Effects of High-Flow-Velocity Stress on Energy Metabolism and Transcription Level of Triplophysa orientalis

Simple Summary

River channel modification and hydropower development can create unusually fast currents that force stream-dwelling fishes to swim harder, spend more energy, and experience greater physiological stress. We examined how a plateau loach (Triplophysa orientalis) responds to strong water flow at the gene-expression level. Fish were exposed for 3 days to a normal current (3 body lengths per second) or a high current (33 body lengths per second) in a controlled circular swimming system. By sequencing muscle RNA and validating key targets with qPCR, we found that high flow changed the expression of a small set of genes (78 in total). High flow altered 78 genes (55 up-regulated and 23 down-regulated). Specifically, multiple genes involved in mitochondrial energy production (oxidative phosphorylation) were strongly down-regulated, whereas genes linked to oxygen handling and limiting reactive oxygen species increased; for example, UCP3 and cygb1 were up-regulated with log2 fold changes of 3.19 and 3.93. At the same time, several genes associated with antioxidant transport and cellular stress protection showed a decreasing trend. Together, these patterns suggest that T. orientalis undergoes metabolic reprogramming under sustained high flow—shifting how energy is produced and allocated to cope with hydrodynamic stress—which could help survival but could constrain energy available for growth.

Abstract

River channel development and hydraulic engineering alter natural flow-velocity patterns, subjecting Triplophysa orientalis to heightened hydrodynamic stress and energy expenditure in high-flow-velocity habitats. Regulating the molecular regulatory mechanisms underlying their adaptation to high-flow velocities provides a basis for species conservation and habitat optimization. Fish were exposed for 3 days to a normal flow velocity (3 BL/s) or a high flow velocity (33 BL/s) in a controlled circular swimming system that maintained a stable current without a deliberate low-flow velocity refuge; fish at 33 BL/s sustained upstream swimming throughout the exposure. RNA-seq differential expression analysis and GO/KEGG enrichment were performed on harvested skeletal muscle, with key genes validated via qPCR. A total of 78 differentially expressed genes (DEGs) were identified between the high-flow-velocity group and the normal-flow-velocity group, including 55 up-regulated genes and 23 down-regulated genes. GO and KEGG analyses revealed that the DEGs were predominantly enriched in mitochondrial energy metabolism and neural regulation, notably oxidative phosphorylation, and were further linked to FoxO and IL-17 signaling. Compared to the normal-flow-velocity group, the high-flow-velocity group exhibited significant down-regulation of multiple oxidative phosphorylation-related subunits, including Mt-co1, Mt-co2, Mt-co3, Mt-nd1, Mt-nd2, Mt-nd4, Mt-nd5, and Mt-atp6. Concurrently, stress response-related genes, such as Selenop, GADD45B, and SIK2, showed a down-regulation trend. These transcriptional changes are consistent with reduced expression of genes involved in antioxidant defense and cellular protection under high-flow conditions, integrating differential-expression and pathway-enrichment results. This decline correlates with the down-regulation of genes associated with antioxidant and stress regulation. Pathways related to energy metabolism show significant enrichment, suggesting enhanced regulation of energy supply and allocation. This pattern indicates metabolic reprogramming characteristics adapted to high-flow-velocity stress.

1. Introduction

Environmental factors significantly regulate fish health and growth [1]. In practical research, environmental factors affecting fish growth can be broadly categorized into biotic and abiotic factors [2,3]. Biotic factors include stocking density [4], interspecies relationships, and plankton abundance [5]; abiotic factors primarily encompass temperature, flow velocity, and salinity [4,6]. With economic development, dam construction and land development have modified river channels, thereby further altering flow velocities in river ecosystems [7,8]. Stream-dwelling fish species whose life histories are closely tied to flow velocity [9] may accelerate their swimming frequency to maintain habitat homeostasis when flow velocity exceeds optimal ranges [10]. This significantly increases energy expenditure and heightens the risk of oxidative stress [11]. For example, fish can regulate their energy expenditure based on water flow velocity [12] and turbulence generated by obstacles such as rocks and riverbeds affects fishes’ swimming behavior [13] which leads to a significant increase in energy expenditure.

