The Lamellae-Free-Type Pseudobranch of the Euryhaline Milkfish (Chanos

Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology, Part A

journal homepage: www.elsevier.com/locate/cbpa

The lamellae-free-type pseudobranch of the euryhaline milkﬁsh (Chanos chanos)isaNa+,K+-ATPase-abundant organ involved in hypoosmoregulation

Sheng-Hui Yang a, Chao-Kai Kang a,Hsiu-NiKungb,Tsung-HanLeea,c,⁎ a Department of Life Sciences, National Chung Hsing University, Taichung 402, Taiwan b Department of Anatomy and Cell Biology, College of Medicine, National Taiwan University, Taipei 100, Taiwan c Department of Biological Science and Technology, China Medical University, Taichung 404, Taiwan article info abstract

Article history: In teleosts, the pseudobranch is hemibranchial, with a gill-like structure located near the first gill. We hypothe- Received 30 September 2013 sized that the pseudobranch of the milkfish might exhibit osmoregulatory ability similar to that of the gills. In this Received in revised form 10 December 2013 study, the obtained Na+,K+-ATPase (NKA) activity and protein abundance profiles showed that these parame- Accepted 17 December 2013 ters were higher in the pseudobranchs of the seawater (SW)- than the freshwater (FW)-acclimated milkfish, op- Available online 31 December 2013 posite the situation in the gills. The pseudobranch of the milkfish contained two types of NKA-immunoreactive cells, chloride cells (CCs) and pseudobranch-type cells (PSCs). To further clarify the roles of CCs and PSCs in Keywords: + + − Pseudobranch the pseudobranch, we investigated the distributions of two ion transporters: the Na ,K , 2Cl cotransporter Osmoregulation (NKCC) and the cystic fibrosis transmembrane conductance regulator (CFTR). NKCC on the basolateral mem- Na+ K+-ATPase brane and CFTR on the apical membrane were found only in pseudobranchial CCs of SW-acclimated individuals. Na+ K+,2Cl− cotransporter Taken together, the results distinguished NKA-IR CCs and PSCs in the pseudobranch of milkfish using antibodies Cystic fibrosis transmembrane conductance against NKCC and CFTR as markers. In addition, increases in the numbers and sizes of CCs as well as in NKA ex- regulator pression observed upon salinity challenge indicated the potential roles of pseudobranchs in hypo- fi Milk sh osmoregulation in this euryhaline teleost. © 2014 Elsevier Inc. All rights reserved.

1. Introduction free of lamellae along the opercular edge (i.e., close to the afferent artery) are observed (Mattey et al., 1980). In the teleosts the pseudobranch exhibits a gill-like structure on ei- The epithelium of the pseudobranch contains pavement cells, chlo- ther side of the opercular epithelium near the first gill arch. This struc- ride cells (CCs; i.e., ionocytes), and the unique pseudobranch-type ture displays only one row of parallel filaments, similar to a modified cells (PSCs). In the covered-type pseudobranch of trout and grass carp, hemibranch (Laurent and Dunel-Erb, 1984). Due to variations in its the CCs have been replaced by PSCs (Hamidian and Alboghobeish, size, form and location as well as its presence or absence in different 2007). The PSCs resemble branchial ionocytes, with a high density of fish species, the pseudobranch has been an object of investigation for mitochondria and tubular system in bass (Dicentrachus labrax); mullet a long time. Bertin (1958) described three morphological types of the (Liza ramada); rainbow trout (Salmo gairdneri), smelt (Osmerus pseudobranchs in teleosts: (i) the lamellae-free type: pseudobranchs esperlangus) and golden orfe (Leuciscus idus)(Mattey et al., 1978; with distinguishable filaments and completely free lamellae that con- 1980; Laurent and Dunel-Erb, 1984; Fischer-Scherl and Hoffmann, tact the water, e.g., in bass (Dicentrachus labrax); (ii) the lamellae 1986). In addition, PSCs exhibit relatively high Na+,K+-ATPase (NKA) semi-free and the covered type: pseudobranchs covered by the opercu- activity in the pinfish (Lagodon rhomboides)andChinooksalmon(Onco- lar membrane and connective tissue, e.g., in grey mullet (Liza ramada, rhynchus tshawytscha), similar to branchial ionocytes (Dendy et al., semi-free) and rainbow trout (Salmo gairdneri, covered); and (iii) em- 1973; Quinn et al., 2003). In contrast, ultrastructural differences be- bedded type: pseudobranchs that are completely reduced and embed- tween the pseudobranchial PSCs and CCs in fresh water (FW) and sea- ded in the connective tissues, e.g., in carp (Cyprinus carpio)(Mattey water (SW) teleosts have been described and proved that they were et al., 1980; Hamidian and Alboghobeish, 2007). In the semi-free two different types of cells (Mattey et al., 1978; Laurent and Dunel- pseudobranch of the mullet, some fusion areas on the buccal edge, Erb, 1984; Fischer-Scherl and Hoffmann, 1986). The physiological relevance of the teleost pseudobranch as a rem- nant of a reduced gill arch is not yet clear. Numerous hypotheses have ⁎ Corresponding author at: Department of Life Sciences, National Chung Hsing, been put forth regarding the physiological roles of the pseudobranch, University, 250, Kuo-Kuang Road, Taichung 402, Taiwan. Tel.: +886 4 2285 6141; fax: fi +886 4 2287 4740. butdirectcon rmatory evidence is lacking (Bridges et al., 1998; E-mail address: [email protected] (T.-H. Lee). Mölich et al., 2009). Among various proposed functions of the

