Introduction

Species living in the intertidal zone and estuaries, such as crustaceans, experience fluctuating oxygen levels1. Crustaceans utilize hemocyanin as the primary oxygen-carrying molecule2. Generally, crustaceans display a strong tolerance to hypoxic environments by regulating oxygen consumption when ambient O₂ levels decrease3. During hypoxia, burrowing crayfish (Cambarus truncatus) increase the ventilation rate of the gill3. Semiterrestrial crab (Neohelice granulate) perform an escape behavior from a hypoxic seawater into tidal zone or land, and also perform anaerobic metabolism with the accumulation of lactate in hemolymph4,5. Some species like the mediterranean green crab (Carcinus aestuarii) can increase the oxygen-binding affinity of hemocyanin during hypoxia6.

The deep-sea7 and shallow-water hydrothermal vent environments are characterized by extreme conditions, including high temperatures, high hydrogen sulfide content, low pH, variable oxygen concentrations, and even hypoxia (Table S1). Some species, such as deep-sea vent mussels (Bathymodiolus thermophilus) may exhibit a degree of hypoxia tolerance, allowing them to endure fluctuations in oxygen availability and periods of low oxygen in hydrothermal vent environments8. Deep-sea hydrothermal vent bivalves (Gigantidas platifrons) and shallow-water hydrothermal vent crustaceans (Xenograpsus testudinatus) both form symbiotic relationships with chemosynthetic bacteria, which provide a primary source of nutrition through the oxidation of hydrogen sulfide9,10. Another deep-sea hydrothermal vent crabs (Bythograea thermydron) not only regulates its oxygen consumption rate but also accumulates lactate in hemolymph which indicates the occurrence of anaerobic metabolism11,12. Taken together, a hydrothermal vent ecosystem might be considered as a “natural laboratory” to understand versatile physiological adaptations to extreme conditions, particularly high temperature, low pH, high sulfide stress and low oxygen.

The shallow-water hydrothermal vent crabs, X. testudinatus (xtcrab; Xenograpsidae), resides in the hydrothermal vent region of Kueishan Island in Northeastern Taiwan13. According to previous studies14–20 the hydrothermal vent region where vent xtcrabs reside was characterized by normal seasonal seawater temperatures (20.83–28.09 ℃), lower pH (6–8), lower dissolved oxygen (DO = 4.15–9.5 mg/L), and higher sulfide content (0–3500 µM) (Table S1) compared with the non-hydrothermal vent region around Kueishan Island which exhibited temperatures of 19.93–29.27 ℃, pH 8.01–8.17, DO 7.83–8.29 mg/L, and non-detectable sulfide14. Portunidae is a family of crabs that inhabit diverse marine environments, including estuaries, rocky shores, and coral reefs in intertidal and subtidal zones21. Thranita danae (tdcrab, Portunidae) inhibits the intertidal rocky shore, where variables such as dissolved oxygen, sulfide content (< 300 µM), and pH (6.9–8.3) fluctuate daily22. Both xtcrab and tdcrab inhabit dynamic and variable environments, with the vent conditions being more extreme (Table S1). Studying the physiological differences between xtcrabs and tdcrabs offers valuable insights on adaptations to extreme environments.

Altered cellular respiration is one of the primary physiological responses to hypoxia. Sulfide decreases the water oxygen content and its toxicity in mitochondrial functions derives the electron from “cytochrome c” resulting in the termination of electron transport chain and the interruption of aerobic respiration22,23,24,25. Our previous study revealed the presence of duplicated sulfide: quinone oxidoreductase (SQR1 and SQR2) paralogs in vent xtcrabs14. Among them, xtSQR1 is predominantly expressed in the gill and showed increased expression in the high-sulfide hydrothermal vent environment, suggesting a specific role in detoxifying environmental sulfide into less toxic compounds.14 We propose that acclimation to high-sulfide hydrothermal vent environments is partly facilitated by duplicated SQR paralogs in xtcrabs, which may function in sulfide detoxification and support aerobic metabolism. However, our understanding of the respective aerobic and anaerobic pathways in the xtcrabs under the sulfide-rich and hypoxic hydrothermal vent environment remains limited.

In this study, we hypothesized the importance of switching between aerobic and anaerobic metabolism, as part of the metabolic adaptation of xtcrabs in response to high sulfide stress and serve hypoxia. We measured the hemolymph lactate levels in xtcrabs and tdcrabs, which is considered an indicator of anaerobic metabolism. To examine aerobic metabolism, we focused on cytochrome c (CYTC), a highly conserved protein that plays a crucial role in cellular respiration. Cytochrome c is primarily involved in the mitochondrial electron transport chain, where it shuttles electrons between Complexes III and IV, to facilitate the production of ATP26. We measured gill levels of CYTC transcripts and of mitochondiral Complex IV (COXIV, also named cytochrome c oxidase) enzymatic activity, as indicators of aerobic metabolism. In addition, we also measured transcript levels and enzymatic activity of gill SQR as the facilitative factor for aerobic metabolism and removal of sulfide toxicity in high sulfide environment. The survival of xtcrabs in hypoxia and their metabolic responses were tested through laboratory experiments. Comparisons were made with a non-vent, intertidal crab species, tdcrab. The phylogenetic relationship between xtSQR paralogs and tdSQR paralogs was further compared. Most importantly we performed field studies, including transferring xtcrabs from the hydrothermal vent habitat to normal seawater aquarium, or conversely transfer of normal seawater acclimated xtcrabs to the hydrothermal vent region or the non-hydrothermal vent region of Kueishan Island.

Results

Hypoxic tolerance

All aquarium xtcrabs exhibited the activities and survived under hypoxia (DO = 0.19 mg/L) up to 16 h. All control species tdcrabs exhibited the activities and survived only up to 8 h (Fig. 1a). Meanwhile, all xtcrabs and tdcrabs survived in normoxia seawater for 24 h (data not shown). These data demonstrated that vent xtcrabs tolerated hypoxic stress greater than intertidal tdcrabs.

Fig. 1
figure 1

Survival of hypoxia and antimycin A challenge in the hydrothermal vent xtcrabs. (a) Aquarium xtCrabs (Xenograpsus testudinatus) and control tdcrabs (Thranita danae) collected from their original habitats (hydrothermal vent and intertidal zone, respectively) were maintained for one month in the normal seawater aquarium for acclimation. Then aquarium xtcrabs and control tdcrabs were subjected to a severe hypoxic challenge (Dissolved oxygen, DO = 0.19 mg/L seawater continuously flushed with nitrogen gas. DO of the normoxia group = 8.31 mg/L). The activity levels of xtcrabs and tdcrabs, including movement, touching reaction, and body calibration were checked every 4 h over 24 h period (n = 9 crabs/group). (b) LC50/96 hr of antimycin A was tested in aquarium xtcrabs. The aquarium xtcrabs were exposed to various antimycin A concentrations (0 ppb, 1 ppb, 10 ppb, 100 ppb, 500 ppb, 1000 ppb) for 96 h. Survival was recorded every 24 h for each group (n = 9 xtcrabs/group). (c) Oxygen consumption rate (mg O2 g− 1 hr− 1) in xtcrab under different antimycin A treatments for 30 min were examined. Means ± SD (n = 3 xtcrabs/group). Asterisks indicate the significant difference between 0 ppb and 400 ppb group by t-test (*: p < 0.05). (d) Oxygen content in a sealed water chamber for each xtcrab was recorded under 0 ppb antimycin A treatment for 30 min. (e) Oxygen content in a sealed water chamber for each xtcrab was recorded under 400 ppb antimycin A treatment for 30 min.

