Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The fish branchial epithelium : an immunological approach to ion transport protein localization Wilson, Jonathan Mark 1999

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-487377.pdf [ 17.49MB ]
Metadata
JSON: 831-1.0089799.json
JSON-LD: 831-1.0089799-ld.json
RDF/XML (Pretty): 831-1.0089799-rdf.xml
RDF/JSON: 831-1.0089799-rdf.json
Turtle: 831-1.0089799-turtle.txt
N-Triples: 831-1.0089799-rdf-ntriples.txt
Original Record: 831-1.0089799-source.json
Full Text
831-1.0089799-fulltext.txt
Citation
831-1.0089799.ris

Full Text

T H E FISH BRANCHIAL EPITHELIUM: A N I M M U N O L O G I C A L A P P R O A C H T O I O N T R A N S P O R T P R O T E I N L O C A L I Z A T I O N B y "Jonathan Mark Wilson B . S c , University of British Columbia, 1993 M.Sc . , University of British Columbia, 1995 A THESIS S U B M I T E D I N P A R T I A L F U L F I L M E N T O F T H E R E Q U R E M E N T S F O R T H E D E G R E E OF D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E S T U D I E S D E P A R T M E N T OF Z O O L O G Y We accept this thesis as conforming J^rTfTSTequired standard T H E U N I V - E R S I T Y OF B R I T I S H C O L U M B I A 1999 © Jonathan Mark Wilson, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of 2 O O L O ey The University of British Columbia Vancouver, Canada Date / 7 + u D t C /79<? DE-6 (2/88) ABSTRACT In my thesis the cellular and subcellular distributions of various ion transport proteins were determined in the fish branchial epithelium using an immunological approach employing non-homologous antibodies. The viabilities of current models of transepithelial ion transport in the gills of freshwater, and seawater fishes and an unusual amphibious air-breathing fish were tested. M y thesis work primarily addresses the mechanisms involved in the translocation of N a + and Cl" and their counterions, H + and HCO3", respectively and of N H / . In freshwater fishes (tilapia: Oreochromis mossambicus; trout: Oncorhynchus mykiss) the vH + -ATPase colocalizes apically with an epithelial N a + channel (ENaC)-like protein, however, the distribution of these proteins is restricted to pavement cells (PVCs) in tilapia while being found in both P V C s and mitochondria-rich ( M R ) cells in trout. A n N a + / H + exchanger-like (NHE) protein is also identified in the gi l l of the freshwater tilapia. In tilapia and coho salmon {Oncorhynchus kisutch), an apical band 3-like C I V H C C V anion exchanger (AE) is localized to M R cells demonstating the presence of a freshwater-type chloride cell (involved in C l " uptake). In seawater fishes, the chloride cell has been well characterized as the site of active C l " elimination. In my observations of seawater fishes (O.kisutch; turbot: Scophthamus maximus) and the brackishwater, air-breathing mudskipper fish (Periophthalmodon schlosseri), ion transport proteins involved in active C l " elimination are localized to chloride cells ( N a + , K + -ATPase, Na + :K + : 2C1" cotransporter ( N K C C ) and apical cystic fibrosis transmembrane receptor (CFTR)-l ike anion channel). The A E and N H E proteins that are potentially involved in acid-base regulation are localized to the chloride cell apical crypt and accessory cell, respectively, in the stenohaline turbot. However, in the euryhaline coho the apical A E of the freshwater M R cell is not observed in the seawater acclimated fish. i i In the mudskipper fish, I have shown that N H E , N a + , K + - A T P a s e , and carbonic anhydrase in gi l l M R cells contribute to active ammonia ( N H ^ ) excretion in pharmacological studies. The branchial epithelium contains an unusual abundance of M R cells and immunolocalization studies show the presence of N H E - 2 and 3-like isoforms, C F T R , and C A associated with the apical crypt and N a + , K + - A T P a s e and N K C C with the tubular system of these cells. A n d so it goes.... i i i TABLE OF CONTENTS A B S T R A C T i i T A B L E OF C O N T E N T S iv LIST OF T A B L E S v i i LIST OF F I G U R E S v i i i L IST OF A B B R E V I A T I O N S x i i A C K N O W L E D G E M E N T S xiv C H A P T E R 1. G E N E R A L I N T R O D U C T I O N 1 1.1 Basic description of the fish gi l l 1 1.1.1 Branchial epithelium 3 1.1.2 Mitochondria-rich cells 4 1.1.3 Pavement cells 6 1.2 The importance of environment on ion transport protein expression 6 1.2.1 Freshwater N a C l uptake models 7 1.2.2 Seawater N a C l elimination model 8 1.2.3 Models of gi l l N H / elimination 8 1.3 Other iono acid-base regulatory epithelia 9 1.4 Ion transport protein properties 11 1.4.1 N a + , K + - A T P a s e 12 1.4.2 vH + -ATPase 14 1.4.3 N H E 16 1.4.4 E N a C 17 1.4.5 A E 18 1.4.6 N K C C 19 1.4.7 C F T R 20 1.4.8 Others 21 1.5 In vitro surrogate models 21 1.6 A n d so it goes 22 C H A P T E R 2. G E N E R A L M A T E R I A L A N D M E T H O D S 24 2.1 Tissue fixation for immunolabeling 24 2.2 Immunocytochemistry 25 2.2.1 Immunofluorescence microscopy 25 2.2.2 Immunoelectron microscopy 25 2.3 Antigen retrieval 26 2.4 S D S - P A G E and Western Analysis 27 2.5 Antibodies 28 2.5.1 vH + -ATPase 28 2.5.2 E N a C 29 2.5.3 N a + , K + - A T P a s e 29 2.5.4 A E 30 2.5.5 N H E 31 2.5.6 C F T R 31 2.5.7 N K C C 32 2.6 In vitro ATPase assay 32 iv C H A P T E R 3. FRESH WATER FISH GILL 34 3. S U M M A R Y 34 3.1 I N T R O D U C T I O N 35 3.1.1 Sodium uptake models 35 3.1.2 Chloride uptake model 37 3.2 M A T E R I A L A N D M E T H O D S 39 3.2.1 Animals 39 3.2.2 Tissue Sampling and Fixation 40 3.2.3 Tissue preparation for Western analysis and ATPase assay 40 3.2.4 ATPase Activity 43 3.2.5 Immunocytochemistry, Western analysis and Antibodies employed...43 3.3 R E S U L T S 43 3.3.1 Tilapia 43 3.3.2 Trout 46 3.4 D I S C U S S I O N 70 C H A P T E R 4. SEAWATER FISH GILL '. 81 4. S U M M A R Y 81 4.1 I N T R O D U C T I O N 82 4.1.1 Sodium Chloride elimination model 82 4.1.2 Mechanisms of acid-base regulation 83 4.1.3 Seawater versus freshwater 84 4.2 M A T E R I A L A N D M E T H O D S 86 4.2. i Animals 86 4.2.2 Salinity acclimation experiment 86 4.2.3 Fixation and immunolabeling 87 4.2.5 Antibodies employed in this study 87 4.2.6 Statistical Analysis 87 4.3 R E S U L T S 88 4.4 D I S C U S S I O N 105 C H A P T E R 5. MUDSKIPPER GILL 110 5. S U M M A R Y 110 5.1 I N T R O D U C T I O N I l l 5.2 M A T E R I A L A N D M E T H O D S 114 5.2.1 Animals 114 5.2.2 Measurements of ammonia fluxes in water and air 115 5.2.3 Inhibitor Studies 115 5.2.4 Surface p H measurements 116 5.2.5 Acclimation to different ambient salinities 117 5.2.6 Plasma analytical procedures 117 5.2.7 ATPase Activity 117 5.2.8 Tissue fixation for Routine T E M 118 5.2.9 Tissue fixation for immunocytochemistry 118 5.2.10 Immunocytochemistry and Antibodies Employed 118 5.2.11 Statistical Analysis 119 5.3 R E S U L T S 119 5.3.1 Ammonia fluxes and inhibitors 119 5.3.2 G i l l ultrastructure 121 5.3.3 Immunocytochemistry 125 5.4 D I S C U S S I O N 165 C H A P T E R 6 G E N E R A L D I S C U S S I O N 179 R E F E R E N C E S 187 A P P E N D I C E S 212 VI LIST of TABLES Table 3.1 Profiles of Ottawa and Vancouver tap waters in which rainbow trout were reared 41 Table 5.1 Plasma total ammonia concentrations ( m M + S E M ) in P. schlosseri exposed to 50%SW (n=6) or 2 m M NH4CI in 50% seawater in the presence of either 0.1 of 0.01 m M ouabain (n=5 and 6, respectively). The asterisk indicates a significant difference from the control value (PO.05) 128 Table 5.2 Net ammonia and urea flux rates (umol N • kg" 1 fish • h"1), and plasma [ N H / ] (mM), PNH3 (torr), and urea (mM) of P. schlosseri acclimated to salinities of 5, 15 and 25%o for 2 weeks (n = 6). Total in vitro ATPase activity (umol A D P • mg"1 protein • h"1) in crude mudskipper gi l l homogenates from fish acclimated to 5 and 25%o S W (n - 5 and 6, respectively). N a + , K + - A T P a s e (1 m M ouabain) and vH + -ATPase activities ( lOOmM KNO3) were determined from inhibitor sensitivities. L ike characters indicate no statistically significant difference. (PO.05) : -. 129 Table 5.3 Surface p H measurements of the skin and gills of Periophthalmodon schlosseri during emersion. Asterisk (*) indicates significant difference from all other groups, n = 4, P<0.05 130 v i i LIST of FIGURES FIGURE 1.1 Basic description of the teleost fish gi l l 2 FIGURE 3.1 Double labeled sections on the tilapia gi l l epithelium from efferent leading (a,c) and afferent trailing ends (b,d) of the filament using a rabbit polyclonal antibody against the A-subunit of vH + -ATPase (a,b) and mouse monoclonal antibody against the a subunit of N a + , K + - A T P a s e (c,d). Western blots of crude gi l l tissue homogenate probed with the vH + -ATPase (e) and N a + , K + -ATPase (f) antibodies 48 FIGURE 3.2 A n SDS treated section of the tilapia gi l l filament trailing edge duel labeled for A E l t (A) and N a + , K + - A T P a s e (C) using a rabbit polyclonal antibody generated against trout erythrocyte band 3 ( A E l t ) and a mouse monoclonal antibody specific of the a subunit of N a + , K + - ATPase, respectively 50 FIGURE 3.3 Western blot of lanes loaded with tilapia red blood cell homogenate (rbc), saline perfused crude gi l l tissue homogenate (gill), and loading buffer (blk) probed with the A E l t antibody 52 FIGURE 3.4 Indirect immunofluorescence labeling of pENaC in the tilapia afferent lamellar and filament epithelium and a Western blot of crude gi l l homogenate probed with the PENaC antibody 54 FIGURE 3.5 Western blot of tilapia gi l l tissue homogenate probed with antibodies generated against the a subunit of bovine E N a C (A) and biochemically purified bovine amiloride-sensitive N a + channel complex (B) 56 FIGURE 3.6 Immunolocalization of N H E 2 (B,D) in the tilapia gi l l using a rabbit polyclonal antibody. The section from the afferent area of the filament has been duel label with antibody 597 specific for N H E 2 (B) and a5 for N a + , K + - A T P a s e (A) 58 FIGURE 3.7 Western blots of a tilapia gi l l membrane preparation separated on a sucrose step gradient 60 FIGURE 3.8 Immunolocalization of vH + -ATPase , N a + , K + - A T P a s e and E N a C in trout gi l l tissue 62 FIGURE 3.9 Immunolocalization of vH + -ATPase in freshwater trout g i l l using the immunogold technique 64 FIGURE 3.10 Western blots of rainbow trout crude gi l l homogenate probed with antibodies against (a) a subunit of the N a + , K + - A T P a s e , (b) A-subunit of the v H + -ATPase, (c) P-subunit of the E N a C , and (d) a subunit of the E N a C 66 v i i i FIGURE 3.11 Dose response curve of rainbow trout ATPase activity (mU/h) to bafilomycin A l and K N 0 3 . Membranes prepared from crude gi l l homogenates using differential centrifugation were used for the measurements of ATPase activity. D M S O was used as a control, n = 3 68 FIGURE 3.12 Illustrations of the freshwater tilapia (a) and rainbow trout (b) branchial epithelium cell types ( M R cells and P V C s ) to summarize the immunolocalization data 79 FIGURE 4.1 Double labeled sections on the turbot gi l l epithelium using a rabbit polyclonal antibody (597) against the N H E - 2 fusion protein (FITC, green) and mouse monoclonal antibody against the a subunit of N a + , K + - A T P a s e (Texas Red) 91 FIGURE 4.2 Double labeled section of the seawater adapted coho gi l l epithelium using a rabbit polyclonal antibody (597) against the N H E - 2 fusion protein (a) and mouse monoclonal antibody against the a subunit of N a + , K + - A T P a s e (b) 93 FIGURE 4.3 A n SDS treated section of the turbot gi l l epithelium duel labeled for A E l t and N a + , K - ATPase using a rabbit polyclonal antibody generated against trout erythrocyte band 3 ( A E l t ) (A) and a mouse monoclonal antibody specific of the a subunit of N a + , K + - A T P a s e (B), respectively 95 FIGURE 4.4 Mean changes in sodium and chloride ion concentrations in the plasma of coho salmon in freshwater (time 0) and after a gradual increase in external salinity for 8 days. Changes in salinity of the holding water are plotted on the lower graph. n= 5-9 97 FIGURE 4.5 Changes in the vH + -ATPase and N a + , K + - A T P a s e activities in the gil l homogenates of coho salmon during sea water acclimation as determined by N E M and ouabain sensitive activity, n = 5 99 FIGURE 4.6 Immunohistochemistry of gi l l from coho salmon (O. kitsutch) adapted to either fresh water (upper and lower panels A B C , GHI) or sea water (middle panel DEF) showing the distributions of the band 3-like anion exchanger (AEl t ;A ,D) and N a + , K + - A T P a s e (B,F) 101 FIGURE 4.7 Paired fluorescent (A ,C) and phase contrast (B,D) micrographs of fixed-frozen sections of gills from O. kisutch held in freshwater (A,B) and after 2 days (C,D) exposure to a progressive increase in external salinity. The sections were immunolabeled for vH + -ATPase using a peptide antibody derived from the A-subunit of bovine brain vH + -ATPase 103 FIGURE 4.8 Illustration of cell types of the branchial epithelium of a seawater fish summarizing the immunolocalization data 108 ix FIGURE 5.1.1 The effects of the N a + / H + exchange inhibitor amiloride (0.1 m M ) on (a) net ammonia ( J A M M ; umol • kg" 1 • h"1) and (b) net acid fluxes (JACID = J T A + J A M M ; uEq • kg" 1 • h"1) in mudskippers in 50% S W 131 FIGURE 5.1.2 The effects of the carbonic anhydase inhibitor acetazolamide (0.1 m M ) on (a) net ammonia ( J A M M ; umol • kg" 1 • h"1) and (b) net acid fluxes (JACID = JTA + J A M M ; uEq • kg" 1 • h"1) in mudskippers in 50% S W 133 FIGURE 5.1.3 The effects of changes in boundary layer p H by 5 m M H E P E S p H 7.0 and p H 8.0 on net ammonia ( J A M M ; umol • kg" 1 • h"1) (a, c respectively) and net acid fluxes (JACID = J T A + J A M M ; uEq • kg" 1 • h"1) (b, d respectively) in mudskippers in 50% S W 135 FIGURE 5.1.4 The effects of the vH + -ATPase inhibitor 100 m M K N 0 3 and 100 m M KC1 (control) on net ammonia (a,c respectively) ( J A M M ; umol • kg" 1 • h"1) and net acid fluxes (b,d respectively) (JACID = JTA + J A M M ; uEq • kg" 1 • h"1) in mudskippers in 50% S W over a 3 h period 137 FIGURE 5.1.5 The effect of the specific N a + , K + - A T P a s e inhibitor ouabain (0.1 m M and 0.01 m M ) on net ammonia fluxes in P.schlosseri in 2 m M NH4CI in 50%SW. n= 6 and 5 139 FIGURE 5.2.1 A low magnification light micrograph of a cross section through a gi l l filament (a). Superimposed black boxes indicate the general area of the higher magnification micrographs (b-e) 141 FIGURE 5.2.2 A n electron micrograph (a) showing a lamellar mitochondria-rich (MR) cell anchored to the basal lamina (arrow) opposite the blood spaces defined by a pillar cell (PiC) 143 FIGURE 5.2.3a,b Electron micrographs of the apical plasma membrane domain of two lamellar M R cells with accessory cell ( A C ) processes 145 FIGURE 5.2.4 Electron micrographs of branchial filament rich (FR) cells 147 FIGURE 5.2.5 Light (a) and electron (b,c) micrographs of cross sections through the opercular epithelium with its intraepithelial capillaries (a; arrows) 149 FIGURE 5.3.1 Indirect immunofluorescence (a) and phase-contrast (c) microscopy showing the distribution of N a + , K + - A T P a s e in the gills of the mudskipper 151 FIGURE 5.3.2 Immunogold localization of N a + , K + - A T P a s e in the mudskipper gi l l lamellar M R cell using the oc5 antibody and a secondary antibody conjugated to 20nm colloidal gold particle 153 x F I G U R E 5.3.3 Indirect immunofluorescence and phase-contrast microscopy showing the distributions of N a + , K + - A T P a s e (a,b), C F T R (c,d), and N K C C (e,f) in cells of the lamellar epithelium 155 F I G U R E 5.3.4 Indirect immunoperoxidase staining of fixed-frozen sections of mudskipper gi l l tissue. Sections were either labeled with the carbonic anhydrase (a,b) or vH + -ATPase (c) antisera 157 F I G U R E 5.3.5 N a + / H + exchanger 3 (NHE3) distribution in the gills of fixed-frozen sections of mudskipper gi l l using indirect immunofluorescence and phase-contrast microscopy 159 F I G U R E 5.3.6 Projection from a z-stack of 50 confocal images showing the NHE2- l ike distribution in the mudskipper gi l l lamellae using indirect immunolabeling with the rabbit polyclonal antibody 597 161 F I G U R E 5.3.7 Western blots of mudskipper gi l l tissue crude homogenate probed for (a) a subunit o f N a + , K + - A T P a s e , (b) C F T R protein, (c) A-subunit of v H + -ATPase, (d) N H E - 2 (Ab 597), and (e) N H E - 3 (Ab 1380) 163 F I G U R E 5.4 Illustration of the mudskipper gi l l M R cell and its proposed role in active N H 4 + elimination 177 xi LIST OF ABBREVIATIONS A C Accessory cell A E Anion exchanger A E l t An ion exchanger 1 (band-3 protein) trout A P Alkaline phosphatase B C I P 5 -bromo-4-chloro-3 -indolyl phosphate B S A Bovine serum albumin B S C Bumetanide sensitive cotransporter (a.k.a N K C C ) C C Chloride cell C D Collecting duct C F T R Cystic fibrosis transmembrane regulator ddFLO Double distilled water DIDS 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid D M S O Dimethylsulfoxide D P C Diphenylamine-2-carboxylate E C L Enhanced chemiluminescence E N a C Epithelial Sodium Channel F ITC Fluorescein isothiocyanate F R cell Filament-rich cell Glut Glutaraldehyde H Hypothesis H E P E S N-[2-hydroxethyl] piperazine-N'-[2-ethane sulfonic acid] H R P Horse radish peroxidase JACID Net acid flux JAMM Net ammonia excretion M R cell Mitochondria-rich cell M T A L medulla thick ascending limb M W Molecular weight N B T Nitroblue tetrazolium N E M N-ethylmaleimide x i i N G S Normal goat serum N H 3 Gaseous or unionized ammonia N H 4 + Ionized ammonia or ammonium N H E N a + / H + exchanger N K A N a + / K + - A T P a s e N K C C N a + : K + : 2C1—contransporter N M S Normal mouse serum N P P B 5-nitro-2-(3-phenylpropylamino) benzoic acid N R S Normal rabbit serum P A G E Polyacrylamide gel electrophoresis P B S Phosphate buffered saline P F A Paraformaldehyde P L P Paraformaldehyde-lysine-periodate P V C Pavement cell rt Room temperature R T - P C R Reverse Transcriptase-Polymerase Chain Reaction SDS Sodium dodecyl sulfate SEI Sucrose, E D T A , Imidazole buffer SITS 4-acetomido-4'-isothiocyanatostilbene-2,2'-disulfonic acid T A Titratable acidity T E M Transmission electron microscopy T E P Transepithelial potential difference T P B S Tween-20 in Phosphate buffered saline vH + -ATPase vacuolar (V-type) proton ATPase x i i i A C K N O W L E D G M E N T S I am indebted to my supervisor, Dave Randall, who I have had the pleasure of knowing for the past seven or so years. He has provided me with opportunities and shown me that there is much to appreciate in life. This Ph.D. has not been so much about studying ion regulation in fishes but getting out and seeing the world, to see how other people from diverse backgrounds study ion regulation in fishes. I have had the opportunity to return to the womb, so to speak, to the land of my birth, Hong Kong, to work on the mudskipper, an air-breathing amphibious fish (if you believe such things exist). I was there to see the Pearl in the Crown returned to the Motherland. I also had the pleasure to collaborate with Alex K - Y . Ip of the National University of Singapore on project "mudskipper". I have lived in the picturesque capital of the New Europe, Strasbourg, to work in a place where they work on penguins and I worked on fish. I was there to see the French win the World Cup (Allez les Blues!). To my French host, Pierre Laurent, my palate w i l l forever remember the country potatoes. A n d here at home (Vancouver), I owe a debt of gratitude to Wayne V o g l of the Department of Anatomy who has always left the door open for me to return from my expeditions. When I started, my son, Matthew was far from making his first cell division and now (many cell divisions later) he w i l l be 3 years old. H o w can any of this make sense to him? M y wife, Cristina, has been very understanding during this unsettled period but I did spend many an afternoon with her in MongKok. M y parents, sisters, and brother have also been supportive in my pursuit occasionally providing floor space for me to rest my head. During the course of my Ph.D. studies I have received funding from the University of British Columbia through numerous graduate fellowships and the European Commission though a Marie Curie Training and Mobil i ty of Researchers ( T M R ) grant. I also would like to thank the Society for Experimental Biology and the European Society for Comparative Physiology and Biochemistry for subsidizing my travel to conferences. So as you can see, even i f I didn't get another degree, these last four years w i l l not have been for naught. A n d so it goes. xiv Chapter 1. GENERAL INTRODUCTION In this general introduction I w i l l describe the g i l l and its role in active ion regulation as well as provide some background information on the ion transport proteins involved. In the following chapters on freshwater, and seawater fishes and the mudskipper I w i l l introduce in more detail the different models for N a + , and C l " uptake and elimination and ammonia excretion, respectively. 1.1. Basic description of the fish gill The fish g i l l is designed to accommodate gas exchange, ion regulation and the elimination of nitrogenous waste (ammonia) (see reviews by Piiper and Scheid 1984; Karnaky 1998; Walsh 1998; Evans et al. 1999). To optimize the exchange between the internal (blood) and external (water) environments, the gi l l has a large surface area with a short diffusion distance between blood and water, which is well perfused (receiving the full cardiac output), and well ventilated. The high perfusion and ventilation ensure diffusion gradients are maximized. The typical teleost fish g i l l is composed of g i l l arches with paired rows of filaments extending in a laterocaudal direction (Figure 1). Regularly spaced rows of lamellae on both upper and lower sides of the filament are arranged perpendicular to the long axis of the filaments. These thin lamellae account for the greatest surface area of the g i l l and provide an interface ideal for gas exchange. Blood, delivered to the lamellae from the afferent arteries, flows through the lamellar blood spaces and is collected by the efferent arteries to enter the systemic circulation (arterio-arterial circulation). The water flow between lamellae is counter current to the blood flow within the lamellae. The water flow is forced over the filaments by having them in a sieve-like arrangement whereby the tips of filaments from adjacent gi l l arches come into close contact. The thicker filament epithelium is generally associated with ion regulation due to the presence of a 1 F I G U R E 1.1 Basic description of the teleost fish gi l l . The gills are found in the mouth between the buccal and opercular cavities. The four pairs of gi l l arches have paired rows of filaments extending in a laterocaudal direction. Regularly spaced rows of lamellae on both upper and lower sides of the filament are arranged perpendicular to the long axis of the filaments. Blood, delivered to the lamellae from the afferent arteries flows through blood spaces within the lamellae and is collected by the efferent arteries to enter the systemic circulation (arterio-arterial circulation). The water flow between lamellae is counter current to the blood flow within the lamellae. The water flow is forced over the filaments by having them in a sieve-like arrangement whereby the tips of filaments from adjacent gi l l arches come into close contact. (Modified from Eckert et al. 1997). The micrographs show the dominant cell types of lamellar and filament epithelia, the pavement cell and M R cell, respectively. They also illustrate the short distance from the blood (rbc; red blood cell) to water present in lamellae. 2 population of ionocytes ( M R cells) and has an arterio-venous circulation. The arterio-venous or secondary circulation is served by a central venous sinus ( C V S ) that runs the length of the filament and drains into the branchial vein. Blood is delivered to the C V S via anastomoses on the efferent and/or afferent vessels, (see reviews by Hughes 1984; Laurent 1984). 1.1.1 Branchial epithelium The branchial epithelium can be sub-divided into the lamellar and filament epithelia on the basis of location and appearance. The lamellar epithelium that covers the lamellae is pseudo-stratified and composed mainly of squamous epithelial cells commonly referred to as pavement (PVC) or respiratory cells. Mucocytes or goblet cells are not commonly found within the lamellar epithelium. Modif ied endothelial cells called pillar cells define the lamellar blood spaces. A s their name would imply, pillar cells have a pillar-like shape and have thin flanges that extend to contact neighbouring pillar cell flanges (Newstead 1967). Separating the endothelium and epithelium is the basal lamina. The filament epithelium covers the filament leading and trailing edges and the spaces between lamellae (interlamellar spaces). This epithelium is much thicker than the lamellar epithelium and is composed of cells similar to the pavement cell ( P V C ) as well as columnar cell, mucocytes and M R cells. The M R cells tend to be more numerous toward the trailing edge o f the filament within the interlamellar space and on the trailing edge (van der Heijden et al. 1997). However, fish living in extremely ion poor water also have M R cells on their lamellae (Laurent et al. 1985; Avel la et al. 1987). Two cell types that are likely directly involved in ion regulation are the M R cell and P V C (Laurent 1984). Mucocytes indirectly contribute to ion regulation through the mucus they secrete, which modifies boundary layer properties of the epithelium (Handy 1989; Randall et al. 1991; Shephard 1992). M u c h work has been done describing the morphology of the M R cells 3 and P V C s and identifying morphological correlates to changes in environmental or physiological conditions (reviews by Laurent 1984; Laurent and Perry 1991; Pisam and Rambourg 1991; Goss et al. 1995). 