1095-6433/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cbpa.2013.12.018 16 S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25 pseudobranch, evidence of a potential osmoregulatory role has been (FW) ([Na+]0.22mM;[K+] 0.04 mM; [Ca2+] 0.68 mM; [Mg2+] reported (Epstein et al., 1967). In contrast, the pseudobranchial NKA 0.28 mM; [Cl−] 0.14 mM; pH = 7.6 ± 0.1) or SW ([Na+] 482.97 mM; activity of juvenile Chinook salmon did not increase 10 days after trans- [K+] 11.38 mM; [Ca2+] 15.34 mM; [Mg2+] 67.87 mM; [Cl−] fer from FW to SW. It was therefore concluded that the pseudobranch 572.89 mM; pH = 7.6 ± 0.1), at 28 ± 1 °C with a daily 12 h photope- differs from the gill in terms of the regulation of NKA activity (Quinn riod for at least four weeks before the experiments. Feeding was termi- et al., 2003). nated 24 h prior to the experiments. The water was continuously Maintaining a stable internal environment is important for verte- circulated through fabric-floss filters, and the fish were fed twice a day brates to survive in a variety of habitats. In response to changes in envi- with the Milkfish Feed (Uni-President Aquatic Products). The experi- ronmental conditions, the ion-transporting epithelia modulate ion mental fish were reviewed and approved by the Institutional Animal fluxes. The strategy for successful adaptation from hypertonic to hypo- Care and Use Committee (IACUC) of the National Chung Hsing Universi- tonic environments is to switch from active hypoosmoregulation (ion ty (IACUC Approval No. 97–79) to T.–H. L. excretion) to active hyperosmoregulation (ion uptake) to counter the passive diffusion fluxes resulting from concentration gradients 2.2. Tissue collection (Lin et al., 2006; Tang et al., 2010; Tang et al., 2011). In contrast, during adaptation from hypotonic to hypertonic environments, Fish were treated with 100–200 mg/L concentration of neutralized the hypoosmoregulatory ability of fish is increased to maintain MS-222 (Sigma–Aldrich, St. Louis, MO, USA) within two minutes to be ionic homeostasis (Franklin et al., 1992; Lee et al., 1996; Uchida et al., anesthetized over crushed ice. Following anesthesia, the pseudobranch 1996; Kang et al., 2010). The gills, which are directly exposed to the ex- and operculum of the fish were sampled immediately after killing by ternal medium, are the major osmoregulatory organs of teleosts. The decapitation. In addition, the first gill arch was excised. To obtain paraf- main ion transport system found in epithelial ionocytes is characterized fin and cryo-sections, the sampled pseudobranchs and gills were fixed by the presence of a high abundance of mitochondria and an extensive in 10% neutral buffered formalin at 4 °C overnight. tubular system in the cytoplasm. This tubular system is continuous with the basolateral membrane, providing a large area for expression of the 2.3. Histological examination of the paraffin sections primary ion transport protein NKA, which is the driving force for secondary ion transporters to absorb ions from FW and secrete ions to The fixed tissues were dehydrated through a graded ethanol series SW (Marshall, 2002; Evans et al., 2005; Hwang and Lee, 2007; Hwang (50, 70, 80, 95, 100% ethanol) and cleared in xylene for 3 h. Then, the et al., 2011). According to previous studies (Evans et al., 2005; Hwang samples were infiltrated and embedded in paraffin (Merck, Darmstadt, and Lee, 2007), the model of chloride (Cl−) secretion in gill ionocytes Germany). Serial sections of the paraffin blocks with a thickness of of SW teleosts has been depicted as involving a basolateral Na+,K+, 2–4 μm were mounted on 0.03% poly-L-lysine (Sigma)-coated glass 2Cl− cotransporter (NKCC) and an apical chloride channel, the cystic fi- slides and stained with Gill's hematoxylin and eosin (HE) reagent brosis transmembrane conductance regulator (CFTR). (Merck), followed by mounting with cover slips. The sections were ob- The milkfish, Chanos chanos, which is the sole species belonging to served under an optical microscope (BX50; Olympus, Tokyo, Japan), and the family Chanidae of the order Gonorynchiformes, exhibits adaptation micrographs were obtained using a cooled CCD camera (DP72; Olym- to a remarkably wide range of salinities, from 0 to 158‰, but is relatively pus) with CellSens, standard version 1.4 software (Olympus). stenothermal. This euryhalinity makes the milkfish an excellent subject for studies on physiological and behavioral processes associated with 2.4. Cryosectioning salinity adaptation (Ferraris et al., 1988; Swanson, 1998). Previous studies have reported that the milkfish is an excellent osmoregulator that Fixed samples were washed three times with phosphate-buffered generally maintains its muscle water contents and plasma osmolality saline (PBS; 137.00 mM NaCl, 2.68 mM KCl, 10.14 mM Na2HPO4, within a narrow physiological range upon salinity challenge (Tang 1.76 mM KH2PO4, pH = 7.4), then stored in 100% methanol at − et al., 2009, 2010). In addition, NKA expression (i.e., the distribution, 20 °C. Prior to cryosectioning, the pseudobranchs were gradually im- abundance of mRNA and protein, and activity of NKA) has been studied mersed in PBS. The tissue was then mounted in O.C.T. (optimal cutting in the gills of milkfish acclimated to environments with different sa- temperature) compound (Tissue-Tek, Sakura, Torrance, CA, USA) at − linities. A marked enhancement of NKA activity and protein abun- 20 °C for 2 h. Longitudinal and cross-sections of the pseudobranchs dance, derived from a significant increase in the number of lamellar were cut at a thickness of 4–6 μm using a cryomicrotome (CM3050S, NKA immunoreactive (NKA-IR) cells, has been observed in the gills Leica, Heidelberger, Nussloch, Germany) at −25 °C. The cryosections of FW-adapted milkfish compared to SW-adapted individuals were finally placed on 0.03% poly-L-lysine-coated slides and kept in (Chen et al., 2004; Lin et al., 2003, 2006; Tang et al., 2010, 2011). slide boxes at −20 °C until immunofluorescence staining. The pseudobranch of the milkfish exhibits gill-like filaments, and the lamellae located near the first gill arch are assumed to play potential 2.5. Antibodies roles in osmoregulation. Therefore, this study aimed to investigate the morphological type of the milkfish pseudobranch and figure out the The following primary antibodies were used in the present study: correlation between its structure and osmoregulatory ability in SW- (1) anti-NKA α subunit: a mouse monoclonal antibody, α5 (Develop- and FW-acclimated milkfish. mental Studies Hybridoma Bank, Iowa City, IA, USA), raised against the α-subunit of avian NKA; (2) anti-NKCC: a mouse monoclonal anti- 2. Materials and methods body, T4 (Developmental Studies Hybridoma Bank), raised against the C-terminus of human NKCC; (3) anti-NKA α subunit: a rabbit monoclo- 2.1. Experimental conditions and fish nal antibody, ab76020 (Abcam, Cambridge, UK), a synthetic peptide corresponding to residues near the N-terminus of human Sodium Potas- Juvenile milkfish (Chanos chanos) with a body mass of 15 ± 0.9 g sium ATPase α subunit; (4) anti-CFTR: a mouse monoclonal antibody, and total length of 10 ± 0.2 cm were obtained from a fish farm in cen- MAB25031, 24–1 (R & D System, Boston, MA, USA) directed against tral Taiwan. Seawater (SW; 35‰) was prepared from local tap water 104 amino acids at the C-terminus of the human CFTR; and (5) anti- with appropriate amounts of synthetic sea salt (Aquarium Systems, actin: a rabbit polyclonal antibody, sc1616 (Santa Cruz, CA, USA), raised Mentor, OH, USA). Milkfish were reared first in brackish water (15‰) against the C-terminus of human actin, for immunoblotting as a loading ([Na+] 156.11 mM; [K+] 5.72 mM; [Ca2+] 9.29 mM; [Mg2+] control. The secondary antibodies employed in the immunoblotting as- 30.34 mM; [Cl−] 270.60 mM; pH = 7.6 ± 0.1), then in fresh water says were horseradish peroxidase (HRP)-conjugated goat anti-mouse S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25 17