Full size image

Antimycin A is a potent inhibitor of mitochondria by interrupting the electron transport chain27,28. Antimycin A was used to induce the cellular hypoxia in xtcrabs with various concentrations (up to 1000 ppb) for 96 h (Fig. 1b). All xtcrabs survived in 0, 1, and 10 ppb treatments up to 96 h, but the survival rates decreased with the increase of antimycin A concentrations (100, 500, and 1000 ppb) and exposure time (up to 96 h), showing dose-dependent decreasing survivals. Based on the survival rates and their relative concentrations of antimycin A, we estimated the LC50/96 hr was 265 ppb in xtcrabs (Fig. 1b).

To further assess the effects of antimycin A in xtcrabs, oxygen consumption rate was measured to validate the interference to the aerobic respiration of xtcrabs under antimycin A treatments for the durations of 30 min. The oxygen consumption rate was significantly decreased in the 400 ppb group (0.28 ± 0.08 mg O2 g− 1 hr− 1) compared to the 0 ppb group (0.70 ± 0.14 mg O2 g− 1 hr− 1; p < 0.05; Fig. 1c). The oxygen water content of the 0 ppb group decreased faster within 30 min (Fig. 1d) compared to the oxygen water content of the 400 ppb group (Fig. 1e), illustrating that antimycin A could interrupt the mitochondrial electron transport chain in xtcrabs. Taken together, these data showed that xtcrabs had a higher tolerance to severe hypoxic environments than control tdcrabs.

Regulation of lactate level under various environmental conditions

The results of 12 h hypoxia challenge showed that the lactate level observed in xtcrabs were significantly increased in 1 h (10-fold, p < 0.001), 3 h (108-fold, p < 0.001), 6 h (146-fold, p < 0.001), and 12 h (144-fold, p < 0.001) compared to 0 h of immersion in hypoxic seawater (Fig. 2a). In contrast, the lactate level observed in tdcrabs were significantly increased only after 12 h (7.8-fold, p < 0.001; Fig. 2b).

Fig. 2
figure 2

Lactate levels in hemolymph under vent habitat, non-vent aquarium, and hypoxia environment in the xtcrabs. To study the effects of severe hypoxic environments on respiration metabolism, aquarium xtcrabs and control tdcrabs were immersed in the nitrogen gas-induced severe hypoxic seawater (DO = 0.19 mg/L) for 12 h, respectively. During 12 h of severe hypoxic seawater exposure, hemolymph samples were collected at 0, 1, 3, 6, and 12 h (n = 8 xtcrabs/group; n = 4 tdcrabs/group). (a) Lactate levels in hemolymph of xtcrabs under severe hypoxia for different periods. (b) Lactate levels in hemolymph of tdcrabs under severe hypoxia for different periods. To investigate the respiration metabolism of hydrothermal vent in xtcrabs, habitat xtcrabs transferred to normal seawater on the boat and then transferred to the lab aquarium. Hemolymph samples were collected for several intervals including habitat xtcrabs, transferred habitat xtcrabs immersed in normal seawater for 10 min, 30 min, 60 min, 6 h, 12 h, and 1 month (n = 4 xtcrab/group). (c) Lactate levels in hemolymph of habitat xtcrabs transferred to normal seawater for different periods. To further investigate the effects of hydrothermal vent environments on respiration metabolism, aquarium xtcrabs were transferred to station-1 (hydrothermal vent region, station-1 xtcrabs) or station-2 (sulfide-free region with normal seawater, station-2 xtcrabs) for 2 h with cages. After transfer, hemolymph samples were obtained immediately on the boat. Meanwhile, hemolymph samples of habitat xtcrabs and aquarium xtcrabs (0 point) were also collected on the boat for further comparison (n = 8 xtcrab/group). (d) Comparison of lactate levels in hemolymph among different xtcrabs of habitat xtcrabs, aquarium xtcrabs, station-1 xtcrabs, and station-2 xtcrabs. Different letters indicate significant differences among groups by one-way ANOVA (p < 0.05). Asterisks indicate significant difference between groups by t-test (*: p < 0.05; **: p < 0.01; ***: p < 0.001). The bar represents mean ± SD (standard deviation).

Full size image

To investigate that sulfide-rich hydrothermal vent environments would induce the anaerobic metabolism to xtcrabs, habitat xtcrabs were immersed in sulfide-free normal seawater for different time periods. The results showed that the lactate level was significantly decreased in the 30 min group, then dramatically dropped in the 60 min group, and remained constant in the 6 h, 12 h, and 1 month groups (p < 0.05; Fig. 2c). Lab-maintained aquarium xtcrabs were transferred back to sea for 2 h either in the hydrothermal vent sulfide-rich seawater environment (station-1 xtcrabs; Fig. 3; Table S1) or normal seawater environment (station-2 xtcrabs; Fig. 3). The lactate level in hemolymph samples were significantly higher in habitat xtcrabs and station-1 xtcrabs while not in station-2 xtcrabs and aquarium xtcrabs (p < 0.05; Fig. 2d). These results revealed that the increased lactate levels in hemolymph clearly indicate the crabs were exposed to the hypoxic environment. Furthermore, under hypoxic or hydrothermal vent environments, anaerobic metabolism rapidly switched on within 1 h in xtcrabs.

Fig. 3
figure 3

Location for the collection of vent xtcrabs off Kueishan Island in the Northeast of Taiwan. The habitat for xtcrabs of shallow-water hydrothermal vents is located on the Southeast of Kueishan Island (Northeastern Taiwan) at a depth of about 10–20 m (station-1; 24°50ʹ31ʺN, 121°57ʹ6ʺE). In contrast, no hydrothermal vent is present on the other side of the island (station-2; 24º50ʹ558ʺN, 121º56ʹ212ʺE). With the advantages of these unique environments, we can collect xtcrabs from their habitat (habitat xtcrabs) and culture them at the lab aquarium for at least 1 month (aquarium xtcrabs). For the field transfer experiment, the aquarium xtcrabs were transferred to the hydrothermal vent environments station-1 (station-1 xtcrabs) or station-2 with sulfide-free normal seawater (station-2 xtcrabs) for further comparison. For comparison, we collect control tdcrab from intertidal habitat (Keelung coast, 25°08ʹ09.1ʺN, 121°48ʹ10.4ʺE) and culture at the lab aquarium for at least 1 month (control tdcrabs). White square indicates the habitat for xtcrabs (KI, Kueishan Island). Black triangle indicates the habitat for tdcrabs (Keelung).