1.1.2 Mitochondria-rich (MR) cells The M R cells generally make up less than 10% of the epithelium yet are generally considered to be the dominant site of branchial ion regulation (Karnaky 1986; Jiirss and Bastrop 1994; Zadunaisky 1996). In addition to having two unique intracellular membrane systems (the subapical tubulovesicular system and an extensive tubular system) these cells are rich in mitochondria, as their name implies (Phillpot 1980; Pisam and Rambourg 1991). The tubulovesicular system is a collection of irregular tubules and vesicles found in the subapical region of the cell . They are likely involved in the transport of glycoproteins from the Golgi apparatus to the apical membrane (Pisam et al. 1980). The tubular system is found throughout the cytoplasm of the cell closely associated with the mitochondria except in the Golgi region and in close proximity to the apical membrane. The tubular system is composed of anastomosing tubules of constant diameter that are in continuity with the basolateral membrane. The tubular system can be differentiated from the tubulovesicular system and the endoplasmic reticulum through the use of extracellular markers (lanthanum, ruthenium red, and horseradish peroxidase; Phillipot 1980) and the ferrocyanide-reduced osmium staining technique (Pisam 1981). The tubular system is also associated with high concentrations of N a + , K + - A T P a s e (Karnaky et al. 1976; Hootman and Phillpot 1979; Witters et al. 1996). Pisam and co-wokers (1987, 1990) have been able to elegantly demonstrate that there are at least two sub-types of M R cells (a and (5 cells) in freshwater fishes (Lebistes reticulatus, Gobio gobia, Cobitus taenia). The a - M R cell is found at the base o f the secondary lamellae 4 closely associated with the arterio-arterial circulation and is elongate and pale (less electron dense). The apical membrane is flat and not elaborated. The tubulovesicular system is poorly developed while the tubular system forms a regular meshwork throughout the cytoplasm. The (3-M R cell is found closer to the centre of the interlamellar space and is associated with the arterio-venous circulation (central venous sinus). The P-cell is darker (more electron dense) and ovoid or cuboidal in shape with a slightly concave and wavy apical membrane. The tubular system forms an irregular meshwork and the tubulovesicular system has additional electron-dense bodies (using a magnesium-lead stain; Pisam et al. 1987). In other species the appearance of the apical membrane can be highly variable ranging from flat and unornamented to sponge-like (Perry et al. 1992). Morphological variation is also introduced by cells undergoing the processes of apoptosis and necrosis (Wendelaar Bonga and van der Meij 1989). In the M R cell of seawater fishes, the tubular system is more developed and the apical membrane tends to be invaginated forming a crypt. Higher N a + , K + - A T P a s e activities are associated with the seawater type M R cell. The seawater type M R cell is typically referred to as the chloride cell (see chapter 4). In fish transferred from freshwater to seawater, it is the P - M R cell that develops into the chloride cell (Pisam et al. 1987). Closely associated with the apical region of the chloride cell is a small accessory M R cell with a poorly developed tubular system and fewer mitochondria. This accessory cell sends cytoplasmic projections into its larger neighbour that emerge at the apical crypt. The tight junctions between chloride and accessory cells have fewer strands and are considered leaky (Sardet et al. 1979) and the accessory cell projections increase the length of leaky junction. This leaky paracellular pathway is important in the N a + efflux pathway (see chapter 4 for more details). Interestingly, these accessory cells are 5 found in euryhaline fishes living in freshwater although the number of projections is much-reduced (Pisam etal. 1988, 1989). 1.1.3 Pavement cells (PVC) The pavement cells are generally characterized by a flattened appearance, the presence of concentric microridges on their apical surface (however, not in all species), lack of a tubular system, and relatively few mitochondria (Laurent 1984). However, there have been reports of mitochondria-rich P V C s in the brown bullhead (Ictalurus nebulosus; Goss et al. 1992; 1994). 1.2. The importance of environment on ion transport protein expression In the preceding section I have mentioned that there are some morphological and biochemical differences between M R cells of freshwater and seawater fishes. Teleost fishes are osmoregulators maintaining their internal milieu relatively constant (NaCl ~150mM) despite the large extremes in the salinity of their environment. Seawater (NaCl ~450mM) represents a hyperosmotic environment and confronts the animal with large inward ion gradients resulting in salt gain and dehydration (osmotic water loss) while freshwater (NaCl ~0.5mM) represents the opposite extreme: salt loss and water gain. To achieve an ionic homeostasis animals must be able to excrete and take up ions against large gradients in marine and freshwater environments, respectively. A l so marine fishes drink seawater to replace osmotically lost water while freshwater fishes get r id of their excess water by producing copious amounts of urine (see review byKarnaky 1998). Since the demands for ion homeostasis are diametrically opposed in marine and freshwater fishes so too must be the expression and arrangement of ion transport proteins which mediate the transepithelial ion movements. So the freshwater fish g i l l w i l l be organized for salt uptake while the seawater fish gi l l w i l l be organized for salt elimination. The mechanisms for 6 N a C l elimination by the seawater type M R cell are well characterized while the exact mechanism of N a C l uptake in freshwater is still being contested (Karnaky 1998; Perry 1997). Movement of fishes between these salinity extremes represents a dramatic yet natural event encountered by many species. Some euryhaline estuarine and tidepool fishes face such changes on a daily basis (tidal cycles; e.g. killifish) while others may do so only a few times in their lives (for spawning; e.g. salmonids and anguillids) while some never venture out of their salinity regime (stenohaline marine and freshwater fishes). 1.2.1 Freshwater NaCl uptake models Measurements of transepithelial potentials (TEP) in freshwater fishes predict active uptake of both N a + and C l " (Potts 1984). The uptake of N a + and C l " is not directly coupled but involves N a + / H + and C I V H C C V exchange processes. Krogh originally proposed these mechanisms in 1937 yet the exact nature of the exchanges has still yet to be fully elucidated. The N a + / H + exchange is thought to be the result of an apical vaculolar-type proton-ATPase ( v H + -ATPase) driving N a + uptake via a N a + channel (ENaC) rather than by direct coupling via a N a + / H + exchanger (NHE) (Avella and Bornancin 1989; L i n and Randall 1995). It has largely been accepted that C l " uptake is facilitated by an apical C I 7 H C O 3 " anion exchanger ( A E ; Goss et al. 1995). The exit step of C l " uptake into the blood is not known, although a basolateral C l" channel and a K : C 1 " cotransporter have been suggested (Wright 1991). The use of H + and HCO3" as counterions implies a role for these exchange processes in acid-base balance (Heisler 1993). The freshwater mechanism of N a C l uptake w i l l be explored in more depth in chapter 3 and additional information on the individual ion transport proteins can be found in the section 1.4 (below) on ion transport protein properties. 1.2.2 Seawater NaCl elimination model The model for N a C l elimination in sea water centres on the chloride cell-accessory cell complex and is the same as that applied to the dogfish rectal gland and avian salt gland (Silva et al. 1977; e.g. Marshall et al. 1998). In euryhaline marine fishes, the T E P predicts active C l " elimination (Potts 1984). The basolateral N a + , K + - A T P a s e creates a N a + gradient to drive C l " entry via a N a + : K + : C 1 " cotransporter into the chloride cell (Eriksson et al. 1985). Potassium cycles through a K + channel (Suzuki et al. 1999). The accumulated C l " exits the cell apically via a C F T R (Cystic fibrosis transmembrane regulator)-like anion channel down its electrochemical gradient (Singer et al. 1998). N a + on the other hand, accumulates in the extracellular spaces and leaks out paracellularly down its electrochemical gradient (Sardet et al. 1979). The N a + equilibrium potential is generally near the measured transepithelial potential (+20mV TEP) (Evans 1979; Potts 1984). Acid-base balance in marine fishes is achieved by N a + / H + and C l T H C C V exchange processes but in this case driven by the inward N a C l gradients (Heisler 1993; Claiborne 1998). The cell type distribution of these exchanges is unknown. The vH + -ATPase , which may be involved in acid-base regulation in freshwater fishes, is down regulated in fish transferred to seawater (L in and Randall 1995). Chapter 4 w i l l focus on the seawater fish gi l l ion transport processes in more detail while section 1.4 (below) provides more details on the ion transporters. 1.2.3 Models of gill NH/ elimination The majority of teleost fishes use ammonia as a nitrogenous waste product for excretion that takes place across the g i l l (see review by Walsh 1998). Generally, with the high ventilatory water flows across the gi l l and low ambient ammonia concentration, N H 3 partial pressure (PNH3) gradients can account for the total ammonia efflux (Cameron and Heisler 1983). When fish are in 8 high environmental ammonia and the gradients for both N H 4 and PNH3 are directed inwards then the efflux component may be active. However, the analysis is complicated by boundary layer p H effects that may involve the v H + - A T P a s e - E N a C and NH3 movement rather than N a V N H / exchange (Wilson and Taylor 1994). Yet, there are some fish species that clearly must be using active NFLf1" elimination because they can tolerate extremely high environmental ammonia levels and can still maintain an ammonia efflux (30mM NH4CI, Periophthalmodon schlosseri; Randall et al. 1999). There have been a number of ion transport proteins that have been shown to be able to substitute N H 4 + ions for H + or K + ions and potentially could facilitate transbranchial N H 4 + movements. Notably the N a + / H + exchanger (NHE) , N a + , K + - A T P a s e and N a + : K + : C l" cotransporter ( N K C C ) (Evans and Cameron 1986). This model w i l l be examined in greater detail in chapter 5. 1.3 Other model iono acid-base regulatory epithelia The models of transepithelial ion transport and cell type composition of 'tight' epithelia like turtle/toad bladder, frog skin, cortical collecting duct have been applied to fresh water fish gil l and common mechanisms assumed to be present. For example, the vH + -ATPase -epithelial sodium channel model for N a + uptake was proposed by Ave l l a et al. (1989). Higher vertebrate distal renal epithelia and amphibian skin and bladder show some structural and functional similarities to the freshwater fish gi l l (turtle bladder, Stetson and Steinmetz, 1985; amphibian skin, Ehrenfeld et al. 1990; Larson 1991; collecting duct, Brown et al. 1988). These tight or high resistance epithelia have tight junctions (zonulae occludens) which minimize paracellular movements (leaks) of ions and water and allow for the generation or maintenance of steep ion gradients across the epithelia. Ion transport proteins within the apical and basolateral membrane domains of the epithelial cells constitute the principal pathways of transepithelial ion fluxes. 9 These epithelia also have in common, populations of M R cells that are involved in active N a + and C l " uptake in addition to acid-base regulation. Three functional types ( a (A) , p (B), and y) of M R cells have been described in these epithelia (Larsen 1991). M R cells in these epithelia are also characterized by high levels of intracellular carbonic anhydrase activity (Brown et al. 1983; Stetson and Steinmetz 1985; Lonnerholm and Wistrand 1984). C A catalyzes the hydration of C O 2 to provide H + and HCO3" for ion transport proteins (vH + -ATPase and A E ) . The a type or acid excreting M R cell has been described in collecting duct (CD) and urinary bladder as well as amphibian skin. A n apical vH + -ATPase (Brown et al. 1988; Gluck and Nelson 1992) and basolateral C17HC0 3 " exchanger ( A E 1 ; Verlander et al. 1988; Alper et al. 1989) characterize these cells. Morphologically these cells have numerous apical membrane microplicae and subapical vesicles associated with rod shaped intramembranous particles and immunoreactivity to various subunits of the vH + -ATPase complex. The intramembranous rod shaped particles are interpreted as being the transmembrane domain of the vH + -ATPase complex (Brown et al. 1987a,b; Stetson and Steinmetz, 1985). The P type M R cell, found in C D and urinary bladder epithelia, is involved in HCO3" elimination and electroneutral C l " uptake. A n apical C I T H C C V antiporter is responsible for this exchange ( A E 2 , Alper et al. 1997). H C O 3 " generated from catalyzed C O 2 hydration provides the driving gradient for C l " uptake while the generated H + s are pumped across the basolateral membrane by a vH + -ATPase (Brown et al. 1988). Morphologically, these cells have simple apical microvil l i , a darker cytoplasm, and an amplified basolateral membrane domain associated with intramembranous rod shaped particles (Madsen and Tisher 1985). The y type M R cell has been described in toad skin (Larsen et al. 1992) and Chinese crab gi l l (Onken and Putzenlechner 1995). C l " uptake by these cells is mediated by an apical A E 10 driven by an apical vH + -ATPase at low mucosal C l " concentrations. However, in toad skin at high mucosal C l " concentrations an apical C l " channel, activated by apical membrane depolarization, mediates C l " uptake. C l " leaves through basolateral C l " channels. These cells have also been shown to have a significant apical N a + conductance (see Larsen 1991). The principal or granular cell is another cell type that can be differentiated from the M R cells. These cells are involved in active N a + uptake via apical amiloride-sensitive N a + channels (Brown et al. 1989) and a basolateral N a + , K + - A T P a s e (Madsen and Tisher 1985; Sabolic et al. 1999). The frog skin M R cell apical vH + -ATPase hyperpolarizes the apical membrane creating a favourable electrochemical potential for N a + uptake while in C D , the epithelium does not function as a syncitium and N a + uptake is driven by the basolateral N a + , K + - A T P a s e . These cells also have a basolateral N a + : K + : C 1 " cotransporter which is involved in cell volume regulation. The seawater fish gi l l is a leaky epithelium and involved in N a C l elimination. Other functionally similar epithelia include the avian salt gland and the shark rectal gland (Riordan et al. 1994). A common model is used to describe N a C l elimination in these epithelia (see Section 2.2 above). The mammalian kidney medulla thick ascending limb ( M T A L ) is involved in carrier mediated movement of ammonia (Palliard 1998). 1.4 Ion transport protein properties and evidence for their presence in the fish gill. The translocation of ions across the plasma membrane against electrochemical gradients can be achieved through the input of energy in the form of A T P . In animal cells, the most important plasma membrane ATPase is N a + , K + - A T P a s e which is ubiquitously expressed, while a vacuolar type proton ATPase (vH + -ATPase) is emerging as an important contributor in some epithelia. These primary active ion transporters are not only important in direct translocation of 11 ions but also in creating standing gradients for coupled secondary transport of other ions against otherwise unfavourable conditions. Coupling can be in the same direction, (symport or cotransport) or in the opposite direction, (antiport or exchange). In the following section, I w i l l give a brief description of some of the transporters involved in active ion transport which I have immunolocalized and follow with the evidence for these transporters in the fish g i l l . 1.4.1 Na,it-ATPase The sodium pump or N a + , K + - A T P a s e is a ubiquitous basolateral plasma membrane protein in epithelial cells, however, in cells involved in transcellular ion transport, levels of N a + , K + - A T P a s e expression are higher (see reviews by Skou and Esmann 1992; Blanco and Mercer 1998). This ATPase is important in maintaining low intracellular N a + and high intracellular K + and the negative intracellular membrane potential. The potential difference is created by the movement of 3 N a + out and 2 K + in for each A T P consumed by the pump. The N a + gradient is also used to drive secondary N a + coupled transporters (eg N a + / H + exchanger (NHE) for H + excretion out of the cell , or N a + :K + :2C1" cotransporter ( N K C C ) for C l " uptake into the cell; see below). N a + , K + - A T P a s e belongs to the P-type class of ATPases having a phosphorylated intermediate state which makes it sensitive to inhibition by vanadate. However, N a + , K + - A T P a s e activity can be specifically inhibited by ouabain. The sodium pump is composed of an a and P subunit and possibly a third y-subunit. The a subunit (~112kDa) is responsible for catalytic and transport properties of the pump and is the site of cation, A T P and ouabain binding. The glycosylated p subunit (40-60kDa) is essential for pump activity and is involved in modulation o f N a + and K + affinities as well as having a chaperone role in a-subunit folding and insertion in to the plasma membrane. In mammals there are four a ( a l , a2 , a3 and a4) and three 12 P-subunit ( p i , p2 and P3) isoforms that have been identified (Blanco and Mercer 1998). These isoforms exhibit tissue specific patterns of expression and the different ocP-heterodimer combinations contribute to the variation in N a + pump enzymatic properties. The y-subunit (8-14 kDa) is not required for normal N a + , K + - A T P a s e activity, although there is accruing evidence that it can modify N a + , K + - A T P a s e activity. In the fish g i l l , N a + , K + - A T P a s e has probably been the most studied ion transport protein from enzyme kinetics, to mechanism of regulation, to subunit expression (McCormick 1995). N a + , K + - A T P a s e has been localized to the M R cells using [ 3H] ouabain autoradiography (Karnaky et al. 1976) and anthroylouabain fluorescent staining (McCormick 1990), enzyme histochemistry (Hootman and Philpott 1979; Conley and Mallatt 1987) and immunohistochemistry (eg Witters et al. 1996; Ura et al. 1996). The use of in vito assays of ouabain sensitive ATPase activity are commonplace in determining gi l l N a + , K + - A T P a s e activities in fishes (see McCormick 1995). Levels of N a + , K + - A T P a s e activity are closely correlated with numbers of seawater-type M R cells. The association of N a + , K + - A T P a s e with the tubular system of M R cells makes it a good cell type specific marker. It has been assumed that N a + , K + - A T P a s e in P V C s is below the level of detection by [ 3H] ouabain binding and immunohistochemistry rather than being completely absent (Hootman and Phillpot 1979; Witters et al. 1996; Ura et al. 1996). The oc-subunit (Kawakami et al. 1985; Schorock et al 1991; Kisen et al. 1994; Cutler et al. 1995a) and the p-subunit (Noguchi et al. 1986; Appel et al. 1996; Culter et al. 1995b; Cutler et al. 1997) have been sequenced in a number of fish species. There is generally high sequence similarity amongst vertebrates (Pressley 1992). In the g i l l , a l and p i isoforms are expressed while the p2 and p3 have been shown to be absent (Cutler et al. 1995a,b, 1997). Lee et al. (1998) 13 have also shown immunoreactivity of a l and a3-like subunits in the g i l l o f fresh and sea water acclimated tilapia. 1.4.2 vlt-ATPase Another plasma membrane protein involved in energizing active ion transport is the electrogenic vacuolar proton ATPase (vH + -ATPase or V-type ATPase; named after the H + -ATPase originally found in plant and fungi vacuolar membranes) (reviewed by Harvey and Wieczorek 1997; Nelson and Harvey 1999). In animals, the vH + -ATPase is found in lysosomes, coated vesicles, and chromaffin granules, in addition to the plasma membrane of specialized H + excreting cells (eg Harvey 1992). The vH + -ATPase creates an electric potential difference (A\|/) across the cell membrane in which it resides by coupling the hydrolysis of A T P to the translocation of H + out of the cytoplasm (across the plasma membrane or into cellular compartments). The proton-translocation can be used for direct acidification or the A\\i used for secondary ion movements by channels or by sym or anti-porters. The vH + -ATPase is specifically inhibited by bafilomycin A i and concanamycin A in nanomolar concentrations (Bowman et al. 1988; Drose and Altendorf 1997), N-ethylmaleimide ( N E M ) in micromolar concentrations and nitrate (NO3") and other oxidizing agents in millimolar concentrations (Moriyama and Nelson 1987; Dschida and Bowman 1997). The vH + -ATPase is a complex heteromultimeric protein composed of a cytoplasmic, peripheral catalytic ( V i ) domain and a membrane spanning ( V 0 ) domain (reviews by Finbow and Harrison 1997; Nelson and Harvey 1999). For example the kidney brush boarder vH + -ATPase (~580kDa) is composed of subunits 70, 56, 45, 42, 33, 31, 15, 14, and 12 k D a in size (Wang and Gluck 1990). Multiple 56 kDa B-subunit isoforms with differential expression patterns have been found (Puopolo et al. 1992; van Hil le et al. 1994) as well as microheterogeneitity within the 14 31 kDa E-subunit (Wang and Gluck 1990; Hemken et al. 1992). These differences may explain the enzymatic differences between brush boarder and microsomal vH + -ATPase found by Wang and Gluck (1990) in kidney and differences found between and within other tissues and species. The 70 kDa A-subunit is the most highly conserved of the vH + -ATPase subunits (Finbow and Harrison 1997) and has two characteristic cysteine residues (254 and 532) which also are involved in N E M binding and redox modulation of vH 4 -ATPase activity (Feng and Forgac 1994; Oluwatosin and Kane 1997). There are however, also A-subunit isoforms (van Hil le et al. 1993). Less is known about the other subunits (Finbow and Harrison 1997). The vH + -ATPase complex has been visualized by freeze fracture and high-resolution electron microscopy as studding of vH + -ATPase membranes (Brown et al. 1992). The stud is the V i domain or portasome of the complex. Immunological techniques have been used in a number of studies to determine the cellular and subcellar distributions of the vH + -ATPase in a number of epithelia (in a variety of species) involved in acid-base regulation and/or ion regulation (e.g. mammalian kidney: Brown et al. 1988; male reproductive tract: Breton et al. 1996; frog skin: Kle in etal. 1997). In the fish g i l l , vH + -ATPase has been localized to the epithelium using immunohisto-chemistry (70kDa A subunit: L i n et al. 1994; Wilson et al. 1997; 31 k D a E-subunit, Sullivan et al. 1995; Perry and Fryer 1997). Bafilomycin A i sensitive ATPase activity has been measured in vitro in gi l l homogenates (Lin et al. 1993) and has been used in vivo to look at the role of the vH + -ATPase (Morgan and Iwama 1999; Fenwick et al. 1999). The use o f other inhibitors also indicates the presence of a proton pump in fish gills (Klungs0yr 1987; L i n et al. 1993; Justensen et al. 1993). In addition, studding of subapical vesicles has been observed in gi l l epithelial P V C s (Laurent et al. 1994). Recently, the B subunit has been cloned independently by Perry et al. 15 (1999), Pelster and Niederstatter (1999) and M . Sundin (University of Odensa, D K personal communication). They all report a high degree of sequence identity with the mammalian forms. 1.4.3 Na/It exchanger (NHE) The sodium proton exchanger (NHE) or antiporter catalyzes the electroneutral exchange of N a + for H + in a 1:1 ratio (Bianchini and Pouyssegur 1994). In vertebrates, there are at least 6 different N H E isoforms ( N H E 1 , 2, 3, 4, 5, p; Tse et al. 1993). The N H E - 1 is a ubiquitous basolateral isoform with the important housekeeping functions of cell p H i and volume regulation. The N H E - 2 , N H E - 3 , and N H E - 4 isoforms participate in transcellular N a + movements and are expressed in epithelia and have cellular and tissue specific distributions (Hoogerwerf et al. 1996; Bookstein et al. 1997; Chambrey et al. 1997). The N H E - 2 and N H E - 3 isoforms reside in the apical plasma membrane while the N H E - 4 has a basolateral location. The N H E - 5 isoform is specifically expressed in the brain (Klanke et al. 1995). Amiloride and its analogues are very potent and specific inhibitors of sodium transport including both N a + / H + exchanger and N a + channel (Benos 1982; Kleyman and Cragoe 1988). The amiloride analogues 5-(N,N-dimethyl) amiloride ( H M A ) and 5-(N,N-hexamethene) amiloride ( D M A ) are commonly used specific inhibitors of the N H E . The N H E isoforms vary in their amiloride sensitivities. There also is evidence that N H V can substitute for H + in exchange for N a + by the N H E in kidney tubule (Kinsella and Aronson 1980; Nagami 1988; Blanchard et al. 1998). Although both N H E - 2 and N H E - 3 isoforms are expressed in this kidney tubule segment, the N H E - 3 is present in greater amounts and believed to mediate the apical N H 4 + movement (Palliard 1998). In the trout erythrocyte a p N H E isoform has been characterized (Borgese et al. 1994). The p N H E is catacholamine stimulated and is important in erythrocyte p H i regulation during periods of acidosis (i.e. during exercise) that would otherwise decrease hemoglobin-oxygen 16 affinity. In the sculpin (Myoxocephalus octodecimspinosus), the p N H E and NHE2- l ike isoforms have been identified using molecular techniques (Reverse Transcriptase-Polyerase Chain Reaction ( R T - P C R ) ; Claiborne et al. 1999). N H E l - l i k e immunoreactivity has also been demonstrated in g i l l homogenates of sculpin and kil l if ish (F. heteroclitus) adapted to freshwater and seawater. Externally applied amiloride has been shown to reduce N a + uptake by 84-94% (Perry and Randall 1981; Wright and Wood 1985), however with no effect on net acid or ammonia fluxes (Avella and Bornancin 1989; L i n and Randall 1991). N H E has also been proposed to be involved in facilitating N H / elimination (Na + /NH4 + exchange) but unequivocal evidence is lacking (Wright and Wood, 1985; Wilson et al. 1994). The cellular distribution ( M R cells vs P V C ) of these N H E isoforms in the gil l is still unknown although the N H E l - l i k e and (3NHE isoforms are probably ubiquitously distributed to the basolateral membrane of both epithelial cell types. 