IgG and anti-rabbit IgG (111-035-003 and 115-035-003, respectively, (Fig. S1). The mean numbers and sizes of cells were calculated from six Jackson ImmunoResearch, West Grove, PA, USA). For immunofluo- individuals (n = 6) of the SW or FW group. rescence staining, the secondary antibodies used were DyLight- 549-conjugated affinity purified goat anti-rabbit IgG and DyLight- 2.9. Preparation of tissue homogenates 488-conjugated affinity purified goat anti-mouse IgG (Jackson ImmunoResearch). Pseudobranchs and gills were excised and immediately stored at −80 °C prior to use. For use in experiments, tissues were thawed and 2.6. Immunohistochemical staining (IHC) mixed with 250 μL of SEID buffer (150 mM sucrose, 10 mM EDTA, 50 mM imidazole and 0.1% sodium deoxycholate, pH 7.5) containing The paraffin sections of pseudobranchs and gills were deparaffinized protease inhibitors (v/v 25:1, Roche, Mannheim, Germany). Homogeni- with xylene, rehydrated in a graded series of alcohol, and incubated zation was performed in a 2 mL tube applying a homogenizer with 3% H2O2 for 10 min to inactivate endogenous peroxidase. Then, (PT1200E, Kinematica AG, Lucerne, Switzerland) at maximal speed for the sections were immunohistochemically stained for 2 h at room 20–30 s on ice. Following homogenization, the homogenate was centri- temperature with a monoclonal primary antibody (α5), followed by fuged at 5500 g for 15 min at 4 °C. Protein concentrations were deter- staining with a commercial HRP polymer conjugate kit (AEC, mined with BCA protein assay reagents (Thermo, Rockford, IL, USA) SuperPictureTM, Invitrogen, Camarillo, CA, USA) for visualization. A using BSA (Thermo) as a standard. 1:250 dilution was used for the detection of NKA. Negative control experiments in which PBS replaced the primary antibody were also con- 2.10. Immunoblotting ducted to confirm the positive results. Immunoblotting was performed as described by Tang et al. (2010), 2.7. Double immunofluorescence staining and confocal microscopy with some modifications. Aliquots containing 30 μg of protein from pseudobranch homogenates were heated at 65 °C for 25 min and frac- For double immunofluorescence staining, cryosections were rinsed tionated via electrophoresis in SDS-containing 7.5% polyacrylamide with PBS 3 times for 3 min and then incubated in blocking buffer, gels, together with a pre-stained protein marker (Fermentas, Hanover, consisting of 5% bovine serum albumin (BSA; Bio basic, Markham, MD, USA). The separated proteins were transferred from unstained Canada) in PBS, for 0.5 h at room temperature. The cryosections were gels to 0.45 μm PVDF membranes (Millipore, Bedford, MA, USA) using incubated with the primary monoclonal antibody against the NKA α a tank transfer system (Mini Trans-Blot; Bio-Rad Laboratories, Hercules, subunit (ab76020), diluted 1:250 in blocking buffer, for 2 h at room CA, USA). The blots were blocked for 2 h with 5% (wt./vol.) nonfat dried temperature. Following incubation, the cryosections were washed for milk/PBST (phosphate buffer saline with Tween 20; Merck) to minimize 5 min 3 times with PBS, exposed to the appropriate secondary antibody nonspecific binding, then incubated overnight at 4 °C with primary an- (DyLight-549 goat anti-rabbit) (Jackson Immuno Research), diluted tibodies (α5, 1:4000; T4, 1:500; or actin, 1:2000) diluted in PBST with 1:200 in blocking buffer, at room temperature for 1 h, and then washed 1% BSA (Bio basic) and 0.05% sodium azide (Sigma). The blots were sub- for 5 min 3 times with PBS. After the first staining, the cryosections were sequently washed in PBST, followed by a 1 h incubation with HRP- incubated with the primary monoclonal antibody either against NKCC conjugated secondary antibodies (Jackson) diluted in PBST (1:10,000 (T4), diluted 1:50, or CFTR, diluted 1:200, in blocking buffer overnight for anti-mouse IgG or anti-rabbit IgG). Finally, the blots were developed at 4 °C. Following incubation, the cryosections were washed for 5 min with an ECL kit (ImmobilonTM Western, Millipore) for HRP-conjugated 3 times with PBS, exposed to the appropriate secondary antibody systems, and the signals were detected using the ChemiDoc (DyLight-488 goat anti-mouse) (Jackson Immuno Research), diluted XRS + image system (Bio-Rad). The obtained data were analyzed 1:500 in blocking buffer, at room temperature for 1 h, and then washed with Image Lab software (version 3.0, Bio-Rad). The results were con- several times with PBS. Subsequently, the cryosections were mounted verted to numerical values for comparison of relative protein abundance. with a DAPI Fluoromount-G solution (Southern Biotech, Birmingham, Pseudobranch homogenates mixtures from five SW-treated individuals AL, USA), covered with cover slips, and examined under a fluorescent were used as an internal control among different immunoblots for the microscope (Olympus BX50). Within 3 h after staining, micrographs calibration of relative abundance. were obtained via confocal laser scanning microscopy (Fluoview FV1000, Olympus) using Fluoview Ver. 3.1 Viewer software (Olympus) 2.11. Specific activity of NKA or the fluorescent microscope to detect immunolocalization for subse- quent immunoreactive cell counting and morphometric analyses. Determination of NKA activity was carried out using the 96-well microplate method according to Tang et al. (2010), with some modifica- 2.8. Measurement of cell number and cell size tions. Aliquots of the pseudobranchial and branchial homogenate suspensions, prepared as described above, were used for the determina- Following double staining of longitudinal cryosections with the NKA tion of protein concentrations and NKA enzyme activities. The reaction and NKCC primary antibodies, the immunoreactive cells in the medium (final concentration: 100 mM imidazole–HCl buffer, pH 7.6, pseudobranch cryosections were observed and photographed using a 125 mM NaCl, 75 mM KCl, 7.5 mM MgCl2) was prepared according to fluorescent microscope (Olympus BX50). For immunoreactive cell Hwang et al. (1988). Then, 10 μL of the supernatant, 50 μL of 5 mM oua- counting and morphometric analyses, micrographs were analyzed bain (specific inhibitor of NKA) or deionized water, and 100 μLof using Image-Pro software (Plus 5.1). Cells showing immunoreactivity 10 mM Na2ATP were added to 340 μL of the reaction medium. NKA en- for both NKA and NKCC in the filaments (F) and NKA-immunoreactive zyme activity was defined as the difference between the amounts of in- (IR) cells in the lamellae (L) were counted separately. For each individ- organic phosphates liberated in the presence and absence of ouabain in ual, three different filaments were randomly selected for quantification, the reaction mixture. The reaction mixture was incubated at 28 °C for and a fixed length (350 μm) was measured in each filament to stan- 20 min, followed by immediate transfer to an ice bath for 10 min to dardize the counting of cells in which NKA and NKCC colocalized in stop the reaction (Cheng et al., 1999). Sardella et al. (2008) demonstrat- the filaments. Additionally, a fixed lamellar length (100 μm) was mea- ed that the specific NKA activity in fish gills measured at environmental sured in each filament to standardize the counting of NKA-IR cells in temperature is correlated with the level of in vivo activity, while that the lamellae. A total of ninety cells were selected for quantification of measured at 35 °C provides an estimate of the total amount of function- the sizes of the NKA- and NKCC-IR cells in the filaments, or of the al NKA in the studied organ (the total NKA capacity). Since the milkfish NKA-IR cells in different lamellae from either the SW or FW individuals is stenothermal, 28 °C would be the more pertinent criteria for running 18 S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25 the in vitro assays. Hence, the NKA enzyme assays were incubated at 28 °C in this study. The concentrations of inorganic phosphate were measured according to Doulgerakia et al. (2002). The formation of unreduced phosphomolybdate obtained in the present study is directly proportional to the amount of inorganic phosphate. The reagent used for color development consisted of 1% Tween-20 and 1% ammonium molybdate in 0.9 M H2SO4. The reaction mixtures and color reagent were mixed at a 1:1 (v/v) ratio and incubated for 3 min on ice, prior to detection using a microplate reader (VERSAmax, Molecular Devices, Sunnyvale, CA, USA) at 405 nm. Each sample was measured in triplicate.

2.12. Statistical analysis

The values are presented as the mean ± S.E.M. (standard error of the mean). Statistical signiﬁcance was compared using Student's t-test. The signiﬁcance level was set at P b 0.05.

