The regulation of oxidative phosphorylation pathway on Vibrio alginolyticus adhesion under adversities

Abstract Vibrio alginolyticus is one of the most important pathogens in mariculture and leading to heavy losses. After treatment with Cu2+, Pb2+, and low pH, the expression of oxidative phosphorylation pathway genes, including coxA, coxB, coxC, ccoN, ccoO, and ccoQ, was found commonly downregulated by RNA‐seq as well as quantitative real‐time PCR. RNAi significantly reduced the expression of coxA, coxB, coxC, ccoN, ccoO, and ccoQ in V. alginolyticus. Compared with the wild‐type strain, the adhesion abilities of RNAi strains of V. alginolyticus were significantly impaired, as well as their cytochrome C oxidase activity. ccoQ appeared to be more important in the regulation of bacterial adhesion in these target genes, while ccoO was relatively weak in the regulation of the adhesion. Meanwhile, the changes of temperature, salinity, pH, and starvation affected coxA, coxB, coxC, ccoN, ccoO, and ccoQ expression remarkably. These findings indicated that: the oxidative phosphorylation pathway is a critical regulator of adhesion in V. alginolyticus; coxA, coxB, coxC, ccoN, ccoO, and ccoQ regulate the bacterial adhesion in response to environmental changes such as temperature, salinity, pH, and starvation.

indicated that the oxidative phosphorylation could be affected by salinity. The pH could significantly change the activity of oxidative phosphorylation (Korzeniewski, 2015). As Monternier et al. (2015) described, the activity of oxidative phosphorylation was sensitive to starvation.
Vibrio alginolyticus, a moderately halophilic gram-negative bacillus, is considered to be an important opportunistic pathogens in mariculture Luo et al. (2016). Vibrio alginolyticus was associated with epidemics of maricultured animals, such as coral reefs (Xie et al., 2013), shellfish (Mechri, Monastiri, Medhioub, Medhioub, & Aouni, 2017), fish (Cao et al., 2018), and crustaceans (Gomez-Gil et al., 1998), and resulted in heavy economy losses. It has also been reported to cause disease in human beings during warm periods which may lead to otitis and wound infections (Campanelli, Sanchez-Politta, & Saurat, 2008). The adhesion of pathogen to the surfaces of host is one of the key steps in the initial stage of infection , and many pathogenic bacteria have been found to have strong adhesion ability to their hosts (Qin, Lin, Chen, Xu, & Yan, 2016;Lin et al., 2017), and the bacterial adhesion was regulated by different genes . Environmental conditions were proved to be capable of affecting the adhesion ability of bacteria . The temperature, salinity, pH, and starvation play the important roles of environmental conditions in the process of bacterium to adhere to fish mucus (Yan, Zhao, Wang, Zou, & Chen, 2010).
Although there are a few literatures about the molecular mechanism of V. alginolyticus adhesion in recent years, there are few literatures about the molecular mechanism of V. alginolyticus adhesion regulation under diverse environments. Furthermore, no study has been reported about the relationship between oxidative phosphorylation pathway and the adhesion of pathogenic bacteria. For further understanding of the V. alginolyticus adhesion regulation, the RNAseq was carried out on V. alginolyticus which received the treatments of low pH, Pb 2+ , and Cu 2+ .
In the present study, the bioinformatics analysis about the oxidative phosphorylation pathway was carried out, following by quantitative real-time PCR (qRT-PCR), RNA interference (RNAi), and in vitro adhesion assay. The purpose of this study is to reveal the relationship between bacterial adhesion and oxidative phosphorylation pathway under adversities.
For further validation of the RNA-seq, treatments with low pH (pH 5), Pb (100 mg/L), and Cu (50 mg/L) were carried out on V.
alginolyticus. The control group was cultured in LB broth (added with 2% NaCl, pH = 7) (Guo et al., 2018). There were six replicates for each treatment. After overnight culture at 28°C, the bacterial cell were collected and used for RNA extraction.
In order to explore the influence of temperature, V. alginolyticus was grown at 4, 15, 28, 37, and 44°C, respectively, in LB broth .
For evaluation of the effect of salinity, V. alginolyticus was grown in LB broth with salinities as 0.8%, 1.5%, 2.5%, 3.5%, and 4.5%, and then washed with PBS with corresponding salinities and adjusted to OD 560 ≈ 0.3.
For evaluation of the adhesion of starved cells, V. alginolyticus was adjusted to OD 560 ≈ 0.3 in normal PBS and kept starvation at 28°C for 1, 3, 5, and 7 days, respectively (Yi et al., 2008).