1.4.4 Apical epithelialNa+ channel (ENaC) The epithelial sodium channel (ENaC) facilitates the uptake of N a + across the plasma membrane following its electrochemical gradient. The E N a C is a hetero-oligomer composed of three subunits (a,p,y; 85-95 kDa) that have been identified in a number of different tissues and species (Canessa et al. 1994; Garty and Palmer 1997). There is also an amiloride blockable N a + channel that has been purified from bovine renal papillae and A 6 cells (Benos et al. 1995). It is not entirely clear i f this purified N a + channel represents an entirely different E N a C or a complex incorporating subunits of the aPyENaC (Garty and Palmer 1997). A s mentioned earlier, N a + uptake via the channel is sensitive to amiloride and its analogues, notably benzamil (Benos 1982; Kleyman and Cragoe 1988). 17 In the fish g i l l , the only evidence for the presence of an apical N a + channel conies from the equivocal experiments using water borne amiloride (Perry and Randall 1981; Wright and Wood 1985). In these experiments it is not possible to distinguish between the inhibition of N H E or ENaC-vH + -ATPase . Since there are conditions where the N H E mechanism cannot explain N a + uptake the alternative model has been used. Recently, Fenwick et al. (1999) have reported that N a + uptake in larval tilapia is sensitive to bafilomycin A l . 1.4.5 Cl/HCOi anion exchanger (AE) The chloride-bicarbonate antiporter mediates the 1:1 electroneutral exchange of chloride and bicarbonate across the plasma membrane (reviewed by Alper 1991). The A E is involved in intracellular p H and volume regulation as well as transepithelial acid-base transport. There are three A E that have been identified to date ( A E 1 , A E 2 , and A E 3 ; Alper 1991; Brosius et al. 1997). The erythrocyte CT/HCO3" exchanger (AE1) or band 3 protein (the most abundant transmembrane protein and is recognized as the third band on S D S - P A G E ) was the first A E to be characterized (Kopito and Lodish 1985). Band-3 related proteins have also been found in non-erythroid tissues in the basolateral membrane domain, notably in a subpopulation of kidney cortical collecting duct intercalated cells which are involved in acid excretion (A-type cells; Wagner et al. 1987; Verlander et al. 1988; Kudrycki and Shull 1989). The A E 1 is sensitive to the disulfonic stilbene inhibitors of C l " transport DIDS (4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid) and SITS (4-acetamido-4'-isothiocynato-stilbene-2,2'-disulfonic acid; Cabantchik and Rothstein 1972; Schuster et al. 1986). A E - 2 has been identified in a number of non-erythroid tissues in either an apical or basolateral location and is thought to be primarily involved in bicarbonate secretion (Martinez-Anso et al. 1994; Stuart-Tilley et al. 1994; Alper et al. 1997). In the kidney cortical collecting duct, A E 2 is found apically in a subpopulation o f intercalated cells 18 involved in base excretion (B-type cells; Alper et al. 1997). A E 2 is relatively insensitive to stilbene inhibition (Schuster et al. 1986; Tago et al. 1986). The A E - 3 has only been found expressed in excitable tissues (brain and cardiac muscle; Brosius et al. 1997). In the fish g i l l , evidence for a C l " / HCO3" exchanger comes from studies using the stilbene inhibitors SITS and DIDS. In a number of studies, this inhibitor has been shown to reduce C l " influx and cause an alkalosis; results consistent with C l " /HCO3" exchanges (Perry et al. 1981). Goss and Wood (1990) using a two substrate model were able to demonstrate a 1:1 C l " / H C 0 3 " exchange ratio. Recently, Claiborne et al. (1997) have demonstrated that a DIDS sensitive mechanism exists that contributes to acid-base regulation in a seawater fish. Sullivan et al. (1997) were able to demonstrate in situ hybridization of an oligonucleotide probe for mammalian kidney A E 1 (Kudrycki and Shull 1989), however, Northern analysis was not done. There is also the possibility that anion exchange is driven by C I T H C C V -ATPase (Bornancin et al. 1980). This mechanism is however not widely accepted as mitochondrial contamination is a likely source of membrane associated activity (Claiborne 1998). 1.4.6Na* .Jt :2Cr cotransporter (NKCC) The N a + : K + : 2C1" cotransporter ( N K C C ) or bumetanide sensitive cotransporter (BSC) makes use of the inward sodium gradient to drive the electroneutral cotransport of K + and 2 Cl" . There are two N K C C isoforms, N K C C 1 (BSC2) and N K C C 2 (BSC1) , which are found expressed in the apical and basolateral membrane domains, respectively, of sections of the kidney tubule. In specialized salt secreting organs, like the dogfish rectal gland and the bird nasal salt gland, the N K C C 2 plays an important role in the movement of C l " into the cell against its electrochemical gradient. The N K C C also contributes to cell volume regulation. The N K C C is 19 sensitive to the loop diuretics furosamide and bumetanide (see reviews by O'Grady et al. 1987; Payne and Forbush 1995). In the fish g i l l , the N K C C plays a similar role in the seawater chloride cell as in the special elasmobrach and avian salt glands. In marine teleosts, C l " excretion has been shown to be sensitive to basolaterally applied furosemide and bumetanide (Erikkson et al. 1985). In the elasombranch, N K C C m R N A has been shown to be expressed (Xu et al. 1994). Recently, C. Cutler (personal communication) has cloned a N K C C in a teleost fish. 1.4.7 CFTR anion channel In mammalian tracheal epithelia, colonic cells and pancreatic duct cells, the movement of C l " out across the apical membrane is facilitated by a small conductance, voltage insensitive, c A M P activated C l " channel (Cohen 1994). This anion channel has been identified as a cystic fibrosis transmembrane regulator(CFTR; Anderson et al. 1991; Bear et al. 1992). Bicarbonate is another anion which makes use of this channel and may represent an important mechanism for HCO3" secretion and acid-base regulation (Smith and Welsh 1992; Poulsen et al. 1994; Hogan et al. 1997; Lee et al. 1998; Clarke and Harline 1998). In addition the C F T R is involved in the regulation of other apical membrane channels (e.g. E N a C ; Schwiebert et al. 1999). The C F T R channel is inhibited by diphenylamine-2-carboxylate (DPC) and 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB) but insensitive to 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS). In fishes, a CFTR- l ike C l " channel has been found to be expressed in the gills and is involved in the same pathway for C l " elimination as found in the elasmobranch and bird salt glands (Riordan et al. 1994). The fish C F T R has been cloned in the kill if ish Fundulus heteroclitus and shows low homology with human and shark forms (59 and 60% identity, 20 respectively; Singer et al. 1998). Patch-clamp recordings have identified a C l " channel with similar properties to the C F T R in seabass (Dicentrarchus lahrax) primary culture respiratory-like cells (Duranton et al. 1997) and overnight primary cultures o f ki l l i f ish opercular epithelial cells (Marshall et al. 1995). 1.4.8 Others There are also numerous other ion channels, co-transporters and antiporters and ATPases present in the branchial epithelium that are involved in the transepithelial movements of ions however, since I have not immunolocalized them I have not described them in detail (above). I w i l l however, now mention some of them and review recent publications. A n inwardly rectifying K + channel had been cloned and immunolocalized to branchial M R cells in the eel (Suzuki et al. 1999). There is postulated to be a basolateral K + : C 1 " cotransporter in fish M R cells but there is no work on this subject (Wright 1991; Mount et al. 1998; see Chapter 3 disscusion). The group of F l ik and Wendelaar-Bonga has extensively studied the ion transport proteins involved in Ca"1^ uptake. They have characterized a basolateral plasma membrane (PM) C a ^ ATPase, and N a + /Ca"1^ exchanger (Fl ik et al. 1995). There is an apical Ca""" channel that is not well characterized. 1.5 Study of fish ion regulation using in vitro models The complex anatomical organization of the gi l l has made it a difficult organ to study using traditional electrophysiological techniques. To date there is but one in situ measurement of potential differences across apical or basolateral membranes of the branchial epithelium (filament P V C ; Clarke and Potts 1998b) and intracellular concentrations of ions have only been measured using x-ray microanalysis (Morgan et al. 1994; Morgan and Potts 1995). However, intracellular ion concentrations have been measured using fluorescent dyes in isolated branchial cells (L i et al. 1997). Complicating matters is the thinness of the lamellar P V C s (l-15um), the 21 presence of the M R cell tubular system that infiltrates the cell that is easily damaged by electrode insertion, and the presence of a mucus coat covering the epithelium (Shepard 1992; Sandbacka et al. 1998). The use of surrogate model epithelia, which can be mounted in Ussing chambers, to study transport properties of the branchial epithelium under rigorous conditions has allowed the elucidation of the mechanism of N a C l transport in the seawater fish (see chapter 4; Karnaky et al. 1977; Foskett and Scheffeyl982; Marshall 1985). Such models also are yielding some interesting results that might lead us to reevaluate our current thinking. The seawater model for N a C l elimination is based largely on work using the opercular epithelium M R cell surrogate model, yet studies using a primary culture of seabass (Dicentrarchus labrax) branchial P V C s show similar transport properties (active C l " efflux; Avel la and Ehrenfeld 1997; Duranton et al. 1997; Avel la et al. 1999). In primary cultures of freshwater fish branchial P V C s active C l " uptake can be demonstrated, yet there is no active N a + uptake (Wood et al. 1998). The finding of active C l " uptake by P V C s is unusual because in vivo studies have indicated that M R cells are the sites of C l " uptake and that active N a + uptake should be performed by P V C s (see review by Goss et al. 1995). It seems that the P V C primary culture is only useful for studying permeability properties of the branchial epithelium (Gilmour et al. 1999). Perhaps showing greater promise is the opercular epithelia preparation of freshwater acclimated tilapia (O.niloticus) which is capable of unidirectional N a + and C l " (and Ca**) uptake (Burgess et al. 1998). Unfortunately, however, like in the freshwater primary cultures, these influx rates are dwarfed by large efflux rates. 1.6 Thesis problem In the gi l l we have an organ that is very plastic, adapting to changing environmental conditions to meet the needs of whole animal ion regulation. From the many decades of work on 22 this organ we have become fully aware of its importance (to the fish) but many o f the details still remain unknown. Surrogate models have helped address some of our questions but also have created others. Ion transport proteins facilitate the transepithelial transport of ions and their patterns of expression determine the ion regulatory function of the epithelium. The following hypotheses, formed from current models of ion regulation, are posed and tested in this thesis using immunological techniques to localize ion transport proteins, which form the basis of the different models. In freshwater teleost fishes, the following hypotheses are tested, 1.) Hypothesis (//): Branchial P V C s are involved in vET-ATPase driven N a + uptake via an E N a C and 2.) H: Branchial M R cells are involved in C l " uptake via an apical A E protein. In marine teleost fishes, the following hypotheses are tested, 1.) H: Branchial P V C s are involved in active Cl" excretion and 2.) H: Branchial P V C s are involved in acid-base regulation via apical N H E and A E proteins. Finally, in the terrestrial mudskipper fish, the following hypothesis is tested, H: The branchial epithelium is the site of active N H 4 + elimination. Immunolocalization techniques enable the determination of the cellular and subcellular distributions of proteins. Since many of the ion transport proteins are common to many vertebrates ranging from human to fishes and even invertebrates and unicellular forms of life, the use of non-homologous antibodies may be useful in the study of the fish gi l l epithelium. Antibodies make very specific, high affinity probes for the detection of proteins of interest. In the next chapter on General Materials and Methods a list o f the antibodies used and some background information w i l l be provided. In the following chapters, the different models of ion regulation in freshwater, seawater fishes and the mudskipper w i l l be addressed. 23 Chapter 2. GENERAL MATERIALS AND METHODS 2.1 Tissue fixation for immunolabeling G i l l tissue was immersion fixed using fixatives designed for post embedding immunolabeling at either light or electron microscopic levels. A . For light level investigations excised gi l l arches were fixed in 3% paraformaldehyde (PFA) / phosphate buffered saline (PBS) p H 7.4 or Bouin's solution (24% formalin, 5% acetic acid, 71% saturated aqueous picric acid, p H 2.2) for 2h at rt or overnight at 4°C. Following fixation tissues were rinsed in P B S and 10% sucrose/ P B S and either frozen in liquid nitrogen or processed for paraffin embedding (Paraplast, Fisher Scientific). B . For immuno-electronmicroscopy smaller pieces of tissue (pairs o f filaments) were fixed in either 2% paraformaldehyde, 7 5 m M L-lysine, l O m M sodium m-periodate (PLP) fixative (McLean and Nakane 1974) or by a two step 3% P F A , 2 0 m M ethylacetimidate, PBS and 3% P F A , 0.1% Glut, P B S fixation procedure (Tokuyasu and Singer 1976). Following fixation either overnight at 4°C or 1-2 h at rt, tissue was rinsed in P B S and free aldehyde groups were quenched in a 5 0 m M N H 4 C I solution for 20min. The tissue was then dehydrated through a decreasing temperature (rt to -20°C) ethanol series (1 h at 30%, 50%, 70%, 95% and 3x 100% EtOH) , gradually infiltrated with resin (EtOH:Unicryl ; 2:1,1:1,1:2 for 30 min at -20°C and 100% Unicryl 2 x l h and overnight at - 20°C) and transferred to Beem caps for embedding (BioCe l l Intl. U K ; Scala et al. 1992). The resin was polymerized by U V light at -10°C over 3 days. Note: In Chapter 5 Material and Methods there is a section on tissue fixation and processing for routine electron microscopy. 2.2 Immunocytochemistry 2.2.1 Immunofluorescence microscopy Cryosections (5-10 ixm) were cut on a cryostat at -20°C, collected onto either 0.01% poly-lysine (Sigma) coated slides or electrostatically charged slides (SuperFrost Plus, Fisher Scientific) and fixed in acetone at -20°C for 5 minutes. Sections were then air dried. Paraffin sections (5 um) were collected onto charged slides, air dried at 37°C overnight, and dewaxed though a series o f xylene baths and rehydrated through an ethanol series finishing in P B S . Sections were circled with a hydrophobic barrier (DakoPen, Dako D K ) and sections were blocked with 5% N G S / 0 . 1 % B S A / T P B S , pH7.4 for 20 min. Incubation with primary antibody diluted 1:50-1:200 in 1%NGS/ 0.1% B S A / T P B S , p H 7.3 for 1-2 hour at 37°C. Sections were then rinsed in 0.1% B S A / T P B S or P B S followed by incubation with secondary antibody conjugated to fluorescein isothiocyanate (1:50 F ITC, Chemicon Intl. Inc), Texas Red (1:100, Molecular Probes or Jackson Lab), or Cy3 (1:200, Sigma) for 1 h at 37°C. Following rinsing with 0.1% B S A / T P B S , sections were mounted with glycerofPBS or VectaShield mounting media and viewed on a Ziess AxioPhot photomicroscope with the appropriate filter sets. 2.2.2 Immunoelectron microscopy Ultrathin sections were made on a Reichart ultramicrotome and collected onto either formvar or formvar-carbon coated nickel grids. Following air-drying, sections were rehydrated by floating grids on drops of P B S . Grids were then transfered to drops of diluted 1° antisera (1:100) and incubated at rt for l h . Grids were then rinsed and transferred to drops of diluted 2° antisera (1:100) conjugated to colloidal gold (10 or 20nm, Sigma, Chemicon) and incubated for l h at room temperature. Grids were rinsed with P B S and fixed for 10 min in 1% glut/PBS and rinsed again in ddH^O before counter staining with lead citrate (Reynolds 1963) and saturated 25 uranyl acetate. Sections were viewed on a Phillips 300 T E M and photographed with Kodak E M plate fi lm 4489. 2.3 Antigen Retrieval A number of antigen retrieval techniques were employed to enhance immunoreactivity of tissue sections with various antisera used (see Antibodies below). These techniques expose epitopes masked during tissue fixation or cause the refolding of proteins in such a way that they now possess epitopes not normally present. Two unmasking techniques that were most useful included the preincubation of sections with 1%SDS (sodium dodecyl sulphate) in P B S p H 7.3 (Brown, et al. 1996) and heating sections in ddfkO (100°C) (reviews by Werner et al. 1996; Taylor et al. 1996). Enzymatic treatment using trypsin was also attempted (Curran and Gregory 1977; Wilson et al. 1997) but abandoned in favour of the other techniques. The SDS pre-treatment was conducted on sections mounted on charged slides only (Brown et al. 1996). Following dewaxing of paraffin sections or air drying of cryosection, the sections were circled with a hydrophobic marker and flooded with 1% SDS in P B S , p H 7.3 for 5 min at rt. The sections were then rinsed in a stream of P B S and put through a series of P B S baths (Coplin jars; 5,10,15min) with gentle agitation. The sections tended to be very hydrophobic so care had to be taken when applying the blocking buffer so as to make sure the sections did not dry. The standard immunolabeling protocol was then followed (see section 2.2.1 on Immunofluorescence microscopy). The heat denaturation technique was conducted on paraffin and cryo-sections mounted on either charged or poly-l-lysine coated slides. After, their respective dewaxing or air drying, slides were placed in boiling water for 10 to 30 min. The slides were placed within a coplin jar submerged within a beaker of boiling water on a hot plate to avoid bubbles disturbing the 26 sections. Sections were then air-dried and circled with a hydrophobic pen, the sections blocked and the standard immunolabeling protocol followed through (see section 2.2.1 on Immunofluorescence microscopy). 2.4 SDS-PAGE and Western Analysis Tissue was thawed in ice-cold SEI buffer (300 m M sucrose, 2 0 m M E D T A , lOOmM imidazole, p H 7.3) and filaments were scraped from the arch with a blunt razor blade. Filaments were then homogenized with a Potter-Elvehjem tissue grinder on ice (1000 rpm, 10 strokes). The homogenate was centrifuged at 2000 g and the supernatant discarded. The pellet was resuspended in 2.4 m M deoxycholate in SEI buffer and re-homogenized. Following a second centrifugation at 2000 g the supernatant was saved. Total protein was measured by either the Bradford (Bradford 1976) or Lowry (Lowry et al. 1951) methods using a B S A standard and the homogenate diluted to lug-uf 1 in Laemmli's buffer (Laemmli 1970). Proteins were separated by polyacrylamide gel electrophoresis ( P A G E ) under denaturing conditions described by Laemmli (1970) using a vertical mini-slab apparatus (Bio-Rad, Richmond C A ) . Proteins were transferred to either Immobilon P membranes ( P V D F polyvinylidene difluoride; Millipore) or nitrocellulose membranes using a semidry transfer apparatus (Bio-Rad; Blattler et al. 1972; Towbin et al. 1979). Blots were then blocked in 3% skim milk/ T T B S (0.05% Tween 20 in Tris Buffered Saline: T r i s -HCl ; 500 m M N a C l 5 m M KC1, p H 7.5). Blots were incubated with primary antisera for 1 h at rt or overnight at 4°C. Following a series of washes with T T B S , blots were incubated with either goat anti-rabbit or anti-mouse H R P (horse radish peroxidase) or A P (alkaline phosphatase) conjugated antibody (Sigma). Bands were 27 visualized by enhanced chemiluminescence with H R P ( E C L ; Amersham) or B C I P / N B T (5-bromo-4-chloro-3-indolyl phosphate/ nitroblue tetrazolium) reaction with A P . 2.5 Antibodies The antibodies used in this thesis are described in detail below and are also found listed in Appendix 1. A list o f species' crossreactive is found in Appendix 2. 2.5.1 v H + - A T P a s e The V-ATPase was immunolocalized using a rabbit polyclonal antibody raised against a synthetic peptide corresponding to a sequence from the catalytic 70 k D a A-subunit of the bovine V-type H + -ATPase complex ( C S H I T G G D I Y G I V N E N ; Sudhof et al. 1989) (Protein Service Laboratory, University of British Columbia). This sequence is wel l conserved from plants, and fungi. The peptide was conjugated to keyhole limpet haemocyanin ( K L H ) using a maleimide linker (Pierce) and (300p.g) mixed with Freund's complete adjuvant (Sigma). N e w Zealand White rabbits were immunized by subcutaneous injection and boost injections (300pg) with incomplete Freund's adjuvant followed at biweekly intervals with test bleed. The rabbits were terminally bled and the sera collected. A pre-immunization serum was collected for use as a control. Sera were tested by peptide E L I S A and Western blot analysis o f gi l l homogenates. A polyclonal antibody raised against the same peptide (Sudhof et al. 1989) has been used to identify the vH/1"-ATPase A-subunit distribution in mammalian kidney (Madsen et at. 1991; K i m et al. 1992). Using rat kidney as a positive control tissue I found identical results as those published above. A rabbit polyclonal antibody raised against a peptide sequence from the E-subunit of the vH + -ATPase was kindly donated by S.F Perry and J .N. Fryer (University of Ottawa). This is the same antibody used by Sullivan et al. (1995). I was, however, unable to reproduce the results of 28 Sullivan et al. (1995) or Perry and Fryer (1997) on rainbow trout with the aliquot of antisera provided. 2.5.2 ENaC (Epithelial Na + channel) The following epithelial N a + channel antibodies were kindly provided by Dale Benos (Department of Biophysics and Physiology, University of Alabama at Birmingham). The amiloride sensitive sodium channel was immunolocalized using a rabbit polyclonal antibody raised against a synthetic peptide corresponding to amino acids 411 to 420 of the p subunit of the human epithelial N a + channel clone (PnENaC; C G E K Y C N N R D F ; D.J . Benos, unpublished, McDonald et al. 1995). The IgG used was purified on a protein A column (0.8 mg • ml" 1 in glycine/tris buffer p H 7.5). The a subunit bovine E N a C (a-bENaC) rabbit polyclonal antibody was raised against a full length fusion protein (Ismailov et al. 1996) generated from the c D N A a bENaC clone (Fuller et al. 1995). The IgG was purified from whole serum on a protein A column (1.4 mg • ml" 1 glycine/tris buffer pH7.8). A rabbit polyclonal antibody raised against biochemically purified bovine renal papilla amiloride-sensitive epithelial N a + channel complex ( a E N a C ; Sorscher et al. 1988). The IgG fraction was purified from whole serum on a protein A column (1.48 mg • ml" 1 glycine/tris buffer p H 7.4). This antibody has been used in a number of different studies to localize both low and high amiloride sensitive N a + channels in a number of different tissues (eg. Smith et al. 1993; Brown et al. 1989) 2.5.3 Na+,K+-ATPase G i l l N a + , K + - A T P a s e was immunolocalized using a monoclonal antibody specific for the a subunit of chicken N a + , K + - A T P a s e (Takeyasu et al. 1988). The antibody (a5) developed by 29 D . M . Fambrough (Johns Hopkins University, M D ) was obtained from the Developmental Studies Hybridoma Bank maintained by the University of Iowa Department of Biological Sciences, Iowa City, I A 52242, under contract NOl-HD-7-3263 from the N I C H D . The antibody was purchased as culture supernatant (0.9 mg • ml" 1). This antibody is now in routine use for identifying g i l l M R cells (Witters et al. 1996; Lee et al. 1998; Dang et al. 1999; Piermarini and Evans 1999). 2.5.4 A E (Cl" / H C 0 3 A n i o n exchanger) Polyclonal antibodies were generated against rainbow trout erythroid band 3 protein (AE1) purified by SDS polyacrylamide electrophoresis (under reducing conditions) (Cameron et al. 1996). This antibody has been shown to crossreact with trout and lamprey erythrocyte preparations by Western analysis (Cameron et al. 1996). Unpurified rabbit serum was kindly provided by Bruce Tufts (Queens University, Kingston ON) . I attempted to use a rabbit polyclonal antibody generated against the native 43-kDa fragment cytoplasmic domain of human erythroid band 3 (Appell and L o w 1981; Verlander et al. 1988) (Philip L o w Department of Chemistry, Purdue University, West Lafayette, Indiana). This antibody has been used to immunolocalize A E in the basolateral membrane of collecting duct a -type intercalated cells (Wagner et al. 1987; Verlander et al. 1988; Drenckhahn et al. 1987; Madsen, et al. 1991; K i m et al. 1992). However, it was not found to be crossreactive with fish tissues (gill tissue or erythrocytes). I also tested a monoclonal antibody generated against a synthetic peptide ( N R S L A G Q S G Q G K P R ) corresponding to amino acids 871 to 884 in the deduced primary structure for human kidney A E 2 (Martinez-Anso et al. 1994; Eduardo Martinez-Anso Department of Medicine and Liver Unit, University Clinic and Medical School, University of 30 Navara, Pamplona, Spain). This amino acid sequence corresponds to the Z-loop, characteristic of non-erythroid A E . This antibody has been shown to cross react with the luminal membrane of shark rectal gland tissue (George et al. 1998). I was unable to obtain cross reactivity with teleost fish gi l l tissue either by Western analysis or immunohistochemistry with either the human A E 1 or A E 2 antibodies. Rat kidney tissue was used as a positive control. 