3. Results

3.1. Morphology of the pseudobranchs of seawater (SW)- and fresh water (FW)-acclimated milkﬁsh

The pseudobranch of the milkfish includes a pair of gill-like structures located near the first gill arch, which contains a row of parallel filaments attached to the opercular epithelium just behind the eyes (Fig. 1A, B). The pseudobranch is covered by part of the first gill from a side view (illustrated by schematic drawing of Fig. 1C). The average length, height and filament numbers of the pseudobranch were 10.7 ± 0.9 mm, 5.0 ± 0.8 mm and 30.7 ± 2.4, respectively. Measure- ments of morphometric indexes normalized to body length revealed that the morphology of the milkfish pseudobranch did not differ significantly between SW- and FW-acclimated individuals (Table S1). Paraffin sections of the pseudobranch stained with hematoxylin and eosin (HE) exhibited the gill-like structure. In longitudinal sections, the arrange- ment of lamellae alternated in each filament. The pseudobranch-type cells (PSCs) are located beneath the pavement cells along parallel lamellae and are associated with the blood compartment separated by pillar cells in both SW- and FW-acclimated milkfish (Fig. 2A, B). In cross- sections, the pseudobranch filaments were attached to the operculum membrane (Fig. 2C, D). Between the pavement cells and PSCs appears a white empty space which is not observed in gills (Fig. 2). Anatomical and histological observations indicated that the pseudobranch of the milkfish was the lamellae-free type (illustrated in Fig. 3).

3.2. Na+,K+-ATPase (NKA) expression in the pseudobranchs of the milkﬁsh acclimated to SW and FW

Immunohistochemical staining of the longitudinal and cross- sections of both SW- and FW-acclimated fish revealed that NKA- immunoreactive (IR) cells were distributed along the edge of the parallel lamellae of the pseudobranch. Despite the expression detected in the lamellar epithelium, within the filaments, the NKA-IR cells were more abundant in the inter-lamellar epithelium close to the afferent artery Fig. 1. Illustration of the milkfish pseudobranch. (A) A pair of pseudobranchs located in the of the pseudobranch in SW-acclimated individuals than the FW- operculum near the first gill arch (ventral side view). (B) The pseudobranchs are located acclimated fish (Fig. 4). In milkfish gills, immunohistochemical staining behind the eyes and exhibit a gill-like structure, with a row of parallel filaments attached of NKA showed that most of the immunoreactive cells occurred in the to the opercular epithelium. (C) The schematic drawing shows the location of the fi pseudobranch in the operculum. Arrows indicate the pseudobranch; PSL: pseudobranchial inter-lamellar epithelium close to the afferent artery of laments in length (mm), PSH: pseudobranchial height (mm). SW-acclimated individuals. In contrast, the longitudinal and cross- sections of gills showed NKA-IR cells throughout almost the whole gill epithelium, except some regions of lamellae close to the efferent artery of filaments in FW-acclimated fish (Fig. S2). In the pseudobranchs, ap- (113 kDa) in the pseudobranchs revealed a single immunoreactive proximately 2-fold higher NKA activity was observed in SW- band. The profiles of pseudobranchial NKA α-subunit protein expres- acclimated individuals than in FW-acclimated fish. In the gills, however, sion showed significantly higher expression in SW-acclimated milkfish an opposite expression pattern was observed, as the FW-acclimated in- than in the FW group, while the relative NKA protein abundance in the dividuals exhibited more than 2-fold higher NKA activity than the SW- FW-acclimated fish was approximately 3-fold higher than that in the acclimated fish (Fig. 5). Immunoblotting of the NKA α-subunit protein SW group in the gills (Fig. 6). S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25 19

Fig. 2. Histological observations of the milkfish pseudobranch. Paraffin sections stained with hematoxylin and eosin (HE) were obtained from SW- and FW-acclimated milkfish. (A, B) Lon- gitudinal section, (C, D) cross-section. Arrows, pseudobranch-type cells; arrowheads, chloride cells; F, filaments; L, lamellae; C, cartilage; OM, the operculum membrane; AFA, afferent filament artery; EFA, efferent filament artery; SW, seawater; FW, fresh water. (A, B) scale bar = 20 μm, (C, D) scale bar = 100 μm.

3.3. Comparisons of chloride transporters between the PSCs and chloride to the FW group (Fig. 7). Double immunofluorescence staining of the cells (CCs) NKA α-subunit and NKCC in paraffin sections of pseudobranchs collect- ed from SW- and FW-acclimated milkfish showed that NKCC was The relative protein expression of the Na+,K+,2Cl− cotransporter colocalized to NKA-IR cells in the filament epithelium. More NKCC and (NKCC, 125–220 kDa, a major band was noted at 125 kDa and another NKA-IR cells were found in SW-acclimated milkfish than FW- one at 220 kDa) in the pseudobranchs and gills of the SW- and FW- acclimated individuals. In addition, more NKCC and NKA-IR cells oc- acclimated milkfish was detected with a monoclonal antibody (T4). curred in the inter-lamellar epithelium close to the afferent artery of The obtained profiles revealed significantly higher NKCC protein abun- the filament and the epithelium of the operculum membrane in dance in the pseudobranchs and gills of SW-acclimated fish compared pseudobranchs of SW-adapted fish than FW-acclimated individuals (Fig. 8). On the other hand, double immunofluorescence staining of the NKA α-subunit and cystic fibrosis transmembrane conductance regulator (CFTR) in frozen cross-sections of the milkfish pseudobranchs showed that CFTR was localized in the apical regions of NKA-IR cells in the filament epithelium of SW-acclimated milkfish, whereas CFTRs were not found in the NKA-IR cells of the lamellar epithelium (Fig. 9). Therefore, the spherical NKA-IR cells, which were mainly located in the filaments close to the afferent artery and the epithelium of the operculum membrane and exhibited both NKCC and CFTR staining in SW- acclimated milkfish, constituted CCs, similar to the ionocytes found in gills. On the other hand, the flat penta- or hexagonal cells located basi- cally under the pavement cells along the lamellae, exhibiting only NKA staining, were considered PSCs (Figs. 8, 9). Comparisons between the CCs and PSCs in the pseudobranchs of the SW- and FW-acclimated milkfish showed that only the CCs exhibited NKCC and CFTR staining in the basolateral and apical membranes, respectively, of SW- acclimated individuals. To further analyze the characteristics of the pseudobranchs, the numbers and sizes of CCs/PSCs were counted. The average numbers Fig. 3. A schematic illustration of the lamellae-free-type pseudobranch of the milkfish and sizes of CCs were increased approximately 10-fold and 2.5-fold, re- (modified from Mattey et al., 1980). F, filaments; L, lamellae; C, cartilage; CVS, central ve- fi nous sinus; V, vein; OM, the operculum membrane; AFA, afferent filament artery; EFA, ef- spectively, in the SW milk sh compared to the FW group, mainly in the ferent filament artery. filament epithelium close to the afferent artery. Meanwhile, in FW- 20 S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25