| Functional classification and enrichment analysis for differential expression genes
As our previously reported, the Blast2GO was applied to get Gene Ontology (GO) annotation of the unigenes, while WEGO was applied for GO functional classification analysis of all genes and understand the distribution of gene functions at the macro level. GO terms with Bonferroni corrected p value ≤ 0.05 were identified as significantly enriched in differential expression genes (DEGs) (Sun, Leo et al., 2018;Zhang et al., 2018).
Software of Blastall against KEGG database (http://www.genome.jp/kegg/) was used for the KEGG pathways annotation. Q value ≤ 0.05 was taken as a threshold, while pathways fulfilling this were regarded as significantly enriched pathways in DEGs (Huang, Liu et al., 2018).

| Preparation of mucus
Healthy marine cultured large yellow croakers from Ningde (Fujian, China) were used for mucus collection via our previously described procedure . The skin was first washed with sterile PBS (0.01 mol/L, pH 7.2). Then, the skin mucus was harvested by scraping off the surface of the skin. The mucus was then homogenized in PBS, removed particulate materials by centrifugation twice (20,000 g, 4°C, 30 min), filtered through 0.45-and 0.22μm pore size filters successively, and finally, adjusted in PBS to 1 mg protein/ml.

| Total RNA extraction and reverse transcription
Trizol (Invitrogen) was used to extract the total RNA from the bacterial cells following the manufacturer's protocol. A Rever Aid Mu-MLV cDNA synthesis kit (TransGen Biotech) was used to synthesize firststrand complementary (c) DNA from 2.0 mg total RNA following the manufacturer's protocol (Huang, Zuo, & Jiang, 2019 (Qin et al., 2014), heat shock was introduced to transform recombinant plasmids into E. coli SM10. The recombinant plasmids were then transferred from E. coli SM10 into V. alginolyticus by conjugation. The empty pACYC184 vector was taken as the control. Chloramphenicol (34 μg/ml) was used to screen stable silenced clones (Huang, Xu et al., 2018).

| qRT-PCR assay
A QuantStudio™ 6 Flex real-time PCR system (ABI) was used to carry out qRT-PCR with the SYBR green I fluorescent dye (Dongsheng) in accordance with the manufacturer's instructions (Liu et al., 2017).

| Determination of cytochrome C oxidase activity
The procedure for determining the cytochrome C oxidase (CoxC) activity of V. alginolyticus was performed according to the instructions of Sigma-Aldrich Cytochrome C Oxidase Activity Assay kit. Prior to starting the assay, the multifunctional microplate reader was incubated at room temperature and set the absorbance of 550 nm. Then, the multifunctional microplate reader was used to read data once every 10 s interval within 1 min. Finally, the activity of CoxC of each experimental group was calculated according to the formula in the instructions.

| In vitro adhesion assay
According to the procedure described by Kong et al. (2015), in vitro adhesion assay was carried out. A quantity of 50 μl of mucus was evenly spread on a glass slide (22 × 22 mm) and fixed by methanol for 20 min. Then, 1 ml of bacterial suspension (10 8 CFU/ml) was placed on the mucus-coated glass slides, following by incubating humid at 28°C for 2 hr, and washing five times with PBS. The adherent bacteria were then fixed for 30 min by 4% methanol, dyed for 3 min by crystal violet, and counted with a microscope (×1,000).
Each group was conducted for five times, and 20 fields of view were selected randomly. Negative control was performed using PBS instead of bacterial suspension.