2.5.5 NHE (Na+ / H + exchanger) Rabbit polyclonal antibodies generated against glutathione S-transferase fusion proteins incorporating the last 87 amino acids of N H E 2 (Tse et al. 1994) and 85 amino acids of N H E 3 (Hoogerwerf et al. 1996). The N H E 2 antibody (597) and two polyclonal antibodies (1380 and 1381) against N H E 3 were the kind gift of Mark Donowitz (Departments of Medicine and Physiology, Gastrointestinal Division, The Johns Hopkins University School of Medicine, Baltimore M D ) . Antibody specificity has been determined in N H E expressing PS 120 cell lines. These antibody have been used in a number of different studies for immunolocalization and Western analysis of N H E 2 and 3 (He et al. 1997; Lee et al. 1998; Levine et al. 1993; Sun et al. 1997). Three monoclonal antibodies generated against a maltose binding protein fusion protein that contained the carboxyl terminal 131 amino acids of N H E 3 were also tested. (Biemesderfer et al. 1997). I was unable to demonstrate crossreactivity o f gi l l tissue by either Western analysis or immunohistochemistry with these three commercially available monoclonal antibodies (Chemicon Intl. C A ) . 2.5.6 CFTR (Cystic Fibrosis Transmembrane Regulator) A commercial monoclonal antibody specific to human C F T R (165-170 kDa) was purchased from NeoMarkers Inc. ( C A ) . The antibody was raised against a full length human 31 C F T R recombinant protein and the I g M purified from ascites fluid by ammonium sulphate precipitation (0.2 mg • ml" 1 l O m M P B S p H 7.4, with 0.2% B S A and 15 m M sodium azide). This antibody did not cross react with dogfish rectal gland D F T R cryosections and has only been reported to cross react with human tissue. 2.5.7 N K C C ( N a + : K + : 2 C 1 " cotransporter) G i l l N a + : K + :2C1" cotransporter (145- 205 kDa) was immunolocalized using monoclonal antibodies against shark rectal gland N K C C (J3) (Lytle et al. 1992) and human colonic N K C C 1 (T4) (Lytle et al. 1995). The antibodies (T4, J3) developed by Christian Lytle (Division of Biomedical Sciences, University o f California, Riverside, C A ) were obtained from the Developmental Studies Hybridoma Bank maintained by the University of Iowa Department of Biological Sciences, Iowa City, I A 52242, under contract N O l - H D - 7 - 3 2 6 3 from the N I C H D . The antibodies were both purchased as cell culture supernatant (0.8 mg • ml" 1). Fixed-frozen dogfish (Squalus acanthias) rectal gland tissue was used as a positive control. The T4 monoclonal antibody has greater species crossreactivity than J3 whose immunoreactivity is limited to the shark tissues (Lytle et al. 1995). 2.6 In vitro ATPase assay N a + , K + - A T P a s e and vfT-ATPase activities (umol P i • mg"1 protein • h"1) in gill homogenates were determined at 22.5°C in a plate reader (Thermomax, Molecular Devices Corp., C A ) using a protocol modified from McCormick (1993). In short, ATPase activities were measured using the coupled-enzyme method of Scharschmidt et al. (1979). ATPase activity was measured under three conditions for each sample in triplicate 1.) 200ul reaction mixture + 50ul salt solution + 5 p i homogenate, 2.) 200ul reaction mixture containing 1.0 m M ouabain (Sigma) + 50ul salt solution + 5ul homogenate 3.) 200ul reaction mixture containing lOOmM KNO3 + 32 50ul salt solution + 5ul homogenate. The reaction solution contained 5 0 m M Imidazole, 2.8 m M PEP (phosphoenolpyruvate), 0.22 m M N A D H (nicotinamide adenine dinucleotide, reduced), 0.7mM A T P (adenosine triphosphate), 4U/ml L D H (lactic dehydrogenase) and 5 U / m l P K (pyruvate kinase) and the salt solution contained 50 m M Imidazole, 189 m M N a C l , 10.5 m M M g C l and 42 m M KC1. N a + - K + ATPase and vH + -ATPase activities were defined as specific inhibition by 1.0 m M ouabain and 100 m M KNO3, respectively. However, in chapter 4, v H + -ATPase activity in g i l l crude membrane preparation was measured as N E M (N-ethylmaleimide) sensitive activity according to L i n and Randall (1993). Additional KNO3 and bafilomycin A l dose response curves were constructed. Since D M S O was used to dissolve the bafilomycin A l and it interferes with the assay, an equivalent amount of D M S O was added to paired wells for use as a control. The bafilomycin A l stock concentration was determined photometrically at 280nm using its extinction coefficient of 12100 M / c m and subtracted from a D M S O blank (Werner and Hagenmaier 1984). 33 Chapter 3. THE FRESHWATER FISH GILL. 3 SUMMARY Teleost fishes, living in fresh water, engage in active ion uptake for ion homeostasis. Current models of N a C l uptake involve N a + uptake via an apical arniloride-sensitive epithelial N a + channel (ENaC) energized by an apical vH + -ATPase or alternatively by an arniloride-sensitive N a + / H + exchange (NHE) protein and apical Cl" uptake mediated by an electroneutral, SlTS-sensitive C l / H C C y anion exchange (AE) protein. Using non-homologous antibodies, I have determined the cellular distributions of these ion transport proteins to test the predicted models. N a + , K + - A T P a s e was used as a cellular marker for differentiating branchial epithelium mitochondria-rich (MR) cells from pavement cells (PVCs) . In both the freshwater tilapia (Oreochromis mossambicus) and trout (Oncorhynchus mykiss) the vH + -ATPase and E N a C co-localized to P V C s , although in the trout apical membrane labeling is also found in M R cells. In the freshwater tilapia, apical A E immunoreactivity is found in the M R cells. Thus, a freshwater-type M R chloride cell exists in teleost fishes. The N H E - l i k e immunoreactivity is associated with the accessory cell type in freshwater fishes in addition to a small population of P V C s in tilapia. 34 3.1 I N T R O D U C T I O N In freshwater fish, the active uptake of sodium and chloride is necessary for ionic homeostasis. The consequence o f living in a hypoosmotic medium and having a large surface area (the gill) subject to large outward ion gradients is the continual loss of salts by passive diffusion and gain of water by osmosis. The fish manages the water gain by producing copious amounts of urine however this also adds to ion loss problems, as the kidney is not capable of reabsorbing all the salts. Active branchial ion uptake is crucial in ion homeostasis. The uptake of N a + and C l " is achieved by electroneutral N a + / H + and CI7HCO3" exchange mechanisms (Krogh 1939). 3.1.1 Sodium uptake models Sodium uptake is mediated by an amiloride-sensitive N a + / H + exchange which is either directly or indirectly coupled and shows saturation kinetics (Wright 1991; Potts 1994). It is difficult to explain N a + uptake using the directly coupled carrier mediated N a + / H + exchange (NHE) mechanism because of the absence of physiologically relevant gradients to drive the exchange process under many natural conditions (reviewed by L i n and Randall 1995). The N a + levels in the epithelium (6.4-16.5 m M in isolated P V C s and M R cells; L i et al. 1997) are typically greater than in the freshwater environment ( < l m M ) where the sodium chemical gradient is insufficient to drive N a + / H + exchange (Avella and Bornancin 1989). Wright (1991) calculated that a p H gradient of 0.3 units would be required to drive N a + uptake via a N H E but acid excretion and N a + uptake have been measured at water pHs well below that o f the epithelium (pHi 7.4 Wood 1991). However, a N a + / H + exchanger could explain N a + uptake by a fish in water of 0.5 m M [Na + ] , p H 8.0 (Wright 1991). The amiloride concentration typically used (10"4 M ) does not distinguish between the N H E and E N a C routes o f N a + uptake (e.g. Wright and 35 Wood 1985; Clarke and Potts 1998a). The N H E specific inhibitor, amiloride analogue D M A , had no effect on N a + uptake rates in isolated gi l l filament preparations (Clarke and Potts 1998b). The alterative model employs an apical electrogenic vacuolar-type proton ATPase and an amiloride sensitive N a + channel. This indirectly coupled exchange has been proposed for frog skin and several other iono acid-base regulating epithelia (Harvey 1992; Stetson and Steinmetz 1985). The frog skin model (Ehrenfeld et al. 1985) for sodium uptake has also been proposed by Avel la and Bornancin (1989) to operate in freshwater fish. This model has been further substantiated by the work of L i n and Randall (1991,1993) and co-workers (1994). The basis for the mechanism is an apical (V-type) proton-ATPase which pumps protons across the apical membrane creating a membrane potential capable of driving passive sodium influx via an amiloride sensitive channel against its concentration gradient. The protons are provided from C O 2 hydration catalyzed by intracellular carbonic anhydrase (e.g. Rahim et al 1988; L i n and Randall 1991; Clarke and Potts 1998a,b). The accumulated H C 0 3 " is thought to exit the cell v ia a basolateral C l " / HCO3" exchanger with an associated C l " channel, however, experimental evidence is lacking. N a + movement across the basolateral membrane is facilitated by N a + , K + - A T P a s e (Richards and Fromm 1970; Payan et al. 1975). L i n and Randall (1993) initially predicted the plasma membrane H + -ATPase was a P-type rather than the V-type H + -ATPase based on the pharmacological properties of the enzyme. The relatively low sensitivity of the ATPase to the vH + -ATPase specific inhibitor bafilomycin A l and oxyanion NO3" can be explained by the level of purification of their membranes (Drose and Altendorf 1997; Dschida and Bowman 1997). The immunolocalization of V-type H + -ATPase to the apical membrane of pavement cells (Lin et al. 1994; Sulivan et al. 1995) and correlative data from x-ray microanalysis (Morgan, et 36 al. 1994, Morgan and Potts 1995) and morphometric-ion flux studies (reviewed by Goss et al. 1995) suggest that sodium uptake is performed by pavement cells. However, L i n et al. (1994) also reported apical vH + -ATPase labeling in M R cells. In addition, high levels of N a + , K + -ATPase activity were associated with the M R cell and the levels in P V C s were below detection (Witters et al. 1996). Apart from the information that an arniloride-sensitive N a + channel exists on the apical surface, the localization of N a + channels to the P V C or M R cell apical membrane has yet to be demonstrated. 3.1.2 Chloride uptake model Evidence for an epithelial C l " / HCO3" exchanger in fish gills comes from kinetic, pharmacological, and correlative morphological studies (reviewed by Goss et al. 1995). C l " influx has been shown to be stimulated by infusion of HCO3" (Kerstetter and Kirschner 1972) and there is a good correlation between the rate of C l " uptake and base secretion (de Renzis and Maetz 1973). More recently, analysis using two substrate enzyme kinetics has further suported a 1:1 C l " / H C 0 3 " exchange mechanism (Wood and Goss 1990). The use of the anion transport inhibitor SITS (4-acetamido-4'-isothiocynato-stilbene-2,2'-disulfonic acid; Cabantchik and Rothstein 1972) in the external bathing medium resulted in a reduction of C l " influx after both short and longer term treatments (66% Perry et al. 1981, 71%) Perry and Randall 1981, repectively). Interestingly, no effect of short term SITS exposure was seen on net proton efflux (Lin and Randall 1991). N o studies on a basolateral SITS sensitive anion exchanger have been conducted. However, apically applied thiocyanate (SCN"; non-specific anion transport inhibitor) has been shown to inhibit C l " uptake while S C N ' injection resulted in no change in C l " influx (de Renzis 1975). Inhibition of carbonic anhydrase activity by acetazolamide also reduced C l " influx (Maetz and Garcia-Romea 1964). 37 It is, however, somewhat unclear how thermodynamically feasible C I 7 H C O 3 " exchange would be as a mechanism for C l " uptake in the freshwater fish gi l l (Wright 1991; Perry 1997). X -ray microanalysis would indicate total intracellular C l " levels of 4 0 m M (Morgan et al. 1994) however intracellular C l " activities would likely be much lower. Wright (1991) calculated that HCO3" driven C l " uptake is possible assuming similar [Cl"]j as the frog skin (5-20mM) and [HC0 3 "] i similar to plasma levels (5mM) and water [Cl"] o f 0.5 m M and [HC0 3 "] of 1 m M . It should be noted that [HCCVJj has not been measured in the fish g i l l and intracellular carbonic anhydrase and cellular acid excreting mechanisms ( N H E and vH*-ATPase) may be important in elevating intracellular H C 0 3 " levels to make C17HC0 3 " exchange a more thermodynamically feasible mechanism for C l " uptake. In the freshwater fish g i l l , C l " uptake is thought to occur through the M R cells. Evidence for this location is based on correlative studies of M R cell fractional surface area and C l " fluxes (review by Goss et al. 1995). In addition, various disturbances resulting in changes in intracellular ion concentrations as measured using x-ray microanalysis tend to support this idea of a freshwater chloride cell (Morgan et al. 1994; Morgan and Potts 1995). It is possible that the mechanism for C l " uptake by the M R cell is similar to that proposed for the toad skin (Larsen 1991) and Chinese crab (Onken et al. 1991, 1995). In the freshwater adapted crab (Eriocheir sinensis) transcellular SITS sensitive chloride influx has been shown to be driven by a V-type ATPase (Riestenpatt et al. 1995). In this animal, chloride influx was sensitive to externally applied bafilomycin A i but insensitive to serosally applied ouabain. Larsen (1991) has given these cells the name y-cells. In this chapter I w i l l demonstrate the presence and branchial distribution of the amiloride sensitive sodium channel, N a + / H + exchanger, and SITS sensitive anion exchanger 38 immunologically through the use of non-homologous antibodies. If the v H -ATPase - E N a C model is to be valid then co-localization must be demonstrated. Further clarity of the cell types involved w i l l be provided by the specific labeling of M R cell by a N a + , K + - A T P a s e antibody (Witters et al. 1996). The specificities of the antibody probes used w i l l be verified by Western analysis. In addition, the sensitivity o f the gi l l vH + -ATPase activity to bafilomycin A l and KNO3 as measured by L i n and Randall (1993) wi l l be readdressed using purified g i l l membrane preparations. Data w i l l be presented on two freshwater species of fish: the rainbow trout (Oncorhynchus mykiss) and the tilapia (Oreochromis mossambicus) although data on the freshwater coho salmon (Oncorhynchus kisutch) is also presented in chapter 4. 3.2 M A T E R I A L S A N D M E T H O D S 3.2.1 Animals Four adult tilapia, Oreochromis mossambicus, were obtained from the SkekKipMei fish market, Kowloon, Hong Kong. The fish were kept in glass aquaria within a recirculation system and fed commercial fish food. The fish were approximately 500 g and were kept less than one week. The water temperature was 26°C and lighting conditions were not controlled. Water composition (mM): N a + 0.435; C l " 0.457; Cd^ 0.395; p H 8.1. Adult tilapia were also obtained from the Cambie Seafood Market in Richmond. These animals were approximately 1kg and were kept less than 2 days in the lab and not fed. These animals were only used for ATPase activity measurements. Rainbow trout, Oncorhynchus mykiss, were obtained from local suppliers in the Vancouver and Ottawa areas (Canada) and maintained under local conditions (Table 3.1). Fish were fed commercial trout chow and kept under a natural photoperiod. 39 3.2.2 Tissue Sampling and Fixation Fish were quickly netted and kil led by a blow to the head. The second gi l l arch from both left and right sides was excised and immersion fixed in 3% P F A / P B S at ambient temperature (26°C) for 2h. Tissue was then rinsed in P B S and then either frozen (liquid nitrogen) or processed for conventional paraffin embedding (HistoPrep, Fisher Scientific). Fixed-frozen tissue was stored at - 7 0 ° C until required for sectioning. Additional gi l l tissue was also freeze-clamped (liquid nitrogen) and stored frozen for later Western analysis and measurement of ATPase activities. Some fish had their gills perfused for 5 min with ice-cold heparinized saline via the cannulated bulbus arteriosis to remove blood. The use of paraffin embedded tissue allowed large areas to be sectioned. Basically, sections incorporated leading and trailing edges as well as lamellar and interlamellar filament epithelia. 3.2.3 Tissue preparation for Western analysis and ATPase assay Most tissue for Western analysis was prepared using the crude membrane method of Zaugg (1982). Briefly, g i l l tissue was scraped with a microscope slide into ice-cold SEI buffer (0.3 M sucrose, 0.02 M E D T A , 0.1 M imidazole p H 7.3). Tissue was then homogenized using a Potter-Elvehj em tissue grinder. The homogenate was centrifuged at 2000g for 10 min, the supernatant was discarded and the pellet resuspended in SEID (2.4 m M deoxycholate in SEI buffer) with the homogenizer. The homogenate was again centrifuged at 2000g for 10 min and the supernatant saved. Total protein was measured using either the Bradford (1976) or Lowry (1951) method and a B S A standard. A more refined method was used for preparing membranes for the ATPase assay and some Western analysis. Saline perfused gi l l tissue was prepared by differential centrifugation using a protocol adapted from Fl ik and Verbost (1994), and Dubinsky and Monti (1986). The 40 Table 3.1 Profiles of Ottawa and Vancouver tap waters in which rainbow trout were reared. m M Ottawa Vancouver [Na +] 0.15 0.02 [Of] 0.15 0.01 [ C a - ] 0.45 0.03 p H 7.5 5.8-6.4 Bindonera/ . 1994 GVRDwaterworks 41 gills were perfused with cold heparinized modified Cortland's saline (Mommsen and Hochachka 1994) for 5 min and the gi l l arches excised. The tissue was scraped with a glass microscope slide onto an ice-cold piece o f glass (-2.5 g material from 150 g fish). The scrapings were then suspended in erythrocyte lysis buffer (9 parts 0.17 M NH4CI: 1 part 0.17 M Tr i s -HCl , pH7.4, lOpg/ml aprotinin; 40 ml for approximately 2.5 g scraped tissue) for 20 minutes at room temperature. Intact branchial cells and pieces of tissue were collected by low-speed spin (200 g for 10 min at 4°C), and the supernatant discarded. The pellet was resuspended in 20-40 ml hypotonic buffer (25mM N a C l , I m M HEPES-Tr i s p H 8.0, 1.0 m M D T T , lOpg/ml aprotinin) and homogenized using a loose fitting dounce homogenizer (20-40 strokes). A low-speed spin followed to remove nuclei and cellular debris (550 g for 10 min at 4°C) . The supernatant was decanted and centrifuged at 13 000 g for 10 min at 4°C. The mitochondrial pellet was discarded. The supernatant was decanted and centrifuged at 33 000 g for 45 min at 4°C and the membrane pellet saved. The resulting pellet was resuspended with a 23 gauge needle in 0.5-1.0 ml of suspension buffer ( lOOmM mannitol, 5 m M H E P E S p H 7.6, 10 pg/ml aprotinin) and either frozen or layered on a sucrose step gradient (5-25% sucrose, 1 M K B r 10 m M H E P E S , p H 7.4). A n attempt was made to isolate apical membranes from basolateral membrane fractions using a sucrose gradient. The membrane suspension was layered on top of the sucrose step gradient (25, 20, 15, 10 and 5% sucrose) and centrifuged at 100 000 g for 120 min at 4°C. The layers were removed by puncturing the side of the tube with a 23 g needle and 10 ml syringe. Membrane fractions were resuspended in +5 ml suspension buffer and centrifuged at 150 000 g for 15 min. Pellets were resuspended in 200ul suspension buffer and 50 p i aliquots taken for total protein measurement and remainder frozen (-70°C). Total protein was measured using a modified Bradford (1976) method using a B S A standard (Simpson and Sonne 1982). 42 3.2.4 A TPase Activity vH + -ATPase activities (umol P i • mg"1 protein • h"1) in trout gi l l membrane prepared by differential centrifugation were determined at 22.5°C in a plate reader (Thermomax, Molecular Devices Corp., C A ) modified from McCormick (1993). Dose response curves were also constructed for KNO3 (0 to 100 m M ) and bafilomycin A l (0.01 u M to l u M ) . See General Materials and Methods for additional details of the assay. 3.2.5 Immunocytochemistry, Western analysis, and Antibodies Employed The distribution of band 3-like anion exchanger (AE) , epithelial N a + channel (ENaC), N a + / H + exchanger (NHE2,3) , vH + -ATPase A and E subunits, N a + , K + - A T P a s e , and Na + :K + :2C1" cotransporter ( N K C C ) were determined by indirect immunofluorescence and immunoperoxidase methods in paraffin and fixed frozen gi l l tissue as well as unfixed frozen tissue. Antibodies against C F T R , and A E 2 were not crossreactive with epithelia. See General Materials and Methods for additional information. 3.3 R E S U L T S 3.3.1 T i l ap ia vH*-ATPase The rabbit polyclonal anti-peptide antibody to the A subunit of the V-type proton ATPase cross reacts with a population of squamous epithelial cells covering the lamellae and filament on the up stream or efferent (leading) side (Figure 3.1 A ) . There is an absence of immunoreactive epithelial cells towards the downstream or afferent (trailing) side of the filament (Figure 3.IB). Neither pillar cells, erythrocytes, nor mucocytes show any labeling. In Western blots bands in the high 70 kDa M W range are recognized (Figure 3.IE). Na+X-ATPase 43 The distribution of N a + , K + - A T P a s e as determined using the mouse monoclonal antibody to the a subunit o f the N a + , K + - A T P a s e is mainly restricted to a population o f cells concentrated on the afferent side of the filament (Figure 3 . ID, 3.2C, 3.6A). These cells are frequently found on the trailing edge and in the interlamellar spaces of the filament epithelium. They also are found on the lamellae although generally towards the base. There are clearly fewer immunoreactive cells toward the efferent side of the filament (Figure 3.1C). Immunopositive cells are strongly labeled and ovoid or cuboidal in appearance and sometimes associated with an apical crypt. Nucle i also can be made out as a negative image against the cytoplasmic labeling. Mucocytes that have a similar shape as M R cells and which also appear in the trailing edge of the filament epithelium show no labeling (Figure 3.6A). Control labeling with normal mouse serum and buffer also produces negligible levels of fluorescence (Figure 3.2D). In Western blots a 116 kDa and a weaker 78 kDa band are recognized (Figure 3.IF). CT / HCO3 anion exchanger (AE) The polyclonal antibody generated against trout erythrocyte A E 1 ( A E l t ) crossreacts strongly with tilapia erythrocytes regardless o f the fixation conditions used (Figure 3.2A). However, in order to achieve crossreactivity with epitopes in the epithelium a SDS pre-treatment of sections is required. In SDS treated sections, the antibody cross reacts with the apical region of cells that are also strongly immunoreactive for N a + , K + - A T P a s e (Figure 3.2C). There are, however, N a + , K + - A T P a s e immunoreactive cells not associated with apical A E l t . The number of double-labeled cells is greatest in the afferent region of the filament epithelium. Control incubations of sections with normal rabbit serum result in negligible levels of fluorescence (Figure 3.2B). In Western blots, the antibody crossreacts with a 110 k D a band from saline perfused gi l l tissue as well as erythrocytes (Figure 3.3). A n additional band in the 70 kDa M W 44 range is the result of antibody crossreactivity with a contaminate in the electrophoresis set-up which is likely of microbial origin (see Marshall and Williams 1984). Epithelial Na+ channel In the tilapia, the rabbit polyclonal antibody raised against the P subunit of the human epithelial N a + channel specifically labels a population of squamous cells in the epithelium on the efferent side o f the filament (Figure 3.4A). Both lamellar and filament epithelial cells stain in this region. The labeled cells are in the same location as those that were positive for the vH+-ATPase but do not extend as far toward the afferent side of the filament (Figure 3.4A). Pillar cells and erythrocytes show no immunoreactivity. Normal rabbit serum IgG (Figure 3.4B,C) and buffer control sections were also negative. Western blots of tissue homogenates recognize bands in the 98kDa M W range (Figure 3.4D). I was unable to successfully use either the a b E N a C or biochemically purified amiloride sensitive N a + channel antibodies for immunohistochemical localization of ENaC. This was despite trying a number of antigen retrieval techniques and fixation protocols. However, in Western blots crossreactivity can be demonstrated (Figure 3.5A,B). The a b E N a C and E N a C antibodies both recognize bands in the 74 kDa size range. Na+ /H* Exchanger (NHE) In the freshwater tilapia g i l l , the rabbit polyclonal antibody against N H E - 2 crossreacts with cells in both the lamellar and filament epithelia (Figure 3.6B,D). In the afferent region, immunoreactive cells are found predominantly in the interlamellar space of the filament epithelium. These cells are round in appearance and frequently associated with N a + , K + - A T P a s e immunoreactive cells (Figure 3.6A). In contrast, immunoreactive cells in the efferent region are 45 squamous and not associated with N a + , K + - A T P a s e immunoreactive cells (Figure 3.6D). Pillar cells, mucocytes, and erythrocytes show negligible levels of immunoreactivity. In Western blots the N H E - 2 antibody crossreacts with a doublet in the 87 kDa M W range reported for this protein (Figure 3.7). However, there are also immunoreactive bands in the 56, 60 and lOOkDa M W ranges. A comparison of separated purified membrane homogenates on a 5-25% sucrose gradient probed for N H E - 2 and N a + , K + - A T P a s e (as a basolateral membrane marker) indicates that the 87 kDa bands are not in the basolateral fraction of the sucrose gradient fraction. The antibody against the N H E - 3 isoform (Ab 1380) crossreacts with unidentified material in the basal portion of the epithelium that is also recognized by the normal rabbit serum control. This unidentified material has the appearance of a grape-like cluster. 