Fig. 4. Immunohistochemical staining of Na+,K+-ATPase (NKA) in the pseudobranch of SW- and FW-acclimated milkfish. The NKA-immunoreactive (IR) cells are distributed in the filament (F) as well as lamellae (L) of the pseudobranch. Numerous NKA-IR cells were found in the epithelium of the inter-lamellar region close to the afferent filament artery (AFA) in the seawater (SW)-acclimated individuals. (A, B) Longitudinal section and (C, D) cross-section of the pseudobranch. Arrows, pseudobranch-type cells; arrowheads, chloride cells; F, filaments; L, lamellae; C, cartilage; FW, fresh water. Scale bar = 20 μm. acclimated fish, the sizes of PSCs located in the lamellae became larger genera Gymnarchus and Cobitis (Wittenberg and Haedrich, 1974). Dif- than those of SW-acclimated individuals, but there was no significant ferent types of pseudobranchs have been observed according to the difference in the average numbers of PSCs between the SW and FW SW or FW habitat of fish. The sizes, forms and external structures of groups (Figs. 10, S1). the pseudobranchs vary widely within orders, families and species, though the reason for this is not clear (Laurent and Dunel-erb, 1984). 4. Discussion In many FW teleosts, the pseudobranch is small and is covered by the opercular epithelium. This type is referred to as a “glandular The pseudobranch is found in almost all fishes among the 238 pseudobranch” (i.e. belong to Types III in the Bertin's classification) be- known teleostean families, with the exception of a few species belong- cause its lamellae exhibit no contact with the external medium. On the ing to order Anguilliformes, suborder Siluroidei, and all species of other hand, the pseudobranch is a much more impressive organ (i.e. belong to Types I in the Bertin's classification) in the great majority of marine fish (Wittenberg and Haedrich, 1974; Laurent and Dunel-erb, 1984). In this study, however, the morphometric analyses revealed that the morphology of milkfish pseudobranchs was not significantly different between individuals acclimated to SW and FW (for more than one month) (Table S1). It can be speculated that the divergence of different type pseudobranchs is a phenomenon that has arisen through long-term evolution, whereas temporary morphological adaptation of this structure is not observed under acclimation to environments with different salinities. All pseudobranchs contain typical pseudobranch-type cells (PSCs) resting on the basal lamina, which have never been observed in the gills of teleosts (Laurent and Dunel-Erb, 1984). Bertin (1958) described three morphological types of pseudobranchs in teleosts. Accordingly, the milkfish pseudobranch is more likely to be of type-Ι, the lamellae- free-type pseudobranch, with distinguishable filaments and completely free lamellae that contact the water. Mattey et al. (1980) reported that Fig. 5. Na+,K+-ATPase (NKA) activity in pseudobranchs and gills of milkfish acclimated to both chloride cells (CCs) and PSCs can be found in the lamellae-free seawater (SW) and fresh water (FW). Higher NKA activity was found in the pseudobranch and semi-free types of pseudobranchs based on their location and of SW fish compared to the FW group (N = 4). In contrast, the NKA activity in the gills of shape. In covered- or embedded-type pseudobranchs, however, only FW milkfish was significantly higher than in the SW fish (N = 4). The asterisks indicate PSCs can be found along the lamellae, as observed in the pseudobranchs significant differences between the SW- and FW-acclimated milkfish. Different letters indicate significant differences between the pseudobranchs and gills in the SW (capital let- of Chinook salmon (Quinn et al., 2003) and rainbow trout (Kern et al., ters) or FW (lowercase letters) group (Student's t-test, P b 0.05). 2002). The PSCs have been reported to display numerous mitochondria S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25 21

Fig. 7. Protein expression of NKCC in the pseudobranchs and gills of milkfish acclimated to Fig. 6. Protein expression of the NKA α-subunit in the pseudobranchs and gills of milkfish environments with different salinities. (A) A representative immunoblot probed with a acclimated to seawater (SW) and fresh water (FW). (A) A representative immunoblot monoclonal antibody (T4) showing an immunoreactive band at approximately probed with a monoclonal antibody (α5) showing a single immunoreactive band of ap- 125–220 kDa. (B) The relative protein abundance of the immunoreactive bands obtained proximately at 113 kDa. (B) The relative protein abundance of the immunoreactive for the SW and FW groups indicated that the NKCC contents in both the pseudobranchs bands obtained for SW and FW milkfish was significantly higher in the pseudobranchs and gills were higher in the SW than the FW group (N = 6). The asterisks indicate signif- of fish in the SW group compared to the FW group, whereas that in gills of the FW fish icant differences between the SW- and FW-acclimated milkfish. Different letters indicate was significantly higher than in the SW group (N = 6). The asterisks indicate significant significant differences between the pseudobranchs and gills in the SW (capital letters) differences between the SW- and FW-acclimated milkfish. Different letters indicate signif- or FW (lowercase letters) group (Student's t-test, P b 0.05). icant differences between the pseudobranchs and gills in the SW (capital letters) or FW (lowercase letters) group (Student's t-test, P b 0.05).

profiles showed that these parameters were also higher in the in the cytoplasm (Dendy et al., 1973; Mattey et al., 1978) and exhibit pseudobranchs of SW milkfish compared to FW individuals. Therefore, high Na+,K+-ATPase (NKA) activity in several studied teleostean spe- our data suggested that the pseudobranchs of SW-acclimated milkfish cies (Kern et al., 2002; Quinn et al., 2003). required higher NKA activity and protein expression to establish a The lamellae-free- and semi-free-type pseudobranchs of teleosts ex- hypo-osmoregulatory ability. In addition, our results revealed that the hibit epithelial CCs and have therefore been suggested to be involved in number of pseudobranchial CCs was decreased in FW-acclimated fish, osmoregulatory processes (Laurent and Dunel-Erb, 1984; Hamidian and whereas the number of branchial ionocytes in FW-acclimated milkfish Alboghobeish, 2007). In addition, the pseudobranchial CCs equipped was increased compared to the SW group (Fig. S1). Our data indicated with accessory cells have been demonstrated in a marine teleost Mullus that pseudobranchial CCs did not proliferate abundantly along the la- surmuletus (Laurent and Dunel-Erb, 1984). However, no direct evidence mellae, as observed in the gills, due to occupation by PSCs. Hence, of a potential osmoregulatory role of the pseudobranch has been report- fewer pseudobranchial CCs for carrying out ion uptake were found in ed. In this study, two types of NKA-IR cells with distinct locations and the FW group. Accordingly, we suggested that the role of the shapes within the lamellae-free-type pseudobranch of a teleost were pseudobranchs in hyper-osmoregulaton was limited compared to that described for the first time. In the pseudobranch of the milkfish, CCs of the gills of the milkfish. were mainly located in the filaments, especially close to the afferent ar- The pseudobranch might contribute differentially to osmoregula- tery, while PSCs were located in the lamellae. Previous studies have tion among different species, as after transfer from FW to SW, shown that the PSCs within covered-type pseudobranchs exhibit rela- pseudobranchial NKA activity increases in euryhaline killifish but tively high NKA activity, with the enzyme localizing along the outer had not changed in juvenile Chinook salmon at 10 days post-transfer edge of the parallel lamellae, which is opposite the localization observed (Epstein et al., 1967; Quinn et al., 2003). On the other hand, CCs are gen- in gill ionocytes (i.e., chloride cells or mitochondria-rich cells; Laurent erally present in lamellae-free-type pseudobranchs, while in the and Dunel-Erb, 1984; Quinn et al., 2003). In the lamellae-free-type covered- or embedded-type, CCs are replaced by PSCs (Hamidian and pseudobranch of the milkfish, however, we found that NKA mainly local- Alboghobeish, 2007). Because CCs are thought to be the major sites ized to the basolateral membrane of PSCs. of ionoregulation, different types of pseudobranchs might play dif- NKA is considered a marker of gill ionocytes because it is a primary ferent roles. We speculate that the lamellae-free or semi-free-type active transport pump providing the major driving force for ion trans- pseudobranchs containing CCs for hypo-osmoregulation are mainly port (Marshall, 2002; Hirose et al., 2003; Hwang et al., 2011). Greater found in marine or euryhaline fish, whereas covered- or embedded- numbers of NKA-IR CCs were found in the pseudobranchs of SW- than type pseudobranchs, which do not contain CCs carrying out hyper- FW-acclimated milkfish. The NKA protein abundance and activity osmoregulation, are most likely observed in stenohaline FW fish. 22 S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25

Fig. 8. Double immunofluorescence staining with antibodies for the NKA α-subunit and NKCC in paraffin cross-sections of the pseudobranchs of SW- and FW-acclimated milkfish. Greater numbers of NKCC-immunoreactive (IR) cells were found in the filaments of the SW-acclimated milkfish (A, C, E) than the FW-acclimated individuals (B, D, F). F, filaments; L, lamellae; AFA, afferent filament artery. Scale bar = 100 μm.