RNA-Seq offers advantages such as high sensitivity and reliability, enabling rapid detection of gene expression changes [14]. This technology has been widely applied in studies investigating aquatic organism responses to environmental stressors [15]. For example, both AMPK mRNA expression and Na⁺/K⁺ ATPase activity in turbot gills increased progressively, suggesting an effective protective response [16]. Similarly, a muscle transcriptomic study in rainbow trout showed that energy metabolism-related genes were strongly induced at maximal endurance swimming speed [17]. In zebrafish skeletal muscle, sustained swimming stimulation markedly up-regulated genes associated with white muscle growth and development [18]. Because skeletal muscle is the main tissue responsible for propulsion during sustained swimming and a major site of energy demand, it is a suitable target for assessing metabolic and stress-related transcriptional responses to flow-velocity challenges. Collectively, RNA-Seq analyses have enabled the functional characterization of diverse genes. Mitochondrial respiratory complexes I, III, IV, and V are core components of electron transport and oxidative phosphorylation [19,20]. Key mtDNA-encoded subunits include Mt-nd1, Mt-nd2, Mt-nd4, Mt-nd5, Mt-co1, Mt-co2, Mt-co3, Mt-atp6, and Mt-cyb [21]. We briefly summarize the known functions of several genes used as markers in this study. UCP3 provides mild uncoupling that lowers membrane potential and electron leak, which limits ROS at the cost of ATP yield [22]. Selenoprotein P (Selenop) participates in selenium transport and supports antioxidant systems such as glutathione peroxidase [23]. Cytoglobin (cygb1) supports tissue oxygen storage and transport and can scavenge ROS, helping maintain oxygen balance under hypoxia or high metabolic load [24]. GADD45B mediates transcriptional responses to DNA damage and stress [25], SIK2 integrates energy and inflammatory signals and contributes to metabolic reprogramming [26], and Tpt1 (TCTP) promotes cytoprotection, autophagy, and anti-apoptosis [27]. Together, these genes outline a continuous mechanism from ATP demand to electron-transport flux to ROS generation to oxygen transport and DNA repair responses [28]. This study provides new insights into the molecular mechanisms underlying fish behavior and offers practical implications for fish breeding and aquaculture. However, research utilizing transcriptomics to investigate the adaptive mechanisms of fish under high-flow conditions remains limited [29], and it remains unclear which genes and pathways govern energy metabolism and stress defense during sustained high-velocity exposure.

The Qinghai–Tibet Plateau is one of the planet’s most significant biodiversity hotspots, as the world’s largest and highest plateau [30,31]. The genus Triplophysa constitutes a major component of the plateau fish fauna [32], exhibiting high adaptability to the extreme environmental conditions of the plateau, such as low temperatures, low oxygen levels, and fluctuating flow velocity [33,34]. Triplophysa orientalis is a bottom-dwelling cypriniform fish belonging to the suborder Cobitoidea [35]. It is characterized by its relatively small average body length and abundant population, which supports regional fishing activities. This species inhabits the slow-flowing sections of streams. As an important economic fish species in the Yangtze River main and tributary streams, as well as the Yellow River basin, the wild population of the T. orientalis has severely declined due to ecological degradation caused by land development and illegal fishing [36,37]. Dam construction and operation can cause pronounced spatial heterogeneity and rapid temporal fluctuations in flow velocity across regulated river sections [38]. In spillway and discharge zones, flow velocity may increase sharply and exceed the suitable range for flow-sensitive fish species [39]; in many dammed sections, velocities have been reported to be 5 to 15 times higher than the optimal range [40]. Therefore, investigating the effects of flow velocity on the T. orientalis to elucidate its adaptation mechanisms under high-velocity conditions holds critical significance for protecting this species and establishing suitable habitats [41].

This study investigates the effects of high flow velocity on energy allocation and oxidative stress using the transcriptome of T. orientalis. The objectives are to identify differentially expressed genes in response to high flow velocity, characterize their functional categories, and elucidate regulatory changes in mitochondrial phosphorylation and antioxidant defense. This research provides insights into the molecular mechanisms underlying flow adaptation in fish responding to flow velocity variations, offering guidance for balancing the rapid development of hydropower projects with environmental conservation efforts.