3.3.2 Trout In the rainbow trout, the distribution of the vH + -ATPase and N a + , K + - A T P a s e have been previously described using the same antibodies (L in et al. 1994; Witters et al. 1996, respectively). Western analysis data has also been presented by L i n et al. (1994) for the v H + -ATPase A subunit. The trout reared in Vancouver tap water (ion-poor) have (A-subunit) vH + -ATPase widely distributed through out the branchial epithelium (Figure 3.8 A , B ) . This pattern of labeling is similar to that reported by L i n et al. (1994). However, immunoreactivity is also observed with branchial mucocyte mucin granules. There are many N a + , K + - A T P a s e immunoreactive cells throughout both the lamellar and filament epithelia. The epithelial N a + channel p subunit has an identical staining pattern as the vH + -ATPase in trout (Figure 3.8C). Western analysis reveals immunoreactivity of the ct5 N a + , K + - A T P a s e antibody with a band o f approximately 116kDa, the 46 vH + -ATPase A-subunit anti-peptide antibody around 70kDa and the P and a subunits of the E N a C with bands at approximately 98 and 74 kDa, respectively (Figure 3.10). The gi l l tissue of three of the four Ottawa trout examined have a discontinuous apical distribution of vH + -ATPase . This pattern of labeling is similar to that reported by Sullivan et al. (1995). The number o f N a + , K + - A T P a s e immunoreactive cells is also not as prolific as observed in the Vancouver trout. The fourth Ottawa trout has a vH + -ATPase labeling pattern similar to that observed in Vancouver trout. Immunogold labeling of the vH + -ATPase is associated with the apical plasma membrane of both M R cells and P V C (Figure 3.9). There is also clustered subapical labeling of electron-dense areas suggestive of vesicle labeling. The density of labeling decreases toward the basal portion of the epithelial cells. I was unable to get the vH + -ATPase E-subunit anti-peptide antibody used by Sullivan et al. (1995) to crossreact with tissue sections or Western blots. A l s o the N H E - 2 and N H E - 3 antibodies 597 and 1380, respectively, did not crossreact. The antibody against trout erythroid band-3 protein crossreacts only with erythrocytes and never with the branchial epithelium. A number of antigen retrieval techniques were used but all yielded negative results (1% SDS/PBS , trypsin, and heat). 47 F I G U R E 3.1 Double labeled sections on the tilapia g i l l epithelium from efferent leading (A ,C) and afferent trailing ends (B,D) of the filament using a rabbit polyclonal antibody against the A -subunit of vH + -ATPase ( A , B ) and mouse monoclonal antibody against the a subunit o f N a + , K + -ATPase (C,D) . The arrow indicates a lone M R cell in the efferent filament epithelium. Western blots of crude gi l l tissue homogenate probed with the vH + -ATPase (E) and N a + , K + - A T P a s e (F) antibodies. M W stds: 205, 112, 87, 69, 56, 39, and 33 kDa. Arrows indicate immunoreactive bands. Scale Bar= 50 um 48 FIGURE 3.2 A n SDS treated section of the tilapia gi l l filament trailing edge duel labeled for A E l t (A) and N a + , K + - A T P a s e (C) using a rabbit polyclonal antibody generated against trout erythrocyte band 3 ( A E l t ) and a mouse monoclonal antibody specific of the a subunit of N a + , K + - A T P a s e , respectively. The corresponding phase-contrast image is in panel (E). Arrows indicate cells immunopositive for both A E l t and N a + , K + - A T P a s e . Arrowheads indicate erythrocytes (immunopositive for A E l t only). A similarly treated section (F; phase-contrast) was incubated with normal rabbit serum (B) and normal mouse serum (D) for use as controls for (A) and (C), respectively. Scale bar = 50 pm V 50 51 F I G U R E 3.3 Western blot of lanes loaded with tilapia red blood cell homogenate (rbc), saline perfused crude gi l l tissue homogenate (gill), and loading buffer (blk) probed with the A E l t antibody. The arrow indicates an immunoreactive band with an apparent M W of 116 kDa. The arrowhead indicates a non-specific reaction with a contaminant of the system. 52 rbc gill blk - i * * I • FIGURE 3.4 Indirect immunofluorescence labeling o f (3ENaC in the tilapia afferent lamellar and filament epithelium (A). This is the same region as in Figure 3.1A,C. Paired immunofluorescence (B) and phase contrast (C) images of a control section incubated with normal rabbit serum at an equivalent dilution indicate low non-specific labeling. A western blot of crude gi l l homogenate probed with the pENaC antibody (D). M W stds: 205, 112, 87, 69, 56, 39, and 33 kDa. Scale Bar = 50 um 54 55 F I G U R E 3.5 Western blot of tilapia g i l l tissue homogenate separated on a 10% polyacrylamide gel and probed with antibodies generated against the a subunit of bovine E N a C (A) and biochemically purified bovine arniloride-sensitive N a + channel complex (B). Both antisera recognize a 74 k D a band. 56 A B 2 5 0 -9 8 -6 4 - d *Py1 5 0 -3 6 - ,.. 57 F I G U R E 3.6 Irnmunolocalization of N H E - 2 (b,d) in the tilapia gi l l using a rabbit polyclonal antibody. The section from the afferent area of the filament has been duel label with antibody 597 specific for N H E - 2 (b) and a5 for N a + , K + - A T P a s e (a). The corresponding phase-contrast image is shown in (c). Arrows indicate N a + , K + - A T P a s e positive cells with apical crypts and arrowheads indicate labeled cells without apical crypts. The crossed arrows indicate cell immunoreactive for N H E - 2 . The inset (d) shows N H E - 2 labeling of squamous epithelial cells in the efferent region. Scale bar = 50 um. 58 F I G U R E 3.7 Western blots of a tilapia gi l l membrane preparation separated on a sucrose step gradient (c). Lanes were loaded with lOpg and separated on a 10% polyacrylamide gel and transferred to P V D F membranes and probed with antibodies against the N H E 2 isoform using the polyclonal antibody 597 (a) and N a + , K + - A T P a s e using antibody oc5 (b). Lane 1 was loaded with membrane preparation loaded onto the sucrose gradient. The lane numbers (2-8) refer to the layers collected from the sucrose gradient. Molecular weight standards: 205, 112, 87, 69, 56, 38.5 and 33.5 kDa. Large arrowheads indicate bands of interest, while smaller arrowheads indicate other bands. 60 a 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 = 1 - • § ' c 61 F I G U R E 3.8 Immunolocalization o f vH + -ATPase , N a + , K + - A T P a s e and E N a C in trout gi l l tissue. Indirect double immunofluorescence labeling of vH + -ATPase (green, FITC) and N a + , K + - A T P a s e (orange, Cy3) in unfixed-frozen (A) and fixed-paraffin embedded sections (B). In (A), the asterisks indicate mucus labeling, the arrows double labeled cells and the arrowheads cells immunoreactive for only N a + , K + - A T P a s e . In (B), the arrows indicate double-labeled cells and the arrowheads indicate cells immunoreactive for only vH + -ATPase or N a + , K + - A T P a s e . In micrograph (C) apical immunoreactivity for phENaC is indicated by arrows in unfixed-frozen sections of trout g i l l . Scale bars (A) 100, and (B, C) 25 um 62 \ 7 t 63 F I G U R E 3.9 Imrniinolocalization of vH + -ATPase in freshwater trout g i l l using the immunogold technique. There is apical labeling in branchial pavement cells (a) as wel l as M R cells (b,c). Subapical labeling is also observed clustering in electron dense areas suggestive of vesicular location. The intensity of immunogold labeling decreases towards the basolateral membrane as seen in the M R cell (c). Scale bar s (a,b) 0.5 pm and (c) 2pm. 64 6 5 FIGURE 3.10 Western blots of rainbow trout crude gi l l homogenate loaded onto 10% (a,c,d) or 12%) (b) polyacrylamide gels, transferred to P V D membrane and probed with antibodies against (a) a subunit o f the N a + , K + - A T P a s e , (b) A-subunit o f the vH + -ATPase , (c) P-subunit of the E N a C , and (d) a subunit of the E N a C . M W markers (a) 205, 112, 87, 69, 56, 38.5 and 33.5 kDa; (b,c,d) 250, 98, 64, 50, 36, and 30 kDa. Arrowheads indicate bands of interest. 66 a b e d 67 F I G U R E 3.11 Dose response curve of rainbow trout ATPase activity (mU/h) to bafilomycin A l and K N O 3 . Membranes prepared from crude gil l homogenates using differential centrifugation were used for the measurements of ATPase activity. Bafilomycin A l (solid circles) and D M S O (hollow circles) used as a control, n = 3. 68 • Bafilomycin 0.01 0.1 Bafilomycin A 1 (^M) b 35 -, 69 3.4 D I S C U S S I O N This is the first study to identify the distributions of the apical C17HC0 3 " and N a + / H + exchangers and epithelial sodium channels (ENaC) in the branchial epithelium of a fish. In the two freshwater species examined, the patterns of immunolabeling do show some similarities, but also have some major differences. Na+,Ki-ATPase antibody as a MR cell marker The use of antibodies to the N a + , K + - A T P a s e can be used as a reliable tool for identifying M R cells (Witters et al. 1996; Ura et al. 1996). The a5 monoclonal antibody recognizes a highly conserved epitope within the a-subunit of the N a + , K + - A T P a s e . It has been used to immunolocalize N a + , K + - A T P a s e in animals from invertebrates to mammals (eg Takeyasu et al., 1988; Witters et al. 1995; Zeigler 1997; Barradas et al. 1999; Lignon et al. 1999; Sabolic et al. 1999). In fishes, the a5 antibody makes a good tool to identify M R cells because the unique tubular system of these cells is rich in N a + , K + - A T P a s e in both fresh and seawater fishes although to varying degrees (McCormick 1995). Labeling is generally absent from the P V C but considered to be present at levels below detection sensitivity by immunohistochemistry (or autoradiography) since N a + , K + - A T P a s e is generally considered to be a ubiquitous basolateral membrane protein. The freshwater Chloride cell and the apical anion exchanger The apical localization of the A E in the tilapia M R cell provides convincing evidence that these presumed freshwater chloride cells are aptly named. In seawater fishes, the evidence that branchial M R cells are involved in the active C l " efflux has been until now far more convincing than the evidence that C l " uptake occurs via the freshwater M R cell (Perry 1997). The tilapia M R cell apical A E is immunogenically related to the erythrocyte band-3 o f trout (Figure 3.). This is 70 unlike the pattern seen in higher vertebrates where non-erythroid band-3 like proteins are restricted to the basolateral membrane domain of acid excreting cells (A or ot-type cells of bladder and C C T ; Drenckhahn et al. 1987; Alper et al. 1989). In these higher vertebrates the apical C I 7 H C O 3 " exchanger is an A E 2 isoform (Alper et al. 1997) which is not immunogenically related to band-3 (AE1) and shows less sensitivity to disulfonic stilbenes (SITS; Cohen et al. 1978). In the trout, apically applied SITS has been shown to inhibit C l " uptake and cause an alkalosis; results consistent with an apical C I 7 H C O 3 " exchanger (Perry et al. 1981; Perry and Randall 1981). Sullivan et al. (1996) were also able to localize band-3 m R N A by in situ hybridization using a 28-mer oligonucleotide probe (Kudrycki and Shull 1989) to the filament interlamellar epithelium which is typically populated by M R cells. Thus unlike higher vertebrates, it appears that fishes make use of a band-3 like A E for apical C l " / HCO3" exchange. There was no crossreactivity o f fish tissue with an antibody against mammalian A E 2 (Martinez-Anso et al. 1994) raised against a peptide derived from the Z-loop (highly conserved region among non-erythroid A E and also found in the A E 1 o f fish). A comparison of the amino acid sequences of the synthetic peptide and that o f trout A E 1 Z-loop region show very little similarity. It would thus appear that i f the gill-type band-3 and the erythroid band-3 are indeed closely related then no cross reactivity would be expected. I would like to note that immunolabeling of gi l l tissue with the trout A E l t antibody was only possible after pre-treating the sections with detergent (SDS). In retrospect, such a treatment makes sense as the polyclonal antibody was generated against band-3 protein purified under denaturing conditions ( S D S - P A G E ) . Thus the conformational changes that occur in the native protein with denaturation and subsequent refolding (upon removal of SDS) result in different epitopes. The polyclonal antibody, which is composed of a battery o f different antibodies that 71 recognize different epitopes on the band-3 protein, is capable of recognizing the erythrocyte band-3 protein in a number of different species. These epitopes may not be present in the native protein of either the gi l l epithelium or erythrocyte and thus account for the negative results observed without pretreatment. Erythrocytes of both species showed strong crossreactivity regardless of pre-treatment. It also is possible that SDS pretreatment exposes epitopes masked by tissue fixation (protein crosslinking), however, heat denaturation did not enhance reactivity. So why is it that the antibody raised against trout band-3 does not crossreact with the branchial epithelium band-3 like protein in trout when it does in tilapia? One possible explanation is that the A E expressed in the trout gi l l is an isoform(s) that does not have the epitopes recognized in both fish erythrocytes and tilapia g i l l . Interestingly, salmonids belong to a tetraploid lineage and thus have more gene copies (Wittbrodt 1998). Na+ uptake mechanisms The co-localization o f the vftT-ATPase and E N a C in both tilapia and trout greatly increases the viability of this N a + uptake model in freshwater fishes. This is also the first study to even identify the distribution of the E N a C in a vertebrate lower than an amphibian (Harvey 1992). However, it is interesting that the co-localized vH + -ATPase and E N a C are found exclusively in P V C s in a particular area of the branchial epithelium in one species (tilapia) and found in a mixed population of P V C s and M R cells throughout the branchial epithelium in the other (trout). It seems that the tilapia exclusive P V C distribution follows the predicted distribution (Goss et al. 1995; Perry 1997) while the trout does not. It is not clear in the trout i f N a + uptake differs in P V C s and M R cells. It maybe that the higher N a + , K + - A T P a s e activities associated with the M R cell are required to aid N a + uptake. There is nothing to preclude the possibility that both N a + and C l " uptake mechanisms can be found in a subpopulation of branchial 72 M R cells. In the trout, N a + influx has been shown to be sensitive to SITS and C l " uptake to amiloride (Perry and Randall 1981). A case in point is the amphibian skin M R cells, which have both apical E N a C and A E proteins along with the vH + -ATPase (y-cell; Larsen 1991). I was only able to identify the distribution of the E N a C using the polyclonal antibody against the P-subunit while the a-subunit antibody was negative. Perhaps it is not surprizing that the P-subunit of the E N a C was recognized in fish while the a-subunit was not since there is greater amino acid identity within the p than the a-subunit isoform groups within higher vertebrates (human vs frog; P 79%: a 59%; Garty and Palmer 1997). The finding of N H E - 2 crossreactivity in the branchial epithelium of the freshwater tilapia was not expected. However, Wright (1991) calculated that it is possible to operate a N H E for N a + uptake in water with greater than 0.5 m M [Na +] and a p H greater than 8. Also since the tilapia is euryhaline, the N H E may be expressed in fresh water in expectation of movement into a more saline environment where it could potentially operate for acid excretion or N a + uptake. Claiborne et al. (1999) have identified a N H E 2-like sequence in the marine sculpin while they appeared to be unable to confirm the presence of a N H E - 3 homolog (Blackston et al. 1997). I was also unable to detect any N H E - 3 crossreactivity. The localization of N H E - 2 to subapical cells adjacent to M R cells is interesting and again may be due to the tilapia being a euryhaline fish. Based on the location of these cells it appears they may be a freshwater-type accessory cell. These cells have been found in a number of freshwater euryhaline fishes (Pisam et al. 1988, 1989). The function of the accessory cell is poorly understood and the function of the N H E - 2 in such an arrangement is not known. This finding w i l l be discussed in more detail in chapter 4 which focuses on seawater fishes. vlf-ATPase E subunit and A subunit distributions 73 Within the literature there are two conflicting reports regarding the distribution of the vH + -ATPase in the freshwater rainbow trout (O. mykiss) by L i n and co-workers (1994) and Sullivan and co-workers (1995). L i n and co-workers (1994) using an anti-peptide antibody directed against the A-subunit (Sudhof et al. 1989) found extensive apical staining of the branchial epithelium and concluded that both epithelial pavement cells and M R cells expressed the vH + -ATPase . Sullivan and co-workers (1995) using an anti-peptide antibody directed against the E-subunit (Hemken et al. 1992) found a small population of lamellar cells with apical immunoreactivity. On the basis of immunogold studies they concluded that labeling was restricted to pavement cells. Recently Evans et al. (1999) have suggested that the differences may be due to the rearing conditions of the fish (see Table 3.1). Vancouver tap water has a much lower p H and ionic strength (pH 6.4-6.8; [Na +] O.OlmM) compared to Ottawa tap water (pH 8; [Na +] 0.5mM). Laurent et al. (1985) have shown that such ion poor conditions result in the proliferation of branchial M R cells implying a greater need for ion uptake. With a central role in driving N a + uptake, increased vH + -ATPase would be predicted under such conditions as seen in the Vancouver trout. This does explain the more extensive distribution seen in Vancouver trout but does not explain why the vH + -ATPase should be found in M R cells in addition to P V C s in Vancouver trout. I have been using the same anti-A subunit antibody as L i n et al. (1994) and can confirm their observation but unfortunately my attempts at getting the anti-E subunit antibody used by Sullivan et al. (1995) to work were unsuccessful. However, I have been able to look at g i l l tissue collected from Ottawa trout and can report a similar labeling pattern as observed by Sullivan et al. (1995) using the A-subunit antibody. Additionally, I have also done double labeling 74 experiments and can clearly show that some of the apical v H -ATPase labeling is associated with N a + , K + - A T P a s e immunoreactive cells ( M R cells). This observation would contradict the immunogold studies of Sullivan et al. (1995), but in their paper they admit their observations were not exhaustive and did not entirely exclude the possibility that a subpopulation of M R cells may also apically express the vH + -ATPase . It may also be that different isoforms of the E -subunit are expressed in P V C s and M R cells. Studies of the mammalian kidney have shown that there are heterogenous forms of the E subunit which have different tissue and membrane distributions (Hemken et al. 1992). The antibody that Sullivan et al. (1995) employed was generated against the same synthetic peptide used by Hemken et al. (1992) in the development of a battery o f monoclonal antibodies. The immunologically determined distribution of the E subunit was not the same for all o f the antibodies suggesting that different variants of the E subunit may exist and possibly impart functional variation on the vH + -ATPase . Immunolocalization of the 56 (B) and 70kDa (A) subunits was found not to differ in the rat kidney (Brown et al. 1987a). Comparing results obtained on trout obtained from Ottawa but using the A-subunit probe suggests that the immunoreactivity observed by Sullivan et al. (1995) is localized to g i l l M R cells and not only P V C as had been suggested. There is also an interesting hypothesis put forward by Feng and Forgac (1994) that reducing conditions favour vH + -ATPase activity or conversely that oxidizing conditions inhibit activity. When the A subunit cystein (Cys) 254 is oxidize it forms a disulphide bond with Cys 532 inactivating the ATPase. It has been argued that the 'redox modulation' allows the v H + -ATPase to be active in vesicle acidification under the reducing condition of the cytoplasm but when the vH + -ATPase cycles through the plasma membrane it is inactive. It may also explain why the plasma membrane vH + -ATPase is frequently found in M R cells where reducing 75 conditions can be maintained by mitochondrial cytochrome c oxidase (Harvey and Weiczorek 1997). However, how does the vH + -ATPase manage to operate in the g i l l , which of course is the central gas exchange organ in fishes and subjected to high P 0 2 ? The mitochondria may act as reducing agents within close proximity but it is unclear what may be going on in P V C s . In the brown bullhead, Goss et al. (1992, 1994) have found P V C s which have many mitochondria and which show an increase in microvillar density with hypercapnia. Hypercapnia has been shown to increase vH + -ATPase activity (Lin and Randall 1993) and protein level expression (Sullivan et al. 1995). These cells were distinguished from ' M R cells' by the absence of a tubular system in the brown bullhead (Goss et al. 1992, 1994). In tilapia which have only upstream pavement cells expressing the vH + -ATPase , mitochondria may be numerous, however, in the trout immunogold analysis of P V C s by Sullivan et al. (1995), P V C s were characterized by their absence of mitochondria. I also have used mitochondria density as a characteristic for distinguishing M R cells from P V C s and have observed apical immunogold labeling in cells with few mitochondria. Immunocytochemistry places limitations on tissue fixation or morphological preservation in favour of preservation of antigenicity making the identifcation o f the tubular system within the cell uncertain. Alternatively, the trout vfT-ATPase might possess an A-subunit isoform that has a reduced sensitivity to oxidizing conditions. Nitrate (NO3") acts as an oxidizing agent inhibiting the vH + -ATPase by promoting the formation of disulfide bonds (Dschida and Bowman 1995). The degree o f ' N C V inhibition of total ATPase activity in purified membranes is similar to bafilomycin A l . The question of redox modulation of the teleost gi l l vH + -ATPase wi l l probably be resolved once the A-subunit is cloned. 76 Upstream acidification in tilapia In the tilapia there is an interesting upstream distribution o f the H + excreting mechanisms (vH + -ATPase and NHE-2 ) . One possible explanation of this arrangement may be to aid ammonia unloading by boundary layer acidification (Randall et al. 1991). This seems a more plausible explanation rather than some specific condition necessary for sodium uptake. Acidification of the boundary layer by both H + excretion and hydration of respiratory CO2 (—> H + + HCO3") aids ammonia excretion by maintaining the transbranchial ammonia partial pressure (PNHS) gradient by removing NH3 by protonation to form N H 4 + . The transbranchial PNH3 gradient can account for the majority of the total ammonia efflux (Cameron and Heisler 1983) and the disruption of the boundary layer p H effect by addition of buffer to the water inhibits ammonia excretion (Wright et al. 1989; Wilson et al. 1994). The contribution of CO2 hydration to boundary layer acidification would be greatest downstream, decreasing upstream as Pco2 levels of transiting blood decrease. Presumably, upstream acidification of the boundary layer by the vH^-ATPase and N H E would aid ammonia elimination whilst downstream alkalinization by Cl" / HCO3" exchange would not interfere. The absence o f such a relationship in trout maybe related to a greater need for N a + uptake, H + efflux and/or boundary layer acidification in these animals. Summary Figure 3.12 a and b summarize the immunolocalization data collected from the tilapia and trout, respectively. In the freshwater adapted tilapia and trout there is a clear co-localization of the E N a C with the vH + -ATPase . However, in the trout the vH + -ATPase -ENaC colocalize to N a + , K + - A T P a s e rich M R cells as well as pavement cells whereas in the tilapia the vH + -ATPase-E N a C colocalize exclusively to pavement cells. In the tilapia it is possible to demonstrate the presence of an apical A E with N a + , K + - A T P a s e rich M R cells. The N H E - 2 immunolocalized to 77 subapical ovoid cells in the interlammelar space o f the filament epithelium as well as to pavement cells in the same region as the vH + -ATPase -ENaC. In the tilapia we have a near complete picture o f the organization of N a C l uptake and acid-base regulatory mechanisms, whereas in the trout the picture is far from complete. The absence o f crossreactivity should not be taken to mean that that particular transporter is not present in the gi l l . There are numerous technical reasons for the absence of cross reactivity. However, the partial picture o f the trout gi l l is important in highlighting the point that the organization of the gi l l for iono and acid-base regulation need not be uniform among fishes and that generalizations need to be taken with some caution. There are of course at least 25,000 different species of teleost fishes and our understanding of the few we have studied is still far from complete. 78 F I G U R E 3.12 Illustrations of the freshwater tilapia and coho salmon (a) and rainbow trout (b) branchial epithelium cell types ( M R cells and P V C s ) to summarize the immunolocalization data. In the tilapia (and coho) there is immunological evidence for an apical C I 7 H C O 3 " exchanger in some M R cells (as defined by high N a + , K + - A T P a s e immunoreactivity) or freshwater-type chloride cells (fwCC) and the vH + -ATPase -ENaC in P V C s . In the trout there is evidence of a mixed population of P V C s and M R cells with the vH + -ATPase and E N a C . The N H E was found apically in branchial P V C s in tilapia but was absent from trout. The data on the freshwater coho salmon is presented in chapter 4. 