have read and agreed to
the published version of the manuscript.

Funding: This research was founded by the National Natural Science Foundation of China (31960097).

Institutional Review Board Statement: All specimen collections in this study were conducted in strict
compliance with Chinese laws and were formally reviewed and approved by the Research Ethics
Committee of Guizhou Normal University (Approval No. 20221100004). All experimental procedures
strictly adhered to internationally recognized animal welfare guidelines and ethical standards for
biological research.

Informed Consent Statement: Not applicable.

Data Availability Statement: All sequences generated during this study have been deposited in
GenBank (Accession numbers: PV411066−PV411294, PV411932−PV412184, PV540427−PV540681,
PV546889−PV547139).

Acknowledgments: We are grateful to Qian Tang, Qi Luo, Lei Deng, Qian Duan, Leishan Wang, and
Tingting Zhu for their help with sample collection.

Conflicts of Interest: The authors declare no conflicts of interest.

Abbreviations
The following abbreviations are used in this manuscript:

QTP Qinghai–Tibet Plateau

Mya Million years ago
ILS Incomplete lineage sorting

References
1. Lu, M.; Zhang, H.S.; An, H.M. Chloroplast DNA-based genetic variation of Rosa roxburghii in Southwest China: Phylogeography
and conservation implications. Hortic. Plant. J. 2021, 7, 286–294. [CrossRef]
2. Wang, S.J.; Li, Y.B. Problems and development trends about karst rocky desertification. Adv. Earth. Sci. 2007, 22, 573–582.
3. Qiu, Y.X.; Fu, C.X.; Hans, P.C. Plant molecular phylogeography in China and adjacent regions: Tracing the genetic imprints
of quaternary climate and environmental change in the world’s most diverse temperate flora. Mol. Phylogenet. Evol. 2011, 59,
225–244. [CrossRef]
4. Ji, S.Q. Taxonomic Study on the Limacodidae from Southwest China (Lepidoptera: Limacodidae). Master’s Thesis, South China
Agricultural University, Guangzhou, China, 2018.
5. Feng, Q.X. Preliminary Study on Fishes and Their Characteristics in the Pearl River System. Freshw. Fish. 1985, 3, 14–22.
6. Yang, L.; Mayden, R.L.; He, S. Population genetic structure and geographical differentiation of the Chinese catfish Hemibagrus
macropterus (Siluriformes, Bagridae): Evidence for altered drainage patterns. Mol. Phylogenet. Evol. 2009, 51, 405–411. [CrossRef]
7. Perdices, A.; Cunha, C.; Coelho, M.M. Phylogenetic structure of Zacco platypus (Teleostei, Cyprinidae) populations on the upper
and middle Chang Jiang (=Yangtze) drainage inferred from cytochrome b sequences. Mol. Phylogenet. Evol. 2004, 31, 192–203.
[CrossRef] [PubMed]
8. Zeng, Q.Q.; Sun, Y.X.; Zhong, H.; Yang, C.H.; Qin, Q.B.; Gu, Q.H. Population genomic evidence for the diversification of Bellamya
aeruginosa in different river systems in China. Biology 2023, 12, 29. [CrossRef]
9. Lin, B.J. Genetic Diversity Analysis of Hynobius maoershanensis. Master’s Thesis, Guangxi Normal University, Guangxi, China,
2015.
10. Huang, H.Y.; Bu, R.P.; Xie, H.; Hou, S.B.; Wu, Z.J. Habitat selection by Hynobius maoershanensis during its breeding period. Acta.
Ecol. Sin. 2019, 39, 6443–6451.
11. Che, J.; Zhou, W.W.; Hu, J.S.; Yan, F.; Papenfuss, T.J.; Wake, D.B.; Zhang, Y.P. Spiny frogs (Paini) illuminate the history of the
Himalayan region and Southeast Asia. Proc. Natl. Acad. Sci. USA 2010, 107, 13765–13770. [CrossRef]
12. Harrison, T.; Copeland, P.; Kidd, W.; Yin, A. Raising Tibet. Science 1992, 255, 1663–1670. [CrossRef]
Animals 2025, 15, 1202 18 of 20