Full size image

Regulation of glucose level under various environmental conditions

In the immersion of hypoxia experiment, the glucose level of xtcrabs was significantly decreased in 3 h compared to 0 h (p < 0.05), but returned to initial levels at 6 h and 12 h (Fig. S1a) while the glucose level of tdcrabs showed no significant difference among groups (p > 0.05; Fig. S1b). In the experiment of habitat xtcrabs transferred to sulfide-free normal seawater, the glucose level of xtcrabs in each group showed no significant difference (p < 0.05; Fig. S1c). In the immersion of hydrothermal vent, station-1 and station-2 experiment, the glucose level of xtcrabs showed no significant difference (p > 0.05; Fig. S1d). Overall, glucose level showed no change and remained stable in all experiments. Taken together, these indicate that xtcrabs are able to maintain constant hemolymph glucose levels under hypoxic or hydrothermal vent environments.

Molecular identification of xtcrab CYTC and development of xtCYTC qPCR analysis

De novo assembly of our on-going draft transcriptomic database (SRR33664475) and BLAST search using human CYTC (PV554173) allowed us to obtain a potential CYTC sequence in xtcrabs. Using the specific primers designed from the potential CYTC sequence, we identified and cloned one CYTC sequence, named xtCYTC, from the gill of xtcrabs. The results of phylogenetic analysis showed that xtCYTC was clustered with other CYTC sequences from crustaceans (Fig. S2). To develop xtCYTC quantitative real-time PCR (qPCR) analysis, specific qPCR primers were designed and applied for qPCR reactions (Table S2). After the qPCR reaction, the amplified products of the xtCYTC DNA were sequenced, which confirmed the specificity of the xtCYTC qPCR primers.

Tissue distribution of xtCYTC transcripts and regulation of xtCYTC transcripts, xtCYTC protein level, and COXIV enzymatic activity by high sulfide-hypoxia environment

An upregulation of CYTC expression was observed in the posterior gill (pGi), which showed a significant increase in habitat xtcrabs compared to that in aquarium xtcrabs (6.7-fold, p < 0.05) (Fig. 4a). To further investigate the potential correlation between the expression of xtCYTC transcripts in pGi and the high sulfide-hypoxia hydrothermal vent environment, the results of xtCYTC transcripts in gills showed a significant increase in the habitat xtcrabs compared to that in the aquarium xtcrabs (2.6-folds, p < 0.001; Fig. 4b).

Fig. 4
figure 4

xtCYTC transcripts, xtCYTC protein, and xtCOXIV enzymatic activity expressions under various environments. (a) Tissue distribution of xtCYTC transcripts in various tissues collected from habitat xtcrabs and aquarium xtcrabs (n = 6 crabs/group). aGi, anterior gill; pGi, posterior gill; St, stomach; DG, digestive gland; In, intestine; Mu, muscle; He, heart. (b) xtCYTC transcripts in pGi of habitat xtcrabs and aquarium xtcrabs (n = 36 crabs/group). (c) xtCYTC protein expression as measured by Western blot (with xtCYTC antiserum) in the pGi of habitat and aquarium xtcrabs. Selected blot images are depicted above the graphs. Actin was used as a reference protein. Quantification of signal intensity was performed with ImageJ software (n = 3 xtcrabs/group). (d) COXIV enzymatic activity in habitat xtcrabs and aquarium xtcrabs (n = 6 crabs/group). Different letters indicate one-way ANOVA in xtCYTC tissue distribution (p < 0.05). EF1A was used as an internal reference to normalize the gene expression levels. The highest value among groups is set at 100, and each value is normalized to the ratio (each value/highest value) x100. Asterisks indicate significant difference between groups by t-test (*: p < 0.05; **: p < 0.01; ***: p < 0.001). The bars represent mean ± SD (standar deviation).

Full size image

The specificity of anti-xtCYTC was confirmed by a Western blot (WB) analysis with peptide preadsorption (Fig. S3). WB with anti-xtCYTC showed a major band of approximately 12 kDa, which is in agreement with the predicted molecular size (Fig. S3). Moreover, the absent immunoblot signal of CYTC was observed using the preadsorbed antisera (Fig. S3). These data suggested that the polyclonal antisera of anti-xtCYTC successfully identified the CYTC protein in xtcrabs. To compare the patterns of xtCYTC protein expression between habitat xtcrabs and aquarium xtcrabs, the immunoblot signal was significantly increased in habitat xtcrabs compared to the aquarium xtcrabs group (p < 0.05; Fig. 4c). After normalizing the xtCYTC signals to Actin levels, the habitat xtcrabs exhibited 6.8-fold higher gill expression of xtCYTC protein compared to the aquarium group (Fig. 4c). Original complete blots and multiple brightness of WB analysis (Fig. S4a-4c) demonstrated the valid of the data. xtCOXIV activity analysis showed no significant difference between the habitat xtcrabs and aquarium xtcrabs (p > 0.05; Fig. 4d).

Molecular identification of xtcrab SQR paralogs, tdcrab SQR paralogs, and phylogenetic analysis

Our previous study obtained the full coding DNA sequence (CDS) of xtcrabs and further demonstrated that the expression of duplicated and subfunctionalized SQRs is strongly up-regulated by hydrothermal vent environments in xtcrabs14. For further comparison of SQRs between vent xtcrabs and control tdcrabs, the full CDS of tdSQRparalogs was obtained. According to previous studies29,30 we annotated three putative FAD-binding domains (I, II, and III), one sulfide oxidation domain, and one quinone reduction domain in the SQR sequences of xtcrabs and tdcrabs (Fig. S5a). The identity of xtSQR1 and xtSQR2 was 86.36%, and the identity of tdSQR1 and tdSQR2 was 92.48% (Fig. S5b). These data suggest that xtSQR1 and xtSQR2 paralogs were more divergent compared to tdSQR1 and tdSQR2 paralogs.

To investigate the phylogenetic relationships between xtSQR paralogs and tdSQR paralogs, the partial sequences of cp.SQR paralogs (Cancer pagurus, cp.SQR1 and cp.SQR2) and esSQR (Eriocheir sinensis), the full-length sequences of xtSQR paralogs14 and tdSQR paralogs were used together with other SQR sequences of metazoan species from the NCBI database for the phylogenetic analysis. The phylogenetic analysis showed that xtSQR paralogs and tdSQR paralogs clustered with SQR sequences from crustaceans (Fig. 5a). About 36% species of crustaceans (4 species from 11 total species) had duplicated SQR paralogs. In contrast, there is no SQR duplication occurred in vertebrates and other invertebrates (Fig. 5a). Among the crab SQR sequences, the xtSQR1 and xtSQR2 in xtcrabs were quite distant in the phylogeny tree; while tdSQR1 vs. tdSQR2 in tdcrabs, cbSQR1 vs. cbSQR2, and cp.SQR1 vs. cp.SQR2 in other crustaceans were clustered more closely (Fig. 5a). Moreover, both xtSQR paralogs and tdSQR paralogs did not cluster with the Cancer borealis/C. pagurus SQR paralogs (Fig. 5a).