| Data processing
Results were showed as mean ± SD and statistically analyzed with spss 18.0. Differences were determined by analysis of one-way ANOVA followed by Dunnett's multiple comparison tests. A value of p < 0.05 was considered statistically significant.

| DEGs screening by RNA-seq
Results of previous study  showed that the low pH, Pb 2+, and Cu 2+ could significantly reduce the adhesion ability of V. alginolyticus by 56.58%, 39.26%, and 37.41%, respectively. Therefore, RNA-seq was performed on V. alginolyticus which was treated with low pH, Pb 2+ and Cu 2+ .
Differential expression genes analysis yielded 1,791, 1,085, and 1,637 DEGs in the low pH-, Pb 2+ -and Cu 2+ -treated groups and compared with the control group. Then, GO and KEGG analysis were also performed according to the previously described by Huang, Liu et al. (2018) in this research, which showed that the oxidative phosphorylation pathway was worthy for further research to validate its role in adhesion regulation.

TA B L E 1 siRNA sequences
There were six commonly downregulated DEGs in the oxidative phosphorylation pathway, and these DEGs were significantly changed in the Cu 2+ , Pb 2+, and low pH groups (Figure1-3). Therefore, they appeared to be associated with the change of environmental factors. These six DEGs were subunit I of aa 3 -type cytochrome c oxidase gene (coxA), subunit II of aa 3 -type cytochrome c oxidase gene (coxB), subunit III of aa 3 -type cytochrome c oxidase gene (coxC), subunit I of cbb 3 -type cytochrome oxidase gene (ccoN), subunit II of cbb 3 -type cytochrome oxidase gene (ccoO), and subunit IV of cbb 3type cytochrome oxidase gene (ccoQ), respectively.
These reinforced the reliability of the sequencing data.

| Influence of transient gene silencing
After being treated with siRNAs, the gene expression levels of V.
alginolyticus were detected at 1, 3, 6, 9, and 12 hr. The expression levels of these target genes were normalized against the control (scrambled) siRNA treatments. The expression of these genes decreased significantly at 1-6 hr. After 6 hr, coxA and ccoQ were no longer changed significantly, while coxB, coxC, ccoN, and ccoO were remain significantly changed at 12 hr ( Figure 5a). The decreased expression of these genes indicated proper work of these siRNAs.
Our previously study results indicated that the siRNA was uptaken into V. alginolyticus at a relatively high efficiency (Huang, Hu et al., 2015). The V. alginolyticus adhesion ability with and without RNAi treatments were compared. In vitro adhesion assay showed that RNAi significantly reduced V. alginolyticus adhesion (Figure 5b), while the downregulation of adhesion was alleviated as time went by. After RNAi, the trend of qRT-PCR and in vitro adhesion assay results were quite similar. These results showed that RNAi of these genes could significantly impair the V. alginolyticus adhesion ability.

TA B L E 3 Primers used in quantitative real-time PCR
Primer Sequence 2.91-fold) at 1 hr. Simultaneously, the adhesion of V. alginolyticus was significantly reduced by 1.97-, 2.76-, 2.28-, 1.97-, 2.32-, and 3.02-fold, respectively. Thus, ccoQ appeared to be more important in the regulation of bacterial adhesion in these target genes, while ccoO gene was relatively weak in the regulation of the adhesion.

| qRT-PCR results of DEGs in the oxidative phosphorylation pathway after stable gene silencing
The results in Figure 6 showed that the expression levels of coxA, coxB, coxC, ccoN, ccoO, and ccoQ were decreased significantly in stable silenced strains by 1.52-, 2.10-, 3.00-, 1.85-, 2.35-, and 1.17-fold, respectively. The expression level of coxC gene was the most impaired among six genes in the oxidative phosphorylation pathway.

| Adhesion assay of V. alginolyticus after stable gene silencing
The adhesion ability of the control V. alginolyticus and stable silenced clones were shown in Figure 7.

| Enzyme activity of cytochrome C oxidase of Vibrio alginolyticus after stable gene silencing
The results in Figure 8 showed that the control V. alginolyticus dis-