2. Materials and Methods

2.1. Rationale for the Contrast (3 vs. 33 BL/s)

Field measurements of in situ water-flow velocity (m/s) at the microhabitats where T. orientalis individuals were observed were standardized to body-length units (BL/s). These data indicate that T. orientalis experiences a wide range of hydrodynamic conditions in situ, with a substantial portion of observations clustering around 30–35 BL/s [42]. We therefore bracketed an ecologically relevant spectrum by selecting 3 BL/s as a near-refuge, normal-flow condition close to the literature-reported suitable range for Triplophysa of 1–3 BL/s [42,43] and 33 BL/s as an upper-envelope strong-current condition that can occur in regulated reaches, such as spillway or discharge zones, while remaining ethically sustainable over a 3-day exposure [44]. This endpoint design maximizes contrast and statistical power for detecting flow-responsive pathways, while subsequent KEGG/GO results can be interpreted as bounding the molecular adjustments from near-refuge to strong-current conditions.

2.2. Ethics Statement

Institutional Animal Care and Use Committee Statement: All animal procedures were approved by the Ethics Committee of the Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences (Approval No. LAECFFRC-2025-06-16).

2.3. Experimental Fish and Acclimation

T. orientalis were obtained from the Fisheries Research Institute, Sichuan Academy of Agricultural Sciences (Chengdu, China). The formal experiments were conducted in Qianwei County (Leshan, China). Prior to experimentation, healthy, disease-free T. orientalis with comparable body size (3.5 ± 0.5 cm) were selected and acclimated in a culture system for 7 days to adapt to the experimental conditions. During the experimental period, water temperature, dissolved oxygen (DO), and pH were monitored using a portable water-quality analyzer (Hach, Loveland, CO, USA). Water temperature was maintained at 14–16 °C, DO was ≥7 mg/L, and pH was 7.2–7.4.

2.4. Experimental Apparatus

Experiments were conducted in a circular swimming track system with adjustable flow (2.0 m × 1.5 m × 1.5 m; L × W × H). Flow was driven by a stainless-steel propeller (Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences, Wuxi, China). The water level was maintained above the propeller, and activating the pump generated a stable circular current within the system. Target velocities were achieved by adjusting the pump output. Instantaneous flow velocity was measured using a portable current meter (LS1206B; Laisida Information Technology Co., Ltd., Fuzhou, China) [45]. Sixteen points were selected on two orthogonal transects; velocity at each point was measured three times, and the average across points (mean of triplicates per point) was used as the system-wide velocity (Figure 1A,B).

Figure 1.

Schematic of the circular swimming track system and behavioral states of Triplophysa orientalis under graded flow velocities. (A) Overall view of the custom-built circular swimming track used for the flow challenge. (B) Diagram of the flow circuit and the swim chamber; the unit was placed in a water bath for temperature control, and flow was generated by a motor-driven propeller/pump. (C) Representative photograph of T. orientalis. (D) Flow-velocity settings (Groups A–H; 3–39 BL/s) and the predominant behavioral state observed at each velocity, including station holding, sculling, and upstream swimming.

2.5. Experimental Design and Sampling

A total of 640 healthy T. orientalis of comparable body size (0.4 ± 0.1 g; 3.5 ± 0.7 cm) (Figure 1C) were randomly assigned to 24 annular raceway systems (20 fish per system). The total sample size of 640 fish corresponds to the full design of eight flow-velocity treatments with four system-level replicates per treatment and 20 fish per system (8 × 4 × 20), and because only 24 systems were available, the 32 system replicates were completed in sequential experimental blocks. Based on Shi et al., who reported that 1–3 body lengths per second (BL/s) is a commonly suitable flow range for Triplophysa [43], eight flow treatments were established to examine flow effects: 3, 6, 9, 15, 21, 27, 33, and 39 BL/s. Each treatment included four system-level replicates, and the trial lasted 3 days. Throughout the experiment, water quality and feeding regimes in all systems were maintained consistently with those during acclimation. For downstream molecular analyses, skeletal muscle was sampled from individual fish without pooling, with five individuals from the 3 BL/s group and four individuals from the 33 BL/s group used for RNA-seq, and the same biological replicates and RNA samples were used for qPCR validation. During the experiment, we observed the behavioral states of T. orientalis across all flow rate treatment groups. At 3 and 6 BL/s, the fish predominantly maintained a station-holding posture; at 9, 15, and 21 BL/s, they frequently adopted a sculling posture; while at 27 and 33 BL/s, the fish sustained an upstream swimming posture. However, when flow velocity reached 39 BL/s, T. orientalis could no longer sustain upstream swimming and exhibited signs of exhaustion (Figure 1D). After 3 days, all T. orientalis were netted and anesthetized with tricaine methanesulfonate (MS-222; Sigma-Aldrich, St. Louis, MO, USA) at 100 mg/L. Skeletal muscle was dissected. Each RNA-seq and qPCR sample represents skeletal muscle from a single individual fish; libraries and qPCR reactions were prepared independently for each individual, so A_1–A_5 and G_1–G_4 are true biological replicates rather than technical replicates. The annular raceway system was the experimental unit for velocity exposure, and for molecular sampling, we selected fish across independent system-level replicates, with no more than one fish taken from any single system to avoid pseudoreplication. Samples were snap-frozen in liquid nitrogen and stored at −80 °C until analysis. Although eight flow treatments were established and all groups were assessed under the same experimental conditions, we focused the transcriptomic comparison on 3 BL/s and 33 BL/s because they represent a suitable velocity and a high, yet sustainable, strong-current endpoint over 3 days; 39 BL/s induced clear exhaustion and was therefore not used for transcriptomic profiling. This endpoint selection maximizes contrast for detecting flow-responsive pathways while remaining biologically and ethically appropriate.