Fig. 5
figure 5

Multidisciplinary investigations of SQR from sequences to enzymes. (a) Phylogenetic analysis of SQR amino-acid sequences from various species. SQR sequences of xtcrab and tdcrab were marked with the bottom line. Numbers indicate bootstrap values (% of 1000 replicates). pGi samples collected from habitat xtcrabs and aquarium xtcrabswere used for the analysis of xtSQR1 transcripts and SQR enzymatic activity. (b) xtSQR1 transcripts in pGi of habitat xtcrabs and aquarium xtcrabs (n = 10 xtcrabs/group). (c) SQR enzymatic activity of pGi in habitat xtcrabs and aquarium xtcrabs (n = 3 crabs/group). (d) SQR enzymatic activity in habitat xtcrabs and tdcrabs. (e) SQR enzymatic activity in aquarium xtcrabs and tdcrabs (n = 3 crabs/group). EF1A was used as an internal reference to normalize the gene expression levels. The highest value among groups is set at 100, and each value is normalized to the ratio (each value/highest value) x100. Asterisks indicate significant difference between groups by t-test (**: p < 0.01; ***: p < 0.001). The bar represents mean ± SD (standard deviation).

Full size image

Regulation of xtSQR1 transcripts and comparison of SQR enzymatic activity between xtcrabs and tdcrabs

qPCR results showed that xtSQR1 transcripts in gills were significantly increased in the habitat xtcrabs compared to that in the aquarium xtcrabs (p < 0.05; Fig. 5b). SQR enzymatic activity was also significantly higher in habitat xtcrabs compared to that in the aquarium xtcrabs (p < 0.01; Fig. 5c). These data also agree with our previous study that the high sulfide-hypoxia hydrothermal vent environment would up-regulate the SQR expression level (transcripts and protein)14.

SQR enzymatic activity showed a significantly higher level in habitat xtcrab compared to control tdcrab (1.5-fold, p < 0.01; Fig. 5d). Moreover, SQR enzymatic activity showed no significant difference between aquarium xtcrabs and control tdcrabs (p > 0.05; Fig. 5e). These data further support the strong ability of vent xtcrabs to up-regulate of gill xtSQR enzymatic activity in the face of sulfide-rich vent environment.

The localization of xtCYTC and xtSQR-expression cells in in xtcrab gills

Bright field vision under microscopy with DAPI nuclei staining showed the structure of gill filaments (Fig. 6a). NKA-antibody was used to annotate the basolateral membranes of pillar cells (Fig. 6b and e) as previous studies14,31. Cytoplasm of pillar cells and hemocytes in the gill filament exhibited detectable levels of xtSQR (Fig. 6c and e, and 6f) and xtCYTC (Fig. 6d and f). Notably, except for the absence of NAK signal in hemocytes, all hemocytes expressed xtSQR and xtCYTC, and all pillar cells expressed NKA, xtSQR, and xtCYTC, according to sections that were co-stained with DAPI-SQR-NKA (Fig. 6e) and DAPI-SQR-CYTC (Fig. 6f). Remarkably, we found co-localization of xtSQR and xtCYTC in each of more than 100 SQR-positive cells examined from six xtcrabs.

Fig. 6
figure 6

The localization of xtCYTC-expression cells in the gills. (a) Bright Field under microscopy and cell nuclei staining with DAPI (4’, 6-diamidino-2-phenylindole; blue) were used for the observation of gill structure. (b) Immunofluorescence staining with NA+/K+-ATPase (NKA) antibody was performed to localize NKA (red) expression in the gill. (c) Immunofluorescence staining with xtSQR antibody was performed to localize xtSQR (green) expression in the gill. (d) Immunofluorescence staining with xtCYTC antisera was performed to localize xtCYTC (green) expression in the gill. (e) The DAPI (blue), xtSQR (green), and NKA (red) merged figure showed the co-localization of xtSQR and NKA signals in the gill. (f) The DAPI (blue), xtSQR (red), and xtCYTC (green) merged figure showed the co-localization of xtSQR and xtCYTC signals in the gill. HC: hemal channel; GF: gill filament; Hc: hemocyte; Pi: pillar cell. (n = 6 xtcrabs).

Full size image

Discussion

For the acclimation of a hypoxia environment, several strategies have been evolved in crustaceans, such as avoidance behaviors to escape from the hypoxic environment, increasing gill ventilation rates, adjusting the oxygen affinity of hemocyanin, and the onset of anaerobiosis, as evidenced by increased hemolymph lactate levels3,4,5,6.

Here, we focused on the aerobic and anaerobic metabolism for the xtcrabs adapted to the hydrothermal vent extreme environment. With the comprehensive approaches including biochemical, molecular, and enzymatic analyses, our findings revealed that the rapid activation of anaerobic metabolism and the maintenance of aerobic metabolism simultaneously occurred in vent xtcrabs. We combined the favorable factors for aerobic respiration, including xtSQR and xtCYTC, with lactate, a major end product of anaerobic respiration. Our results demonstrated that these biochemical parameters (xtSQR and xtCYTC) were strongly up-regulated by the hydrothermal vent environment.

We examined the hypoxia tolerance of aquarium xtcrabs and control tdcrabs exposed to hypoxia in aquaria (DO = 0.19 mg/L seawater continuously flushed with nitrogen gas). All xtcrabs and tdcrabs exposed to the hypoxic condition survived at least 8 h of exposure, and both species exhibited routine behaviors. However, the xtcrabs showed greater ability to tolerate severe hypoxia compared to tdcrabs. Following 16 h of hypoxic exposure, the xtcrabs fully survived in the hypoxic condition, while tdcrabs experienced mass mortality. This suggests that the xtcrabs possessed specialized physiological acclimations to prolonged severe hypoxia. Generally, environmental hypoxia refers to the phenomenon where the concentration of dissolved oxygen in marine and estuarine waters is below 2–3 mg/L and in freshwater below 5–6 mg/L, while severe hypoxia is defined as ≤ 0.5 mg/L32,33. Our previous study recorded that the dissolved oxygen levels ranged from 4 to 7 mg/L in the hydrothermal vent region of Kueishan Island14. The dissolved oxygen levels in the habitat of the deep-sea hydrothermal vent crab, Bythograea thermydron, were recorded to range from 0 to 1.76 mg/L34. According to our present data, vent xtcrabs could survive in the severe hypoxic environment.

Antimycin A is a fungal resource-metabolites of mixture, which is the inhibitor of the mitochondrial electron transport chain, and widely used as a piscicide to control the fish populations in USA27,28,35. We examined the antimycin A tolerance and confirmed the LC50/96 hr was 265 ppb in aquarium xtcrabs. Our findings also revealed that antimycin A reduced the oxygen consumption rates in xtcrabs, suggesting that aerobic electron transport was inhibited under antimycin A treatment. Previous studies recorded the LC50/96 hr of antimycin A in various species including American paddlefish (Polyodon spathula, 0.0017 ppb), rainbow trout (Oncorhynchus mykiss, 0.04 ppb), black bullhead catfish (Ameiurus melas, 45 ppb), and a freshwater amphipod (Gammarus fasciatus, 0.0089 ppb)36,37. This suggests that xtcrabs have well-developed underpinning cellular mechanisms to tolerate hypoxia, caused by the ambient hypoxic water and cellular hypoxia induced by an inhibitor of electron transport chain.