79 (a) Tilapia and Coho salmon Water Blood AE NKA v-ATPase NHE NKCC (b) Rainbow trout Water so Chapter 4. THE SEAWATER FISH GILL 4. S U M M A R Y Teleost fishes l iving in seawater engage in active ion elimination for ion homeostasis. The accepted model of N a C l elimination involved branchial M R chloride cells that have been shown to be involved in active C l " excretion via a basolateral N K C C (Na + :K + :2C1" cotransporter) and apical C F T R (Cystic Fibrosis Transmembrane Regulator)-like anion channel, energized by a basolateral N a + , K + - A T P a s e while N a + moves paracellularly. Acid-base regulation is believed to be mediated by an electroneutral C I T H C C V anion exchanger (AE) and an N a + / H + exchanger (NHE) . Using non-homologous antibodies, I have determined the cellular distributions of these ion transport proteins to test the predicted models in the stenohaline turbot (Scophthalmus maximus) and euryhaline coho salmon (Oncorhynchus kisutch; which was also acclimated to freshwater for comparison). N a + , K + - A T P a s e was used as a cellular marker for differentiating branchial epithelium mitochondria-rich ( M R ) cells from pavement cell s (PVCs) . Branchial M R cells have an apical CFTR- l ike anion channel and basolateral N K C C and N a + , K + - A T P a s e . The N H E - l i k e immunoreactivity is associated with the accessory cell type in both seawater fishes and the A E - l i k e protein with the M R cell apical crypt in turbot. However, the apical AE- l ike protein is not found in the seawater-type chloride cell o f the coho salmon even though it is found in the freshwater counterpart. In the coho salmon it is possible to demonstrate the loss of the freshwater ion uptake mechanisms (vH + -ATPase and A E ) in sea water adapted fishes. The specificity of the N H E - 2 antibody (597) for N H E - l i k e proteins in the gi l l may be questionable (based on Western blot results of multiple band immunoreactivity) but it may still make a useful marker for the accessory cell type. 81 4.1 I N T R O D U C T I O N 4.1.1 Mechanisms of NaCl elimination Fish l iving i n seawater achieve ionic homeostasis through the active elimination of N a C l by the gills (Keys 1931a,b). The model of N a C l elimination in teleost fishes (Silva et al. 1977; Sardet et al. 1979) emerged at about the same time as the avian and shark salt gland models (recent reviews by Evans 1993; Karnaky 1998; Marshall and Bryson 1998; Riordan et al. 1994). Briefly, N a C l elimination by the seawater fish takes place via the mitochondria-rich 'chloride' and 'accessory' cell complex. These M R cell types are found in intimate contact sharing an apical crypt (reviewed by Pisam and Rambourg 1991). The smaller, peripheral accessory cell has less developed membrane systems than the larger chloride cell. Leaky tight junctions are found only between chloride cells and accessory cells (Sardet et al. 1979). The chloride cell is associated with high levels of basolateral N a + , K + - A T P a s e (Karnaky et al. 1976; Hootman and Phillpot 1979) which maintains an inward N a + gradient that is used to drive C l " uptake by a Na + :K + : 2C1" cotransporter (Eriksson et al. 1985). The accumulated intracellular C l " moves across the apical plasma membrane via a CFTR- l ike channel down its electrochemical gradient (Foskett and Scheffey 1982; Singer et al. 1997). N a + on the other hand, which is accumulated in high concentrations in the interstitial space between chloride and accessory cells by the N a + , K + -ATPase, leaks out paracellularly via leaky tight junctions down its electrochemical gradient. (Sardet et al. 1979; Potts 1984). M u c h of the mechanistic understanding of transepithelial ion transport has come from work using in vitro skin preparations (Degnan et al. 1977; see recent review by Marshall and Bryson 1998; Marshall 1995). Notably, it was the vibrating probe experiment on a tilapia opercular epithelium preparation by Foskett and Scheffey (1982) that provided definitive evidence that the M R chloride cell was the source of the outward C l " current. On the basis of 82 similarities in morphology and expression of key enzymes (e.g. N a , K -ATPase) the branchial M R or chloride cells have been ascribed the same function as their opercular epithelium counterparts. Although the opercular membrane preparations have provided convincing evidence that the M R chloride cells are the site of C l " elimination, Avel la and Ehrenfeld (1997; Avel la et al. 1999) have found that primary cultures of sea bass (Dicentrarchus labrax) respiratory or pavement-like cells are capable of active C l " secretion as well . They were unable to find morphological evidence of M R cells in their cultures. Interestingly, on the basis of inhibitor sensitivity o f short circuit current, they concluded that the cultured respiratory-like cells use the same cellular transporters used in the M R cell model of C l " elimination. Basolateral entry of C l " was via a D I D S sensitive C17HCCV exchanger and a bumetanide sensitive N K C C cotransporter energized by ouabain sensitive N a + , K + - A T P a s e . The apical exit was via a D P C and N P P B sensitive, c A M P stimulated C l " channel (Avella and Ehrenfeld 1997). They were also able to provide evidence of an apical C l " channel using patch-clamping techniques (Duranton et al. 1997) and evidence that C l " secretion in these cultured cells is under the hormonal control of arginine vasotocin (AVT)(+), prostaglandin (+) and a (-) and P (+) adrenergic receptors (Avella et al. 1999). Evans et al. (1999) recently estimated from the measured C l " current of the two surrogate model epithelia and relative areas of the two cell types that P V C and C C could contribute roughly equally to total C l " efflux in vivo. Therefore, it appears that some direct observations o f the branchial epithelium chloride cell (CC) , accessory cells (AC) and P V C are warranted to address this blurring picture. 83 4.1.2 Mechanisms of acid-base regulation in fishes (Ct/HCOj and Na+/H+ exchange) Although the N a + / H + exchanger is not thought to operate in the freshwater fish gi l l because of the lack of physiological gradients for driving exchange (Na + uptake), the N a + gradients in marine fish are capable of driving N a + / H + exchange for acid-base regulation ( H + efflux). A s well the favourable inward C l " gradient would be capable of driving HCO3" efflux. The accumulated N a + and C l " could be eliminated by the excretory mechanism described above and has been estimated to be relatively small (-10% of the total unidirectional N a C l efflux; see Claiborne 1998). In support of this model, If1" efflux has been shown to be dependent on external N a + concentration and Claiborne and co-workers (1997) have shown in a marine fish (Myoxocephalus octodecimspinosus) that proton efflux is sensitive to the N H E specific inhibitor 5-N,N-hexamethylene-amiloride. In addition, preliminary findings by Blackston, Claiborne and co-workers (1997) using Northern blot analysis indicate the presence o f m R N A transcripts in M. octodecimspinosus homologous to a human c D N A probe for the N H E - 1 and N H E - 3 isoforms. Claiborne et al. (1999) have more recently confirmed the presence of a (3NHE-like (NHE1) and NHE2- l ike isoforms but not the NHE3- l ike isoform in this species. The N H E - 1 is a basolateral isoform and functions in the housekeeping roles of cell volume and p H i regulation. It is the apical N H E isoforms that are of interest in transbranchial acid-base balance. Claiborne and co-workers (1997) were also able to show that acid excretion is increased by DIDS and low [Cf] in the water. These findings indicate that an apical DIDS sensitive Cl" /HCO3" exchange is occurring. There is virtually nothing known about the distribution of the apical transporters, although basolateral N H E and A E have been found to operate in M R cell p H i regulation (Zadunaisky et al. 1995). 84 1.1.3 Seawater versus freshwater Movement of fishes between marine and freshwater environments implies a switch in ionoregulatory strategy. L i n and Randall (1993) have shown that the vH + -ATPase , which is thought to be involved in N a + uptake in freshwater fishes, is lower in fish acclimated to seawater or solutions with high N a C l . Both vH + -ATPase activity and the cellular distribution are reduced (Lin and Randall 1993; L i n et al. 1994). Concurrent with a downregulation o f freshwater ion uptake mechanisms is an up-regulation of ion excretory mechanisms. G i l l N a + , K + - A T P a s e activities and M R chloride cell numbers greatly increase during sea water acclimation (e.g. Sargent and Thomson 1974; Pisam etal. 19; Marshall et al. 1999; McCormick 1995). In this chapter the ion transport proteins involved in N a C l elimination (Na + ,K + -ATPase , N K C C and C F T R ) are immunolocalized to address the question of whether branchial P V C s could contribute to the efflux. In addition, the ion transport proteins that have been suggested to be involved in acid-base regulation w i l l be immunolocalized. These are the apical N H E 2 and N H E 3-like, N a + / H + exchangers and band 3-like A E (anion exchanger). Finally, changes in ion transport protein expression in fresh water and sea water adapted representatives o f species are addressed, specifically looking at changes in the vH + -ATPase , N a + , K + - A T P a s e , N H E and A E expression patterns. I w i l l present data on the stenohaline marine turbot (Scophthalmus maximus), the euryhaline seawater and freshwater adapted coho salmon (Oncorhynchus kisutch) and finally a limited part of the data on the brackish water mudskipper fish (Periophthalmodon schlosseri). A more extensive data set on the mudskipper w i l l appear in chapter 5 in the study of active N H / elimination by this species. 85 4.2 M A T E R I A L S A N D M E T H O D S 4.2.1 Animals Coho salmon (Oncorhynchus kisutch) o f both sexes were obtained from the Department of Fisheries and Oceans West Vancouver facility which have been reared in either freshwater or seawater. Three to four fishes ranging from 35-50g were sampled on-site from each group for gi l l tissue and intestinal tissue (for an unrelated study). A n additional group of juvenile coho were transported to the Bamfield Marine Station and used in a salinity acclimation experiment. These fish were 157.5+14.5 g (n = 39) and maintained in an outdoor tank in freshwater. Juvenile turbot (Scophthalmus maximus) obtained from a French hatchery were maintained at the Augusto Nobre Labratorio da Foz of the Universade do Porto, Portugal. Fish were fed twice daily, and were control fish used as part of a growth study (Pereira et al. 1999). Adult mudskippers (Periophthalmodon schlosseri) o f both sexes were collected from the Pasir Ris, a small estuary on the eastern coast of Singapore and transported to City University of Hong Kong. In the laboratory, fish in individual plastic aquaria were kept partially submerged in 50% seawater (15 %o) . Every other day the water was changed and the animals were fed goldfish ad libitum. The fish mass ranged from 55 to 120 g. The room air and water temperatures remained constant around 26°C. N o attempts were made to control lighting conditions and a natural photoperiod was followed. 4.2.2 Salinity acclimation experiment In this experiment, coho salmon were gradually acclimated to full strength (32%o) sea water at a rate o f l%o h" 1. Fish (5-9) were randomly netted in freshwater (Oh) and 12h, 24h, 2, 4, and 8 days after the start of the salinity increase. Fish were killed by an overdose of anaesthetic (1:5000 MS-222) and sampled for blood and gi l l tissue. Blood was collected by caudal puncture 86 using a heparinized needle and syringe and plasma was separated and saved. The gi l l arches were excised and divided to be either fixed (3%PFA/PBS) or freeze-clamped and stored in liquid nitrogen. The frozen gi l l tissue was prepared for measurements of ATPase activity (see General Material and Methods section 2.for details). Unfortunately, the fixed tissue was rendered useless by improper storage. Plasma N a + and C l " ion concentrations were measured using an atomic absorption spectrophotometer (Perkin-Elmer 2380) and by coulometric titration (Haake Buchler Instruments digital chloridometer), respectively. G i l l N a + , K + - A T P a s e and vH + -ATPase activities were determined using the microplate method (See General Materials and Methods for details). 4.2.3 Fixation and Immunolabeling Turbot and coho gi l l tissues were fixed by immersion in Bourn's solution and paraffin embedded. Coho gi l l tissue was additionally fixed in 3 % P F A / P B S and either embedded in paraffin or frozen. The standard immunolabeling protocol was followed in addition to the various pre-treament epitope-unmasking techniques. See General Material and Methods for details. 4.2.5 Antibodies employed in this study Sections of turbot and coho gil l tissue were probed with antibodies against N a + , K + -ATPase, N K C C , C F T R , N H E - 2 and 3, A E l t and vH + -ATPase . See General Material and Methods for additional details. 4.2.6 Statistical Analysis Data are presented as mean ± S E M . Statistical differences between the means of the dependent variables were determined by one-way A N O V A and post hoc Student-Neuman-Keuls test. Differences were accepted as significant at the 95% level of confidence (P<0.05). 87 4.3 R E S U L T S Na "/FT exchanger In the branchial epithelium of the stenohaline marine turbot, the polyclonal antibody (597) raised against the N H E - 2 fusion protein immunoreacts with small cells juxtaposed to larger cells showing strong N a + , K + - A T P a s e immunoreactivity (presumed M R cells) in a pair-like fashion (Figure 4.1). Labeling is restricted to the filament epithelium and double labeling of single cells is seldom observed. In addition to being smaller than N a + , K + - A T P a s e immunolabeled cells, N H E - 2 immunoreactive cells display a cytoplasmic labeling pattern with nuclei having negligible immunoreactivity. Lamellar P V C s do not show detectable immunoreactivity, nor do erythrocytes or pillar cells. In both the seawater and fresh water adapted coho salmons' branchial epithelium, a similar labeling pattern as in the turbot is observed (seawater coho; Figure 4.2). N a + , K + - A T P a s e and N H E - 2 immunoreactive cells show alternating labeling and are seldom double labeled. Within the filament epithelium both types of immunoreactive cell tend to be more elongate, however, the N a + , K + - A T P a s e cells remains larger. N a + , K + - A T P a s e cells are also found in the lamellar epithelium while N H E - 2 cells are not. The polyclonal antibody (1380) raised against a N H E 3 fusion protein does not show a specific labeling pattern in either the turbot or coho. A non-specific staining pattern is observed as in the freshwater tilapia. Antigen retrieval techniques do not alter the pattern of labeling relative to normal rabbit serum controls with either o f these antibodies. Band 3-like AE The rabbit polyclonal antibody raised against trout erythrocyte band-3 protein immunolabels the apical crypt region o f the turbot (Na + ,K + -ATPase immunoreactive) M R cells 88 following SDS pretreatment (Figure 4.3). Staining is clearly seen following the edge of the apical membrane of the M R cell. There is also some faint sub-apical labeling of M R cells. Since, SDS pretreatment tends to increase the background signal it is difficult to determine i f other epithelial cells have significant labeling. A comparison of labeling intensity with normal rabbit serum controls does not appear significantly greater in lamellar P V C s while M R cell apical crypt labeling clearly is. In the seawater adapted coho salmon and mudskipper gills, tissue does not immunoreact although erythrocytes show specific labeling (Figure 4.6D). However, in the freshwater adapted coho, M R cells have apical A E immunoreactivity (Figure 4.6A) similar to the freshwater tilapia (see Figure 3.2). Control incubations with normal rabbit serum substituted for A E l t antisera result in weak labeling of erythocytes (Figure 4.6G). NaX-ATPase and NKCC In seawater adapted coho and marine turbot, N a + , K + - A T P a s e is used as a marker for M R cells (Figure 4.1, 4.2B, 4.6E). Immunoreactive cells display an intense signal throughout the cell body, indicative of labeling associated with the invasive tubular system. These ovoid cells tend to be quite large and the basally located nucleus can be visualized as a negative image against the cell body fluorescence. A characteristic population of these immunoreactive cells is found in the filament epithelium and is concentrated towards the efferent side of the filament. A deep apical crypt is commonly associated with these cells. Strong N K C C immunoreactivity is observed using the T4 monoclonal antibody in an identical pattern as the N a + , K + - A T P a s e (Figure 4.3B). Control incubations with normal mouse serum substituted for oc5 antiserum result in weak labeling of erythrocytes (Figure 4.6H). 89 CFTR anion channel C F T R immunoreactivity could not be demonstrated in either o f these species. However, in mudskippers, the CFTR- l ike protein is immunolocalized to the apical crypt o f branchial M R cells. Mudskipper M R cells also have high N a + , K + - A T P a s e and N K C C immunoreactivity associated with their tubular system (basolateral membrane). The C F T R antibody was not immunoreactive with paraffin embedded tissues indicating that the epitope may be heat denatured. Fixed-frozen tissue was used to obtain immunohistochemical labeling. Seawater acclimation During seawater acclimation, plasma N a + ions show a marked temporal elevation and a recovery by day 8 (Figure 4.4). Plasma C l " levels show a less substantial increase. After some initial variability g i l l N a + , K + - A T P a s e activity shows a significant increase while vH + -ATPase activity shows a decreasing trend albeit statistically insignificant (Figure 4.5). vH + -ATPase immunoreactivity of branchial epithelial cells is greater in lamellae of fresh water adapted fishes than in fish transferred to seawater (Figure 4.7). Immunoreactivity of mucocytes is observed in both fresh water and seawater adapted animals although the intensity of labeling is higher in seawater fishes. 90 F I G U R E 4.1 Double labeled sections on the turbot g i l l epithelium using a rabbit polyclonal antibody (597) against the N H E - 2 fusion protein (FITC, green) and mouse monoclonal antibody against the a subunit of N a + , K + - A T P a s e (Texas Red). The N a + , K + - A T P a s e identifies chloride-type M R cells while the N H E - 2 is localized to smaller closely juxtaposed cells. The nuclei of some of these cells can be made out as negative image against the cytoplasmic fluorescence. Scale Bar= 50 um 91 \ F I G U R E 4.2 Double labeled section of the seawater adapted coho gi l l epithelium using a rabbit polyclonal antibody (597) against the N H E - 2 fusion protein (A) and mouse monoclonal antibody against the a subunit of N a + , K + - A T P a s e (B). The corresponding phase contrast image is shown in (C). The labeling pattern is similar to the turbot with alternating cells in the interlamellar filament immunoreacting with each antibody (Figure 4.1). Nuclei of immunoreactive cells can clearly be made out in negative image. The N H E - 2 immunoreactive cells in coho tend be more elongate extending from the surface to the basement membrane and are indicated by arrows. Scale Bar= 50 urn 93 A F I G U R E 4.3 A n SDS treated section of the turbot gi l l epithelium duel labeled for A E l t and N K C C using a rabbit polyclonal antibody generated against trout erythrocyte band 3 ( A E l t ) (A) and a mouse monoclonal antibody (T4) specific for N K C C (B), respectively. The corresponding phase contrast image is shown in the bottom panel (C). Note the high background staining with the A E l t antibody. The arrows indicate strong labeling associated with the apical crypt of N K C C immunoreactive cells. Scale bar = 50 pm 95 I F I G U R E 4.4 Mean changes in sodium and chloride ion concentrations in the plasma of coho salmon in freshwater (time 0) and after a gradual increase in external salinity for 8 days. Changes in salinity of the holding water are plotted on the lower graph. The askerisks indicate mean values significantly different from those measured in freshwater fishes at time zero. n= 5-9 97 Q . Q . E "co CO 30 20 10 0 0 i 8 Time (days) 98 F I G U R E 4.5 Changes in the vH + -ATPase and N a + , K + - A T P a s e activities in the gil l homogenates of coho salmon during sea water acclimation as determined by N E M and ouabain sensitive activity. Fish were sampled at 0, 12h, 24h, 2, 4, and 8 days. The asterisks indicate a significant difference from the zero hour freshwater control, n = 5 99 7 -, 0 1 / 2 1 2 4 Time (days) 100 F I G U R E 4.6 Immunohistochemistry of g i l l from coho salmon (O. kitsutch) adapted to either freshwater (upper and lower panels A B C , G H I ) or seawater (middle panel D E F ) showing the distributions o f the band 3-like anion exchanger ( A E l t ; A , D ) and N a + , K + - A T P a s e (B,E).The corresponding phase contrast images are shown (C and F , respectively). The lower panel shows control labeling using normal rabbit serum (G) or normal mouse serum (H) of a section from freshwater coho g i l l . There is some non-specific labeling of erythrocytes. Scale bar = 50pm. 101 101 FIGURE 4.7 Paired fluorescent (A,C) and phase contrast (B,D) micrographs of fixed-frozen sections of gills from O. kisutch held in freshwater (A,B) and after 2 days (C,D) exposure to a progressive increase in external salinity. The sections were immunolabeled for vH + -ATPase using a peptide antibody derived from the A-subunit of bovine brain vH + -ATPase . In the freshwater fish g i l l (A,B)> asterisks indicate weakly labeled mucocyte and the arrowheads noticeably immunoreactive cells for \¥T-ATPase. In the seawater fish g i l l (C,D), the arrowheads indicate strongly immunoreactive mucocytes. Scale bar = 50 urn. 103 4.4 D I S C U S S I O N In the marine species of teleost fishes investigated, N a + , K + - A T P a s e and N K C C immunoreactivity was greatest in chloride cells (CC) . The levels of immunoreactivity in other cell types were not significant. Unfortunately, C F T R immunoreactivity was only observed in the mudskipper. This can be explained by the fact that fish C F T R generally has low amino acid identity compared to mammalian C F T R homologues (59% killifish: human; Singer et al. 1998). In the mudskipper I could only find immunoreactivity associated with the apical crypt of M R cells and not lamellar P V C . So in these data I do not find any evidence to support the proposed role of the branchial P V C in the active C l " efflux (Avel la et al. 1997). It should be noted that this absence of immunoreactivity is not proof of the absence of these ion transport proteins in the branchial P V C s . It only indicates that levels in the M R cell are higher than in P V C s and that levels in P V C s may be just below the level of detection of the technique. It is also possible that different isoforms of these proteins are involved in transepithelial ion movements in P V C s . Interestingly, apical band 3-like A E immunoreactivity is associated with the turbot M R cell apical crypt as is also seen in the freshwater tilapia and coho salmon. Since apical labeling was not seen in all N a + , K + - A T P a s e labeled M R cells and the C F T R antibody was not crossreactive in this species, it is not possible to say whether these two anion transport proteins would be co-localized to the same M R cells. The finding o f Degnan et al. (1980) of an absence of C17C1" self-exchange in opercular membrane preparations would argue against the presence of both apical transporters. However, it has been shown in the frog skin that both homologous C F T R anion channel and CI7HCO3" exchanger are situated apically in M R cells (Larsen et al. 1999). The finding of Claiborne and co-workers (1997) of apical DIDS sensitivity and low [Cl"] effect on acid excretion indicates presence of a CF-/HCO3" exchanger in the sculpin and it may 105 follow that perhaps branchial M R cells differ from the opercular type observed by Degan et al. (1980). However, in support of Degan et al. (1980) is the finding that the apical A E is absent from the seawater adapted coho M R cell. It appears that the apical exchanger is present only in the freshwater type M R cell o f this species. The comparison of the freshwater and seawater adapted coho, nicely illustrates how the mechanisms for N a + and C l " uptake can be lost upon change in the salinity regime. The v H + -ATPase activity and epithelial distributions both decrease in seawater and the freshwater-type chloride cell is lost in favour of the seawater-type chloride cell (Cl" uptake versus C l " elimination, respectively). The down regulation of the vH + -ATPase in seawater is similar to that reported by L i n and Randall (1993; L i n et al. 1994) in rainbow trout although not as dramatic. The immunoreactive material they observed appears to be mucocyte mucin granules. The morphology and location of the N H E immunoreactive cells suggests that they are the accessory-type M R cell ( A C ; Hootman and Phillipot 1980; Laurent 1984; Pisam and Rambourg 1991). Both the N H E immunoreactive cells and A C have an intimate relationship with larger M R chloride cells (shared apical crypt, and peripheral, slightly superficial juxtaposition), a similar shape (semi-lunar or elongate) and low levels of N a + , K + - A T P a s e activity (Hootman and Phillpot 1980). The finding of similar cells in the eurhyaline freshwater tilapia (chapter 3) and coho salmon is also consistent with the idea of A C immunoreactivity because Pisam and co-workers have been able to identify A C in euryhaline freshwater fish (Salmo salar smolts: Pisam et al. 1988; juvenile O. mykiss: Pisam et al. 1989). The accessory cell has basically been a cell without a function. It is generally inferior to the chloride cell having a less developed tubular system, fewer mitochondria and low levels o f N a + , K + - A T P a s e . The contribution of the A C to N a C l efflux is a leaky tight junction shared with 106 the C C , which facilitates the paracellular N a + movement (Sardet et al. 1979). It is possible that this A C - N H E 2 is involved in the arniloride-sensitive acid excretion reported by Claiborne et al. (1997). However, what is uncertain is why the N H E should have a cytoplasmic distribution. In mammalian renal epithelia the N H E - 3 isoform has been detected in the endomembrane compartment where it is thought to have a role in maintenance o f endosomal p H (Biemesderfer et al. 1995; D 'Souza et al. 1998). It may also be non-specific crossreactivity, but the finding of an apparently specific marker for A C s may have an application of its own even i f we can not explain its function. Generally, electron microscopy is required to identify this cell type but with this antibody light level microscopic identification is simple. Summary Figure 4.8 summarizes the immunolocalization data collected on seawater fishes. The immunodetection results generally corroborate the classical model of active N a C l elimination (Silva et al. 1977). Basolateral N a , K -ATPase and N K C C are associated with the M R chloride cell and a C F T R anion channel with the apical crypt (see Marshall and Bryson 1998; Singer et al. 1998). The presence of N H E 2-like protein within the accessory cell and A E - l i k e protein within the apical crypt complicate the model. However, the apical A E - l i k e protein is not found in the seawater-type chloride cell o f the coho salmon even though it is found in the freshwater counterpart. Also the N H E immunoreactivity may not be specific although it may make a useful marker for the accessory cell type. In the coho salmon it is possible to demonstrate the loss of the freshwater ion uptake mechanisms (vH + -ATPase and A E ) in sea water adapted fishes. 