2.6. Sequence and Filtering of Clean Reads

A cDNA library constructed from pooled RNA from samples of the research species was sequenced with the Illumina NovaseqTM 6000 sequence platform. For RNA-seq, skeletal muscle from individual fish was used as biological replicates (3 BL/s: n = 5; 33 BL/s: n = 4). Using the Illumina paired-end RNA-seq approach, we sequenced the transcriptome, generating a total of 2 million × 150 bp paired-end reads [46]. Reads obtained from the sequencing machines include raw reads containing adapters or low-quality bases, which will affect the following assembly and analysis. Thus, to get high-quality, clean reads, reads were further filtered by Cutadapt (https://cutadapt.readthedocs.io/en/stable/, version: cutadapt-1.9). The parameters were as follows: removing reads containing adapters; removing reads containing more than 5% of unknown nucleotides (N); and removing low-quality reads containing more than 20% of low-quality (Q-value ≤ 20) bases. Then, sequence quality was verified using FastQC (v0.10.1), including Q20, Q30, and GC content. After filtering, cleaned paired-end reads were retained for downstream analyses [47]. The raw sequence data have been submitted to the NCBI Gene Expression Omnibus (GEO) datasets with accession number GSE318193.

2.7. GO or KEGG Enrichment Analysis

GO function or KEGG pathway significant enrichment analysis first maps all significant differentially expressed genes to each GO term in the Gene Ontology database or each KEGG pathway in the KEGG database, and calculates the number of genes in each GO term or KEGG pathway to get the number of genes annotated as a significant difference in a specific GO term or KEGG pathway (S), the number of genes annotated with a significant difference in a GO term or KEGG pathway (TS), and the number of genes annotated as a specific GO term or KEGG pathway (B), total number of genes with GO term or KEGG annotations (TB), and then applies a hypergeometric test to find GO term or KEGG pathway that are significantly enriched in significantly differently expressed genes compared to the entire genome background [48,49].

2.8. Quantitative Real-Time PCR (qPCR) Validation

Based on the RNA-seq results, differentially expressed genes (DEGs) related to oxidative stress and growth were selected for qPCR validation. β-actin was used as the internal reference gene. Primers were designed using Primer 3 (v2.3.7) and synthesized by Lianchuan Biotechnology Co., Ltd. (Hangzhou, China).

2.9. Statistical Analysis

All statistical analyses were performed in SPSS 20.0 (IBM, Armonk, NY, USA). After confirming homogeneity of variances, group differences were evaluated by one-way ANOVA followed by Tukey’s post hoc test (p < 0.05). Pairwise comparisons between two groups were conducted with independent samples t-tests. Trend responses were further assessed using orthogonal polynomial contrasts to identify linear, quadratic, and cubic components. Data are presented as mean ± standard error.

3. Results

3.1. Quality of Library Sequencing

The raw reads of each sample generated by RNA-Seq exceeded 34,743,738 (Table 1). Sample A, whose flow velocity is 3 BL/s, and sample G, whose flow velocity is 33 BL/s, averaged 38,176,916 and 39,233,569 raw reads, respectively. The validity ratio of the raw reads exceeded 98.54%. The GC content occupied 47.77–48.88% of the libraries. The proportion of bases with a mass value ≥ 30 (sequencing error rate < 0.001) in all samples was greater than or equal to 97.15%. The above results verified that the quality of library sequencing met the requirements for differentially expressed gene analysis

Table 1.