To understand the alternation of aerobic and anaerobic metabolism in hypoxic environments, the hemolymph lactate, the end product of the anaerobic metabolism, was measured. Aquarium xtcrabs and control tdcrabs were immersed in severe hypoxic seawater (DO = 0.19 mg/L) for 12 h. Both xtcrabs and tdcrabs showed increased lactate content in hemolymph. However, unlike the rapid increase in xtcrabs (1 h), the lactate level showed a significant increase until 12 h in hemolymph of tdcrabs. These data demonstrated that xtcrabs showed a faster activation of anaerobic respiration than tdcrabs. Our field study showed higher lactate level in habitat xtcrab (6.9 ± 4.2 mmol/L), which have a long term exposure to hydrothermal vent environment, as compared to aquarium xtcrab exposed for one month to normal sea water (1.9 ± 1.0 mmol/L). Furthermore, the time course of the decrease in lactate levels after transfer of habitat xtcrab to normal sea water on the boat, showed a rapid drop already significant after 30 min of transfer. This further highlights the ability of xtcrab to rapidly switch on and off their anaerobic metabolism. Little information is available for the lactate accumulation in the crabs. Anaerobic metabolism is also known to occur along with the accumulation of lactate during prolonged exercise in Necora puber crabs38. In the deep-sea hydrothermal vent crabs (B. thermydron), lactate accumulation was found to occur in the hydrothermal vent environment12. Previous study revealed that vent xtcrabs under acidified conditions (pH 6.5) dropped hemolymph pH unit by 0.25 compared to control (pH 8.0) but hemolymph pH value could restore after 4 h and then stably maintained to pH 7.4. Meanwhile, gill Na⁺/K⁺-ATPase (NKA) and V-type H⁺-ATPase (VHA), the primary ion pumps responsible for acid–base regulation, exhibited marked upregulation at the mRNA, protein, and enzymatic activity levels under acidified conditions31. We postulate that well-developed gill machinery associated with ion transport and acid-base regulation may help buffer their extracellular pH during the onset of anaerobiosis and an accumulation of hemolymph lactate31. This mechanism may imply that xtcrabs have much more capacity for lactate accumulation in hemolymph. The possible excretion of lactate via gill, mucus, or other organs to alleviate the lactate burden could not be excluded. Taken together, the rapid switch-on of the anaerobic metabolism and strong ion regulatory ability in vent xtcrabs may help them conquer the hypoxic environment. Moreover, these data suggest that the high sulfide-hypoxia hydrothermal vent environment in Kueishan Island may induce a rapid switch on anaerobic metabolism in xtcrabs.

Previous studies have reported that hypoxia triggers the activation of gluconeogenesis to maintain aerobic metabolism in various crustacean species39,40. In Chinese mitten crab (Eriocheir sinensis) and Chinese grass shrimp (Palaemonetes sinensis), previous study revealed that a significant increase in glucose content after environmental hypoxia41. However, the glucose content in vent xtcrabs showed no significant difference after severe hypoxia exposure or the field transfer experiment. Glucose serves as the primary energy source for both aerobic and anaerobic metabolism42. Additionally, glucose is synthesized from the end products of anaerobic metabolism, such as lactate, to facilitate recovery after hypoxia43. Our findings suggest that the pathways to maintain glucose supply are well operated under energy-demanding conditions and also the increase of lactate could be the indicator for the experience of hypoxia in xtcrabs. Moreover, in the extreme environment of a hydrothermal vent and at the limits of hypoxia, aerobic and anaerobic metabolism may simultaneously occur and dynamically switch on and off in vent xtcrabs, depending on the concentrations of environmental sulfide and dissolved oxygen. xtCrabs have the strong ability to rapidly turn on the anaerobic metabolism for the supply of metabolic energy. Therefore, maintaining glucose supply and rapidly activating anaerobic metabolism are suggested as the potential strategies for survival and adaptation in vent xtcrabs to the high sulfide-hypoxic extreme environment.

In the present study, we cloned the full-ORF of two SQR gene paralogs from tdcrabs (tdSQR1 and tdSQR2). Both tdSQR1 and tdSQR2 performed the functional domains as conserved SQR. The phylogenetic analysis exhibited that tdSQR1 and tdSQR2, and other paralogs were clustered more closely, while xtSQR1 and xtSQR2 were quite distant. These findings further support the occurrence of SQR duplication and subfunctionalization in vent xtcrabs as suggested in our previous study14. We also revealed that several independent SQR duplication events occurred across the brachyuran lineage but not in other metazoans14. The independent SQR duplication in tdcrabs and xtcrabs suggested a potential impact on the fate of gene mutations. We proposed that the high H2S hydrothermal vent extreme environment might be one of the factors driving the duplication and further subfunctionalization of SQR in vent xtcrabs.

Again, we confirm the sulfide detoxification ability of SQR in vent xtcrabs. The increase of xtSQR1 transcripts/protein and xtSQR enzymatic activity in habitat xtcrabs also agrees with our previous study14. To clarify the xtSQR ability in xtcrabs, intertidal tdcrabs was used as a control species for the comparison of enzymatic activity. SQR activity were significantly increased in habitat xtcrabs compared to aquarium xtcrabs and control tdcrabs. In aquarium xtcrabs, no difference in SQR enzymatic activity was observed between aquarium xtcrabs and tdcrabs. These data highlight the strong ability and high plasticity of xtSQR in xtcrabs under high sulfide hydrothermal vent environment.

Sulfide can inhibit mitochondrial cytochrome c oxidase activity by blocking the electron transport chain in animals44. High concentrations of sulfide decreased the reduction state of CYTC (become oxidized CYTC state) and dissolved oxygen, and further resulted in the termination of aerobic respiration24,25. Mitochondrial respiratory chain is suggested to be critical for the adaptation to the exposure to hydrogen sulfide and sulfide detoxification mediated by SQR45. Therefore, high xtSQR expressions at gene and enzyme levels also help mitochondria to maintain the electron transport chain to further facilitate the activity of aerobic metabolism. Moreover, in short-term acclimation, there might be other potential strategies to overcome sulfide toxicity and hypoxia, such as temporary alternation to anaerobic metabolism to achieve the energy demand.

In the present study, we identified and cloned CYTC gene from xtcrabs. Our phylogenetic analysis showed that xtCYTC was clustered with CYTC from other crustacean. According to our qPCR results, xtCYTC was expressed in all investigated tissues of habitat xtcrabs and aquarium xtcrabs, including anterior gill, posterior gill, stomach, digestive gland, intestine, muscle, and heart, but it showed a significant increase specifically in the posterior gill of habitat xtcrabs. Gills are the major respiratory organs in aquatic crustaceans that perform essential functions in ion transport and osmoregulation46. Cytochrome c, which plays a crucial role in aerobic metabolism, is an electron carrier between Complex III (cytochrome bc1 complex) and Complex IV (cytochrome c oxidase) in the mitochondrial inner membrane to maintain the electron transport chain47. However, the electron transport chain would be interrupted by surrounding sulfide via the deprivation of electron from reduced CYTC23,24,25. As the sulfide detoxification-related xtSQR paralog we revealed in our previous study, xtCYTC also showed upregulation in the high sulfide vent environment. An increase in CYTC, which is associated with a significant increase in cellular respiration, has been confirmed in mouse embryonic fibroblasts under serum starvation up to 12 h48.

