IMMUNOLOCALIZATION OF Na + /K + -ATPase IN BRANCHIAL EPITHELIUM OF CHUM SALMON FRY DURING SEAWATER AND FRESHWATER ACCLIMATION

The Journal of Experimental Biology 201, 3031 3040 (1998) Printed in Great Britain The Company of Biologists Limited 1998 JEB1654 3031 IMMUNOLOCALIZATION OF Na + /K + -ATPase IN BRANCHIAL EPITHELIUM OF CHUM SALMON FRY DURING SEAWATER AND FRESHWATER ACCLIMATION TAKAHITO SHIKANO* AND YOSHIHISA FUJIO Department of Fisheries, Faculty of Agriculture, Tohoku University, Sendai, Miyagi 981-8555, Japan *e-mail: shikano@mtb.biglobe.ne.jp Accepted 2 September; published on WWW 26 October 1998 Immunolocalization of the α-subunit of Na + /K + -ATPase was examined in the gill epithelium of chum salmon (Oncorhynchus keta) fry during acclimation to brackish water (25 salinity) and reintroduction to fresh water. In freshwater fish, strong immunoreactivity was associated with the large spherical cells located on the free surface of the primary lamellae, especially in those found at the base of the secondary lamellae, and with the large spherical cells located on the secondary lamellae.the large spherical cells located near the central venous sinus at the base of the secondary lamellae and in the interlamellar regions, however, showed little or no immunoreactivity. When freshwater fish were acclimated to brackish water, immunoreactivity developed in the large spherical cells near the central venous sinus concomitant with an increase in the hypo-osmoregulatory ability of the fish. In contrast, reintroduction from brackish water to fresh water caused the disappearance of the immunoreactivity in the large spherical cells near the central venous sinus Summary and a reduction in hypo-osmoregulatory ability. During acclimation to brackish water and reintroduction to fresh water, the hypo-osmoregulatory ability of the fish did not correlate with the total number of large spherical cells located on the primary lamellae but was closely correlated with the number of large spherical cells showing strong immunoreactivity for Na + /K + -ATPase. We conjecture that these immunopositive large spherical cells are mature differentiated chloride cells, whereas the immunonegative large spherical cells are young developing chloride cells. The development of immunoreactivity for Na + /K + -ATPase in young chloride cells may be one of the most important factors in the development of hypo-osmoregulatory ability by chum salmon fry. Key words: Na + /K + -ATPase, gill, branchial epithelium, chum salmon, Oncorhynchus keta, acclimation, immunolocalization, salinity tolerance, chloride cell. Introduction Chloride cells in the gills are now recognized as the sites of excretion of excess Na + and Cl in seawater-adapted teleosts (Foskett and Scheffey, 1982; Zadunaisky, 1984). The cells are characterized by a large and columnar appearance, numerous mitochondria and an extensive branching and anastomosing network of tubules (Karnaky et al. 1976a; Philpott, 1980; Pisam, 1981). Chloride cells are also thought to be responsible for active branchial ion absorption in hypo-osmotic environments (Laurent and Dunel, 1980; Perry and Wood, 1985; Avella et al. 1987; Perry and Laurent, 1989; Laurent et al. 1994; Marshall et al. 1997), although this role has not been demonstrated directly. Na + /K + -ATPase plays a central role in ion transport in chloride cells (Marshall, 1995; McCormick, 1995). The enzyme is located on the basolateral tubular network (Karnaky et al. 1976b; Hootman and Philpott, 1979). Many investigators have reported that the abundance and activity of Na + /K + -ATPase in the gills increase when freshwater fish are transferred to sea water (Kamiya and Utida, 1968, 1969; Thomson and Sargent, 1977; Epstein et al. 1980). Thus, the enzyme can effectively be used as an indicator of hypoosmoregulatory competence. Na + /K + -ATPase consists of two subunits, α and β, both of which are necessary for function. The α-subunit is the catalytic part of the Na + /K + exchange mechanism and shows a highly conserved amino acid sequence in diverse vertebrate and invertebrate species (Shull et al. 1985; Kawakami et al. 1985; Takeyasu et al. 1988; Lebovitz et al. 1989). In the gills of teleosts, Ura et al. (1996) and Witters et al. (1996) reported that immunoreactivity for the Na + /K + -ATPase α-subunit occurred in chloride cells. It has also been demonstrated that the intensity of immunoreactivity for the Na + /K + -ATPase α- subunit in branchial chloride cells increases during seawater adaptation (Uchida et al. 1996) and smoltification (Ura et al. 1997) in salmonids. The increase in the intensity of immunoreactivity correlated with an elevation in the activity of the enzyme (Uchida et al. 1996; Ura et al. 1997). Therefore, immunolocalization of the Na + /K + -ATPase α-subunit may be one of the most useful indicators of hypo-osmoregulatory function in branchial chloride cells.

