Starry flounder, Platichthys stellatus, is a truly euryhaline teleost that can acclimate to various environmental salinities ranging from complete freshwater (FW) to seawater (SW). Teleosts undergo osmotic water loading in hypo-osmotic FW environments and excrete a large volume of diluted urine while drinking little ambient water to minimize the decrease in blood osmolality. In contrast, teleosts in SW actively drink water to counterbalance dehydration, and water is absorbed through ion-coupled fluid uptake in the intestine (Grosell, 2011). Excess NaCl is extruded via the gills in parallel with secretion of a low volume of iso-osmotic urine (Sundh et al., 2014). Therefore, the intestine plays a critical role maintaining the water absorption necessary in hyper-osmotic SW.

The Na⁺-K⁺-2Cl⁻ cotransporter (NKCC) is a member of the cation-chloride cotransporter family that mediates electroneutral movement of Na⁺ and K⁺ and is tightly coupled to movement of Cl⁻ across cell membranes. NKCC occurs as a secretory isoform (NKCC1) and an absorptive isoform (NKCC2) (Payne and Forbush, 1994; Xu et al., 1994). In mammals, NKCC2 is expressed mainly in the apical membrane of epithelial cells in the thick ascending loop of Henle where it acts to reabsorb a large amount of NaCl to dilute urine forming in the tubule lumen (Nielsen et al., 1998; Markadieu and Delpire, 2014). NKCC2 dysfunction exerts negative consequences in mammals. In humans, inactivating mutations in the gene coding NKCC2 cause Bartter syndrome type I, which features severe volume depletion, hypokalemia, metabolic alkalosis, and hypercalciuria (Markadieu and Delpire, 2014). Mammalian NKCC2 is the main pharmacological target of loop diuretic drugs used to treat edema (Markadieu and Delpire, 2014).

Piscine NKCC has been suggested to be involved in intestinal salt absorption in several fish species, including European eel, Anguilla anguilla, and Mozambique tilapia, Oreochromis mossambicus (Musch et al., 1982; Trischitta et al., 1992; Li et al., 2014). In addition, a duplicated pair of NKCC2 isoforms (NKCC2α and NKCC2β) has been identified in several teleosts, and they are expressed in different tissues and organs (Cutler and Cramb, 2008; Hiroi et al., 2008; Watanabe et al., 2011). For example, NKCC2α mRNA is expressed only in the renal tissue of European eel, whereas NKCC2β expression is detected in the intestine and urinary bladder. NKCC2β is predominantly expressed in the intestine in SW-transferred Japanese eel, A. japonica, and Mozambique tilapia. Furthermore, the effects of acclimation to different salinities on tissue distribution and expression levels of NKCC2 have focused on euryhaline fish species.

Starry flounder are a candidate aquaculture species to substitute for olive flounder, Paralichthys olivaceus, the most popular marine fish species in Korean aquaculture (Noh et al., 2013). Efforts to increase farming practice efficiency of starry flounder have been emphasized, and exploitation of genetic components related to host defense mechanisms has become important, particularly regarding a deeper understanding of the immune-relevant physiology at the cellular and organismal levels.

Streptococcus parauberis is a causative agent of streptococcosis in starry flounder with symptoms of exophthalmia, eye and jaw hemorrhage, and abdominal distension, as reported in olive founder (Kang et al., 2007; Cho et al., 2008). S. parauberis may originate from a local infection, such as in the intestine (Ferguson et al., 1994).

Despite its non-hematopoietic derivation, the intestinal tract functions as a critical defense barrier to the external environment, in addition to being the site of nutrient digestion and osmoregulation (Pastorelli et al., 2013). Ion transport disturbances in mammals have been associated with cellular dysfunction, intra and extracellular edema, and abnormalities in epithelial surface liquid volume (Eisenhut, 2006). However, the functions of teleost NKCC2 genes during immune or bacterial challenge have not been investigated.

Despite the importance of NKCC2 in intestinal function, the molecular mechanism by which this cotransporter is regulated in health and disease is poorly understood. The objective of this study was to characterize the genetic determinants of the NKCC2 isoform from starry flounder and to scrutinize its expression pattern in response to a S. parauberis challenge.

Apparently healthy starry flounder (weight, 12.4 g) were obtained from a local Korean fish farm. Fish were reared in 600-L tanks with flow-through seawater maintained at 20℃. The fish were fed a commercial diet once daily before the onset of the experiment. Fish were anesthetized with 0.2% 2-phenoxyethanol for sampling, and body weights were measured. The intestine was split into the anterior intestine, posterior intestine (also referred to as the middle intestine), and rectum based on external appearance (Kim et al., 2008).