To further evaluate the contribution of aerobic respiration in the gill of habitat xtcrabs and aquarium xtcrabs, xtCOXIV enzymatic activity was conducted. The results of xtCOXIV enzymatic activity showed that habitat xtcrabs under extreme environment had a similar activity to aquarium xtcrabs. Previous study revealed that COXIV enzymatic activity was reduced under hypoxic conditions (DO = 2 mg/L for 6 h), and gradually increased as hypoxia continued (DO = 1 mg/L for another 6 h) in white shrimp (Litopenaeus vannamei)49. This evidence suggested that the shortage of oxygen would inhibit the activity of aerobic metabolism in crustaceans. Unlike crustaceans, the shortfin molly (Poecilia Mexicana), which lived in the sulfide-spring region, evolved a lower H2S susceptibility of COX to adapt the sulfide environment50. These data suggested that xtcrabs are able to maintain aerobic metabolism even in the high sulfide and hypoxic hydrothermal vent environment. Altogether, we hypothesize that the increased expression xtCYTC (transcripts and protein) may compensate the reduction of CYTC activity due to hypoxia and sulfide-induced oxidized xtCYTC to maintain the xtCOXIV function for the electron transport chain and aerobic metabolism.

In xtcrab, CYTC expression was observed in all hemocytes and pillar cells of the gill filament. Here, we also established the SQR antibody against the xtSQR in xtcrab. Impressively, all observed xtCYTC-positive cells were also xtSQR-positive cells and all observed xtSQR-positive cells were also NKA-positive cells. Our previous study demonstrated the co-localization of SQR and NKA in pillar cells using anti-human SQR14, which agreed with our results in the present study with anti-xtSQR antibody. We also confirmed the localization of xtSQR in the mitochondria by protein fraction analysis14. In mammalian cells, CYTC, which serves as the electron carrier protein, is localized in the mitochondria26,51. Higher protein level and enzymatic activity of xtSQR (in previous14 and the present study) and protein level of xtCYTC (the present study) were found in the vent habitat xtcrabs as compared to the aquarium xtcrabs. Hence, we proposed that the sulfide detoxification via SQR might prevent CYTC from electron deprivation. Moreover, the production of electrons during sulfide detoxification might be recruited to the electron pool in the mitochondrial electron transport chain. These proposed mechanisms may be favorable factors for the maintenance of aerobic metabolism in xtcrabs under the sulfide-hypoxia hydrothermal vent environment.

Conclusion

Our study revealed metabolic strategies that enable the hydrothermal vent crab X. testudinatus to tolerate hypoxic and sulfide-rich conditions. The mechanisms for the physiological adaptation of xtcrabs to high sulfide and low dissolved oxygen in hydrothermal vents were proposed (Fig. 7). The rapid increase in hemolymph lactate levels during hypoxic exposure indicates the rapid activation of anaerobic metabolism, helping the crab endure oxygen depletion—an important physiological adaptation to vent environments. Both low oxygen and high sulfide inhibit aerobic metabolism by reducing CYTC activity and disrupting electron transfer in the electron transport chain. However, the long-term increase of gill xtCYTC expression in vent-adapted xtcrabs may compensate for sulfide inhibition. The maintained enzymatic activity of xtCOXIV, together with high expressions of xtSQR and xtCYTC, and their cellular co-localization, might support the persistence of aerobic respiration for energy production in the vent habitat.

Fig. 7
figure 7

Proposed physiological adaptation of xtcrabs to high sulfide and low dissolved oxygen in hydrothermal vent environment. Hydrogen sulfide suppressed xtCYTC function which serves as the electron carrier between Complex III (cytochrome bc1 complex) and Complex IV (cytochrome c oxidase) in the mitochondria. Therefore, sulfide interrupts the electron transport chain and inhibit aerobic respiration. Additionally, the lack of O2 directly decreases the activities of xtCYTC and aerobic respiration. To cope with high sulfide-hypoxia ecological niches, xtSQR in xtcrab gills are labored for sulfide detoxification and also favored for mitochondrial electron transport chain. Moreover, the increased xtCYTC transcripts may recruit the reduced state of xtCYTC protein to replace sulfide-induced oxidized state of xtCYTC. Hydrogen sulfide easily interacts with the ambient oxygen and then induces a hypoxic environment14. The increase in anaerobic metabolism is rapidly turning on to cope with the decreased energy production from aerobic respiration in ecological niches with lower O2. Meanwhile, vent xtcrabs also try to keep the operation of aerobic metabolism and maintain a certain degree of aerobic respiration (as shown with a similar xtCOXIV enzymatic activities between habitat xtcrabs and aquarium xtcrabs). These multiple strategies may contribute to the acclimation and adaptation of xtcrabs to the hydrothermal vent extreme environment.

Full size image

By integrating ecological, physiological, and molecular approaches, we found that X. testudinatus employs both aerobic and anaerobic pathways concurrently. The ability to rapidly switch to anaerobiosis ensures a continuous energy supply under fluctuating oxygen and sulfide levels. Collectively, these findings suggest that multiple metabolic strategies support the acclimation, adaptation, and survival of X. testudinatus in the extreme, sulfide-rich, hypoxic conditions of shallow-water hydrothermal vents.

Methods

Crab collection and sampling

Vent xtcrabs were collected from the hydrothermal vent region in Kueishan Island, Taiwan by SCUBA divers (station-1, 24°50ʹ31ʺN, 121°57ʹ6ʺE) during 2016–2022 (Fig. 3, Table S1). The individual carapace lengths widths ranged from 15 to 25 mm. Sampled vent xtcrabs were either (i) immediately submerged and anesthetized with 120 µM eugenol in seawater (referred to as “habitat xtcrabs”), hemolymph was collected and posterior gill (pGi) were sampled, (ii) dissected for hemolymph and pGi sampling during field transfer experiment, (iii) maintained in seawater on the boat and transferred to aquaria laboratory at National Taiwan Ocean University (NTOU; referred to as “aquarium xtcrabs”). For comparison, intertidal crabs (control species crab), Thranita danae (tdcrab, Portunidae), were collected by net from the intertidal zone of a rocky seashore in Keelung coast, Taiwan in 2019, 2022 and 2022 (25°08ʹ09.1ʺN, 121°48ʹ10.4ʺE; Fig. 3). The individual sizes ranged from 25 to 35 mm in carapace length and from 50 to 60 mm in carapace width and were transferred to aquaria at NTOU. Aquarium xtcrabs and tdcrabs were held in a respectively 300 L tank with normal seawater and daylight conditions for one month. Crabs were fed with the shrimp food pellets (Omega, Kaohsiung, Taiwan) once a day. The collected samples were frozen in liquid nitrogen and then kept at −80℃ freezer in the field and laboratory experiments.