3032 T. SHIKANO AND Y. FUJIO Current ultrastructural studies have suggested the existence of distinct types of chloride cell in the gills of freshwater teleosts (for a review, see Pisam and Rambourg, 1991). Pisam et al. (1987) identified two types of chloride cell, α- and β-chloride cells, on the primary lamellae of freshwater guppies on the basis of their location, shape and ultrastructure. The α-chloride cells were characterized as elongated cells located at the base of the secondary lamellae in close contact with the basement membrane of the pillar capillary, whereas the β-chloride cells were characterized as ovoid cells observed in the interlamellar regions facing the central venous sinus. Pisam et al. (1987) reported that the α- chloride cells hypertrophied while the β-chloride cells degenerated when guppies were transferred to sea water, suggesting that the α-chloride cells play a significant role in seawater adaptation. In addition to the primary lamellar chloride cells, the secondary lamellae of some freshwater teleosts also contain chloride cells (Laurent and Dunel, 1980; Perry and Wood, 1985; Avella et al. 1987; Perry and Laurent, 1989; Laurent et al. 1994; Witters et al. 1996). Laurent and Dunel (1980) reported that the secondary lamellar chloride cells proliferated when rainbow trout were adapted to deionized water and suggested that these cells play a significant role in hypo-osmotic environments, presumably acting as sites for ion uptake. Although many investigators have examined chloride cell subtypes, there is little useful information about the functional identification of these cell types. Chum salmon Oncorhynchus keta fry have a strong preference for sea water and migrate from river to sea soon after yolk absorption. The transition from a freshwater hyperosmoregulator to a seawater hypo-osmoregulator necessitates an alteration in osmoregulatory mechanisms. Taking advantage of their high osmoregulatory capacity, the present study examines the immunolocalization of the Na + /K + -ATPase α-subunit in the branchial epithelium of chum salmon fry during acclimation to brackish water and reintroduction to fresh water. A functional identification of chloride cell subtypes in the gills is discussed in relation to hypo-osmoregulatory ability. Materials and methods Animals Eyed-stage embryos of chum salmon Oncorhynchus keta were obtained from the Hirose-natori Salmon Hatchery (Miyagi, Japan) in December 1997. They were transported to our laboratory and reared in 180 l aquaria with recirculating fresh water at 10 C. Approximately 200 individuals were maintained per aquarium. After yolk absorption, the fish were fed twice daily with trout pellets. Fry, 40 60 days after hatching and weighing 0.3 0.5 g, were used in the present study. Seawater and freshwater acclimation Chum salmon fry were acclimated from fresh water to 25 brackish water prepared from artificial sea water (Aquasalz, Nissei, Japan) at 10 C. To examine changes in salinity tolerance and the immunolocalization of Na + /K + -ATPase in the gills, fish were sampled 0, 1, 2, 3, 5 and 7 days after transfer to 25 brackish water. Chum salmon fry were also acclimated from fresh water to 25 brackish water for 7 days and then reintroduced to fresh water at 10 C. Fish were sampled 1, 2, 3, 5, 7 and 14 days after reintroduction to fresh water. Salinity tolerance Changes in salinity tolerance were examined during acclimation to 25 brackish water and reintroduction to fresh water. Fish were acclimated to 25 brackish water for 0 7 days and then exposed to 30, 35, 40, 45, 50 or 55 sea water prepared from artificial sea water (Aquasalz, Nissei, Japan). Fish were also reintroduced to fresh water for 1 14 days after acclimation to 25 brackish water for 7 days and then exposed to 30, 35, 40, 45, 50 or 55 sea water. Three individuals were exposed to each salinity and survival rate was examined after 24 h. Salinity tolerance was expressed as the LD 50 of salinity (the salinity at which 50 % of the fish did not survive) calculated on the basis of the survival rate data. The experiments were repeated three times. The standard error of LD 50 was calculated from the actual variation among three replicates. Immunocytochemistry Gill arches were fixed in 4 % paraformaldehyde in 0.1 mol l 1 phosphate buffer (ph 7.2) for 8 h at 4 C. The tissue samples were dehydrated in ethanol and embedded in paraffin wax. Serial sections (5 µm) were cut parallel to the long axis of the primary lamellae at right angles to the secondary lamellae or at the base of the secondary lamellae. The sections were mounted on polylysine-coated glass slides. The sections were dewaxed in xylene, hydrated in ethanol and washed with phosphate-buffered saline (PBS). For immunocytochemical staining, the avidin-biotinylated alkaline phosphatase method (Vectastain ABC-AP kit, Vector Laboratories, Inc., Burlingame, CA, USA) was used according to the manufacturer s recommendation. To reduce nonspecific staining, the sections were treated with 2 % normal goat serum in PBS for 30 min at room temperature (20 22 C). The sections were incubated with primary antibody at a dilution of 1:1000 overnight at 4 C. The primary antibody was a rabbit polyclonal antibody raised against a synthetic peptide corresponding to part of the conserved region of at least three different isoforms of the Na + /K + -ATPase α-subunit, which are encoded by different genes designated α1, α2 and α3 (Ura et al. 1996). The amino acid sequence of the antigen was: Val- Thr-Gly-Val-Glu-Glu-Gly-Arg-Leu-Ile-Phe-Asp-Asn-Leu-Lys- Lys-Cys (Ura et al. 1996). The sites of immunoreactivity were visualized by incubating the sections successively with a solution of biotinylated anti-rabbit IgG for 30 min, ABC-AP reagent for 60 min and alkaline phosphatase substrate solution for 4 min. The immunostained sections were counterstained with Mayer s haematoxylin. As a negative control, the primary antibody was replaced with non-immune rabbit serum at the