Sequencing data, statistics, and quality control.

Sample	Raw_Reads	Valid_Reads	Valid%	Q20%	Q30%	GC%
A_1	37,313,084	36,908,498	98.92	99.49	97.42	48.62
A_2	39,742,922	39,372,370	99.07	99.50	97.40	48.51
A_3	34,743,738	34,340,394	98.84	99.49	97.30	48.65
A_4	39,672,948	39,285,878	99.02	99.47	97.17	48.49
A_5	39,411,890	39,004,116	98.97	99.46	97.44	48.43
G_1	37,019,086	36,478,250	98.54	99.44	97.35	48.88
G_2	40,289,922	39,734,852	98.62	99.42	97.15	47.77
G_3	38,909,246	38,401,572	98.70	99.48	97.45	48.54
G_4	40,715,984	40,124,658	98.55	99.46	97.16	48.51

A: 3 BL/s group; G: 33 BL/s group; Raw_Reads: number of raw reads generated by the sequencer; Valid_Reads: number of reads retained after quality filtering; Valid%: proportion of valid reads (Valid_Reads/Raw_Reads × 100%); Q20%: proportion of bases with Phred quality score ≥ 20 (error rate ≤ 0.01); Q30%: proportion of bases with Phred quality score ≥ 30 (error rate ≤ 0.001); GC%: proportion of guanine (G) and cytosine (C) bases in the sequencing data.

3.2. Analysis of DEGs in the Muscle of T. orientalis: 3 BL/s Versus 33 BL/s

To more accurately assess clustering among samples, we evaluated pairwise correlations based on gene-expression profiles; higher Pearson correlation coefficients indicate better within-group clustering. The correlation analysis revealed strong relationships among biological replicates (Figure 2A). Principal component analysis further showed that flow velocity produced a pronounced separation of muscle transcriptomes: the A (3 BL/s) and G (33 BL/s) groups were clearly divided along PC1, which explained 95.3% of the total variance, whereas PC2 accounted for only 2.1%, indicating good within-group consistency and no evident outliers (Figure 2B). The number of differentially expressed genes (DEGs) is summarized in Figure 2C. In total, 78 DEGs were identified between G and A, including 55 up-regulated and 23 down-regulated genes (Figure 2D). The statistical information on energy metabolism/mitochondrial function, oxidative stress, and stress signaling/apoptosis-related DEGs in T. orientalis muscle at 33 BL/s (high flow velocity) compared with 3 BL/s (normal flow velocity) is presented in Table 2. Notably, several mitochondrially encoded OXPHOS subunits showed very large decreases (log2FC ≈ −6 to −10), indicating an order-of-magnitude suppression of oxidative-phosphorylation transcripts under high flow rather than a single-gene anomaly.

Figure 2.

Transcriptome overview of Triplophysa orientalis muscle under 3 BL/s (Group A) and 33 BL/s (Group G). (A) Pearson correlation matrix of all samples based on gene-expression profiles, showing strong within-group correlations. * p < 0.05, *** p < 0.001. (B) Principal component analysis (PCA) with 95% confidence ellipses; A and G separate clearly along PC1. (C) Volcano plot of DEGs for G vs. A (red, up-regulated; blue, down-regulated). (D) Numbers of DEGs in G vs. A (up-regulated = 55; down-regulated = 23). (E) KEGG enrichment bubble plot for DEGs; representative pathways include oxidative phosphorylation, thermogenesis, Parkinson’s/Alzheimer’s/Huntington’s diseases, FoxO signaling, IL-17 signaling, cardiac muscle contraction, cholesterol metabolism, primary bile acid biosynthesis, aminoacyl-tRNA biosynthesis, and pyruvate metabolism. (F) GO enrichment of DEGs in biological process (BP), cellular component (CC), and molecular function (MF); membrane/integral membrane terms dominate CC, whereas ATP/metal-ion binding and oxidoreductase activities are prominent in MF. For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.

Table 2.

Information statistics of energy metabolism/mitochondrial function, oxidative stress, and stress signaling/apoptosis-related DEGs in the muscle of Triplophysa orientalis at 33 BL/s (Group G) compared with 3 BL/s (Group A).