13. Bao, L.; Wang, N.; Ni, Z.Y.; Lu, T. Influence of the Tibetan Plateau uplift on climate evolution in southwestern China: From the
monsoon perspective. J. Earth. Envir. 2018, 9, 444–454. [CrossRef]
14. Clark, M.; Schoenbohm, L.; Royden, L.; Whipple, K.; Burchfiel, B.; Zhang, X.; Tang, W.; Wang, E.; Chen, L. Surface uplift, tectonics,
and erosion of eastern Tibet from large-scale drainage patterns. Tectonics 2004, 23, 1006–1029. [CrossRef]
15. Yang, J.; Chen, X.; Bai, J.; Fang, D.M.; Qiu, Y.; Jiang, W.S.; Yuan, H.; Bian, C.; Lu, J.; He, S.Y.; et al. The Sinocyclocheilus cavefish
genome provides insights into cave adaptation. BMC Biol. 2016, 14, 1. [CrossRef] [PubMed]
16. Vincenzi, V.; Furin, S.; Fazzini, M.; Zagoršek, T.; Jugovic, J. The importance of hydrogeological, geological and climatic features in
the karst landscape for the protection of water resources and biodiversity. In Biodiversity and Conservation of Karst Ecosystems;
Bužan, E., Pallavicini, A., Eds.; Padova University Press: Padova, Italy, 2014; pp. 85–106.
17. Wen, H.M.; Liu, E.S.; Yan, S.S.; Chang, J.; Xiao, N.; Zhou, J. Conserving karst cavefish diversity in Southwest China. Biol. Conserv.
2022, 273, 109680. [CrossRef]
18. Wang, S.; Zhang, X.; Bai, X. An outline of karst geomorphology zoning in the karst areas of southern China. Mount. Res. 2015, 33,
641–648. [CrossRef]
19. Yuan, D.X. Karstology of China; Yuan, D.X., Ed.; Geological Press: Beijing, China, 1994; pp. 154–196.
20. Zhao, Y.H.; Gozlan, R.E.; Zhang, C.G. Out of sight out of mind: Current knowledge of Chinese cave fishes. J. Fish. Biol. 2011, 79,
1545–1562. [CrossRef]
21. Zhao, Q.Y.; Shao, F.; Li, Y.P.; Yi, S.V.; Peng, Z.G. Novel genome sequence of Chinese cavefish (Triplophysa rosa) reveals pervasive
relaxation of natural selection in cavefish genomes. Mol. Ecol. 2022, 31, 5831–5845. [CrossRef]
22. Mao, T.R.; Liu, Y.W.; Meegaskumbura, M.; Yang, J.; Ellepola, G.; Senevirathne, G.; Fu, C.H.; Gross, J.B.; Pie, M.R. Evolution in
Sinocyclocheilus cavefish is marked by rate shifts, reversals, and origin of novel traits. BMC Ecol. Evol. 2021, 21, 45. [CrossRef]
23. Zhou, J.J.; Qin, Z.X.; Du, L.N.; Wu, H.Y. A New Species of the Blind Cave Loach Genus Protocobitis (Cypriniformes: Cobitidae),
Protocobitis longicostatus sp. nov., from Guangxi, China. Zoolog. Sci. 2024, 41, 210–215. [CrossRef]
24. Aldemaro, R. The Biology of Hypogean Fishes; Aldemaro, R., Ed.; Springer: Dordrecht, The Netherlands, 2001; pp. 13–41.
25. Helfman, G.S. Fish Conservation: A Guide to Understanding and Restoring Global Aquatic Biodiversity and Fishery Resources; Island
Press: Washington, DC, USA, 2007; pp. 485–563.
26. Zhao, Y.H.; Zhang, C.G. Cavefishes: Concept, diversity and research progress. Biodiv. Sci. 2006, 14, 451–460. [CrossRef]
27. Wang, D.Z.; Chen, Y.Y. The origin and adaptive evolution of the genus Sinocyclocheilus. Acta. Hydrobiol. Sin. 2000, 24, 630–634.
[CrossRef]
28. Chen, W.T.; Li, Y.F.; He, Y.J.; Li, H.X.; Li, J. Characterization of two complete mitochondrial genomes of Pterocryptis anomala
(Siluridae) and its phylogeny and cryptic diversity. Biologia 2021, 76, 613–621. [CrossRef]
29. Chu, X.L.; Zheng, B.S.; Dai, D.Y. Fauna Sinica, Osteichthyes: Siluriformes; Scientific Press: Beijing, China, 1999; pp. 79–80.
30. Zhou, J.; Zhang, C. Freshwater Fishes of Guangxi, China; Guangxi People’s Publishing House: Nanning, China, 2006; pp. 386–387.
31. Aljanabi, S.M.; Martinez, I. Universal and rapid salt-extraction of high quality gnomic DNA for PCR-based techniques. Nucleic.
Acids. Res. 1997, 25, 4692–4693. [CrossRef]
32. Tamura, K.; Stecher, G.; Kumar, S. MEGA11: Molecular Evolutionary Genetics Analysis Version 11. Mol. Biol. Evol. 2021, 38,
3022–3027. [CrossRef]
33. Librado, P.; Rozas, J. DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics 2009, 25,
1451–1452. [CrossRef]
34. Garrick, R.C.; Sunnucks, P.; Dyer, R.J. Nuclear gene phylogeography using PHASE: Dealing with unresolved genotypes, lost
alleles, and systematic bias in parameter estimation. BMC Evol. Biol. 2010, 10, 118. [CrossRef] [PubMed]
35. Ronquist, F.; Huelsenbeck, J.P. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 2003, 19, 1572–1574.
[CrossRef]
36. Zhang, D.; Gao, F.L.; Jakovlić, I.; Zou, H.; Zhang, J.; Li, W.X.; Wang, G.T. PhyloSuite: An integrated and scalable desktop platform
for streamlined molecular sequence data management and evolutionary phylogenetics studies. Mol. Ecol. Resour. 2020, 20,
348–355. [CrossRef]
37. Nylander, J. MrModeltest V2. Program Distributed by the author. Bioinformatics 2004, 24, 581–583. [CrossRef]
38. Letunic, I.; Bork, P. Interactive Tree Of Life (iTOL) v4: Recent updates and new developments. Nucleic. Acids. Res. 2019, 47,
256–259. [CrossRef]
39. Jessica, W.L.; David, B. POPART: Full-feature software for haplotype network construction. Methods. Ecol. Evol. 2015, 6, 1110–1116.
[CrossRef]
40. Kimura, M. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide
sequences. J. Mol. Evol. 1980, 16, 111–120. [CrossRef]
41. Excoffier, L.; Lischer, H.E. Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under linux
and windows. Mol. Ecol. Resour. 2010, 10, 564–567. [CrossRef]
Animals 2025, 15, 1202 19 of 20