LD 50 of salinity ( ) Immunolocalization of Na + /K + -ATPase in salmon fry 3033 same dilution. No staining was detected when non-immune rabbit serum was used instead of the primary antibody. The sections were observed with a light microscope equipped with Nomarski differential interference optics. Since chloride cells are large and spherical, the numbers of immunopositive and immunonegative large spherical cells could be measured. The cells could easily be distinguished from respiratory epithelial cells on the basis of their appearance. Cell numbers were counted in sections cut parallel to the long axis of the primary lamellae at right angles to the secondary lamellae. From each individual, the number of large spherical cells was counted in approximately 120 interlamellar regions on the primary lamellae (approximately 450 cells) in six separate and randomly chosen areas and at the free surface on 200 secondary lamellae in six separate and randomly chosen areas. Cell numbers on the primary and secondary lamellae are expressed as the number of cells in 100 interlamellar regions on the primary lamellae and in 100 secondary lamellae in the gill sections, respectively. Values for each experimental group are presented as the means ± S.E.M. of five individuals. Statistics Statistical analysis was performed using one-way analysis of variance (ANOVA) followed by Fisher s protected least significant difference (PLSD) multiple-comparison test. Results Salinity tolerance The LD 50 of salinity increased significantly from 35.8 to 45.2 during acclimation to 25 brackish water over 7 days (P<0.01) and decreased to 34.8 during reintroduction to fresh water over 14 days (Fig. 1). The LD 50 of control fish did not change over the course of the experiment (34.2 36.2 ) and did not differ from that of the experimental group at the beginning of the acclimation experiment. Immunolocalization of Na + /K + -ATPase Immunocytochemical staining of a sagittal section of the gill in a freshwater fish is shown in Fig. 2A. Strong immunoreactivity appeared in the large spherical cells located on the free surface of the primary lamellae mainly close to the base of the secondary lamellae. In addition, immunoreactivity was found in the large spherical cells located on the secondary lamellae at some distance from their point of contact with the primary lamellae. In contrast, the large spherical cells located near the central venous sinus at the base of the secondary lamellae and in the interlamellar regions showed little or no immunoreactivity. Deeper cell layers of the primary lamellae were examined (Fig. 2B). The large spherical cells near the central venous sinus were rarely immunoreactive. Following acclimation of the freshwater fish to 25 brackish water, the number of immunopositive large spherical cells on the primary lamellae gradually increased, whereas the number of immunonegative large spherical cells decreased (Fig. 2C,E). The large spherical cells located on 46 44 42 40 38 36 34 32 0 1 2 3 5 7 8 9 10 12 14 21 Time after transfer (days) the primary lamellae near the central venous sinus gradually developed immunoreactivity and became strongly immunoreactive (Fig. 2D,F). Following reintroduction to fresh water, the number of immunopositive large spherical cells on the primary lamellae decreased (Fig. 3A,C), whereas the number of immunonegative large spherical cells on the primary lamellae near the central venous sinus increased (Fig. 3B,D). Fig. 4 shows changes in numbers of large spherical cells located on the primary and secondary lamellae following acclimation to 25 brackish water and reintroduction to fresh water. In freshwater fish, the number of large spherical cells located on the primary lamellae at the base of the secondary lamellae was 250.5 per 100 interlamellar regions, and the number of these cells in the interlamellar regions was 125.8 per 100 interlamellar regions. No significant changes were observed in the cell numbers during acclimation to 25 brackish water or after reintroduction to fresh water. The number of large spherical cells located on the secondary lamellae, all of which were strongly immunoreactive, decreased significantly from 37.5 to 7.5 per 100 secondary lamellae during acclimation to 25 brackish water over 7 days (P<0.01). The cell number remained significantly lower than that in freshwater fish for 7 days after reintroduction to fresh water (P<0.01) but increased to 35.8 per 100 secondary lamellae 14 days after reintroduction to fresh water. Although no changes were observed in the number of large spherical cells located on the primary lamellae, their immunoreactivity changed significantly during the acclimation experiment (Fig. 5). In regions of primary lamellae at the base of the secondary lamellae, the number of immunopositive large spherical cells increased significantly from 210.5 to 247.0 per 100 interlamellar regions, whereas the number of 25 brackish water Fresh water Fig. 1. Changes in the LD 50 of salinity following acclimation to 25 brackish water and reintroduction to fresh water (filled columns). Fish were acclimated to 25 brackish water at day 0 and reintroduced to fresh water at day 7. Control fish were left in fresh water (open columns). Asterisks indicate values significantly different from the initial value; P<0.01. Values are expressed as means ± S.E.M. The standard error was calculated from the actual variation between three replicates.