F I G U R E 2
The overall scheme for genes involved in the oxidative phosphorylation pathway which were significantly affected by Pb stress. Green and red indicated down-and upregulation, respectively

| Influence of different salinities
The influences of salinities on gene expression were quite different ( Figure 11). The coxB, coxC, ccoN, and ccoO had the highest expression levels at the 0.80% salinity, while the expression levels of coxB, coxC, ccoN, and ccoO were significantly decreased from 1.50% salinity. The expression of coxA was not significantly changed at the 0.80%, 1.50%, and 2.50% salinity, while its expression was remarkably reduced at 3.50% and 4.50% salinity. The expression level of ccoQ displayed an U-shaped trend and reached the lowest expression level at the 2.50% salinity.
The maximum adhesion was achieved at the 0.80% salinity  and this results was consistent with previous studies . The trend of in vitro adhesion assays and qRT-PCR were similar at different salinities except for ccoQ gene.
These results indicated that salinity could influence V. alginolyticus adhesion while ccoQ gene probably be not involved in the regulatory network governing adhesion in response to different salinities.

| Effects of starvation
The results in Figure 12 showed that the expression levels of the six genes were significantly downregulated in a time-dependent manner after starvation. According to our previous research , the viable bacteria number was not significantly changed at first 3 days of starvation, and the adhesion ability of V. alginolyticus was substantially reduced with the time prolonging. The decline of the number of V. alginolyticus adhered to skin mucus was mainly owe F I G U R E 3 The overall scheme for genes involved in the oxidative phosphorylation pathway which were significantly affected by low pH stress. Green and red indicated down-and upregulation, respectively F I G U R E 4 Expression level of coxA, coxB, coxC, ccoN, ccoO, and ccoQ after stress treatments detected by quantitative real-time PCR. Data are presented as the means ± SD, n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05 to the decline of the adhesion ability of V. alginolyticus rather than the declining number of bacteria in the suspension. These results were quite similar to the previous reports (Wang & Leung, 2000), which proved that vibriosis induced by V. alginolyticus was more likely to outbreak in eutrophic seawater rather than oligotrophic seawater . These results probably explain the decreased adhesion ability of starved V. alginolyticus.
Although the expression levels of coxA, coxB, coxC, ccoN, ccoO, and ccoQ of starved V. alginolyticus were impaired at 3 days, the expression levels of coxB and ccoQ had no significantly reduced.