Total RNA was extracted from the intestinal segments using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). After a 15-min DNase I (Invitrogen) treatment, total RNA (1 μg) was reverse transcribed using the SMART^TM RACE cDNA amplification kit (Clontech, Palo Alto, CA, USA) according to the manufacturer’s instructions. The degenerate primers to obtain the cDNA fragment encoding NKCC2 were designed based on available information in other vertebrate species. Reverse transcription-polymerase chain reaction (RT-PCR) analysis was carried out using a degenerate primer set (see Table 1 for primers). The resulting PCR products were electrophoresed on a 2% agarose gel and subcloned into the pGEM^®-T vector (Promega, Woods, WI, USA) for the sequencing analysis. After determining the partial cDNA sequences, the SMART^TM RACE cDNA amplification kit (Clontech) was used to isolate the full-length cDNA according to the manufacturer’s instructions. The 5’-rapid amplification of cDNA ends (RACE) and 3’-RACE products were produced in primary/nested PCR reactions using specific psNKCC2-1F/2F and psNKCC2-1R/2R primer sets for the NKCC2 gene with an adaptor primer in the kit. RACE products were TA-cloned and sequenced as described above. The nucleotide sequences were determined by analyzing more than five clones for the NKCC2 gene to obtain a representative cDNA sequence.

[Table 1.] Oligonucleotide primers used in this study

Oligonucleotide primers used in this study

The open reading frame (ORF) analysis of the NKCC2 cDNA sequence was performed using the NCBI web service ORF finder (http://www.ncbi.nlm.nih.gov/projects/gorf/). The molecular weight and theoretical isoelectric point (pI) values of the deduced amino acid sequence were analyzed using the ExPASy ProtParam tool (http://web.expasy.org/protparam/). To determine the degree of homology of the starry flounder NKCC2 isoform with other vertebrate orthologs, we retrieved the NKCC2 gene cDNA sequences of other species from BLAST and/or Ensembl genome database (http://asia.ensembl.org/index.html) searches. Multiple sequence alignments were generated by CLUSTAL W (Thompson et al., 1994). The phylogenic tree was constructed using Molecular Evolutionary Genetics Analysis (MEGA) program (ver. 5.2) with the neighbor-joining method. GenBank accession numbers or Ensembl codes for the NKCC1/NKCC2 sequences employed for the multiple sequence alignments and the phylogenic tree are shown in Table 2. Reliability of the internal tree branches was assessed by 1,000 bootstrap replications. The topology of the NKCC2 deduced amino acid sequences was predicted using the TMHMMfix software (http://www.sbc.su.se/~melen/TMHMMfix/).

[Table 2.] Protein sequence identities of starry flounder NKCC2 with other orthologs

Protein sequence identities of starry flounder NKCC2 with other orthologs

The brain, esophagus, gill, intestine (anterior and posterior intestines and rectum), head kidney, liver, spleen, and stomach were removed from starry flounder under non-stimulated conditions, and total RNA was extracted from the tissues as described above. After a 15-min DNase I treatment, an aliquot of total RNA (2 μg) was reverse-transcribed using the Superscript II First-strand Synthesis System for RT-PCR (Invitrogen). End-point RT-PCR was performed to determine basal expression of the NKCC2 transcripts. The primer pairs for NKCC2 and 18S rRNA were psNKCC2-3F/3R (amplicon = 152 bp) and 18S rRNA-F/R (amplicon = 313 bp), respectively.

Fish were randomly assigned to two groups, and transferred into one of two test tanks (200 l). Starry flounder were given an intraperitoneal injection of 0.1 ml S. parauberis resuspended in phosphate-buffered saline (PBS, pH 7.4) (Kim et al., 2013). The final dose was 1 × 10³ colony forming units/fish. An unchallenged control was injected with 0.1 ml sterile PBS. After the injection, each group was transferred to one of two 200 l tanks maintained at 20℃. Mortality observed beginning 3 days post injection was about 20% by the end of experiment in the bacterial-challenged fish. The cause of mortality was confirmed by re-isolating S. parauberis from the kidney of dead fish. Three fish from each tank were randomly sacrificed 1, 3, and 5 days post-injection to quantify the NKCC2 transcripts. Fish were anesthetized with 0.2% 2-phenoxyethanol for sampling, and decapitated before removing tissues.