3034 T. SHIKANO AND Y. FUJIO immunonegative large spherical cells decreased significantly from 40.0 to 3.0 per 100 interlamellar regions during the acclimation to 25 brackish water over 7 days (P<0.01). The number of immunopositive large spherical cells decreased to 204.5, whereas the number of immunonegative large spherical cells increased to 39.5 per 100 interlamellar regions during reintroduction to fresh water over 14 days. In the primary lamellae at the interlamellar regions, the number of immunopositive large spherical cells increased significantly from 69.0 to 119.0 per 100 interlamellar regions, whereas the Fig. 2. Sagittal sections of the gills (A,C,E) and level surface sections of the primary lamellae (B,D,F) stained with an antibody to Na + /K + - ATPase in fish maintained in fresh water (A,B), 5 days after acclimation to 25 brackish water (C,D) and 7 days after acclimation to 25 brackish water (E,F). In fresh water, strong immunoreactivity appeared in the large spherical cells located on the primary lamellae (arrowheads) and in the large spherical cells located on the secondary lamellae (asterisk). However, the large spherical cells located near the central venous sinus showed little or no immunoreactivity (arrows). A comparison of B, D and F shows that the large spherical cells located near the central venous sinus developed immunoreactivity following acclimation to 25 brackish water. The bases of the secondary lamellae appear as vertical stripes in B, D and F. Scale bars, 30 µm.

Immunolocalization of Na + /K + -ATPase in salmon fry 3035 Fig. 3. Sagittal sections of the gills (A,C) and level surface sections of the primary lamellae (B,D) stained with an antibody to Na + /K + -ATPase in fish 5 days after reintroduction to fresh water (A,B) and 14 days after reintroduction to fresh water (C,D). Fish were acclimated to 25 brackish water for 7 days and then reintroduced to fresh water. Strong immunoreactivity appeared in the large spherical cells located on the primary lamellae (arrowheads) and in the large spherical cells located on the secondary lamellae (asterisk). The large spherical cells located near the central venous sinus showed a decrease in immunoreactivity following reintroduction to fresh water (B,D), with some cells showing little or no immunoreactivity (arrows in A and C). The bases of the secondary lamellae appear as vertical stripes in B and D. Scale bars, 30 µm. number of immunonegative large spherical cells decreased significantly from 57.0 to 2.0 per 100 interlamellar regions during acclimation to 25 brackish water over 7 days (P<0.01). Over the 14 days following reintroduction to fresh water, the number of immunopositive large spherical cells decreased to 69.0, whereas the number of immunonegative large spherical cells increased to 55.0 per 100 interlamellar regions. Relationship between salinity tolerance and immunolocalization Fig. 6 shows the relationships between the LD 50 of salinity and the numbers of large spherical cells located on the primary lamellae, of large spherical cells located on the secondary lamellae, of immunopositive large spherical cells located on the primary lamellae and of immunonegative large spherical cells located on the primary lamellae during acclimation to 25 brackish water and reintroduction to fresh water. Although the LD 50 of salinity was not correlated with the total number of large spherical cells located on the primary lamellae, it was correlated positively with the number of immunopositive large spherical cells and negatively with the number of the immunonegative large spherical cells. A negative correlation was also observed between the LD 50 of salinity and the number of large spherical cells located on the secondary lamellae. Discussion Recently, we reported that the expression of Na + /K + -ATPase differed significantly between the two types of chloride cell, the α- and β-chloride cells, in the primary lamellae of freshwater guppies, using an immunocytochemical method (Shikano and Fujio, 1998). Our findings revealed that the expression of Na + /K + -ATPase was strong in α-chloride cells and absent or weak in β-chloride cells, indicating functional differences between the two cell types. In the present study of freshwater chum salmon fry gills, strong immunoreactivity was associated

Number of cells per 100 IR Number of cells per 100 IR or SL 3036 T. SHIKANO AND Y. FUJIO Fig. 4. Changes in the numbers of large spherical cells located on the primary lamellae at the base of the secondary lamellae (filled columns), on the primary lamellae at the interlamellar regions (grey columns) and on the secondary lamellae (open columns) following acclimation to 25 brackish water and reintroduction to fresh water. Fish were acclimated to 25 brackish water on day 0 and reintroduced to fresh water on day 7. Cell numbers are expressed as the number of cells in 100 interlamellar regions on the primary lamellae (IR) or on 100 secondary lamellae (SL) in the gill sections. Asterisks indicate values significantly different from the initial value; *P<0.05, P<0.01. Values are expressed as means ± S.E.M. (N=5). 250 200 150 100 50 0 * * 0 1 2 3 5 7 8 9 10 12 14 21 25 brackish water Fresh water Time after transfer (days) with the large spherical cells located on the free surface of the primary lamellae, mainly at the base of the secondary lamellae, and with the large spherical cells located on the secondary lamellae. Since Na + /K + -ATPase is mainly located within chloride cells on the basolateral tubular network (Karnaky et al. 1976b; Hootman and Philpott, 1979), the immunopositive large spherical cells may correspond to chloride cells. In this context, Witters et al. (1996) used immunolocalization of the Na + /K + - ATPase α-subunit to demonstrate that the Na + /K + -ATPasepositive cells corresponded to cells labelled with dimethylamino-styrylmethyl-pyridiniumiodine, a fluorescent mitochondrial probe for chloride cells. In the present study, the large spherical cells located near the central venous sinus at the base of the secondary lamellae and in the interlamellar regions showed little or no immunoreactivity in a freshwater environment. Because most of these cells developed immunoreactivity for Na + /K + -ATPase in fish acclimated to brackish water, the cells may have the same ultimate function as chloride cells but may have different properties in a freshwater environment. As reported by Pisam et al. (1987) in the guppy, Prunet et al. (1994) reported that, in the Atlantic salmon, the α-chloride cells were mainly observed at the base of the secondary lamellae whereas the β-chloride cells were mainly observed in the interlamellar regions facing the central venous sinus. This suggests that the immunopositive and immunonegative large spherical cells on the primary lamellae of freshwater chum salmon fry may correspond to the α- and β-chloride cells, respectively. 