42. Oksanen, J.; Sinpson, G.L.; Blanchet, F.G.; Kindt, R.; Legendre, P.; Minchin, P.R.; O’Hara, R.B.; Solymos, P.; Steven, M.H.H.; Szoecs,
E.; et al. “vegan: Community Ecology Package.” R package version 2.6-4. 2022. Available online: https://CRAN.R-project.org/
package=vegan (accessed on 20 April 2025).
43. Ihaka, R.; Gentleman, R. R: A language for data analysis and graphics. J. Compu. Graph. Stat. 1996, 5, 299–314. [CrossRef]
44. Pritchard, J.K.; Stephens, M.; Donnelly, P. Inference of population structure using multilocus genotype data. Genetics 2000, 155,
945–959. [CrossRef] [PubMed]
45. Earl, D.A.; Vonholdt, B.M. Structure harvester: A website and program for visualizing structure output and implementing the
Evanno method. Conserv. Genet. Resour. 2012, 4, 359–361. [CrossRef]
46. Evanno, G.; Regnaut, S.; Goudet, J. Detecting the number of clusters of individuals using the software structure: A simulation
study. Mol. Ecol. 2005, 14, 2611–2620. [CrossRef] [PubMed]
47. Jakobsson, M.; Rosenberg, N.A. CLUMPP: A cluster matching and permutation program for dealing with label switching and
multimodality in analysis of population structure. Bioinformatics 2007, 23, 1801–1806. [CrossRef]
48. Rosenberg, N.A. DISTRUCT: A program for the graphical display of population structure. Mol. Ecol. Notes. 2004, 4, 137–138.
[CrossRef]
49. Tajima, F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 1989, 123, 585–595.
[CrossRef]
50. Fu, Y.X. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics
1997, 147, 915–925. [CrossRef] [PubMed]
51. Drummond, A.J.; Rambaut, A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol. 2007, 7, 214. [CrossRef]
[PubMed]
52. Chen, W.T.; Hubert, N.; Li, Y.F.; Zhu, S.L. Mitogenomic phylogeny, biogeography, and cryptic divergence of the genus Silurus
(Siluriformes: Siluridae). Zool. Res. 2024, 45, 711–723. [CrossRef] [PubMed]
53. Kappas, I.; Vittas, S.; Pantzartzi, C.N.; Drosopoulou, E.; Scouras, Z.G. A time-calibrated mitogenome phylogeny of catfish
(Teleostei: Siluriformes). PLoS ONE 2016, 11, e0166988. [CrossRef]
54. Rambaut, A.; Drummond, A.J.; Xie, D.; Baele, G.; Suchard, M.A. Posterior Summarization in Bayesian Phylogenetics Using Tracer
1.7. Syst. Biol. 2018, 67, 901–904. [CrossRef]
55. Rambaut, A. FigTree 1.4.3. 2016. Available online: http://tree.bio.ed.ac.uk/software/figtree (accessed on 5 March 2025).
56. Lande, R. Genetics and demography in biological conservation. Science 1988, 241, 1455–1460. [CrossRef]
57. Reed, D.H.; Lowe, É.H.; Biscoe, D.A.; Richard, F. Fitess and adaptation in a novel environment:effect ofinbreeding, prior
environment, and lineage. Evolution 2003, 57, 1822–1828. [CrossRef]