Some studies have investigated the osmoregulatory functions of In the gill epithelium of teleosts, ionocytes express high NKA protein pseudobranchs by comparing NKA activity profiles upon salinity chal- abundance and activity in immunocytochemical studies (Hirose et al., lenge (Epstein et al., 1967; Quinn et al., 2003). However, no pervious 2003; Hwang and Lee, 2007). Prominent increases in Na+,K+,2Cl− study has compared the differences in the abundance of the NKA cotransporter (NKCC) protein expression in branchial ionocytes can in- protein and NKA activity between the pseudobranchs and gills. In the crease the efficiency of Cl− exclusion from the plasma when fish are ac- present study, the NKA protein abundance and activity in the climated to hyperosmotic environments (Kang et al., 2010). In gills, an lamellae-free-type pseudobranch of the milkfish were approximately increased abundance of NKCC protein in response to hyperosmotic chal- 2- to 3-fold higher than those in the gills in SW-acclimated fish, which lenge has been reported in eels (Tse et al., 2006), sea bass (Lorin-Nebel might be because the pseudobranchs exhibit many more NKA-IR PSCs et al., 2006), killifish (Scott et al., 2004), salmon (Pelis et al., 2001; in the lamellae compared to the gills (Fig. S2). On the contrary, the Tipsmark et al., 2002; Hiroi and McCormick, 2007; Nilsen et al., 2007), NKA protein abundance and activity in the gills were approximately tilapia (Wu et al., 2003; Tipsmark et al., 2008) and milkfish (Lin et al., 1- to 2-fold higher than in the pseudobranch in FW-acclimated milkfish 2003; Tang et al., 2011). In the present study, the abundance of NKCC because the gills exhibit numerous NKA-IR ionocytes (CCs) along the la- protein in the pseudobranch was found to be similar to that in the mellae, in contrast to what is observed in the pseudobranch. This find- gills of SW-acclimated milkfish. The colocalization of the NKCC and ing implied that upon salinity challenge, the difference in NKA NKA proteins in the pseudobranchs of SW- and FW-acclimated milkfish expression in comparison with the gills might be a characteristic of further revealed that NKCC protein was localized in the basolateral hypo-osmoregulation in lamellae-free-type pseudobranchs. membrane of pseudobranchial CCs. The monoclonal anti-human NKCC S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25 23

Fig. 9. Double immunofluorescence staining with antibodies for the NKA α-subunit and CFTR in frozen cross-sections of the pseudobranchs of SW-acclimated milkfish. The CFTR-immunoreactive (IR) cells were localized to the apical membranes of NKA-IR cells (arrowheads, chloride cells) in the filaments (A, B) but were not found in the lamellar NKA-IR cells (arrows, pseudobranch-type cells) of the SW- and FW-acclimated milkfish (C, D). F, filaments; L, lamellae; AFA, afferent filament artery. (A, B) scale bar = 50 μm; (C, D) scale bar = 20 μm. antibody T4 cannot distinguish between secretory isoform NKCC1 and analysis conducted in this study demonstrated that a CFTR antibody absorptive isoform NKCC2 in mammals. T4 immunoreactivity in the api- can be used as a marker to trace CCs in the pseudobranch of SW- cal membrane is assumed to be NKCC2 and that in the basolateral mem- acclimated milkfish because the PSCs of SW-acclimated milkfish and brane is likely NKCC1, based on the reacting position of the CCs (Payne the CCs and PSCs of FW-acclimated milkfish were CFTR et al., 1995; Kang et al., 2010). The NKCC detected in the immunonegative. Taken together, our results indicated that the pseudobranchial CCs of the milkfish should be the isoform NKCC1. Sim- pseudobranchial NKA-IR cells, mainly located in the epithelium of the ilar results have been found in the branchial ionocytes of other euryha- filaments and the operculum membrane and exhibiting NKCC and line teleosts such as killifish, salmon and tilapia (Pelis et al., 2001; Wu CFTR staining in SW-acclimated milkfish, were CCs, whereas the cells et al., 2003; Hiroi and McCormick, 2007; Katoh et al., 2008) when accli- under the pavement cells along the lamellae exhibiting only NKA ex- mated to hyperosmotic environments. The distribution of NKCCs ob- pression were PSCs. CCs, but not PSCs exhibited NKCC expression in served in the pseudobranchs of SW milkfish agreed with the currently the basolateral membranes and CFTR expression in the apical mem- accepted model of Cl− secretion via branchial ionocytes in SW fish branes for carrying out hypo-osmoregulation in the pseudobranch of (Evans et al., 2005; Hwang et al., 2011). On the other hand, NKCC pro- SW-acclimated individuals. tein expression was decreased but was still detected in the CCs of the Additionally, the increases in the numbers and sizes of pseudobranchs of FW-acclimated milkfish. These changes might be nec- pseudobranchial CCs observed in the filaments close to the afferent ar- essary for maintaining the hyposmoregulatory endurance of the fish tery in the SW-acclimated milkfish indicated the potential role of the (Kang et al., 2010) or for conducting cell volume regulation (Pelis lamellae-free-type pseudobranch in hypo-osmoregulation upon salinity et al., 2001). Our immunocytochemical staining results indicated that challenge. On the other hand, the average size of PSCs was greater in the the NKCC antibody (T4) could be used as a marker to detect CCs, rather FW- than SW-acclimated fish, but the number per 100 μm in the lamel- than PSCs in the pseudobranchs of milkfish. lae did not change significantly. The larger size of PSCs observed in the Cystic fibrosis transmembrane conductance regulator (CFTR) is a hypoosmotic environment indicated that there was salinity- chloride channel expressed in the apical membranes of the branchial dependent expression of NKA-IR PSCs in the milkfish pseudobranch. ionocytes that are responsible for chloride secretion in teleosts accli- However, the PSCs are not the sites of salt secretion for the following mated to SW (Hirose et al., 2003; Evans, 2008; Kaneko et al., 2008). De- reasons: (1) the PSCs lack NKCC and CFTR, which are the crucial ion tected by the monoclonal anti-human CFTR antibody, the western blot transporters involved in Cl− excretion; and (2) the PSCs are covered of SW bass gills revealed a major band at 138 kDa and the other band by pavement cells, therefore lacking apical surfaces contacting the ex- at 71 kDa (Bodinier et al., 2009). In the pseudobranchs of SW- ternal environment, which is a characteristic of branchial ionocytes for acclimated milkfish, CFTR was also expressed in the apical membranes salt secretion. According to the results of Kern et al. (2002),H+- of CCs, similar to previous findings in the branchial ionocytes of killifish ATPase was immunolabeled on the tubular regions of the PSCs in the (Katoh and Kaneko, 2003), tilapia (Hiroi et al., 2005), milkfish and pseudobranch of the FW rainbow trout (Oncorhynchus mykiss)and pufferfish (Tang et al., 2011). The immunocytochemical staining was sensitive to specific inhibitor of H+-ATPase (bafilomycin A1). 24 S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25