250 200 A 150 100 Fig. 5. Changes in the numbers of immunopositive large spherical cells (filled columns) and immunonegative large spherical cells (open columns) located on the primary lamellae at the base of the secondary lamellae (A) and at the interlamellar regions (B) following acclimation to 25 brackish water and reintroduction to fresh water. Fish were acclimated to 25 brackish water on day 0 and reintroduced to fresh water on day 7. Cell numbers are expressed as the number of cells in 100 interlamellar regions on the primary lamellae (IR) in gill sections. Asterisks indicate values significantly different from the initial value; *P<0.05, P<0.01. Values are expressed as means ± S.E.M. (N=5). 50 0 150 100 50 0 0 1 2 3 5 7 8 9 10 12 14 21 B * * * 0 1 2 3 5 7 8 9 10 12 14 21 25 brackish water Fresh water Time after transfer (days)

Number of cells per 100 IR Number of cells per 100 IR Number of cells per 100 IR Number of cells per 100 SL Immunolocalization of Na + /K + -ATPase in salmon fry 3037 390 385 380 375 370 365 A r=0.272 N=12 P>0.05 360 34 36 38 40 42 44 46 B 40 y=122.669 2.559x r= 0.847 35 N=12 30 P<0.01 25 20 15 10 5 0 34 36 38 40 42 44 46 Fig. 6. Relationships between the LD 50 of salinity and the number of large spherical cells located on the primary lamellae (IR) (A), the number of large spherical cells located on the secondary lamellae (SL) (B), the number of immunopositive large spherical cells located on the primary lamellae (C) and the number of the immunonegative large spherical cells located on the primary lamellae (D) during acclimation to 25 brackish water and reintroduction to fresh water. 360 340 320 300 280 C y=6.874+7.505x r=0.752 N=12 P<0.01 260 34 36 38 40 42 44 46 LD 50 of salinity ( ) 120 100 80 60 40 20 D y=353.784 7.133x r= 0.761 N=12 P<0.01 0 34 36 38 40 42 44 46 LD 50 of salinity ( ) In addition to their presence on the primary lamellae, immunopositive large spherical cells are found on the secondary lamellae of freshwater chum salmon fry, and these cells may correspond to the secondary lamellar chloride cells. Many investigators have suggested a role for the secondary lamellar chloride cells in ion uptake in hypo-osmotic environments on the basis of observations that the cells degenerated following transfer from fresh water to sea water and hypertrophied following transfer to deionized water (Laurent and Dunel, 1980; Perry and Wood, 1985; Avella et al. 1987; Perry and Laurent, 1989; Laurent et al. 1994). Morphological observations showed that the immunopositive large spherical cells on the secondary lamellae degenerated when chum salmon fry were transferred from fresh water to brackish water and reappeared when they were reintroduced to fresh water. However, immunocytochemical observations of freshwater chum salmon fry showed that not only the large spherical cells on the secondary lamellae but also many large spherical cells on the primary lamellae were strongly immunoreactive for Na + /K + -ATPase, which plays a crucial role in ion transport in the gills (Payan et al. 1984; Marshall, 1995). Thus, the immunopositive cells located on the primary and secondary lamellae may play a significant role in ion uptake in fresh water. The possible role of the primary lamellar chloride cells in ion uptake was suggested by studies in the guppy, in which the secondary lamellar chloride cells were absent in both freshwater and seawater environments (Shikano and Fujio, 1998). Acclimation of chum salmon fry from fresh water to brackish water caused the development of an ability to hypoosmoregulate. Morphological observations in the gills showed that there were no changes in the numbers of large spherical cells on the primary lamellae. However, immunocytochemical observations revealed that, although the large spherical cells near the central venous sinus showed little or no immunoreactivity for Na + /K + -ATPase in the freshwater environment, they developed immunoreactivity following acclimation to brackish water. Thus, most of the large spherical cells on the primary lamellae showed strong immunoreactivity 7 days after acclimation to brackish water. In contrast, following reintroduction to fresh water, the large spherical cells near the central venous sinus showed a decrease in immunoreactivity and the fish showed a reduction in their ability to hypo-osmoregulate. Uchida et al. (1996) reported that Na + /K + -ATPase activity in the gills correlated with the intensity of immunoreactivity for the Na + /K + -ATPase α- subunit in branchial chloride cells. Because the ultimate driving force for chloride secretion is the electrochemical gradient for Na + established by Na + /K + -ATPase (Payan et al. 1984; Marshall, 1995), the significant increase in the number of immunopositive cells may be a key point in the development of hypo-osmoregulatory function in the gills. The importance of the immunopositive large spherical cells on the primary lamellae for hypo-osmoregulation was directly indicated by the fact that, although the ability to hypo-osmoregulate did not correlate with the total number of large spherical cells, it correlated closely with the number of immunopositive large spherical cells.