| D ISCUSS I ON
According to our previous reports, the adhesion of pathogens was a very complex process and was also a key step in the virulence F I G U R E 5 Transient RNAi reduced the adhesion of Vibrio alginolyticus. (a) Quantitative real-time PCR analysis of the expression of coxA, coxB, coxC, ccoN, ccoO, and ccoQ after transient gene silencing at 1, 3, 6, 9, and 12 hr. Data are presented as the means ± SD, n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05. (b) The adhesion capacity to mucus of transient silenced V. alginolyticus at 1, 3, 6, 9, and 12 hr. Data are presented as the means ± SD, n = 3. **p < 0.01 versus the control group. *p < 0.05 versus the control group.
of pathogens. In the previous researches of our laboratory, RNAseq was performed on V. alginolyticus which were treated with Cu 2+ , Pb 2+ , and low pH, and these stress conditions reduced the adhesion ability of V. alginolyticus significantly .
Based on the RNA-seq data and bioinformatics analysis, we found two pathways (including the TCA pathway and the flagellar assembly pathway) and three ncRNAs that might be related to adhesion of V. alginolyticus, which has been proved by qRT-PCR, RNAi and in vitro adhesion assays (Huang, Xu et al., 2018). In the present study, the relationship between the oxidative phosphorylation pathway and adhesion ability of V. alginolyticus was investigated.
The results of our RNA-seq analysis showed that there were six commonly downregulated DEGs in the oxidative phosphorylation pathway and the expression levels of these genes were significantly different in all stressed groups. Therefore, we hypothesized that these genes (including coxA, coxB, coxC, ccoN, ccoO, and ccoQ) might be sensitive to environmental stresses. Previous studies have shown that starvation (Monternier et al., 2015), pH (Fletcher, Feizi, Kim, Siewers, & Nielsen, 2015), temperature , and salinity are the important factors influencing distribution, abundance and diversity of aquatic animals . As Monternier et al. (2015) reported, starvation could decrease the activity of oxidative phosphorylation in skeletal muscle. In Pichia anomala cells, the oxidative phosphorylation were highly expressed when it grew in low pH (Fletcher et al., 2015). Seawater-acclimated milkfish upregulated oxidative phosphorylation to produce more energy budget when they were under hypothermal stress . Tine, McKenzie, Bonhomme, and Durand (2011) also described that many genes of oxidative phosphorylation were relatively overexpressed in gill tissue with high salinity. But these studies only showed that the oxidative phosphorylation is sensitive to environmental stress, and have not shown the expression levels of genes involved in the oxidative phosphorylation pathway under the environmental stress. Hence, our present study could complement the defects, while the previous studies also indirectly supported our hypothesis.
In order to examine the relationship between coxA, coxB, coxC, ccoN, ccoO, and ccoQ and adhesion, qRT-PCR, RNAi, bacterial adhesion assays, and CoxC activity assays were performed in this study.
After the transient gene silencing (Figure 5), the expression levels of these genes were significantly impaired and the adhesion ability of V.
alginolyticus was also significantly declined at the same time. These results indicated that coxA, coxB, coxC, ccoN, ccoO, and ccoQ were closely related to adhesion, which supported the results of RNAseq and our hypothesis. After the stable gene silencing, the results F I G U R E 6 Quantitative real-time PCR results of coxA, coxB, coxC, ccoN, ccoO, and ccoQ in the oxidative phosphorylation pathway after stable gene silencing. Data are presented as the means ± SD, n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05 F I G U R E 7 Stable RNAi reduced the adhesion of Vibrio alginolyticus. Data are presented as the means ± SD, n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05 F I G U R E 8 Stable gene silencing reduced the enzyme activity of cytochrome C oxidase of Vibrio alginolyticus. Data are presented as the means ± SD, n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05 in Figure 6 showed that coxC gene had the lowest expression level among the six genes and the expression level of coxA gene was higher compared with coxC. Meanwhile, the expression level of ccoQ gene was the highest among these six genes. But the results in Figure 7 and Figure 8 showed that the adhesion ability and the activity of CoxC in V. alginolyticus had the most impaired after stable silencing F I G U R E 9 Temperature influenced the Vibrio alginolyticus adhesion. Quantitative real-time PCR validation of the expression level of coxA, coxB, coxC, ccoN, ccoO, and