Field transfer experiment

To discover the aerobic and anaerobic metabolism of aquarium xtcrabs encountered the hydrothermal vent environment, two sets of aquarium xtcrabs (n = 6) in the 60 × 25 × 25 cm plastic cage (n = 6 crabs/group/cage) were left at transfer stations (depth = 9 m) for 2 hr. One group of aquarium xtcrabs (n = 6) were transferred back to the hydrothermal vent region of Kueishan Island with a plastic cage (sulfide-rich environment, station-1 24°50ʹ31ʺN, 121°57ʹ6ʺE; Fig. 3, Table S1) and were referred to as “station-1 xtcrabs.” Another group of the aquarium xtcrabs (n = 6) were transferred to non-hydrothermal vent region of Kueishan Island with a plastic cage (sulfide-free environment of normal seawater, station-2: 24º50ʹ558ʺN, 121º56ʹ212ʺE; Fig. 3) and were referred to as “station-2 xtcrabs.” After transfer experiment, the cages were collected, and the experimental crabs were collected for hemolymph samples to analyze hemolymph glucose and lactate levels. For further comparison, ”habitat xtcrabs” were sacrificed at various intervals after transfer to normal seawater (10 min, 30 min, 60 min, 6 h, 12 h, and 1 month) for the collection of hemolymph and analysis of lactate and glucose levels.

Tissue sampling

For tissue sampling, all experimental crabs were submerged in seawater containing 120 µM eugenol for anesthesia. Before sampling, the crabs were recorded for the length of the carapace. For tissue distributions of cytochrome c (xtCYTC transcripts by qPCR), habitat xtcrabs (n = 6) and aquarium xtcrabs (n = 6) were analyzed; tissues including anterior gills (aGi), posterior gills (pGi), stomach (St), digestive gland (DG), intestine (In), muscle (Mu), and heart (He) according to previous study14. pGi from 36 habitat xtcrabs and 36 aquarium xtcrabs were obtained from our previous study further analyze the xtCYTC transcripts by qPCR14. To analyze the xtSQR1 transcripts by qPCR, pGi from 10 habitat xtcrabs and 10 aquarium xtcrabs were collected. To analyze the COXIV enzymatic activity, pGi of 6 habitat xtcrabs and 6 aquarium xtcrabs were collected. To analyze the SQR enzymatic activity, pGi of 3 habitat xtcrabs, 3 aquarium xtcrabs, and 3 control tdcrabs were collected. To analyze the Western blot and immunofluorescence of xtCYTC, 3 habitat xtcrabs and 3 aquarium xtcrabs were collected.

Severe hypoxic experiment

Habitat xtcrabs and tdcrabs were subjected to a hypoxic challenge. The hypoxic environments in the aquarium (direct infusion with nitrogen gas for 30 min continuously flushing from DO = 8.31 mg/L to DO = 0.19 mg/L), was prepared by continuously flushing with nitrogen gas for 24 h in the 2 L aquarium at 25 ℃ with thermostat. An Orion Star™ A329 Portable pH/ISE/Conductivity/DO meter (Thermo Fisher Scientific, Waltham, MA, USA) was used to monitor the DO valu14. Survival was checked every 4 h using several indicators to evaluate the activity of crabs as described in previous study14. During the hypoxic exposure, hemolymph samples were independently collected at 0, 1, 3, 6, and 12 h (n = 8 xtcrabs/group and n = 4 tdcrabs/group for each time point) and used for the analysis of lactate and glucose contents.

Cellular hypoxia induced by antimycin A

Antimycin A is a potent inhibitor of the mitochondrial electron transport chain and is used to induce cellular hypoxia27,28. To test the tolerance of xtcrabs to chemical induced hypoxia, antimycin A (A8674, Sigma-Aldrich, St. Louis, MO, USA) exposure experiment was conducted with aquarium xtcrabs. To investigate the tolerance to hypoxia and 50% lethal concentration (LC50) during 96 h in xtcrabs, several antimycin A concentrations (from 0 ppm up to high dose) were tested (0 ppb, 1 ppb, 10 ppb, 100 ppb, 500 ppb, and 1000 ppb) for 96 h in 2 L aquaria at 25 ℃. Survivals, as described previously, was recorded every 24 h (n = 9 crabs/treatment). The 50% lethal concentration (LC50) of antimycin A was calculated at 96 h.

Measurement of oxygen consumption

To assess the impact of antimycin A on metabolism, oxygen consumption was measured under different concentrations of antimycin A. Before the experiment, seawater was filtered with the 0.22 μm filter to remove the bacterial respiration effect. Aquarium xtcrabs were placed in sealed 500 ml glass chambers with different antimycin A concentrations: 0 ppb, 40 ppb, 120 ppb, 160 ppb, 200 ppb, and 400 ppb. Each chamber contained one crab and three-repeated independent test were performed (n = 3 crabs/group). The oxygen consumption rate was monitored by using the OXY-4 mini-multichannel fiber optic oxygen transmitter (PreSens, Regensburg, Germany). The device set-up followed the manufacturer’s instructions (https://www.presens.de/products/detail/oxy-4-mini). For calibration, 0% standard was made with 1% NaHSO3 (243973, Sigma-Aldrich) seawater, and 100% standard was made with 30 min air pumped seawater. A chamber with only filtered seawater was used as the background control. During the experiment, the temperature was maintained at 25 ℃. To achieve the optimal measurement time, the oxygen consumption of individuals was measured until the air saturation level dropped to 80%. To obtain the mass-specific oxygen consumption rates, the wet weight was recorded and the oxygen consumption rate (mg O2 g− 1 hr− 1) was calculated52.

Measurement of lactate and glucose levels in hemolymph

The hemolymph samples (100 µl) collected from the nitrogen-induced hypoxia experiment and field transfer experiment were used for the analysis of lactate and glucose content. After the collection of hemolymph, samples were centrifuged to 1,000 rpm at 4℃ for 5 min to remove particles. Lactate and glucose levels were detected using the L-Lactate Assay Kit (ab65330, Abcam, Cambridge, MA, USA) and Glucose Assay Kit (MAK263, Sigma-Aldrich) according to the manufacturer’s instructions, respectively.

Cloning of xtcrab cytochrome c (CYTC) cDNA

Posterior gill samples (pGi), which were used to build our in-house transcriptome of xtcrab, were homogenized in TRIzol reagent (Invitrogen, Waltham, MA, USA). Total RNA was extracted following the manufacturer’s protocol. The quality and quantity of total RNA was determined by gel electrophoresis and NanoDrop™ 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), respectively. Total RNA extracts were used for Illumina sequencing, cDNA library construction, and cDNA synthesis.

For RNA-seq, total RNA extracts were performed by Genomics, Inc. (New Taipei city, Taiwan). Paired-end sequencing (150 bp) was performed on a HiSeq 2500 (Illumina, San Diego, CA, USA). Raw reads were uploaded to the Galaxy platform (http://usegalaxy.org) for cDNA library construction. Trimmed reads were generated from raw reads using the Trimmomatic trimming tool (ILLUMINACLIP step, Slide Window = 4:5, Leading = 5, Trailing = 5, minlen = 25) by removing adapters and low-quality reads. All trimmed reads were pooled together for de novo assembly by Trinity (v2.9.1) to get the assemble transcripts dataset as our cDNA library (SRR33664475). Assemble transcripts were used as a local BLAST database to design the primers for gene cloning. Primers are listed (Table S2).