3038 T. SHIKANO AND Y. FUJIO The present study suggests the existence of functionally different types of chloride cell on the primary lamellae. One of the most important problems in chloride cell biology is that of identifying the chloride cell subtypes that are found in the primary lamellae of freshwater teleosts. An important finding from our immunocytochemical study is that immunonegative large spherical cells are observed on the primary lamellae near the central venous sinus, where undifferentiated stem cells and young developing chloride cells can also be observed. Some investigators have reported that chloride cells develop from undifferentiated cells in the deeper cell layers of the primary lamellae (Conte and Lin, 1967; Chretien and Pisam, 1986). Chretien and Pisam (1986) reported that, in guppies, in which both α- and β-chloride cells have been observed (Pisam et al. 1987), young developing chloride cells occur in the interlamellar region facing the central venous sinus and that these cells elongate towards the free surface of the primary lamellae concomitant with the development of their mitochondria and tubular system to become mature differentiated chloride cells. Because the Na + /K + -ATPase is located in the tubular system (Karnaky et al. 1976b; Hootman and Philpott, 1979), the development of this system may be accompanied by an increase in the abundance of Na + /K + - ATPase in the cells. Judging from the immunoreactivity for Na + /K + -ATPase and its location, in both our present study using chum salmon and a previous study of guppies (Shikano and Fujio, 1998), the immunopositive, or α-chloride, cells may be mature differentiated chloride cells, whereas the immunonegative, or β-chloride, cells may be young developing chloride cells. Another important problem in chloride cell biology is why seawater-adapted fish have only one chloride cell type together with accessory cells. Shirai and Utida (1970) reported that, although two types of chloride cell were observed in the primary lamellae of the Japanese eel Anguilla japonica, one of the cell types developed basolateral membrane tubular extensions like those of the other cell type when the fish were transferred to sea water. Doyle and Epstein (1972) reported that an elevation in gill Na + /K + -ATPase activity correlated with the maturation of young chloride cells rather than with an increase in cell numbers. Our immunocytochemical observations in chum salmon fry show that the large spherical cells near the central venous sinus develop immunoreactivity for Na + /K + - ATPase, suggesting that young developing chloride cells are transformed into mature differentiated chloride cells. The morphological and functional changes we have observed indicate that seawater adaptation causes all primary lamellar chloride cells to become highly activated and, therefore, only one cell type, mature differentiated chloride cells, is observed in the primary lamellae of seawater-adapted fish. Accessory cells exhibit many of the fine structural features characteristic of chloride cells, such as numerous mitochondria and a tubular system (Hootman and Philpott, 1980); however, the accessory cells lack appreciable Na + /K + -ATPase activity (Hootman and Philpott, 1980). Thus, Hootman and Philpott (1980) suggested that the accessory cells represent a population of partially differentiated chloride cells. The characteristics of the accessory cells may resemble those of immunonegative chloride cells in chum salmon fry and of β-chloride cells in the guppy. Further research at the ultrastructural level should clarify this issue. In addition to the immunonegative large spherical cells on the primary lamellae, immunopositive secondary lamellar cells disappeared following acclimation of freshwater chum salmon fry to brackish water. Previous work by Laurent et al. (1994) has shown that the secondary lamellar chloride cells are derived from undifferentiated cells in the primary lamellae and migrate to the secondary lamellae. A significant change in the cell turnover rate has been reported in chloride cells following seawater adaptation. It has been reported that the rate of turnover of chloride cells in the primary lamellae was three times greater in sea water than in fresh water in chum salmon (Uchida and Kaneko, 1996) and in guppies (Chretien and Pisam, 1986), indicating that the lifetime of chloride cells is significantly shorter in sea water than in fresh water. Our immunocytochemical observations during the reintroduction of fish from brackish water to fresh water indicated that the number of immunopositive secondary lamellar cells increased significantly after the level of immunoreactivity in the large spherical cells on the primary lamellae had been completely transformed into that of freshwater fish. Furthermore, the