ccoQ under different temperatures. Data are presented as the means ± SD, n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05 F I G U R E 1 0 pH influenced the Vibrio alginolyticus adhesion. Quantitative real-time PCR validation of the expression of coxA, coxB, coxC, ccoN, ccoO, and ccoQ under different pH. Data are presented as the means ± , n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05 F I G U R E 11 Salinity influenced the Vibrio alginolyticus adhesion. Quantitative real-time PCR analysis of the expression of coxA, coxB, coxC, ccoN, ccoO, and ccoQ under different salinities. Data are presented as the means ± SD, n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05 F I G U R E 1 2 Starvation influenced the adhesion of Vibrio alginolyticus. Quantitative real-time PCR analysis of the expression of coxA, coxB, coxC, ccoN, ccoO, and ccoQ after starvation for 1, 3, 5, and 7 days. Data are presented as the means ± SD, n = 6. Means of the treatments not sharing a common letter are significantly different at p < 0.05 of ccoQ gene, while coxA gene was on the heels. After stable silencing of coxC gene, the adhesion ability of V. alginolyticus and the activity of CoxC in V. alginolyticus had the least impaired. The previous reports by Iwata, Ostermeier, Ludwig, and Michel (1995) and Ekici, Pawlik, Lohmeyer, Koch, and Daldal (2012) supported our results.
As Iwata et al. described, the coxA is the biggest of the three subunits, and it is also a more important subunit. The CoxA is associated with CoxB at one side and CoxC at the other side. The coxA contains three redox centers: heme a, heme a 3 , and copper B (Cu B ) (Iwata et al., 1995). Heme a is a low-spin heme with two axial histidine ligands, while heme a 3 is a high-spin heme with one histidine ligand (Michel, Iwata, & Ostermeier, 1999). The fast electron transfer between heme a and the binuclear heme a 3− Cu B redox centers was recently used to explain the high operational oxygen affinity of the oxidase by kinetic trapping of bound oxygen (Ostermeier, Iwata, & Michel, 1996). The heme a 3− Cu B center is the catalytic core for oxygen reduction (Ostermeier et al., 1996). Obviously, the CoxA protein subunit plays an important role in electron translation and protoncoupled electron transfer .
The CoxB is the subunit II of COXs which containing the binuclear Cu A center to receive the electrons from cytochrome C and transfer them to heme a and finally to the binuclear heme a 3− Cu B center (Ostermeier et al., 1996).
As previously described, subunit III of CoxC contains no metal centers, so it has no direct function in electron transfer (Hosler, 2004). The pH and the proton affinity can influence the activity of CoxC. The CoxC has the ability to maintain proton flow to the active site, and it also exerts some structural influence that more directly protects the active site from destructive chemistry during the oxygen reduction cycle (Hosler, 2004). The CoxC connects with CoxA could contribute to optimum proton pumping efficiency (Hosler, 2004;Ostermeier et al., 1996).
Cytochrome C oxidase, a protein complex located in the inner membrane of mitochondria in eukaryotic and many prokaryotic aerobic organisms (Soulimane et al., 2000), is commonly abbreviated to COX (Ostermeier et al., 1996), CcO (Naruta et al., 2001), CytcO (Johansson et al., 2011), and complex IV (Barrientos, Gouget, Horn, Soto, & Fontanesi, 2009). The enzyme is the terminal enzyme of most respiratory chains (Iwata et al., 1995), which catalyzes the reduction of molecular oxygen to water concomitant with the oxidation of reduced cytochrome C and uses the free energy derived from this reaction to pump protons across a membrane (Johansson et al., 2011). Eukaryotic COX is consisted of 11-13 subunits (11 in the yeast Saccharomyces cerevisiae and 13 in mammals) (Barrientos et al., 2009). Prokaryotic COX usually have a less subunits, four subunits as usual (Iwata et al., 1995). However, subunits I (CoxA), II (CoxB), and III (CoxC) of CoxC are highly degree of sequence conserved from bacteria to humans (Hosler, 2004). These three subunits are essential for all of the aa 3 -type bacterial CoxCs (Hosler, 2004). Although there has been no reports on the relationship between coxA, coxB, coxC and bacterial adhesion or virulence, it has been reported that when the activity of mitochondrial CoxC is reduced and the cell adhesion ability also declined (Karu & Pyatibrat, 2011).