107 F I G U R E 4.8 Illustration of cell types of the branchial epithelium of a seawater fish summarizing the immunolocalization data. The seawater M R chloride cell has a basolateral N a + , K + - A T P a s e and N K C C and an apical C F T R anion channel. The band 3-like C I T H C C V exchanger is found in the stenohaline turbot but not in the euryhaline seawater adapted coho salmon or mudskipper. N H E - l i k e immunoreactivity is found in M R accessory cells of both seawater and freshwater fishes. 108 Seawater fish water blood NKA AE CFTR NHE NKCC ID i Chapter 5. THE MUDSKIPPER GILL 5 S U M M A R Y The mudskipper, Periophthalmodon schlosseri, is able to tolerate extremely high environmental ammonia concentrations that would necessitate the active elimination of NFL; + ions to explain the absence o f internal elevations of ammonia. In this chapter, the contribution of various ion transport proteins to net ammonia excretion were determined pharmacologically and their presence in the g i l l M R cells determined using immunolocalization techniques. A component of the ammonia eliminated by P. schlosseri involves an arniloride-sensitive N a + / H + exchanger (NHE) and carbonic anhydrase (CA) and is not dependent on boundary layer p H effects. A n apical C F T R - l i k e (Cystic fibrosis transmembrane regulator) anion channel may be serving as a HCO3" channel accounting for the acid-base neutral effects observed with N H / inhibition. N a + , K + - A T P a s e plays a role in ammonia elimination only against a gradient. Although a vH + -ATPase is present its role in ammonia elimination is not certain. These animals are able to eliminate significant quantities of ammonia when out of water. The findings that branchial epithelium of P. schlosseri has an unusual high density of mitochondria-rich cells that express all o f the above proteins link the animal's ability to actively eliminate ammonia to the gi l l . It also is clear that gas exchange across the gi l l lamellae would be severely handicapped by the long diffusion distances present. The observation of intra-epithelial capillaries in the inner operculum and to a lesser extent in the gil l filament would present more practical sites for gas exchange. 110 5.1 I N T R O D U C T I O N In 1 9 2 9 Homer Smith originally demonstrated that ammonia was the dominant nitrogenous waste product eliminated by fishes and that the major site of elimination was the gills (rather than the kidneys). Studies since then have further corroborated this view but the mechanism(s) of transbranchial ammonia elimination is still debated (reviews by Evans and Cameron 1 9 8 6 ; Walsh 1 9 9 8 ) . Theoretically, ammonia can move across the epithelium by diffusion (NH3 and N H 4 + ) either transcellularly or paracellularly with ion transport proteins facilitating the transcellular movement of NH4 + , as l ipid solubility is low. The difficulty is in determining the importance of the various pathways when ammonia exists in both gaseous (NH3) and ionized ( N H 4 + ) forms in an equilibrium dependent on a number of factors (pH, temperature, and ionic strength) and when N H 4 + versus N H 3 + H + movement cannot be experimentally distinguished. Cameron and Heisler ( 1 9 8 3 ) were able to demonstrate that P N H 3 gradients across the branchial epithelium could account for the majority of total ammonia elimination under normal conditions of p H and low ambient ammonia in the freshwater rainbow trout. However, when fish are subjected to inward N H 3 and N H 4 + gradients they are still able to maintain plasma levels below that of the ambient water. Under these conditions the active exchange of N H 4 + for either external H + or N a + has been suggested (Cameron and Heisler 1 9 8 3 ; Cameron 1 9 8 6 ; Claiborne and Evans, 1 9 8 8 ; Wilson and Taylor 1992). However, it is possible that the contribution of gi l l boundary layer acidification by CO2 and H + transport is an important factor in maintaining outward PNH3 gradients even under these conditions (in vitro Wright et al. 1 9 8 9 ; in vivo Wilson et al. 1 9 9 4 ) . In these studies, the addition of buffers to the water has been used to eliminate the contribution of boundary layer acidification. Boundary layer acidification, facilitates NH3 i l l diffusion by trapping N H 3 by protonation maintaining the P N H 3 gradient ( N H 3 + H ->• N H 4 ). The low ionic conductance of the freshwater fish gi l l indicates that paracellular N H / diffusion is likely a minor component of the total ammonia efflux. The ionic permeability of the gil l of seawater fishes is orders of magnitude larger than freshwater fishes (Evans 1979) as a result of leaky paracellular junctions found between chloride and accessory cells (see Chapter 4 Introduction; Sardet et al. 1979). Thus in seawater fishes paracellular N H / diffusion is an important component to J A M M - The large inward N a + gradient is also thought to contribute to J A M M by driving N a + / N H 4 + exchange. NH .4 + has also been shown to substitute for H + on the N a + / H + exchanger (NHE) protein (Kinsella and Aronson 1981). In addition, Evans et al. (1989) were able to find that 22% of total ammonia elimination in a perfused head preparation of toadfish was sensitive to N a + , K + - A T P a s e inhibition by ouabain in the perfusate. K + and N H 4 + share a similar hydration radius and in in vitro studies NH4+ has been shown to compete with K + for binding sites on N a + , K + - A T P a s e (Mallery 1983; Kurtz and Balaban 1986; Garvin et al. 1985; Towle and H0lleland 1987; W a l l and Koger 1994) and also the N a + : K + : 2 C f cotransporter (Kinne et al. 1986). In order to examine the mechanism of active NH4+ elimination and the role of ion transport proteins I focused my attention on a marine fish that is able to tolerate high environmental ammonia levels (Peng et al. 1998) and that also happens to be an amphibious obligate air-breather (Ishimatsu et al. 1999). This fish is the mudskipper Periophthamodon schlosseri. The tolerance of the mudskipper to environmental ammonia is much higher than that of other fishes. The 96h LC50 for P. schlosseri is 514 p M N H 3 (Peng et al. 1998) compared to literature values ranging from 8.2 to 247 u M for other fishes (Thurston et al. 1983). P. schlosseri has an ammonia tolerance similar to the Lake Magadi tilapia {Oreochromis alcalicus grahami; 1 1 2 Randall et al. 1989) and the air breathing catfish Heteropneustes fossilis (Saha and Ratha 1990). However, these species make use of the ornithine-urea cycle to produce urea as a means of detoxifying ammonia while the mudskipper does not (Peng et al. 1998). Mudskippers make use of free amino acids and have a powerful glutamate dehydrogenase-glutamine synthetase system for ammonia detoxification in the brain (Peng et al. 1998). Interestingly, the mudskipper P. schlosseri shows no response to 3 6 u M NH3 in its bath water. It is able to maintain plasma and muscle ammonia, urea, and T F A A constant in spite of steep inward ammonia gradients of 8 m M NH4 + (36uM NH3). Although ammonia fluxes were not measured during these experiments it seems likely that active elimination of ammonia was occurring since endogenous production was continuing and internal levels were not changing. These animals also have a high terrestrial affinity. A significant problem encountered during terrestrial excursions is the lost capacity for the elimination of ammonia (into water) (Mor i i et al. 1978; Iwata et al. 1981). During terrestrial exposure P.schlosseri remain ammonotelic and increase ammonia, urea and free amino acid stores (Ip et al. 1993), but not enough to account for the >90% inhibition of the excretory pathway for ammonia reported in other species of mudskippers (Mor i i et al. 1978; Iwata et al. 1981). P. schlosseri is more active out of water (Kok et al. 1998) and must, therefore, be producing significant amounts of ammonia. Interestingly, some ammonotelic terrestrial isopods and crustaceans overcome this problem by alkalinizing their gills and eliminating ammonia by volatilization (eg Wieser 1972; Greenaway andNakamura 1991; Speeg and Campbell 1968). Studies by Schottle (1931), L o w and co-workers (1988, 1990), and Yadav and co-workers (1990) all recognize the uniqueness of the gills of P. schlosseri. The gi l l filaments are short and branched and lamellae are thickened and have interlamellar fusions. The total gi l l 113 surface area is less than other mudskippers (Schottle 1931; L o w et al. 1988,1990). These characteristics have been interpreted as adaptations for air exposure to prevent collapse of the lamellae. A mucus coat and microridges also reduce desiccation (Low et al. 1990; Yadav et al. 1990). None of these studies, however, describe the fine structure of the branchial epithelium that would help in understanding gi l l function in this amphibious fish. In this chapter, the mechanism of active ammonia elimination w i l l be investigated by first determining which ion transport proteins are involved in ammonia excretion using a pharmacological approach, and then determining a morphological or cellular basis for ammonia excretion by examining the fine structure of the branchial epithelium. Finally, a pathway for NH4+ elimination w i l l be constructed by localization of ion transport proteins using immunological techniques. In addition, terrestrial ammonia excretion rates as well as body and gi l l surface p H are presented to address the contribution of ammonia volatilization to terrestrial nitrogen elimination. 5.2 MATERIALS AND METHODS 5.2.1 Animals Adult Periophthalmodon schlosseri o f both sexes were collected from the Pasir Ris, a small estuary on the east coast of Singapore and transported to City University of Hong Kong. In the laboratory, fish in individual plastic aquaria were kept partially submerged in 50% seawater (15 %o). Every other day the water was changed and the animals were fed goldfish ad libitum. The fish mass ranged from 55 to 120 g. The room air and water temperatures remained constant around 26°C. N o attempts were made to control lighting conditions and a natural photoperiod was followed. 114 5.2.2 Measurements of ammonia fluxes in water and air Animals were fasted 3 days prior to the start of experiments. A pair of flux measurements was performed on groups of fish kept individually in horizontal 1000 ml polypropylene bottles while half full o f 50% S W and then with no S W . The containers were capped and provided with humidified air and the exiting air was passed through a series of acid traps (0.1N HC1). The animals were subjected to a three-hour partially submerged flux period followed by a twenty-four hour air exposed period. A t the end of the 3h flux, the water was removed and a sample was acidified and stored at 4°C. A t the end of the 24h air exposure, the containers with animals still inside were rinsed with 50%SW. A sample of this wash water was saved and stored as above. Blank runs were also made to account for naturally occurring ammonia in the system. Water and acid-trap samples were assayed for total ammonia (Verdouw et al. 1978) from which ammonia flux rates (JAMM ) were calculated (umol ammonia • kg" 1 fish • h"1). Appropriate standards were prepared with either 50% S W or 0.1 N HC1. A total of 8 animals had paired flux measurements made. 5.2.3 Inhibitor Studies Animals were fasted 3 days prior to the start of experimentation. A series of three flux periods was performed on fishes kept in 500 or 1000 m l of 50% S W . A n initial three hour control flux period was followed by a three hour experimental flux period and finally a three hour recovery flux period. The experimental exposures consisted o f exposing fish to either inhibit carbonic anhydrase with 0.1 m M acetazolamide, N a + / H + exchange with 0.1 m M amiloride-HCl, boundary layer acidification with 5.0 m M H E P E S at either p H 7 or p H 8, or v H + -ATPase with lOOmM K N 0 3 (and lOOmM KC1 as a K + control) in 50%SW. Control and recovery 115 fluxes were conducted in 50% SW. For the first two flux periods water samples were taken at 0, 1, 2,and 3 hours but only at 0 and 3 hours for the recovery period. In initial experiments the N a + , K + - A T P a s e inhibitor ouabain was shown to have no effect on J A M M in 50%SW (T. K o k and K - Y Ip, personal communication). Therefore, an additional series of ammonia flux measurements were made on fish in 2 m M NH4CI in 50% S W followed by 2 m M NH4CI in 50% S W with either 0.1 or 0.01 m M ouabain. Water samples were taken at 0, 1, 2, and 3h during each exposure period. Water samples were assayed for total ammonia concentration (Verdouw et al. 1978) from which ammonia flux rates (JAMM ) were calculated (umol ammonia • kg" 1 fish • h"1). Appropriate standards were prepared with the inhibitors in 50% S W . Titratable acidity (TA) flux was measured according to the method of M c Dona ld and Wood (1981) from the difference between 0 and 3h water samples from control and experimental periods. In short, 100 ml water samples were aerated overnight to remove respiratory CO2 and a 25 ml sub-sample (weighed to 0.0lg) was titrated to p H 4.00 with 0.01 N HC1 using a Radiometer autotitrator and Orion Ross type combination electrode. Aeration continued during titration to remove liberated CO2. The net acid flux (JACID; pEq H + • kg" 1 • h"1) was calculated as the sum of the T A and J A M M , signs considered. 5.2.4 Surface pH measurements Measurements o f the surface p H of the skin and gills were made during air exposure using a solid state micro p H electrode (PHR-146; Lazar Research Laboratories, C A ) coupled to a Radiometer P H M 84 meter (Radiometer-Copenhagan, D K ) . The dorsal surface behind the eyes, and below the dorsal fin, laterally between the base of the pectoral fin and the body wall as well as the gills were measured repeatedly and values averaged for 4 animals. 116 5.2.5 Acclimation to different ambient salinities Groups o f 6 animals were acclimated over two weeks to either 5%o SW, 15%o S W or 25%o SW. Every other day the water was changed and the animals were fed goldfish ad libitum. A t the end of the second week, ammonia and urea flux measurements were made over a 3h period (as described above) and the animals were sacrificed by an overdose of MS-222 and cervical transsection. A 1 ml blood sample was taken by caudal puncture using a heparinized syringe. Whole blood p H was measured using an Orion micro-tip Ross type combination electrode coupled to a Radiometer-Copenhagen P H M 84 meter. Blood P02, Pco2, and p H were also measured on a Corning blood-gas analyzer. The remainder of the sample was centrifuged and the plasma collected and frozen for later analysis of ammonia, urea, and Cl" . 5.2.6 Plasma analytical procedures Plasma total ammonia concentrations (Tamm) were determined using a commercial diagnostic kit ( G L D H / N A D H : Sigma, St. Louis, M O ) and the P N H3 and [NH» + ] calculated from the p H and Tamm values using a rearrangement of the Henderson-Hasselbalch equation and p K ' and NH3 solubility values determined by Cameron and Heisler (1983). Plasma chloride concentrations [Cl"] were determined by coulometric titration (Haake Buchler Instruments digital chloridometer) and plasma urea concentrations were measured by the diacetyl monoxime method (Sigma). 5.2.7 A TPase Activity N a + , K + - A T P a s e and vH + -ATPase activities (umol P i • mg"1 protein • h"1) in crude gi l l homogenates were determined at 22.5°C in a plate reader (Thermomax, Molecular Devices Corp., C A ) modified from McCormick (1993). N a + - K + ATPase and vH + -ATPase activities were 117 defined as specific inhibition by 1.0 m M ouabain and lOOmM K N 0 3 , respectively. See General materials and methods for additional details of the assay. 5.2.8 Tissue fixation for Standard TEM (Transmission electron microscopy) Animals were anaesthetized with MS-222 (3-aminobenzoic acid ethylester; 1:5000) and killed by cervical transection. The gi l l arches were excised as well as pieces of the inner opercular membrane. Pieces of tissue were immersed in fixative containing 1.5% glutaraldehyde, 1.5% paraformaldehyde, buffered with 0 .1M sodium cacodylate (pH 7.4) over night at 4°C. Following primary fixation the tissue was rinsed with 0.1 M cacodylate buffer and post fixed on ice in 1% osmium tetroxide in 0.1 M sodium cacodylate. The tissue was then rinsed with d t ^ O , and stained en bloc with 1% aqueous uranyl acetate. Tissue was embedded i n Epon 812 (PolyScience) following serial dehydration with ethanol followed by propylene oxide. Thin sections (70-90 nm) were made on an Ultracut microtome and mounted on copper grids and counter stained with lead citrate and uranyl acetate. Sections were viewed on a Phillips 300 T E M at 60kV. Semithin sections (1pm) were also made and stained with toludine blue. 5.2.9 Tissue fixation for immunocytochemistry Whole g i l l arches were fixed for immunolocalization studies at both light and em levels using 3 % P F A / P B S and P L P fixatives, respectively, overnight at 4°C. Tissue for immunofluorscence was frozen as well as paraffin embedded. Tissue for immuno-em was embedded in Unicryl . See General Materials and Methods for additional information. 5.2.10 Immunocytochemistry and Antibodies Employed The distribution of N a + , K + - A T P a s e , Na + :K + : 2C1" cotransporter ( N K C C ) , N a + / H + exchanger (NHE2,3), C F T R , carbonic anhydrase ( C A ) and vH + -ATPase was determined by indirect immunofluorescence and immunoperoxidase methods in paraffin and fixed frozen gil l 118 tissue. Antibodies against A E were not crossreactive with epithelial. See General Materials and Methods for additional information. 5.2.11 Statistical Analysis Data are presented as mean ± S E M (n). Ammonia excretion rates from the different treatments were compared using a one way repeated measures A N O V A and post hoc S N K . Net ammonia, titratable acidity and acid fluxes under control and experimental conditions were compared using paired t-tests. The fudicial limit was set at 0.05. 5.3 R E S U L T S 5.3.1 Ammonia fluxes and effects of inhibitors Flux rates for ammonia o f fish kept partially submerged are 486.2 ± 35.0 umol ammonia • kg" 1 fish • h" 1. During emergence the total ammonia excretion rate is significantly reduced to 288.74 ± 29.57. The volatilized component is minor, usually contributing less than 3% to the total (7.75 ± 1.07). Addit ion of the N H E inhibitor (0.1 m M ) amiloride to the water reduces J A M M by 42% but has no effect on JACID which remains around zero (Figure 5.1.1). During the subsequent hours of exposure, J A M M increases slightly and begins to recover after changing the bath to fresh 50% S W without amiloride. The carbonic anhydrase inhibitor (0.1 m M ) acetazolamide reduces J A M M by 48% yet does not have a significant effect on JACID (Figure 5.1.2). J A M M starts to recover during exposure to acetazolamide and after its removal. The addition of 5 m M H E P E S to 50% S W is sufficient to fix the water p H at either 7 or 8 during the 3h experimental period (Figure 5.1.3e). Buffering the water at p H 7.0 does not 119 significantly affect either J A M M or JACID (Figure 5.1.3a,b). However, addition of 5 m M H E P E S p H 8 significantly increases JACID while having no effect on J A M M (Figure 5.1.3c,d). In vivo the vH + -ATPase inhibitor K N O 3 only significantly affects J A M M after the second hour of exposure (Figure 5.1.4a). However, the averaged J A M M value over the 3h period is not significantly different from the initial control (Figure 5.1.4b). Although, JACID is significantly reduced from the initial control, so too is the parallel KC1 control exposure (Figure 5.1.4d). Animals fail to recover J A M M after their experimental treatments. Animals exposed to 2 m M N H 4 C I in 50% S W at p H 7.8 are still able to eliminate ammonia (Figure 5.1.5) at rates comparable to control animals from other experiments (Figure 5.1.3a,c). Addit ion of 0.1 m M ouabain significantly reduces J A M M while 0.01 m M ouabain is without effect. There is also a significant increase in plasma total ammonia in animals exposed to 0.1 m M ouabain (Table 5.1). Animals acclimated to salinities of 5, 15 and 25 %o S W for two weeks show no differences in net ammonia or urea fluxes (Table 5.2a). However, plasma [ N H 4 + ] and P N H 3 are significantly higher in fish acclimated to higher salinity (25%o) (Table 5.2a). In vitro ATPase activities N a + , K + - A T P a s e and vH + -ATPase enzyme activities measured in crude gil l homogenates as specific inhibition with 1 m M ouabain and 100 m M K N O 3 , respectively, are shown in Table 5.2. Significant levels of activity are present in gi l l homogenates but there are no differences in N a + , K + - A T P a s e or vtT-ATPase activities in groups of fish acclimated to the high and low salinities. 120 Surface pH The measurements o f the skin surface p H in air-exposed animals taken from several sites around the body are consistent within and between animals and not significantly different from seawater (Table 5.3). However, the surface p H of the gills is significantly lower than seawater and the other body surfaces measured. 5.3.2 Ultrastructural level examination of the gill epithelium Basic organization of the gill The gills o f P. schlosseri are composed of four pairs of g i l l arches. The first gi l l arch supports a fold of opercular tissue. In addition, P. schlosseri has a pair of pseudobranches. The gi l l filaments are short and occasionally have complicated branching patterns. They are not organized in a sieve-like arrangement of orderly rows seen in most gills. Instead, the filaments dangle in the opercular cavity, which is more often than not filled with air. Filament Along the length of the filament are parallel rows of semi-circular shaped lamellae on both the top and bottom filament edges. A cross section through the filament reveals the afferent ( A V ) and efferent (EV) vessels seen at opposing ends and embedded in loose connective tissue (Figure 5.2.1a). The central venous sinus, nutritive vessels and nerve bundles are found below the lamellar bases (Figure 5.2.le). Blood spaces are also sometimes seen in the spaces between the filament epithelium and A V and E V (Figure 5.2.1a,b,c). The filament (in cross section) tends to be consistent in appearance along its length except during branching when bundles of skeletal muscle are observed between the A V and trailing edge epithelium (Figure 5.2.1c). A cartilaginous support rod runs closely parallel to the afferent vessel (Figure 5.2.1a). Following branching of a filament, the cartilaginous rod disappears. 