appearance of the immunopositive secondary lamellar cells correlated negatively with the hypo-osmoregulatory ability of the fish during acclimation to brackish water and reintroduction to fresh water. These results suggest that the migration of chloride cells to the secondary lamellae occurs when activation of the primary lamellar chloride cells is low. The higher level of activation of the chloride cells on the primary lamellae of seawater-adapted fish might prevent the migration of secondary lamellar chloride cells, although more research is required to clarify this issue. In conclusion, the present study suggests the existence of two functionally different types of chloride cell, immunopositive and immunonegative chloride cells, on the primary lamellae of freshwater chum salmon fry on the basis of the immunolocalization of the Na + /K + -ATPase α-subunit. The important role of the immunopositive large spherical cells in hypo-osmoregulation was directly indicated by the strong positive correlation between the number of these cells and the hypo-osmoregulatory ability of the fish during acclimation to brackish water and reintroduction to fresh water. We conjecture that the immunopositive large spherical cells are mature differentiated chloride cells, whereas the immunonegative large spherical cells are young developing chloride cells. The development of immunoreactivity for Na + /K + -ATPase in the young chloride cells may be one of the most important factors in the development of hypoosmoregulatory ability by chum salmon fry. We express our gratitude to Professor K. Yamauchi, Hokkaido University, Japan, for kindly providing the antibody

Immunolocalization of Na + /K + -ATPase in salmon fry 3039 to Na + /K + -ATPase. We thank the staff at the Hirose-natori Salmon Hatchery for supplying the chum salmon eggs. References AVELLA, M., MASONI, A., BORNANCIN, M. AND MAYER-GOSTAN, N. (1987). Gill morphology and sodium influx in the rainbow trout (Salmo gairdneri) acclimated to artificial freshwater environments. J. exp. Zool. 241, 159 169. CHRETIEN, M. AND PISAM, M. (1986). Cell renewal and differentiation in the gill epithelium of fresh- or salt-water-adapted euryhaline fish as revealed by [ 3 H]-thymidine radioautography. Biol. Cell 56, 137 150. CONTE, F. P. AND LIN, D. Y. (1967). Kinetics of cellular morphogenesis in gill epithelium during sea water adaptation of Oncorhynchus (Walbaum). Comp. Biochem. Physiol. 23, 945 957. DOYLE, W. L. AND EPSTEIN, F. H. (1972). Effects of cortisol treatment and osmotic adaptation on the chloride cells in the eel, Anguilla rostrata. Cytobiologie 6, 58 73. EPSTEIN, F. H., SILVA, P. AND KORMANIK, G. (1980). Role of Na K- ATPase in chloride cell function. Am. J. Physiol. 238, R246 R250. FOSKETT, J. K. AND SCHEFFEY, C. (1982). The chloride cell: Definitive identification as the salt secretory cell in teleost. Science 215, 164 166. HOOTMAN, S. R. AND PHILPOTT, C. W. (1979). Ultracytochemical localization of Na + /K + -activated ATPase in chloride cells from the gills of euryhaline teleost. Anat. Rec. 193, 99 130. HOOTMAN, S. R. AND PHILPOTT, C. W. (1980). Accessory cells in teleost branchial epithelium. Am. J. Physiol. 238, R199 R206. KAMIYA, M. AND UTIDA, S. (1968). Changes in activity of sodium potassium-activated adenosinetriphosphatase in gills during adaptation of the Japanese eel to sea water. Comp. Biochem. Physiol. 26, 675 685. KAMIYA, M. AND UTIDA, S. (1969). Sodium potassium-activated adenosinetriphosphatase activity in gills of freshwater, marine and euryhaline teleosts. Comp. Biochem. Physiol. 31, 671 674. KARNAKY, K. J., ERNST, S. A. AND PHILPOTT, C. W. (1976a). Teleost chloride cell. I. Response of pupfish Cyprinodon variegatus gill Na/K-ATPase and chloride cell fine structure to various high salinity environments. J. Cell Biol. 70, 144 156. KARNAKY, K. J., KINTER, L. B., KINTER, W. B. AND STIRLING, C. E. (1976b). Teleost chloride cell. II. Autoradiographic localization of gill Na/K-ATPase in killifish Fundulus heteroclitus adapted to low and high salinity environments. J. Cell Biol. 70, 157 177. KAWAKAMI, K., NOGUCHI, S., NODA, M., TAKAHASHI, H., OHTA, T., KAWAMURA, M., NOJIMA, H., NAGANO, K., HIROSE, T., INAYAMA, S., HAYASHIDA, H., MIYATA, T. AND NUMA, S. (1985). Primary structure of the α-subunit of Torpedo california (Na + +K + )-ATPase deduced from c-dna sequence. Nature 316, 733 736. LAURENT, P. AND DUNEL, S. (1980). Morphology of gill epithelia in fish. Am. J. Physiol. 238, R147 R159. LAURENT, P., DUNEL-ERB, S., CHEVALIER, C. AND LIGNON, J. (1994). Gill epithelial cells kinetics in freshwater teleost, Oncorhynchus mykiss during adaptation to ion-poor water and hormonal treatment. Fish Physiol. Biochem. 