Predict Function	Gene Name	Mean Normalized Counts (A)	Mean Normalized Counts (G)	p Value	Regulation	Log2FC
Energy metabolism/mitochondrial function	Mt-atp6	13.876	0.050	<0.001	down	−8.08
	Mt-co1	12.628	0.012	<0.001	down	−9.92
	Mt-co2	11.730	0.048	<0.001	down	−8.09
	Mt-co3	14.900	0.042	<0.001	down	−8.44
	Mt-cyb	5.388	0.032	<0.001	down	−7.33
	Mt-nd1	4.472	0.048	<0.001	down	−6.46
	Mt-nd2	3.932	0.000	<0.001	down	−9.80
	Mt-nd4	3.686	0.005	<0.001	down	−8.93
	Mt-nd5	1.172	0.008	<0.001	down	−7.25
	UCP3	0.168	1.095	<0.001	UP	3.19
Oxidative stress-related factors	GADD45B	7.806	6.572	<0.001	down	−1.64
	Selenop	1.358	0.000	<0.001	down	−9.38
	cygb1	0.684	17.972	<0.001	UP	3.93
Stress signaling and apoptosis	SIK2	14.488	5.345	<0.001	down	−1.58
	Tpt1	2.728	0.000	<0.001	down	−8.26

Mean normalized counts (A) and mean normalized counts (G) represent the group means of library-size–normalized read counts for the 3 BL/s group (A, n = 5) and the 33 BL/s group (G, n = 4), respectively. Log₂FC denotes the log₂-transformed fold change for G relative to A (log₂[G/A]) estimated from the count-based differential-expression model; values of 0.00 for mean normalized counts indicate that the transcript was not detected in any replicate of that group after filtering. p values are shown as <0.001 where appropriate. GADD45B is listed only once to avoid redundancy; it is a multifunctional stress-response gene relevant to both oxidative-stress regulation and stress signaling/apoptosis pathways.

3.3. Functional Analysis by KEGG Enrichment and GO Enrichment

KEGG pathway enrichment indicated that the DEGs were mainly involved in pathways related to energy metabolism and neuromodulation, including oxidative phosphorylation, thermogenesis, and several neurodegenerative disease pathways, as well as FoxO signaling, IL-17 signaling, cardiac muscle contraction, cholesterol metabolism, primary bile acid biosynthesis, aminoacyl-tRNA biosynthesis, and pyruvate metabolism (Figure 2E). To further elucidate the biological effects of the flow treatments in T. orientalis, GO enrichment analyses were performed across the molecular function (MF), cellular component (CC), and biological process (BP) categories (Figure 2F). Numerous BP terms were significantly enriched. In the CC category, membrane-related terms together with cytoplasm, nucleus, mitochondrion, and cytosol predominated, whereas in the MF category, ATP binding and metal ion binding were among the most represented terms, alongside oxidoreductase activity, NADH dehydrogenase activity, and cytochrome-c oxidase activity.

3.4. Data Validation by qPCR

The heatmap visualized clear differences in DEGs between the normal-flow-velocity (3 BL/s) and high-flow-velocity (33 BL/s) groups (Figure 3A). Representative genes related to mitochondrial energy metabolism (Figure 3B) and oxidative-stress/inflammatory signaling (Figure 3C,D) were selected for validation. Compared with 3 BL/s, the transcriptome results indicated that multiple oxidative-phosphorylation subunits were significantly down-regulated in the G group, including Mt-co1, Mt-co2, Mt-co3, Mt-nd1, Mt-nd2, Mt-nd4, Mt-nd5, and Mt-atp6, together with Selenop; in contrast, cygb1 and UCP3 were significantly up-regulated, while the stress-response genes GADD45B and SIK2 were down-regulated. The qPCR measurements for these genes were consistent with the RNA-seq results, supporting the reliability of the transcriptomic data and their use in subsequent analyses.

Figure 3.