For cDNA synthesis, pGi was homogenized in Trizol reagent (Invitrogen, Waltham, MA, USA). Total RNA was extracted according to manufacturer’s protocol. The first-stand cDNA was synthesized from 1 µg total RNA by Superscript III reverse transcriptase (Invitrogen, Waltham, MA, USA) according to the manufacturer’s protocol. Full-length (including ORF) of xtCYTC was found and cloned in gill14.

Cloning of TdSQRs

The two paralogs of xtcrab SQR were cloned in our previous study14. For comparison, posterior gill samples collected from tdcrabs were used for RNA extraction and cDNA synthesis following the manufacturer’s protocol. The SQR partial sequences in tdcrabs, tdSQR1 and tdSQR2 which derived from our previous study14 were used as templates to obtain the full length with a rapid amplification of cDNA ends (RACE) kit (SMART RACE cDNA; BD Biosciences Clontech, Franklin Lakes, NJ, USA) (Table S1). The sequence alignment and the identity between SQR paralogs were analyzed with DNAMAN sequence analysis software.

Phylogenetic analyses of CYTC and SQR sequences

The amino-acid sequences of CYTC from various species including full-length of xtCYTC were automatically aligned with MUSCLE, as implemented in Mega1153. The phylogenetic tree was constructed using the neighbor-joining method with the best model (LG + G, Gamma rate = 0.6178) in MEGA1153. The number at each node represents the bootstrap value (% from 1000 replicates). The accession numbers of the sequences used in the analysis are listed (Table S3).

The amino-acid sequences of SQR from various species including full-length of xtSQR paralogs and tdSQR paralogs were aligned and the phylogenetic tree was constructed as described above for CYTC. The accession numbers of the sequences used in the analysis are listed (Table S4).

qPCR analyses of XtCYTC and xtSQR1

The quantitative real-time PCR (qPCR) was used to analyze the gene expression profiles. Elongation factor 1 alpha (EF1A) was used as an internal reference to normalize gene expression levels. We developed xtCYTC and xtSQR1 qPCR analysis according to previous study14. The data were calibrated according to the 2-ΔΔCt method54. The relative expression value of target gene in all samples was normalized to EF1A, and the highest value of the target gene was defined as 100%. Specific primers for xtSQR1, xtCYTC and xtEF1A were listed (Table S2).

Enzymatic activities of SQR and COXIV in xtcrabs and tdcrabs

SQR is suggested as an important enzyme for the detoxification of vent sulfide in xtcrabs14. We measured SQR enzyme in the pGi mitochondria of xtcrab and tdcrabs. Mitochondria were isolated and the SQR enzymatic activity assay was performed according to our previous study14.

COXIV is considered as an important factor for the aerobic respiration involving electron chain transport. xtCOXIV enzymatic activities in the isolated mitochondria were detected using the cytochrome c oxidase assay kit (ab239711, Abcam).

The production of xtCYTC and xtSQR antisera

A polyclonal anti-xtCYTC antisera was induced against to the synthetic peptide (CAQCHTVEAGGKHKTGPN). The synthetic peptide was conjugated with ovalbumin (OVA) and used to immunize the antisera. The preparation of antisera was conducted by commercial company (Yao-Hong Biotechnology, Inc., New Taipei, Taiwan). After the synthetic peptide immunization, the sera from the mice were collected for Western blot analysis and immunofluorescence staining.

A polyclonal anti-xtSQR antisera was induced against to the synthetic peptide (SPKHQYDGYTSCPLVTGYSKCIMAEFDMNLS). The synthetic peptide was conjugated with ovalbumin (OVA) and used to immunize the antisera. The antiserum was purified using an affinity column containing 3 mg of synthetic peptide (Yao-Hong Biotechnology Inc.). The purified antibody was used for Western blot analysis and immunofluorescence staining.

xtCYTC Western blot analysis

The protein extraction, protein concentration determination, and Western blot (WB) analysis of the posterior gills were performed as our previous study14. To validate the antisera specificity, the synthetic xtCYTC peptide (1 µg per ml 1.5% nonfat milk) were incubated with anti-xtCYTC (diluted 1:1,000 with 1.5% nonfat milk) at 37℃ for 1 h as preadsorption reaction WB analysis. To detect the CYTC and Actin proteins, mouse anti-xtCYTC (diluted 1:1,000 with 1.5% nonfat milk) and mouse anti-human Actin antibody (MAB1501, Merck, Darmstadt, Germany; diluted 1:10,000 with 1.5% nonfat milk) were used. For secondary antibody, alkaline phosphate-conjugated goat anti-mouse IgG antibody (31320, Thermo Fisher Scientific; diluted 1:10,000 with 1.5% nonfat milk) were used. For detection, the NBT/BCIP Detection System (B1911, Sigma-Aldrich, St. Louis, MO, USA) was used. The same procedures of specificity validation were conducted with the anti-SQR antibody (diluted 1:1,000 with 1.5% nonfat milk).

Immunofluorescence staining

The tissue fixation, paraffin embedding, antigen retrieval, and immunofluorescence staining of the posterior gills were performed as our previous study14. DAPI (DAPI; 4’, 6-diamidino-2-phenylindole; Vector Laboratories, Burlingame, CA, USA) was used to label the nuclear DNA in all staining. For co-staining of SQR and Na+/K+-ATPase (NKA), the rabbit anti-SQR antibody (diluted 1:200 with 1.5% nonfat milk powder) and the mouse anti-NKA antibody (Developmental Studies Hybridoma Bank, Iowa city, IA, USA; diluted 1:200 with 1.5% nonfat milk powder) were used. Secondary antibodies of Alexa Fluor 488-conjugated goat anti-rabbit IgG (H + L) (A-11034, Thermo Fisher Scientific; diluted 1:200 with 1.5% nonfat milk powder) and Alexa Fluor 546-conjugated goat anti-mouse IgG (H + L) (A-11030, Thermo Fisher Scientific; diluted 1:200 with 1.5% nonfat milk powder) were used for the detection of anti-xtSQR and anti-NKA primary antibodies, respectively. For co-staining of xtSQR and xtCYTC, the rabbit anti-xtSQR antibody (diluted 1:200 with 1.5% nonfat milk powder) and the mouse anti-xtCYTC antisera (diluted 1:200 with 1.5% nonfat milk powder) were used. Secondary antibodies of Alexa Fluor 488-conjugated goat anti-mouse IgG (H + L) (A-11029, Thermo Fisher Scientific; diluted 1:200 with 1.5% nonfat milk powder) and Alexa Fluor 546-conjugated goat anti-rabbit IgG (H + L) (A-11035, Thermo Fisher Scientific; diluted 1:200 with 1.5% nonfat milk powder) were used for the detection of anti-xtCYTC antisera and anti-xtSQR primary antibody, respectively. All signals in the cells of the gill were observed, and images were captured under fluorescence microscopy (Olympus BX53).

Data analysis

Data are shown as mean ± standard deviation (SD). One-way ANOVA followed by a Tukey’s analysis was performed with Statistical Package for the Social Science (SPSS) software to test the significance of differences among three or more groups. Student’s t-test was used to analyze the significance of differences between two groups. Asterisks (p < 0.05: *, p < 0.01: **, p < 0.001: ***) indicate significant difference.