As Lohmeyer et al. (2012) described that the cbb 3 -Cox could replace the function of aa 3 -type cytochrome oxidases when the bacteria lack aa 3 -type cytochrome oxidases. The cbb 3 -type cytochrome oxidase (cbb 3 -Cox), without a Cu A center present, is a proton-pumping terminal oxidase solely in bacteria (Trasnea et al., 2018), and it is crucial for anoxyenic photosynthesis and nitrogen fixation. In addition, the cbb 3 -type cytochrome oxidase is important for many pathogenic bacteria to colonize under the low oxygen tissues. The enzyme has a high apparent affinity for oxygen, and it allows these pathogens to colonize low oxygen containing tissues (Ekici et al., 2012). Reducing molecular oxygen at low oxygen concentration, the enzyme is not based on a particularly fast binding of O 2 , but rather on the irreversibility of the O 2 binding to it (Rauhamäki, Verkhovsky, & Wikström, 2008). The previous studies have shown that the activity of cbb 3 -Cox is dependent on the environmental oxygen concentrations (Ekici et al., 2012). The activity is high under microaerobic, low under fully aerobic, and even lower under fully anaerobic growth condition (Ekici et al., 2012). The cbb 3 -Cox is consisted of CcoN (subunit I), CcoO (subunit II), CcoP (subunit III), and CcoQ (subunit IV) proteins (Ekici et al., 2012;Rauhamäki et al., 2008). CcoO is c-type cytochromes with a monoheme. It conveys electrons to heme b of CcoN and makes strong contacts with the alpha-helices of CcoN (Ekici et al., 2012). CcoP is also a C-type cytochromes with a diheme (Ekici, Jiang, Koch, & Daldal, 2013). It conveys electrons to the heme group of CcoO (Ekici et al., 2012). Meanwhile, the main catalytic subunit of cbb 3 -Cox is CcoN, which is a conserved 12 transmembrane helices containing membrane protein with a binuclear center heme b 3 -Cu (Ekici et al., 2013). CcoQ is a small subunit formed by a single transmembrane helix, and is not present in all cbb 3 -type cytochrome oxidase (Ekici et al., 2012). It does not contain any cofactor, and elimination does not completely abolish cbb 3 -type cytochrome oxidase activity (Ekici et al., 2012), but it is thought to enhance the stability of the cbb 3 -Cox (Ekici et al., 2013 coli, and Pseudomonas stutzeri) (Cukier, 2005;Galván et al., 2018;Guffanti, Bornstein, & Krulwich, 1981;Hempfling & Hertzberg, 1979;Kohlstaedt, Buschmann, Langer, Xie, & Michel, 2017;Kohlstaedt et al., 2016;Melin et al., 2018Melin et al., , 2016Schmetterer et al., 2001;Sun, Benlekbir et al., 2018;Sun, Luo et al., 2018). At the same time, the cbb 3 -Cox is only in bacteria and is the sole Cox in pathogenic bacteria. Up to now, some studies have validated the function of the cbb 3 -Cox in some pathogenic bacteria (such as Campylobacter jejuni, Helicobacter pylori, Neisseria meningitides, and Neisseria gonorrhoeae) (Ekici et al., 2012), whereas there is few studies about CoxC, cbb 3type cytochrome oxidase, and oxidative phosphorylation pathway in V. alginolyticus. Therefore, our present study not only investigated CoxC, cbb 3 -type cytochrome oxidase, and oxidative phosphorylation pathway in V. alginolyticus, but also verified the relationship between the oxidative phosphorylation pathway and V. alginolyticus adhesion at the molecular level and indicated how the environmental factors influenced V. alginolyticus adhesion through oxidative phosphorylation pathway.
In this study, with RNAi, qRT-PCR, and adhesion assays, we investigated the relationships among the six genes, adhesion ability and environmental stresses (including temperature, pH, salinity, and starvation). The results showed that the V. alginolyticus adhesion ability was substantially affected by environmental factors. The six genes in the oxidative phosphorylation pathway showed different responses to different environmental stresses. Whereas, the same environmental stress leads to different changes in expression of different genes. These results indicated that the oxidative phosphorylation pathway played an important role in the V. alginolyticus adhesion and was sensitive to some environmental stresses, which was consistent with previous reports .
In summary, the present study demonstrated that the oxidative phosphorylation pathway was sensitive to environmental factors, and it was related to the adherence of V. alginolyticus.

ACK N OWLED G EM ENT
This work was supported by grants from the National Natural

CO N FLI C T O F I NTE R E S T
The authors declared that they have no conflicts of interest to this work.

E TH I C S S TATEM ENT
All animal experiments were carried out strictly under the recommendations in the "Guide for the Care and Use of Laboratory Animals" set by the National Institutes of Health. The animal protocols were approved by the Animal Ethics Committee of Jimei University (Acceptance No. JMULAC201159).

DATA ACCE SS I B I LIT Y
The data of RNA-seq was deposited in the NCBI SRA (accession number: SRP049226).