This study was carried out in accordance with the principles of the Basel Declaration and the recommendations of the Guide for the Care and Use of Laboratory Animals published by the Animal Care Committee of Anhui Agricultural University (Hefei, China). The protocol was approved by the Animal Care Committee of Anhui Agricultural University.

Experimental large-scale loach (20.3 ± 2.8 g, unsexed) were obtained from a fish market in Hefei (China) and acclimated in laboratory cycling tanks containing aerated freshwater (water temperature 23.0 ± 1.0 °C, dissolved oxygen ≥ 5.0 mg L^-1) for 7 days. During acclimatization, the fish were fed a commercial meal twice daily (crude protein 35%, crude lipid 7%).

After 24 h fasting, fish were exposed to various concentrations (nominally 40, 50, and 60 mmol L^-1; measured values: 39.6 mmol L^-1 pH 9.40, 48.0 mmol L^-1 pH 9.50, and 59.6 mmol L^-1 pH 9.60, respectively) of carbonate alkalinity (as NaHCO₃) for 12, 48, and 96 h in a volume of 10 L at 25.0 ± 1.0 °C using a plastic box (diameter 35.5 cm, height 15.5 cm). The carbonate alkalinity levels were designed according to an acute NaHCO₃ toxicityin this species reported by Wu et al., 2017. The NaHCO₃ solution was completely exchanged every 24 h to maintain the alkaline level. Control fish were collected at the beginning of the experiment (0 h). Three replicate tanks containing four fish were sampled for each exposure period. Fish were not fed during the experimental period.

At the end of each exposure period all the fish were anaesthetized with tricaine methane sulfonate (MS-222, 200 mg L^-1), executed by a blow to the head, and the gills removed by dissection. Until RNA isolation, all specimens were stored at -80 °C.

Total RNA was isolated from samples of large-scale loach gills using an Ultrapure RNA Kit (CoWin Biotech, Beijing) following manufacturer’s instructions. The quality and integrity of the isolated RNA were verified by 2% agarose gel electrophoresis and NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA) and stored at -20 °C. RNA (1 μg) from each sample was reverse transcribed to cDNA using MonScript RTIII Super Mix with dsDNase (Monad Biotech, Wuhan) and a thermal cycler (T100, BIO-RAD, USA) following the manufacturer’s protocol.

Gene-specific primers were used to measure relative gene expression. ( Table 1 ). Quantitative real-time PCR (qPCR) was carried out using AceQ qPCR SYBR Green Master Mix (Vazyme Biotech, Nanjing) and a quantitative thermal cycler (CFX Connect 96, BIO-RAD, USA) in a 20 μl reaction volume. Amplification conditions were as follows: pre-denaturation for 5 min at 95°C, followed by 40 cycles of 10 sec at 95°C and 30 sec at 60°C. Relative mRNA levels were calculated using the 2^-△△Ct method as described by Livak and Schmittgen (2001), with the housekeeping gene β-actin being used as an internal standard to normalize the results.

Values are expressed as mean ± standard error. The variance homogeneity of the data was examined using Levene’s test. One-way (exposure time, carbonate alkalinity concentration) and two-way (exposure time × carbonate alkalinity concentration) analysis of variance (ANOVA) were used to compare mRNA expression levels. Tukey’s multiple tests were performed when significant differences were found at the 0.05 level. Statistical analyses were performed using SPSS software (SPSS Inc., Chicago, IL, USA).

Significant interactions were observed between carbonate alkalinity concentration and exposure time on Rhag, Rhbg, and Rhcg expression in the gills of large-scale loach after various periods of acute exposure (two-way ANOVA, P < 0.05, Table 2 ). Rhag transcript levels increased markedly as exposure time increased (P < 0.05, Figure 1A ). High carbonate alkalinity also induced a significant increase in Rhag expression during the various periods of NaHCO₃ exposure (P < 0.05, Figure 1A ). Up-regulation of the Rhag gene was previously reported in the gills of Anabas testudineus exposed to high levels of ammonia (Chen et al., 2017). Rhag in erythrocytes have been shown to enhance NH₃ efflux from red cells into plasma (Wright and Wood, 2009). Rhag may be present in pillar cells between the lamellar blood space and the gill epithelium, allowing NH₃ to permeate the branchial epithelium (Wright and Wood, 2009). Ammonia excretion was clearly reduced by exposure to high alkalinity (Li et al., 2020). Thus, the observed high level of Rhag transcripts accompanying the blocking of ammonia excretion might be attributed to the development of ammonia tolerance. The various periods of carbonate alkalinity exposure induced significant declines in Rhbg mRNA level (P < 0.05), while Rhbg expression was unaffected by NaHCO₃ concentrations (P > 0.05, Figure 1B ). Similarly, changes in Rhbg transcription in response to ammonia were not observed in Micropterus salmoides  (Egnew et al., 2019), Oncorhynchus mykiss, or Cyprinus carpio  (Sinha et al., 2013). The present results suggest that the function of Rhbg in large-scale loach is not primarily relevant to ammonia elimination during alkalinity stress. A marked up-regulation of Rhcg was observed in the fish after extended alkalinity exposure (P < 0.05, Figure 1C ), but it was unaffected by alkalinity concentrations (P > 0.05, Table 2 ). Rhcg facilitates the movement of NH₃ across the apical gill membrane and its excretion into surrounding water (Wright and Wood, 2009). Our results suggest that the Rhcg presented on the apical side of the branchial epithelium facilitate NH₃ excretion from the gills, in accordance with observations in M. salmoides  (Egnew et al., 2019), Dicentrarchus labrax (Shrivastava et al., 2019), and O. mykiss  (Eom et al., 2020).

There was a significant interaction between carbonate alkalinity concentration and exposure time on aqp1 and aqp3 expression in the gills of large-scale loach (two-way ANOVA, P < 0.05, Table 2 ). Exposure to 60 mmol L^-1 NaHCO₃ induced a significant up-regulation of aqp1 (P < 0.05), while no effect was observed after the various periods of 40 and 50 mmol L^-1 NaHCO₃ exposure (P > 0.05, Figure 2A ). Moreover, aqp1 mRNA levels increased obviously as carbonate alkalinity concentration increased ( Figure 2A ). The physiological function of Aqp1 in teleosts is primarily related to passive water exchange (Deane et al., 2011). Elevated ambient alkalinity resulted in the high transcription of aqp1 in the gills of large-scale loach to facilitate excretion of excess internal water and to maintain osmotic balance. Similar results were observed in the gills of hybrid tilapia after 24 h of alkaline treatment (Su et al., 2020). By contrast, aqp3 expression was found to be significantly lower in the gills of large-scale loach after 48 h of 60 mmol L^-1 NaHCO₃ exposure (P < 0.05), while expression in 40 and 50 mmol L^-1 groups were not statistically different from the control (P > 0.05, Figure 2B ). In addition to affecting water permeabilization, Aqp3 has been shown to transport urea (Cutler et al., 2007). Decreased mRNA transcription of aqp3 was also reported in O. mossambicus during seawater acclimation (Breves et al., 2016). Therefore, the absence of Aqp3 in the gills of large-scale loach may result in the blocking of urea elimination to maintain high internal osmolality.

In summary, branchial ammonia transport-related and osmoregulation-related gene expression in large-scale loach are significantly affected by carbonate alkalinity. The large-scale loach up-regulate Rhag and Rhcg to facilitate NH₃ efflux from the gills during exposure to high alkalinity. Elevated transcription of aqp1 in the gills in order to excrete excess internal water, and down-regulation of aqp3 in order to block urea elimination together maintain appropriate osmolality as an adaptation to alkaline environments.