Validation of flow-responsive genes in the muscle of Triplophysa orientalis (A: 3 BL/s; G: 33 BL/s). (A) Heatmap of selected differentially expressed genes (DEGs) across all samples. (B) qPCR validation of DEGs involved in energy metabolism function (mt-nd1, mt-nd2, mt-nd4/5, mt-co1/2/3, mt-cyb, mt-atp6, ucp3, cygb1). (C) qPCR validation of oxidative-stress and stress-signaling/apoptosis-related DEGs (nox1, nod1, gadd45b, sik2, tpt1, selenop). Bars show mean ± SEM; values marked with an asterisk differ significantly from group A (p < 0.05). (D) qPCR validation of stress-signaling/apoptosis-related DEGs (gadd45b, sik2, tpt1). Bars show mean ± SEM; p < 0.05, p < 0.01 (vs. group A). ** p < 0.01.

4. Discussion

Flow velocity exerted a strong organizing effect on the muscle transcriptome of T. orientalis, as evidenced by clear between-group separation along PC1 and tight within-group clustering (A: 3 BL/s vs. G: 33 BL/s), indicating that hydrodynamic forcing was the dominant driver of gene-expression variance in this experiment [50]. Consistent with this pattern, differential expression analysis identified 78 DEGs (55 up- and 23 down-regulated), providing a focused set of targets for mechanistic interpretation. Because 33 BL/s represents an upper-envelope strong-current condition that may occur during spillway or discharge operations, the observed transcriptomic shifts are interpreted as stress-tolerance or coping responses under extreme hydrodynamic forcing rather than optimal physiological performance.

The overarching signal was the coordinated down-regulation of multiple mitochondrially encoded subunits across Complexes I, III, IV, and V in the high-flow group [20,51], suggesting reduced oxidative phosphorylation capacity and electron-transport flux at 33 BL/s, a state expected when locomotor demand and shear stress reallocate energy from biosynthesis to station-holding and stress defense [52]. In parallel, UCP3 was up-regulated, consistent with a shift toward mild uncoupling to limit mitochondrial membrane potential and electron leak [53], thereby limiting ROS formation at the expense of ATP yield [54,55]. The up-regulation of cygb1, a cytoglobin that supports intracellular O₂ transport [56] and scavenges ROS [57], further suggests compensation for elevated oxygen turnover and redox pressure under high velocity. Conversely, the down-regulation of Selenop, a selenium carrier sustaining glutathione peroxidase systems [58,59,60], together with reduced expression of stress-response mediators (GADD45B, SIK2, Tpt1), indicates a potential shortfall in antioxidant and cytoprotective capacity when flow is high. This gene set depicts a coherent metabolic reprogramming from ATP production toward ROS containment that is typical of fishes experiencing hydrodynamic challenge [52,61,62].

Pathway enrichment aligned with these gene-level signals. KEGG terms were dominated by oxidative phosphorylation and other energy-metabolism modules [63,64], alongside regulatory pathways such as FoxO signaling and immune-stress axes like IL-17 signaling [65,66]. These enrichments support the interpretation that high flow increases energetic demand while limiting mitochondrial throughput and integrating stress-related regulation [67].

In GO, membrane/mitochondrion-rich CC terms and oxidoreductase/ATP- and metal-ion-binding MF terms prevailed [68], highlighting key functional categories that differentiated the 33 BL/s high-flow-velocity group from the 3 BL/s normal-flow-velocity group. Importantly, our qPCR validation confirmed the direction of change between groups for the representative mitochondrial [69], antioxidant, and stress-signaling genes, strengthening confidence in these inferences [47]. The integrated picture is that prolonged exposure (3 d) to 33 BL/s depresses the transcription of core ETC subunits while promoting uncoupling and oxygen-handling responses—changes that together would stabilize redox balance but may constrain ATP availability for growth [52,70]. This mechanism offers a molecular rationale for the well-known trade-off in fishes between swimming at supra-optimal speeds and somatic investment [71]. Although our system and tissue differ, parallels can be drawn with dietary-stress transcriptomics in Micropterus salmoides, where feeding regulation and oxidative-stress pathways were also prominent under a low-fish-meal diet [70]; there, neuroactive ligand-receptor and inflammation-related signaling coincided with altered antioxidant indices [69,70]. Such convergence across stressors underscores the central role of mitochondrial function and redox homeostasis in fish stress physiology [52]. Two caveats deserve mention. Our design contrasted two velocities at the extremes of the treatment range (3 vs. 33 BL/s) to maximize effect detection. The observed responses, therefore, represent endpoints; future work should map intermediate speeds to resolve thresholds and potential non-linearities. Transcriptional down-regulation of mtDNA-encoded ETC genes does not by itself prove reduced respiratory flux [71]; integrating enzyme activities, ATP production, and ROS measurements would directly test the functional consequences implied by the transcriptome [52]. Our interpretations of altered oxidative phosphorylation and antioxidant capacity are based on transcript-level signals and should be viewed as hypotheses about functional change rather than direct evidence of metabolic performance. Future work combining transcriptomics with physiological and biochemical assays, such as oxygen consumption, enzyme activities, ATP content, and oxidative-stress indices, will be required to validate these functional inferences.