13, 353 370. LEBOVITZ, R. M., TAKEYASU, K. AND FAMBROUGH, D. M. (1989). Molecular characterization and expression of the (Na + +K + )- ATPase alpha-subunit in Drosophila melanogaster. EMBO J. 8, 193 202. MARSHALL, W. S. (1995). Transport processes in isolated teleost epithelia: Opercular epithelium and urinary bladder. In Cellular and Molecular Approaches to Fish Ionic Regulation (ed. C. M. Wood and T. J. Shuttleworth), pp. 1 23. New York: Academic Press. MARSHALL, W. S., BRYSON, S. E., DARLING, P., WHITTEN, C., PATRICK, M., WILKIE, M., WOOD, C. M. AND BUCKLAND-NICKS, J. (1997). NaCl transport and ultrastructure of opercular epithelium from a freshwater-adapted euryhaline teleost, Fundulus heteroclitus. J. exp. Zool. 277, 23 37. MCCORMICK, S. D. (1995). Hormonal control of gill Na + /K + -ATPase and chloride cell function. In Cellular and Molecular Approaches to Fish Ionic Regulation (ed. C. M. Wood and T. J. Shuttleworth), pp. 285 385. New York: Academic Press. PAYAN, P., GIRARD, J. P. AND MAYER-GOSTAN, N. (1984). Branchial ion movements in teleosts: The roles of respiratory and chloride cells. In Fish Physiology, vol. XB (ed. W. S. Hoar and D. J. Randall), pp. 325 388. New York: Academic Press. PERRY, S. F. AND LAURENT, P. (1989). Adaptational responses of rainbow trout to lowered external NaCl concentration: contribution of the branchial chloride cell. J. exp. Biol. 147, 147 168. PERRY, S. F. AND WOOD, C. M. (1985). Kinetics of branchial calcium uptake in the rainbow trout: effects of acclimation to various external calcium levels. J. exp. Biol. 116, 411 433. PHILPOTT, C. W. (1980). Tubular system membranes of teleost chloride cells: Osmotic response and transport sites. Am. J. Physiol. 238, R171 R184. PISAM, M. (1981). Membranous systems in the chloride cell of teleostean fish gill; their modifications in response to the salinity of the environment. Anat. Rec. 200, 401 414. PISAM, M., CAROFF, A. AND RAMBOURG, A. (1987). Two types of chloride cells in the gill epithelium of a freshwater-adapted euryhaline fish: Lebistes reticulatus, their modifications during adaptation to salt water. Am. J. Anat. 79, 40 50. PISAM, M. AND RAMBOURG, A. (1991). Mitochondria-rich cells in the gill epithelium of teleost fish: An ultrastructural approach. Int. Rev. Cytol. 130, 191 232. PRUNET, P., PISAM, M., CLAIREAUX, J. P., BOEUF, G. AND RAMBOURG, A. (1994). Effects of growth hormone on gill chloride cells in juvenile Atlantic salmon (Salmo salar). Am. J. Physiol. 266, R850 R857. SHIKANO, T. AND FUJIO, Y. (1998). Immunolocalization of Na + /K + - ATPase and morphological changes in two types of chloride cells in the gill epithelium during seawater and freshwater adaptation in a euryhaline teleost, Poecilia reticulata. J. exp. Zool. 281, 80 89. SHIRAI, N. AND UTIDA, S. (1970). Development and degeneration of chloride cell during seawater and freshwater adaptation of the Japanese eel Anguilla japonica. Z. Zellforsch. 103, 247 264. SHULL, G. E., SCHWARTZ, A. AND LINGREL, J. B. (1985). Amino-acid sequence of the catalytic subunit of the (Na + +K + )-ATPase deduced from a complementary DNA. Nature 316, 691 695. TAKEYASU, K., TAMKUN, M. M., RENAUD, K. J. AND FAMBROUGH, D. M. (1988). Ouabain sensitive (Na + +K + )-ATPase activity expressed in mouse L cells by transfection with DNA encoding the α-subunit of an avian sodium pump. J. biol. Chem. 263, 4347 4354. THOMSON, A. J. AND SARGENT, J. R. (1977). Changes in the levels of chloride cells and (Na + +K + )-dependent ATPase in the gills of yellow and silver eels adapting to sea water. J. exp. Zool. 200, 33 40. UCHIDA, K. AND KANEKO, T. (1996). Enhanced chloride cell turnover in the gills of chum salmon fry in seawater. Zool. Sci. 13, 655 660.

3040 T. SHIKANO AND Y. FUJIO UCHIDA, K., KANEKO, T., YAMAUCHI, K. AND HIRANO, T. (1996). Morphometrical analysis of chloride cell activity in the gill filaments and lamellae and changes in Na + /K + -ATPase activity during seawater adaptation in chum salmon fry. J. exp. Zool. 276, 193 200. URA, K., MIZUNO, S., OKUBO, T., CHIDA, Y., MISAKA, N., ADACHI, S. AND YAMAUCHI, K. (1997). Immunohistochemical study on changes in gill Na + /K + -ATPase α-subunit during smoltification in the wild masu salmon, Oncorhynchus masou. Fish Physiol. Biochem. 17, 397 403. URA, K., SOYANO, K., OMOTO, N., ADACHI, S. AND YAMAUCHI, K. (1996). Localization of Na + /K + -ATPase in tissues of rabbit and teleosts using an antiserum directed against a partial sequence of the α-subunit. Zool. Sci. 13, 219 227. WITTERS, H., BERCKMANS, P. AND VANGENECHTEN, C. (1996). Immunolocalization of Na + /K + -ATPase in the gill epithelium of rainbow trout, Oncorhynchus mykiss. Cell Tissue Res. 259, 575 585. ZADUNAISKY, J. A. (1984). The chloride cell: The active transport of chloride and the paracellular pathways. In Fish Physiology, vol. XB (ed. W. S. Hoar and D. J. Randall), pp. 129 176. New York: Academic Press.