References

Bertin, L., 1958. In: Grasse, P.P. (Ed.), Organes de la respiration aquatique. Traite de Zoologie, Vol. XIII. Masson Publishing, Paris, pp. 1301–1341. Bodinier, C., Boulo, V., Lorin-Nebel, C., Charmantier, G., 2009. Influence of salinity on the localization and expression of the CFTR chloride channel in the ionocytes of Dicentrarchus labrax during ontogeny. J. Anat. 214, 318–329. Bridges, C.R., Berenbrink, M., Müller, R., Waser, W., 1998. Physiology and biochemistry of the pseudobranch: an unanswered question? Comp. Biochem. Physiol. A 119, 67–77. Chen, C.N., Lin, L.Y., Lee, T.H., 2004. Distributions of ionocytes in gills of the euryhaline milkfish, Chanos chanos (Forsskål, 1775). Zool. Stud. 43, 772–777. Cheng, S.X.J., Aizman, O., Nairn, A.C., Greengard, P., Aperia, A., 1999. [Ca2+]determinesthe effects of protein kinases A and C on activity of rat renal Na+,K+-ATPase. J. Physiol. 518, 37–46. Dendy, L.A., Philpott, C., Deter, W., 1973. Localization of Na+,K+-ATPase and other en- zymes in teleost pseudobranch ΙΙ. Morphological characterization of intact pseudobranch, subcellular fractions, and plasma membrane substructure. J. Cell Biol. 57, 689–703. Doulgerakia, A., Papadopoulou-Daifotib, Z., Tsakirisa, S., 2002. Effects of L-phenylalanine on acetylcholinesterase and Na+,K+-ATPase activities in suckling rat frontal cortex, hippocampus and hypothalamus. Z. Naturforsch. 57c, 182–188. Epstein, F.H., Katz, A.I., Pickford, G.E., 1967. Sodium and potassium-activated adenosine triphosphatase of gills: role in adaptation of teleosts to salt water. Science 156, 1245–1247. Evans, D.H., 2008. Teleost fish osmoregulation: what have we learned since August Krogh, Homer Smith, and Ancel Keys. Am. J. Physiol. 295, R704–R713. Evans, D.H., Piermarini, P.M., Choe, K.P., 2005. The multifunctional fish gill: dominant site of gas exchange, osmoregulation, acid–base regulation, and excretion of nitrogenous waste. Physiol. Rev. 85, 97–177. Ferraris, R.P., Almendras, J.M., Jazul, A.P., 1988. Changes in plasma osmolality and chloride concentration during abrupt transfer of milkfish (Chanos chanos) from seawater to different test salinities. Aquaculture 70, 145–157. Fischer-Scherl, T., Hoffmann, R., 1986. Light- and electron- microscope studies on the pseudobranch of the golden orfe, Leuciscus idus. L. J. Fish Biol. 29, 699–709. Franklin, C.E., Davison, W., Forster, M.E., 1992. Seawater adaptability of New Zealand’s sockeye (Oncorhynchus nerka) and chinook salmon (O. tshawytscha): physiological correlates of smoltification and seawater survival. Aquaculture 102, 127–142. Hamidian, G., Alboghobeish, N., 2007. Histological study of pseudobranch in Cteno- pharyngodon idella. AATEX 14. 693–696 (Special Issue). Hiroi,J.,McCormick,S.D.,2007.Variation in salinity tolerance, gill Na+/K+ -ATPase, Na+/ K+/2Cl− cotransporter and mitochondria-rich cell distribution in three salmonids Salvelinus namaycush, Salvelinus fontinalis and Salmo salar.J.Exp.Biol.210, 1015–1024. Fig. 10. Comparisons of the average numbers and sizes of the (A) chloride cells (CCs): NKA Hiroi, J., McCormick, S.D., Ohtani-Kaneko, R., Kaneko, T., 2005. Functional classification of and NKCC-immunoreactive (IR) cells in the filaments, and (B) pseudobranch-type cells mitochondria-rich cells in euryhaline Mozambique tilapia (Oreochromis (PSCs): the lamellae NKA-IR cells in the pseudobranchs of SW- and FW-acclimated milk- mossambicus) embryos, by means of triple immunofluorescence staining for Na+/ fish. Cells of the pseudobranch in paraffin longitudinal sections were counted and mea- K+ -ATPase, Na+/K+/2 Cl- cotransporter and CFTR anion channel. J. Exp. Biol. 208, sured. The units for measurement were set as 350 μminthefilaments and 100 μmin 2023–2036. the lamellae. The asterisks indicate significant differences between SW- and FW- Hirose, S., Kaneko, T., Naito, N., Takei, Y., 2003. Molecular biology of major components of – acclimated milkfish (Student's t-test, P b 0.05). chloride cells. Comp. Biochem. Physiol. B 136, 593 620. Hwang, P.P., Lee, T.H., 2007. New insights into fish ion regulation and mitochondria-rich cells. Comp. Biochem. Physiol. A 148, 479–497. + + However, in the pseudobranch of marine euryhaline milkfish, it is not Hwang, P.P., Sun, C.M., Wu, S.M., 1988. Characterization of gill Na -K -activated adeno- + sine triphosphatase from tilapia Oreochromis mossambicus. Bull. Inst. Zool. Acad. Sin. clear if the PSCs exhibit H -ATPase and get involved in the blood acidi- 27, 49–56. fication or whether PSCs and CCs have different functions in acid–base Hwang, P.P., Lee, T.H., Lin, L.Y., 2011. Ion regulation in fish gills: recent progress in the cel- regulation. Further studies will focus on the ultrastructures and the lular and molecular mechanisms. Am. J. Physiol. 301, R28–R47. fi Kaneko, T., Watanabe, S., Lee, K.M., 2008. Functional morphology of mitochondrion-rich roles of the PSCs in the milk sh pseudobranch. cells in euryhaline and stenohaline teleosts. Aquat. BioSci. Monogr. 1, 1–62. This study investigated the anatomical structure of the milkfish Kang, C.K., Tsai, H.J., Liu, C.C., Lee, T.H., 2010. Salinity-dependence expression of a Na+,K+, pseudobranch and examined the localization of ion transporters within 2Cl- cotransporter in gills of the brackish medaka Oryzias dancena: a molecular corre- fi late for hyposmoregulatory endurance. Comp. Biochem. Physiol. A 157, 7–18. the PSCs and CCs of the pseudobranch. The present study was the rst to Katoh, F., Kaneko, T., 2003. Short-term transformation and long-term replacement of distinguish NKA-IR CCs and PSCs in the pseudobranch of milkfish using branchial chloride cells in killifish transferred from seawater to freshwater, revealed antibodies against NKCC and CFTR as markers. In addition, increases in by morphofunctional observations and a newly established time-differential double fluorescent staining technique. J. Exp. Biol. 206, 4113–4123. the numbers and sizes of CCs as well as in NKA expression of SW- − Katoh, F., Cozzi, R.R., Marshall, W.S., Goss, G.G., 2008. Distinct Na+/K+/2Cl cotransporter acclimated milkfish pseudobranchs indicated the potential role of the localization in kidneys and gills of two euryhaline species, rainbow trout and killifish. lamellae-free-type pseudobranch in hypo-osmoregulation upon salinity Cell Tissue Res. 334, 265–281. challenge in this euryhaline teleostean species. Kern, G., Bösch, S.T., Unterhuber, E., Pelster, B., 2002. Mechanisms of acid secretion in pseudobranch cells of rainbow trout Oncorhynchus mykiss. J. Exp. Biol. 205, 2943–2954. Supplementary data to this article can be found online at http://dx. Laurent, P., Dunel-Erb, S., 1984. The pseudobranch: morphology and function. In: Hoar, doi.org/10.1016/j.cbpa.2013.12.018. W.S., Randall, D.J. (Eds.), Fish Physiology. Academic Press Ltd., Orlando, Florida, pp. 285–325. Lee, T.H., Hwang, P.P., Lin, H.C., Huang, F.L., 1996. Mitochondria-rich cells in the branchial epithelium of the teleost, Oreochromis mossambicus, acclimated to various hypotonic Acknowledgments environments. Fish Physiol. Biochem. 15, 513–523. Lin, Y.M., Chen, C.N., Lee, T.H., 2003. The expression of gill Na+,K+-ATPase in milkfish, The monoclonal antibodies T4 and α5 were purchased from the De- Chanos chanos, acclimated to seawater, brackish water and fresh water. Comp. Biochem. Physiol. A 135, 489–497. velopmental Studies Hybridoma Bank (DSHB) maintained by the De- Lin, Y.M., Chen, C.N., Yoshinaga, T., Tsai, S.C., Shen, I.D., Lee, T.H., 2006. Short-term effects partment of Pharmacology and Molecular Sciences, John Hopkins of hyposmotic shock on Na+/K+-ATPase expression in gills of the euryhaline milkfish, Chanos chanos. Comp. Biochem. Physiol. A 143, 406–415. University School of Medicine, Baltimore, MD 2120521205, and the De- − Lorin-Nebel, C., Boulot, V., Bodinier, C., Charmantier, G., 2006. The Na+/K+/2Cl partment of Biological Sciences, University of Iowa, Iowa City, IA 52242, cotransporter in the sea bass Dicentrarchus labrax ontogeny: involvement in osmo- under Contract N01-HD-6-2915, NICHD, USA. regulation. J. Exp. Biol. 209, 4908–4922. S.-H. Yang et al. / Comparative Biochemistry and Physiology, Part A 170 (2014) 15–25 25