NKCC2 expression levels were assessed by real-time quantitative PCR to examine the potential modulation of NKCC2 gene expression by the bacterial challenge. Preparation Total RNA (from anterior and posterior intestines and rectum) and the cDNA synthesis were performed as described above. The diluted cDNA template was subjected to PCR cycling with a LightCycler^® 480 (Roche Diagnostics, Basel, Switzerland). The 205-bp NKCC2 gene fragment was quantified with the specific q-psNKCC2-F and q-psNKCC2-R primer pair (Table 1). Plasmid DNAs containing the amplified parts of the target mRNAs were prepared as standard samples. The PCR mixture consisted of 0.5 μM of each primer pair and the LightCycler^® 480 SYBR Green I Master (Roche Diagnostics) in a final volume of 20 μl. The transcript copy number was calculated in reference to the parallel amplifications of known concentrations of the respective cloned PCR fragments. The NKCC2 transcript level in each sample was normalized against its own level of the 18S rRNA control (Kubista et al., 2006; Schmittgen and Livak, 2008). Triplicate independent assays per cDNA sample were performed.

Differences among samples were assessed by one-way analysis of variance, followed by Duncan’s multiple range test. All statistical analyses were performed using the SPSS ver. 10.0 software (SPSS, Inc., Chicago, IL, USA), and differences were considered significant at P < 0.05.

The starry flounder NKCC2 cDNA (GenBank accession no. AB645956) was 4,137 bp in length and contained a single ORF of 3,129 bp encoding a polypeptide of 1,043 amino acids (aa). The NKCC2 5′-untranslated region (UTR) and 3′-UTR were 174 and 834 bp, respectively. The calculated molecular mass of the mature peptide was 114.6 kDa with a theoretical pI of 6.47. Multiple sequence alignments of starry flounder NKCC2 with other orthologs revealed that it shared comparably higher identities with piscine NKCC2s at the amino acid level (range, 59.5–88.4%; Table 2). According to the topology prediction, starry flounder NKCC2 contained 12 putative transmembrane domains and a long cytoplasmic C-terminal tail, all of which are characteristic of NKCC. There were four amino acid residues within transmembrane domain 4 that were conserved among NKCC family members (N248, R251, G254, and G268) (Fig. 1).

[Fig. 1.] Alignments of the predicted starry flounder Platichthys stellate NKCC2 amino acid sequence (in boldface) with their corresponding orthologues. The conserved amino acid residues around transmembrane 4 are indicated by arrowheads (N248, R251, G254 and G268). Asterisks and hyphens indicate identical residues and gaps introduced for alignment, respectively. Twelve putative transmembrane (TM) spanning domains are shaded.

A phylogenic tree analysis was performed for the NKCC1 and/or NKCC2 sequences from starry flounder and other fish species and mammals using the neighbor-joining method (Fig. 2). The relationships revealed in the phylogenic tree agreed with known taxonomic appraisals. The starry flounder NKCC2 was located in the same branch as that of olive flounder and was closely related to that of tilapia and Japanese rice fish, Oryzias latipes.

[Fig. 2.] Phylogenic tree of vertebrate NKCC1 and NKCC2 isoforms. The starry flounder sequence cloned in the present study is indicated by thick black arrow. The neighbor-joining method with 1,000 bootstrap replications was used to generate the tree. The length of each branch is proportional to the divergence of the protein sequence from other members of the family. Bootstrap values are shown at each node when the confidence level is above 50%.

Unstimulated starry flounder juveniles were used to investigate NKCC2 transcript levels in different tissues (Fig. 3). The NKCC2 transcript was detected at limited levels in the anterior and posterior intestines and rectum. NKCC2 basal expression levels in all intestinal segments were similar. NKCC2 was not found in the other tissues examined.

[Fig. 3.] Tissue distribution of NKCC2 mRNA in the starry flounder under normal physiological condition. Total RNAs from brain (Br), esophagus (Es), gill (Gi), kidney (Ki), liver (Li), anterior (Ai) and posterior intestines (Pi), rectum (Re), spleen (Sp) and stomach (St) were reverse-transcribed and amplified by PCR. The cDNA of 18S rRNA was also amplified as a normalization control.

The quantitative RT-PCR assay results revealed that the anterior intestinal NKCC2 transcript in the bacterial-challenged group increased significantly only 1 day post-challenge with S. parauberis (Fig. 4A). The rectal NKCC2 transcript expression level on day 3 in the challenged fish was significantly upregulated about threefold relative to that of the unchallenged group, but decreased significantly on day 5 (Fig. 4C). However, unlike anterior intestinal and rectal NKCC2 expression, posterior intestinal mRNA expression was not modulated by the S. parauberis challenge (Fig. 4B). In addition, NKCC2 mRNA levels were not different among the intestinal segments on 5 day post-challenge, and no differences were detected between the control and experimental groups.