58. Frankham, R. Genetics and extinction. Biol. Conserv. 2005, 126, 131–140. [CrossRef]
59. Nevo, E. Evolution of genome-phenome diversity under environmental stress. Proc. Natl. Acad. Sci. USA 1998, 98, 6233–6240.
[CrossRef]
60. Zhang, H.Q.; Yang, Z.T.; Li, J.F.; Xiao, S.B.; Sun, J.J.; Yang, H.R.; Zhao, H.H.; Xie, T.H.; Han, X.P.; Liu, L. Genetic Diversity of
Mitochondrial Cyt b Sequences in Pelteobagrus fulvidraco from the Pearl River. Acta. Sci. Nat. Univ. Sunyatseni. 2015, 54, 102–108.
61. Peng, M.; Han, Y.Q.; Wang, D.P.; Shi, J.; Wu, W.J.; Li, Y.S.; Lei, J.J.; He, A.Y. Genetic diversity analysis of Ptychidio jordani in Xijiang
River flowing through Guangxi Province based on mitochondrial Cyt b gene sequence. S. China Fish. Sci. 2020, 16, 10–18.
62. Zhang, X.J.; Dai, Y.G. Research Progress on Karst Cave Fishes in China. J. Shanghai Ocean Univ. 2010, 19, 364–371.
63. Yang, H.R.; Zhao, H.H.; Liu, L.; Chen, Y.A.; Lin, Q.Z.; Huang, H.H. Genetic variation analysis of D-loop gene sequence of
Squaliobarbus curriculus in Pearl River and Yangtze River basins. J. Shanghai Ocean Univ. 2012, 21, 337–343. [CrossRef]
64. Chiocchio, A.; Rysky, D.E.; Carere, C.; Nascetti, G.; Bisconti, R.; Canestrelli, D. Behavioural underpinning of mito-nuclear
discordances: Insights from fire salamanders. R. Soc. Open Sci. 2024, 11, 241571. [CrossRef] [PubMed]
65. Antonia, G.P.F.; Thomas, R.B.; Longson, P.; Martin, J.G.; Roger, B.; Tomáš, F.; Benjamin, P.N.; Lukas, R.; Ulrich, K.S.; Ole, S.; et al.
Molecular phylogeny of Oreochromis (Cichlidae: Oreochromini) reveals mito-nuclear discordance and multiple colonisation of
adverse aquatic environments. Mol. Phylogenet. Evol. 2019, 136, 215–226. [CrossRef]
66. Kocher, T.D.; Thomas, W.K.; Meyer, A.; Edwards, S.V.; Pääbo, S.; Villablanca, F.X.; Wilson, A.C. Dynamics of mitochondrial DNA
evolution in animals: Amplification and sequencing with conserved primers. Proc. Natl. Acad. Sci. USA 1989, 86, 6196–6200.
[CrossRef]
67. Li, L.; Chen, W.T.; Tang, Y.T.; Zhou, C.J.; He, S.P.; Feng, C.G. Molecular Systematics of the Subfamily Xenocyprinae. Acta. Hydrobiol.
Sin. 2023, 47, 628–636.
68. Magalhães, K.X.; Silva, R.D.F.D.; Sawakuchi, A.O.; Gonçalves, A.P.; Gomes, G.F.E.; Muriel-Cunha, J.; Sabaj, M.H.; Sousa, L.M.
Phylogeography of Baryancistrus xanthellus (Siluriformes: Loricariidae), a rheophilic catfish endemic to the Xingu River basin in
eastern Amazonia. PLoS ONE 2021, 16, e0256677. [CrossRef]
69. Lan, Z.J.; Fan, M.J.; Huang, X.L.; Zhao, J. Population diversity and phylogeography of Hemibarbus labeo and Hemibarbus medius in
South China. Acta. Ecologica. Sini. 2016, 36, 6091–6102.
Animals 2025, 15, 1202 20 of 20