Marshall, W.S., 2002. Na+,Cl-,Ca2+ and Zn2+ transport by fish gills: retrospective review Swanson, C., 1998. Interactive effects of salinity on metabolic rate, activity, growth and os- prospective synthesis. J. Exp. Zool. 293, 264–283. moregulation in the euryhaline milkfish (Chanos chanos). J. Exp. Biol. 201, 3355–3366. Mattey, D.L., Moate, R., Morgan, M., 1978. Comparison of ‘Pseudobranch’ type and Tang, C.H., Tzeng, C.S., Hwang, L.Y., Lee, T.H., 2009. Constant muscle water contents and ‘Chloride’ type cells in the pseudobranch of marine, freshwater and euryhaline tele- renal HSP90 expression reflect the osmotic homeostasis in euryhaline teleosts accli- osts. J. Fish Biol. 13, 535–542. matized to different environmental salinities. Zool. Stud. 48, 435–441. Mattey, D.L., Morgan, M., Wright, D.E., 1980. A scanning electron microscope study of the Tang, C.H., Wu, W.Y., Tsai, S.C., Yoshinaga, T., Lee, T.H., 2010. Elevated Na+/K+-ATPase re- pseudobranchs of two marine teleosts. J. Fish Biol. 16, 331–343. sponses and its potential role in triggering ion reabsorption in kidneys for homeosta- Mölich, A., Waser, W., Heisler, N., 2009. The teleost pseudobranch: a role for precondition- sis of marine euryhaline milkfish (Chanos chanos) when acclimated to hypotonic ing of ocular blood supply? Fish Physiol. Biochem. 35, 273–286. fresh water. J. Comp. Physiol. B. 180, 813–824. Nilsen, T.O., Ebbesson, L.O., Madsen, S.S., McCormick, S.D., Andersson, E., Tang, C.H., Hwang, L.Y., Shen, I.D., Chiu, Y.H., Lee, T.H., 2011. Immunolocalization of chlo- Bjornsson, B.T., Prunet, P., Stefansson, S.O., 2007. Differential expression of ride transporters to gill epithelia of euryhaline teleosts with opposite salinity-induced gill Na+,K+-ATPase α-andβ-subunits, Na+,K+,2Cl− cotransporter and Na+/K+-ATPase responses. Fish Physiol. Biochem. 37, 709–724. CFTR anion channel in juvenile anadromous and landlocked Atlantic salmon Tipsmark, C.K., Madsen, S.S., Ceidelin, M., Christensen, A.S., Cutler, C.P., Cramb, G., 2002. Salmo salar. J. Exp. Biol. 210, 2885–2896. Dynamics of Na+,K+, 2Cl− cotransporter and Na+,K+-ATPase expression in the Payne, J.A., Xu, J.C., Hass, M., Lytle, C.Y., Ward, D., Forbush, B.Ι.Ι.Ι., 1995. Primary structure, branchial epithelium of brown trout (Salmo trutta) and Atlantic salmon (Salmo functional expression, and chromosomal localization of the bumetanide-sensitive salar). J. Exp. Zool. 293, 106–118. Na-K-Cl cotransporter in human colon. J. Biol. Chem. 270, 17977–17985. Tipsmark, C.K., Baltzegar, D.A., Ozden, O., Grubb, B.J., Borski, R.J., 2008. Salinity regulates Pelis, R.M., Zydlewski, J., McCormick, S.D., 2001. Gill Na+-K+-2Cl− cotransporter abun- claudin mRNA and protein expression in the teleost gill. Am. J. Physiol. 294, dance and location in Atlantic salmon: effects of seawater and smolting. Am. R1004–R1014. J. Physiol. 280, R1844–R1852. Tse, W.K.F., Au, D.W.T., Wong, C.K.C., 2006. Characterization of ion channel and transport- Quinn, M.C.J., Veillette, P.A., Young, G., 2003. Pseudobranch and gill Na+,K+-ATPase activ- er mRNA expressions in isolated gill chloride and pavement cells of seawater accli- ity in juvenile chinook salmon, Oncorhynchus tshawytscha: developmental changes mating eels. Biochem. Biophys. Res. Commun. 346, 1181–1190. and effects of growth hormone, cortisol and seawater transfer. Comp. Biochem. Phys- Uchida, K., Kaneko, T., Yamauchi, K., Hirano, T., 1996. Morphometrical analysis of chloride iol. A 135, 249–262. cell activity in the gill filaments and lamellae and changes in Na+,K+-ATPase activity Sardella, B.A., Kültz, D., Cech Jr., J.J., Brauner, C.J., 2008. Salinity-dependent during seawater adaptation in chum salmon fry. J. Exp. Zool. 276, 193–200. changes in Na+/K+-ATPase content of mitochondria-rich cells contribute to Wittenberg, J.B., Haedrich, R.L., 1974. The choroid rete mirabile of the fish eye. II. Distribu- differences in thermal tolerance of Mozambique tilapia. J. Comp. Physiol. B. tion and relation to the pseudobranch and to the swimbladder rete mirabile. Biol. 178, 249–256. Bull. 146, 137–156. Scott, G.R., Richards, J.G., Forbush, B., Isenring, P., Schulte, P.M., 2004. Changes in gene ex- Wu, Y.C., Lin, L.Y., Lee, T.H., 2003. Na+,K+,2Cl− cotransporter: a novel marker for iden- pression in gills of the euryhaline killifish Fundulus heteroclitus after abrupt salinity tifying freshwater- and seawater-type mitochondria-rich cells in gills of euryhaline transfer. Am. J. Physiol. 287, C300–C309. tilapia, Oreochromis mossambicus. Zool. Stud. 42, 186–192.