[Fig. 4.] Relative expression levels of the NKCC2 transcripts in the (A) anterior (Ai), (B) posterior (Pi) intestines and (C) rectum (Re) of starry flounder with S. parauberis by real-time PCR method. The experiment was performed in triplicate, and different letters indicate significant differences at P < 0.05.

In the present study, we focused on the association between intestinal ion channel function and streptococcosis. We identified the cDNA encoding NKCC2 in the intestinal tract of starry flounder based on the deduced amino acid sequence and the comparably higher identities with other teleost fish NKCC2s. The starry flounder NKCC2 is a 1,043-aa protein with a proposed topology that features 12 putative transmembrane-spanning domains and predicted cytosolic amino and carboxyl termini. This structural prediction closely resembles those of other known NKCC2 proteins. In addition, the starry flounder NKCC2 contained highly conserved amino acid residues within transmembrane domain 4, which are important for the NKCC cotransporting function (Moreno et al., 2004; Gamba, 2005). These findings suggest that the starry flounder NKCC2 is a functional cation-Cl⁻ cotransporter involved in ion transport in the intestinal tract.

Starry flounder NKCC2 transcripts were found only in the three intestinal segments (e.g., anterior and posterior intestines and rectum) with moderate expression levels under non-stimulated conditions. Other teleosts NKCC2s are expressed in several tissues, although distribution is not widespread. For example, NKCC2 is preferentially expressed in the intestine and kidney of the marine medaka, O. dancena (Kang et al., 2010), whereas NKCC2 is mainly expressed in the gastrointestinal tract and skeletal muscle and is abundant in the intestine, followed by the gill, esophagus, and stomach of SW-transferred Japanese eel and tilapia (Watanabe et al., 2011; Li et al., 2014). NKCC2 expression in mammals is restricted to the kidney, stomach, and intestine (e.g., jejunum and distal colon) at similar levels (Xue et al., 2009). In contrast, NKCC2s show somewhat different expression patterns in European and Japanese eels, in which NKCC2β expression levels are markedly higher in the anterior and posterior intestines, followed by the rectum (Cutler and Cramb, 2008; Watanabe et al., 2011). The highest expression level in olive flounder is in the posterior intestine (Kim et al., 2013). According to Gregório et al. (2013), it is likely that the intestinal distribution is linked to a specific molecular mechanism often referred to as solutelinked water transport, which has a similar anterior-posterior distribution pattern, but species-specific differences exist. The relative importance of the rectum and distal intestine (not in rectum) in osmoregulation has been demonstrated in marine fish, such as sea bream, Sparus aurata, and Gulf toadfish, Opsanus beta (Sattin et al., 2010; Guffey et al., 2011; Gregório et al., 2013). Thus, although we have not clarified the mechanism behind our findings, our data suggest that the rectum in this species is responsible for final control of intestinal osmoregulatory function.

In this study, only the rectal level of NKCC2 in starry flounder responded significantly to the bacterial challenge, indicating that excess salt may be transported into the rectum. This similar induction pattern was observed in olive flounder, in which the NKCC2 transcript is upregulated in the posterior intestine after S. parauberis injection (Kim et al., 2013). NaCl absorption by intestinal epithelial cells is an efficient finely tuned system with Na⁺/K⁺-ATPase, NKCC, and Na⁺/Cl^- (NCC) cotransporters as follows: electrochemical Na⁺/K⁺-ATPase gradients located in the basolateral membrane of intestinal epithelial cells drive electroneutral ion uptake into cells across the apical membrane through NKCC and/or NCC (Marshall and Grosell, 2006). The predominant increase in NKCC2 expression in the rectum would tip the aforementioned intestinal fluid balance in favor of fluid accumulation, probably indicating that a bacterial pathogen can interfere with osmotic balance in the intestine as well as normal mucosal immune homeostasis. Targeted disruption of this gene in mammals produces salt-wasting nephropathy with high mortality during the first 2 weeks of life (Gamba and Friedman, 2009). In the present study, mortality in the S. parauberis-challenged group was observed along with increased rectal NKCC2 expression (e.g., increase on 3 day), suggesting that the increased NKCC2 level; e.g., disruption of intestinal ion homeostasis, could be directly or indirectly related to resistance against the bacterial injection either by acting alone or in association with other ion-uptake-regulating factors. Further studies, including the epithelial components and experiments over a broader time scale, are required to elucidate the strong relationship between bacterial stimulation and intestinal ion uptake. Our data will facilitate development of gene expression assays for starry flounder ion-homeostasis genes to better understand the relationships between ion transport genes and inflammation or infection.