70. Li, C.; Wang, J.J.; Chen, J.Q.; Schnider, K.; Veettil, R.K.; Elmer, K.R.; Zhao, J. Native bighead carp Hypophthalmichthys nobilis and
silver carp Hypophthalmichthys molitrix populations in the Pearl River are threatened by Yangtze River introductions as revealed
by mitochondrial DNA. J. Fish. Biol. 2020, 96, 651–662. [CrossRef]
71. Liu, T.S.; Zheng, M.P.; Guo, Z.T. Initiation and evolution of the Asian monsoon system timely coupled with the ice-sheet growth
and the tectonic movements in Asia. Quat. Sci. 1998, 3, 194–204.
72. Clark, M.K.; House, M.A.; Royden, L.H.; Whipple, K.X.; Burchfiel, B.C.; Zhang, X.; Tang, W. Late Cenozoic uplift of southeastern
Tibet. Geology 2005, 33, 525–528. [CrossRef]
73. Dettman, D.L.; Fang, X.M.; Garzione, C.N.; Li, J.J. Uplift-driven climate change at 12 Ma: A long δ18O record from the NE margin
of the Tibetan plateau. Earth Planet. Sci. Lett. 2003, 214, 267–277. [CrossRef]
74. Ruddiman, W.F.; Kutzbach, J.E. Late Cenozoic plateau uplift and climate change. Trans. Roy. Soc. Edinb. Earth. Sci. 1990, 81,
301–314. [CrossRef]
75. Lin, X.; Wu, Z.H.; Dong, Y.Y.; Xie, Y.Y.; Liu, H.J.; Li, Z.M. The evolutionary process of Cenozoic Asian monsoon. Int. J. Geomech.
2024, 30, 673–690. [CrossRef]
76. Wambulwa, M.C.; Luo, Y.H.; Zhu, G.F.; Milne, R.; Wachira, F.N.; Wu, Z.Y.; Wang, H.; Gao, L.M.; Li, D.Z.; Liu, J. Determinants of
genetic structure in a highly heterogeneous landscape in southwest China. Front. Plant. Sci. 2022, 13, 779989. [CrossRef]
77. Wang, C.L. Formation and evolution of the Nanling Mountains. Trop. Geomorphol. 1993, 14, 46–52.
78. Yang, H.R.; Liu, L.; Gong, S.Y.; Zhao, H.H.; Sun, J.J.; Chen, Y.Z. Evaluation of the genetic diversity of red-eyed trout in the Pearl
River and Yangtze River systems using CR and Cyt b gene sequence combinations. J. Huazhong Agric. Univ. 2013, 32, 84–91.
79. Yap, S.Y. On the Distributional patterns of southeast-east Asian freshwater fish and their history. J. Biogeogr. 2002, 29, 1187–1199.
[CrossRef]
80. Zong, F.S. Preliminary study on the geology and geomorphology of Lingqu. Carsol. Sin. 1990, 9, 88–93.
81. Yang, J.P.; Li, Y.F.; Zhu, S.L.; Li, J.; Chen, W.T. Mitochondrial diversity and population structure of grass carp (Ctenopharyngodon
idella) in the Pearl River after anthropogenic release. Knowl. Manag. Aquat. Ecosyst. 2022, 423, 14. [CrossRef]
82. Zhou, C.J. Molecular Systematics and Biogeography of Several Groups of Siluriformes. Ph.D. Thesis, Southwest University,
Chongqing, China, 2011.
83. Chen, W.T.; Zhong, Z.X.; Dai, W.; Fan, Q.; He, S.P. Phylogeographic structure, cryptic speciation and demographic history of
the sharpbelly (Hemiculter leucisculus), a freshwater habitat generalist from southern China. BMC Evol. Biol. 2017, 17, 216.
[CrossRef] [PubMed]
84. Zhao, H.H. Pterocryptis anomala. The IUCN Red List of Threatened Species 2011: E.T166056A6180117. 2011. Available online:
https://www.iucnredlist.org/species/166056/6180117 (accessed on 4 March 2025).
85. Wu, L.; Fan, C.; Lan, C.; Yu, J.; Wen, H.; Yang, Q.; Xiao, N.; Zhou, J. A long-ignored unique ecosystem of cavefishes in the southern
karst: Achievements, challenges, and prospects. Environ. Sci. Pollut. Res. Int. 2023, 30, 90489–90499. [CrossRef] [PubMed]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.