In practical terms, these data provide candidate biomarkers for diagnosing flow-induced metabolic stress in lotic aquaculture or river-reach management. They also suggest that maintaining T. orientalis within near-optimal velocities could minimize the need for costly uncoupling and protect mitochondrial capacity for growth.

Our results have direct conservation relevance for Triplophysa orientalis, a rheophilic plateau fish that increasingly experiences flow alteration associated with hydropower operation and river engineering. The observed transcriptomic patterns indicate that sustained exposure to high-velocity conditions entails substantial physiological costs, including depressed expression of oxidative-phosphorylation pathways together with responses related to oxygen handling, antioxidant defense, and mitochondrial uncoupling. Importantly, these changes suggest that even when mortality is not immediately evident, prolonged station-holding at high velocities may reduce energetic efficiency and increase oxidative stress risk, potentially lowering growth, condition, and reproductive investment over time. Therefore, tolerating high velocity should not be equated with an absence of biological impact, and management should consider sublethal physiological constraints revealed by molecular responses.

From a conservation and management perspective, our findings support several practical implications for flow regulation and habitat design. First, maintaining a mosaic of flow velocities within regulated reaches, especially ensuring the availability of low-velocity refugia, may allow individuals to intermittently recover from energetically costly station-holding. Second, operational rules that avoid long-duration, extreme-velocity events could reduce chronic energetic stress. Third, in fish passage or channel restoration projects, incorporating resting areas within high-velocity sections may improve passage success and reduce physiological exhaustion. Finally, the candidate molecular indicators identified here could be integrated into monitoring programs to evaluate the effectiveness of mitigation measures and to define velocity thresholds that balance engineering needs with the physiological capacity of T. orientalis populations.

In summary, high flow (33 BL/s) reorganized the muscle transcriptome of T. orientalis toward reduced oxidative-phosphorylation gene expression, increased uncoupling/oxygen-handling defenses, and broad enrichment of energy- and stress-response pathways. These coordinated shifts likely reflect coping responses associated with ROS control under sustained hydrodynamic challenge, but they may limit ATP available for somatic investment—mechanistically linking flow regime to growth potential in this species.

5. Conclusions

Sustained high flow velocity of 33 BL/s markedly reorganized the skeletal-muscle transcriptome of Triplophysa orientalis, yielding 78 DEGs, with 55 genes up-regulated and 23 genes down-regulated, relative to the normal flow velocity of 3 BL/s. At the gene level, the high-flow group showed coordinated down-regulation of mitochondrially encoded oxidative-phosphorylation subunits, including Mt-co1, Mt-co2, Mt-co3, Mt-nd1, Mt-nd2, Mt-nd4, Mt-nd5, and Mt-atp6, as well as the antioxidant-related gene Selenop, whereas UCP3 and cygb1 were significantly up-regulated. These RNA-seq patterns were further supported by qPCR validation of representative mitochondrial, antioxidant, and stress-signaling genes. At the pathway level, enrichment analyses highlighted oxidative phosphorylation and other energy-metabolism modules together with stress-integration pathways, including FoxO and IL-17, indicating a transcriptomic signature consistent with metabolic adjustment to sustained hydrodynamic challenge and altered redox and oxygen-handling demands.

Author Contributions

X.G.: Methodology, investigation, writing—original draft, and writing—review and editing. H.W.: methodology and data curation. Y.J.: investigation and methodology. Z.J.: investigation. L.Q.: investigation. S.M.: Conceptualization; Data curation; Writing – review & editing; Supervision; Project administration; Funding acquisition. C.S.: supervision, conceptualization, writing—review and editing, and funding acquisition. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The study was conducted according to the guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of the Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding authors.

Conflicts of Interest

All authors declare that there are no conflicts of interest.

Funding Statement

This work was funded by the National Key Research and Development Program of China, grant number 2022YFC3202103.