121 The stratified filament epithelium along afferent and efferent edges consists mainly of squamous and columnar pavement cells with a few large mucocytes. M R cells are not very common along this part of the epithelium but intraepithelial capillaries are sometimes observed along the efferent edge notably during filament branching (Figure 5.2.1b). Associated with the appearance of these intraepithelial capillaries is a system of larger sub-epithelial vessels that deliver and return blood to the venous circulation. Lamellae The lamellae o f P. schlosseri follow the basic piscine plan of blood spaces defined by modified endothelial cell (pillar cell) supporting a pseudo-stratified epithelium of pavement cells (PVC) and mitochondria-rich ( M R ) cells (Figure 5.2.2a). However, the lamellae of P. schlosseri have some unique features not found in any other type of fish g i l l so far described. The blood to water diffusion distance is large (~ 15 u.m) because the epithelium consists of a thick layer o f mitochondria-rich and filament-rich (FR) cells covered by a layer o f pavement cells. In addition, adjacent lamellae are frequently seen fused together and this restricts the water space to a narrow channel (Figure 5.2.2b). Typically the cross-sectional area of the blood space is greater than that for the external milieu. Within the fused areas it is not possible to distinguish from which lamellae the cells arose. The bases of neighbouring lamellae are often not separated by (interlamellar) filament epithelium. Mitochondria-rich (MR) cells and accessory cells The lamellar epithelium is composed almost entirely of large mitochondria-rich cells (Figure 5.2.2a). These cells sit on a thick basal lamina and are either ovoid or cuboidal. The cells are characterized by their abundant mitochondria, by an extensive system of regularly branching and anastomosing tubules that is continuous with the basolateral plasmalemma (tubular system), 122 and by a subapical collection of irregular tubules and vesicles (vesiculotubular system). The subapical area is rich in microfilaments. The apical plasmalemma forms a deep crypt with a variable number of microvil l i . The M R cell sends cytoplasmic projections into the P V C s lining the rim of its apical crypt. Apical crypts appear to be formed as invaginations in the apical surface of single M R cells; that is, apical crypts seldom have more than one M R cell forming their boundaries. In cases where crypts are bordered by more than one cell, thin or shallow tight junctions are observed between the neighbouring M R cells (Figure 5.2.3a,b). Moreover, one of the M R cells is smaller in size, has a more granular cytoplasm and is in a peripheral position relative to the other (Figure 5.2 3a). Multiple cytoplasmic processes from the smaller M R cell make their way through the apical cytoplasmic region of the larger M R cell to the apical membrane to increase the total shallow junction length, in sections, this region appears as a 'mosaic'. This smaller M R cell has been termed the 'accessory ce l l ' and the intimate relationship seen with the larger M R (chloride) cells is typical of marine teleost gills (see review by Laurent 1984). Generally, direct contact between neighbouring M R cells was rarely observed because a special type of cell, the filament rich cell, physically separates neighbouring M R cells. Filament Rich (FR) Cells These cells, that are flattened in appearance and are found between neighbouring M R cells, attach to the basal lamina and have cytoplasmic processes extending between, and sometimes into, neighbouring M R cells (Figure 5.2.4a-d). They extend up to the mucosal pavement cells and have not been observed to contact the external medium. These cells have a denser cytoplasm than M R cells, and have thick bundles of intermediate filaments associated with desmosomes (Figure 5.2.4d), in addition to some vesicles and a few mitochondria. Within 123 the narrow space between M R cells, these filament rich (FR) cells often have an interstitial space created from an invagination of the plasmalemma (Figure 5.2.4a-c). The space is quite irregular with branching and anastomosing cytoplasmic processes within. This extracellular space may form isolated pockets or possibly channels (canaliculi). Pavement cells (PVC) The pavement or squamous epithelial cells cover the majority of the lamellar surface with M R cell apical areas restricted to apical crypts. Thick tight junctions jo in neighbouring P V C s . The surface o f these cells displays a system of microridges. These flattened cells have a dense, granular cytoplasm, Golgi system, numerous mitochondria with sparse, loosely arranged crista, and vesicles of various sizes and densities (Figure 5.2.2b). P V C s and F R cells together encapsulate M R cells, isolating them from one another. Pillar cells The lamellar blood spaces are defined by pillar type cells characteristic of fish lamellar endothelium (Figure 5.2.2b, 3a; Newstead 1967; Hughes and Weibel 1972). Generally, the lamellae have 4-6 roughly parallel blood spaces including the marginal and basal channels. The irregular distribution of pillar cells accounts for the changing position of the channels. Weibel-Palade bodies (osmiophillic) are seen within the cytoplasm of flattened endothelial cells of the basal and marginal channels. The marginal channel at the top of the lamellae is larger than other blood spaces and the basal channel is not deeply embedded within the filament. Opercular membrane Schottle (1931) has identified the opercular membrane as an accessory respiratory organ (ARO) in P. schlosseri and a number of other species of mudskipper. The inner operculum is well vascularized, and numerous papillae and furrows increase the surface area. Lining these 124 papillae are capillaries found within the stratified epithelium (Figure 5.2.5). The capillaries pass through the epithelium basal lamina and retain their own sheath of basal lamina separating them from the true epithelial cells. The blood spaces are separated from the opercular cavity by an endothelial cell, basal lamina, a capillary associated epithelial cell and pavement cell layer (Figure 5.2. 5c). The surface of the pavement cell is covered by pronounced microridges. The pavement cells of this epithelium are very similar to those found in the gi l l . These P V C s are electron-dense, having mitochondria with sparse crista, and have a range of vesicles. Mucocytes are quite common in this epithelium and can often be found along side a capillary. M R cells are generally rare. This epithelium is very similar to that found along some sections of the filament leading edge (see Figure 5.2.1b). Nerve bundles often are seen in the papillae in association with blood vessels. The blood vessels appear to contain a system o f valves for controlling blood flow. 5.3.3 Immunohistochemistry NaX-ATPase Figure 5.3.1 shows the abundance of N a + , K + - A T P a s e immunoreactivity associated with the lamellar epithelium. Immunopositive cells are also found in the epithelium of the gi l l arch. Immunoreactivity for N a + , K + - A T P a s e is restricted to the large epithelial M R cells (Figures 5.3.2a). The labeling is throughout the cell body with the exception of the nucleus and apical crypt area. Immunogold labeling for N a + , K + - A T P a s e reveals that the intracellular staining observed by immunofluorescence is restricted to elements of the tubular system and there is no labeling associated with the apical crypt (Figure 5.3.3a,b). In Western blots, the cc5 antibody cross reacts strongly with a band with an apparent M W of 116kDa (Figure 5.3.7a). 125 NKCC The N K C C essentially has an identical distribution as N a + , K + - A T P a s e (Figure 5.3.3e,f). N K C C immunofluorescence labeling shows a diffuse cytoplasmic distribution while the higher resolution immunogold technique indicates that labeling is associated with the M R cell tubular system. N o labeling is associated with the apical crypt. However, I was unable to demonstrate immunoreactivity with Western blots using the J3 and T4 N K C C antibodies. CFTR The mouse monoclonal antibody against the human cystic fibrosis transmembrane regulator ( C F T R ) crossreacts specifically with the M R cell apical crypt (Figure 5.3.3c,d). In sections, labeled crypts appear as either a ring or " U " shape indicative of cross or longitudinal sections through the crypt, respectively. This antibody was only useful with fixed-frozen tissue and not fixed-paraffin embedded tissue. N o immunolabeling is associated with the basal portion of the M R cell , pavement cells, pillar cells, mucocytes or erythrocytes. Western blots indicate an immunoreactive band with an apparent M W of 150 k D a (Figure 5.3.7b). Carbonic anhydrase Immunoperoxidase staining for carbonic anhydrase in fixed-frozen sections is restricted to the apical crypt region of the M R cells (Figure 5.3.4a,b). Staining is not associated with the rest of the cell . Erythrocytes also show immunoreactivity while no immunoreactivity is associated with pavement cells, or pillar cells. Incubation of control sections with normal rabbit serum at an equivalent dilution results in insignificant levels of labeling. vH*-ATPase The vH + -ATPase was immunolocalized using a polyclonal antibody directed against the A-subunit of the vacuolar proton pump complex. Immunoperoxidase staining shows that the 126 vH + -ATPase is restricted to the apical crypt region o f M R cells (Figure 5.3.4c). It should be noted that some difficulty was encountered in consistently reproducing these results. Western blots show an immunoreactive band with an apparent M W of 78 k D a (Figure 5.3.7c). Na+/H* Exchanger The distribution of apical isoforms of the N a + / H + exchanger (NHE-2 , and 3) were determined in the mudskipper g i l l using antibodies Ab597 and A b l 3 8 0 specific for N H E - 2 and N H E - 3 , respectively. Both N H E isoforms are immunolocalized to the apical crypts of M R cells (Figure5.3.5 and 5.3.6). From low power photomicrographs, most crypts can be seen to be labeled (Figure 5.3.5a). The N H E - 3 antibody Abl381 was not useful for tissue localization. In Western blots of crude membrane preparations, immunoreactivity is seen with bands in the predicted M W ranges (~85 kDa), however, additional major bands are also recognized at lower M W s (Figure 5.3.7 d,e). 127 Table 5.1 Plasma total ammonia concentrations ( m M + S E M ) in P. schlosseri exposed to 50%SW (n=6) or 2 m M NH4CI in 50% seawater in the presence of either 0.1 of 0.01 m M ouabain (n=5 and 6, respectively). The asterisk indicates a significant difference from the control value (P<0.05). 0 m M NH4CI in 50% S W 2 m M NH4CI in 50% S W NO Ouabain 0.1 m M Ouabain 0.01 m M Ouabain 114.6 ± 9 . 6 245.3 ± 52.9* 97.1 ± 12.6 128 -4-» O O w ST T 3 • s .2 O 4-» £ cu oo o o u '— + CD to . rt fc! -o 2 T > % ^ rt -*-» •a i s rt U U S > OT • S3 ffi PH "t" . co . *" PM — CO rt « I! 5 ^ S i 3 1-5 C/3 OO "1 o 1^- +1 IT! o o o i n 2 +1 CN ^ 2 +1 CO C N °°. © CN + | O N o o o © +1 O N £ J 2 + 1 CO 00 O N o +1 IT) ^ +1 ,X CN £ +1 O r - 1 OO o o o -CN C N +1 CO + | _<- c^ ^ +1 CO C N +1 i n o " o i n o —! O +1 o CN O © +1 o o vb O N 2 + L O N + | c3 _ , , » r t _ ^ >^ <^  co ^ ^ ° i O N O N ^ o o _ .• c--3S in o S ° - i i i 3 fc «<2 O N O N o o o +1 +1 +1 o O N o o O o o +1 +1 +1 r - -o O N o CN o o o i—i +1 +1 +1 o CN CO •<* CN o C N +1 + 1 +1 CO C N r -o c O C N • v o CN o CO CN O m C N O o o o +1 +1 +1 m CO i n CO i n O N O i n r-^ o IT) e e m C N 129 Table 5.3 Surface p H measurements of the skin and gills of Periophthalmodon schlosseri during emersion. Asterisk (*) indicates significant difference from all other groups, n = 4, P<0.05. Locat ion Surface p H Lateral (base of pectoral fin) 7.73 ± 0.19 Dorsal body 7.78 ± 0 . 1 8 Dorsal head region 7.98 ± 0 . 1 5 G i l l 7.14 ± 0 . 0 3 * 50% S W 7.78 ± 0.08 130 F I G U R E 5.1.1 The effects of the N a V H + exchange inhibitor amiloride (0.1 m M ) on (a) net ammonia ( J A M M ; umol • kg" 1 • h"1) and (b) net acid fluxes (JACID = JTA + J A M M ; pEq • kg" 1 • h"1) in mudskippers in 50% SW. The flux series (a) consists of an initial 3h-control flux (Cl -3) followed by 3h exposure (E l -3 ) and recovery flux (R3) periods. The J A M M is calculated for each hour during the control and exposure periods and only over the 3h period during the recovery. In (b) JACID is calculated from the 3h T A and J A M M values. Values marked with like characters in (a) are not significantly different and the asterisks in (b) indicate a significant difference from the corresponding initial 3h-control value. Note the significant reduction in J A M M following addition of amiloride and the recovery of J A M M with its removal and the absence of changes in JACID- n = 6. P< 0.05 131 X J3 L_ CO "c o E E < a 800 - j 700 -600 -CD 500 -co 400 -'c o 300 -E E 200 -co 100 -o E —* 0 a X a i i r C1 C2 C3 b b X i l l E1 E2 E3 ab R3 Control 0.1 mM Amiloride X •1000 Titratable acidity (TA) Net Ammonia Flux (J A m m ) Net Acid Flux (=TA + J™ J 132 F I G U R E 5.1.2 The effects of the carbonic anhydase inhibitor acetazolamide (0.1 m M ) on (a) net ammonia ( J A M M ; umol • kg"' • h"1) and (b) net acid fluxes (JACID = JTA + J A M M ; P-Eq • kg" 1 • h"1) in mudskippers in 50% SW. The flux series (a) consists of an initial 3h-control flux (Cl -3) followed by 3h exposure (El -3) and recovery flux (R3) periods. The J A M M is calculated for each hour during the control and exposure periods and only over the 3h period during the recovery. In (b) JACID is calculated from the 3h T A and J A M M values. Values marked with like characters in (a) are not significantly different and the asterisks in (b) indicate a significant difference from the corresponding initial 3 h control value. Note the significant reduction in J A M M following addition of acetazolamide and the recovery of J A M M with time and the absence of changes in J A C I D - n = 6. P < 0.05 133 w 800 — 700 ^ 600 * 500 co .5 if o E 400 300 200 100 0 J ab d d c ^ r T ab C1 C 2 C 3 E1 E2 E3 R3 -1000 Control 0.1 mM Acetazolamide Titratable acidity (TA) Net Ammonia Flux (J A m m ) Net Acid Flux (=TA + J A m m ) 134 F I G U R E 5.1.3 The effects of changes in boundary layer p H by 5 m M H E P E S p H 7.0 and p H 8.0 on net ammonia (JAMM; umol • kg" 1 • h"1) (a, c respectively) and net acid fluxes (JACID = JTA + J A M M ; uEq • kg" 1 • h"1) (b, d respectively) in mudskippers in 50% S W . The flux series (a, c) consists of an initial 3 h control flux (Cl -3 ) followed by 3h exposure (El -3) and recovery flux (R3) periods. The J A M M is calculated for each hour during the control and exposure periods and only over the 3h period during the recovery. In (c, d) JACID is calculated from the 3h T A and J A M M values. Water p H values during the initial 3h control flux and the following 3h H E P E S flux period at 0 ,1 and 3 h are shown in (e). The asterisks in (d) indicate a significant difference from the corresponding initial 3 h control value, n = 6. P< 0.05 135 CO ' c o £ E CO o E x zz u_ CO ' c o E E < a 600 500 400 300 200 100 0 Ctrl HEPES pH 7 n i X Ctrl 600 -, 500 400 300 -I 200 100 0 X X X X I X HEPES pH8 X X . X C1 2 3 E1 2 3 R3 x C1 2 3 E1 2 3 R3 Control b 2000 1500 1000 500 Ctrl HEPES pH 7 d 2000 1500 ^ 1000 °? 500 cr L U 0 -500 1000 Ctrl HEPES pH8 * T ' A C I D A M M Hepes 7.0 Hepes 8.0 5.0 mM HEPES I Q _ i _ CD +-> CO 136 F I G U R E 5.1.4 The effects of the vH + -ATPase inhibitor 100 m M K N 0 3 and 100 m M KC1 (control) on net ammonia (a,c respectively) ( J A M M ; umol • kg" 1 • h"1) and net acid fluxes (b,d respectively) (JACID = JTA + J A M M ; uEq • kg" 1 • h"1) in mudskippers in 50% S W over a 3 h period. The flux series consists of an initial 3 h control flux (Cl -3) followed by 3h exposure (El -3) and recovery flux (R3) periods. The J A M M is calculated for each hour during the control and exposure periods and only over the 3h period during the recovery. In (b, d), JACID is calculated from the 3h T A and J A M M values. Values marked with like characters are not significantly different in (a,c) and the asterisks indicate a significant difference from the corresponding initial 3 h control value in (b, d). Note that the effect o f 100 m M KNO3 on J A M M is not apparent until the 3 r d h of exposure and no recovery is seen, n = 6. P< 0.05 137 200 -, 000 800 600 400 200 0 Ctrl 100 mM Recovery KNCL 1000 a T acl I T_abc be be X C1C2C3 E1E2E3 R3 •1000 -J c 800 600 400 200 -J 0 Ctrl 100 mM Recovery KCI 1000 T ab a b j : J L ab ababJ^ J-JL 500 CT LU -500 -A C1C2C3 E1E2E3 R3 •1000 Ctrl 100 mM KNO, 'TA A M M ' A C I D Ctrl 100 mM KCI 138 F I G U R E 5.1.5 The effect of the specific N a + , K + - A T P a s e inhibitor ouabain (0.1 m M and 0.01 m M ) on net ammonia fluxes in P.schlosseri in 2 m M NH4CI in 50%SW. n= 6 and 5 for 0.1 m M and 0.01 m M ouabain treatment groups, respectively. Note that the ammonia excretion rate is significantly reduced in animals exposed to 0.1 m M ouabain while 0.01 m M is without effect. P< 0.05 139 2 mM NrtCI 4 2 mM NH4CI Ouabain • | 600 -o 1h 2h 3h 4h 5h 6h Time (h) I I 0.01 mM Ouabain (n=6) 0.1 mM Ouabain (n=5) 140 F I G U R E 5.2.1 A low magnification light micrograph of a cross section through a gi l l filament (a). Superimposed black boxes indicate the general area of the higher magnification micrographs (b-e). Note (a) the large afferent ( A V ) and efferent (EV) arteries and the oblong cross sectional appearance of the filament. A n asterisk indicates a cartilaginous support rod. The lamellae, which are arranged in parallel rows perpendicular to the long axis of the filament, are seen in sagital section along the top and bottom sides of the filament (a, superimposed white boxes). In higher magnification light micrographs intraepithelial capillaries (b, arrows) can be seen at the efferent edge while being absent at the afferent edge (c). However, skeletal muscle is observed close to the afferent edge (c). The numerous mitochondria of the mitochondria-rich cells found in the lamellae are also visible (d). Also note the distance from the marginal channel to the apex of the lamellae (~15pm) (a, d). The central venous sinus (CVS) , a nerve fascicle (arrow) and an arteriole (asterisk) are located just beneath the base of the lamellae (e). Scale bar (a) 200pm, (b-e) 50pm. 141 F I G U R E 5.2.2 A n electron micrograph (a) shows a lamellar mitochondria-rich (MR) cell anchored to the basal lamina (arrow) opposite the blood spaces defined by a pillar cell (PiC). The outer epithelium is composed of dense pavement cells (PVC) . Note the P V C s numerous vesicles and vacuous mitochondria. The M R cell is flanked by two filament rich cells (asterisk) and connected to the surface by a deep apical crypt (AC) with numerous microvil l i . A light micrograph of a sagital section through a gi l l filament (b). The secondary lamellae appearing in cross section are tightly packed together restricting the water space (arrows) between adjacent lamellae (=interlamellar space). Some pillar cell defined blood spaces are indicated by arrowheads. The efferent or afferent artery occupies the lower part of the figure. Scale Bars (a) 5 pm (b) 50pm. 143 F I G U R E 5.2.3a,b Electron micrographs o f the apical plasma membrane domain of two lamellar M R cells with accessory cell ( A C ) processes. Arrows indicate leaky or thin tight junctions formed between M R cells and accessory cells. The tight junctions between pavement ( P V C ) and M R cells are much thicker (arrowhead).FR: filament rich cell; M : mitochondria. Scale bar = l p m . 145 /4b F I G U R E 5.2.4 Electron micrographs of filament rich (FR) cells. These epithelial cells are found attached to the basal lamina, and compressed between mitochondria-rich cells extend to contact apical pavement cells. They do not themselves make contact with the external mileux. The F R cells are characterized by thick bundles o f intermediate filaments (a, d: arrowheads) associated with desmosomes (d, arrows). Another noticeable feature is the enfolding and elaborate invaginations of the plasmalemma forming pockets or canaliculi like structures (a, b, c: arrows). Scale bar = 1 urn. 147 F I G U R E 5.2.5 Light (a) and electron (b,c) micrographs o f cross sections through the opercular epithelium with its intraepithelial capillaries (a; arrows). Note the similarities in appearance with the gi l l filament efferent edge epithelium (Figure lb) . The capillaries pass though the basal lamina ( B L ) of the stratified epithelium (a), remaining separated from epithelial cells by their retained sheath of basal lamina (b). Red blood cells ( R B C ) can be seen inside the capillary lumens ( C A P ) defined by endothelial cells ( E N D O ) . The blood-water/air diffusion barrier consists of the endothelial cell, the basal lamina, a capillary associated epithelial cell and an epithelial pavement cell (c; < 1pm). Microridges cover the surface cells suggesting mucous adherence may be important in preventing desiccation. Scale Bar = (a) 50um (b) 5pm (c) 1 pm 149 F I G U R E 5.3.1 Indirect immunofluorescence (a) and phase-contrast (c) microscopy showing the distribution of N a + , K + - A T P a s e in the gills of the mudskipper. The mouse monoclonal antibody a5 was used to immunolabel a 5pm fixed-frozen section of 3 % P F A / P B S fixed g i l l tissue. In the two gi l l filaments in cross section, staining is most intense in the cells of the gi l l lamellar epithelium. A section in which a5 was substituted with normal mouse serum in the immunolocalization procedure serves as a control for comparison (b,d). Scale bar = 100pm 151 152. F I G U R E 5.3.2 Immunogold localization of N a + , K + - A T P a s e in the mudskipper gi l l lamellar M R cell using the a5 antibody and a secondary antibody conjugated to 20nm colloidal gold particle. Labeling is greatest in the lower two-thirds of the M R cell (a) and is associated with the tubular system (ts) (b; arrows). N o labeling is associated with the M R cell apical crypt (ac), pavement cells ( P V C ) , pillar cells (P iC) or red blood cells (rbc). Scale bar = (A) 2 pm and (B) 0.5pm 153 F I G U R E 5.3.3 Indirect immunofluorescence and phase-contrast microscopy showing the distributions of N a + , K + - A T P a s e (a,b), C F T R (c,d), and N K C C (e,f) in cells of the lamellar epithelium. Micrographs are taken from the bases of lamellae and the asterisks indicate the alternating lamellar blood spaces. Arrows indicate the location of M R cell apical crypts. N a + , K + -ATPase and N K C C essential have an identical diffuse cytoplasmic distribution confined to lamellar M R cells. C F T R staining is restricted to M R cell apical crypts region. Scale bar = 25 um 155 F I G U R E 5.3.4 Indirect immunoperoxidase staining of fixed-frozen sections of mudskipper gi l l tissue. Sections were either labeled with the carbonic anhydrase (a,b) or vH + -ATPase (c) antisera. Intense labeling is associated with the apical crypt region o f lamellar M R cells. Negligible reactivity was observed in control section incubated with normal rabbit serum (not shown). Scale bar = 20 um 157 15* F I G U R E 5.3.5 N a + / H + exchanger 3 (NHE3) distribution in the gills o f fixed-frozen sections of mudskipper gi l l using indirect immunofluorescence and phase-contrast microscopy. The rabbit polyclonal antibody 1380 was used to immunolabel a 5pm fixed-frozen section of 3 % P F A / P B S fixed gi l l tissue. In low magnification micrographs, a punctate labeling pattern is associated with the lamellar epithelial cells (a,b). Higher magnification of the boxed area reveals that this staining is specific for M R cell apical crypts (c,d; arrows). The asterisks indicate non-specific fluorescence of material associated with the afferent and efferent vessels as is also evident in the final control fluorescence and phase micrograph pair incubated with normal rabbit serum (e,f). Scale bars = (a,b,e,f) lOOum and (c,d) 10p.m. 159 F I G U R E 5.3.6 Projection from a z-stack of 50 confocal images (0.2um z-steps) showing the N H E 2 distribution in the mudskipper gi l l lamellae using indirect immunolabeling with the rabbit polyclonal antibody 597. Staining is associated with the apical crypts of lamellar M R cells and is visible as rings or U shapes of fluorescence in sections, (b) is a high magnification of the outlined area in (a). The distribution of N H E 2 is essential identical to N H E 3 using A b l 3 8 0 (Figure 5.3.5). Scale bar = (a) 25 um and (b) 10 um. 161 F I G U R E 5.3.7 Western blots of mudskipper gi l l tissue crude homogenate (lOpg per lane) separated on 10% polyacrilamide gels probed for (a) a subunit of N a + , K + - A T P a s e , (b) C F T R protein, (c) A-subunit o f vH + -ATPase , (d) N H E - 2 (Ab 597), and (e) N H E - 3 (Ab 1380). M W markers: 205, 112, 87, 69, 56, 38.5 and 33.5 kDa. Large arrowheads indicate bands of interest while smaller arrowheads indicate other bands. 163 5.4 DISCUSSION In the amphibious air-breathing mudskipper fish Periophthalmodon schlosseri, ion transport proteins in gi l l M R cells contribute to active ammonia ( N H 4 + ) excretion. In vivo pharmacological studies show that a significant proportion of the total ammonia efflux ( J A m m ) is sensitive to inhibition of the N a + / H + (NFfV") exchanger (NHE) by amiloride, and carbonic anhydrase ( C A ) by acetazolamide. The elimination of boundary layer acidification by the addition of 5 m M H E P E S buffer p H 8.0 to the 50% S W was without effect on J A m m indicating that NH3 trapping is not an important component of total ammonia excretion. These animals are also capable of excreting ammonia against inward N H / and NH3 gradients (2mM NH4CI), however, a portion of J A m m is then sensitive to N a + , K + (NH 4 + ) -ATPase inhibition by ouabain. Figure 5.4 summarizes the data collected and the proposed pathway o f transbranchial N H 4 + movement. The branchial epithelium contains an abundance of mitochondria-rich cells with deep apical crypts and an extensive tubular system continuous with the basolateral membrane. Immunological localization studies show the presence of N H E 2 and 3 -like isoforms, C F T R , and C A associated with the apical crypt and N a + , K + - A T P a s e and Na + :K + : 2C1" cotransporter ( N K C C ) with the tubular system of these M R cells. The fact that we are able to demonstrate the presence of these transporters within the branchial M R cells strongly indicates that these gil l M R cells are involved in active ammonia elimination. Ammonia elimination during air exposure is significantly greater in P. schlosseri than in other species of mudskipper (50% vs <10% aquatic rate; M o r i i et al. 1978; Iwata et al. 1981). Initially it was thought that mudskippers might use ammonia volatilization as a means of excreting waste nitrogen during land excursions. It has been shown that some terrestrial 165 crustaceans and isopods use this mechanism (eg Greenaway and Nakamura 1991; Weiser 1972). However, this study indicates that volatilization of ammonia is a minor component o f total ammonia excretion, accounting for less than 3% of the total. Alkalinization o f the body surface is necessary to enhance volatilization of ammonia by favouring the formation of gaseous N H 3 . Our measurements of surface p H indicate that there is no special alkaline area of the mudskipper body. In fact, the gi l l surface p H was acidic relative to seawater and other parts of the body surface. It is probable, based on these p H measurements, that ammonia volatilization is from the general body surface rather than from the gills. The amount of volatilization is either not sufficient to acidify the body surface or the animal is able to absorb any acid formed on the general body surface, maintaining p H at about seawater levels. Other evidence indicates that the gills are probably involved in the active elimination of ammonium ions involving Na + ,K + (NH4 + )-ATPase and N a + / H + (NH}"1") antiporter. Consistent with the finding of N H » + elimination during air exposure is the observed blood alkalosis during air exposure (Kok et al. 1998; Ishimatsu et al. , 1999). Boundary layer acidification In P. schlosseri ammonia excretion is not dependent on boundary layer p H . This would indicate that PM3 gradients are not as important a mechanism of total ammonia elimination as in freshwater fishes. A l so the contribution of the vH + -ATPase to ammonia elimination is only significant after 2 h. The facilitation of NH3 diffusion by trapping of NH3 ( N H 3 + H + - » NH4+) through the acidification of the boundary layer by hydration of respiratory C O 2 or direct H + excretion ( N H E or vH + -ATPase) is important in freshwater fishes (Wright et at. 1989; Wilson et al. 1994; Salama et al. 1999) and likely important in seawater fishes as well . The mudskippers kept in 50% S W (pH 8), were able to acidify their bath water through the addition of respiratory 166 C 0 2 to a final p H that was around p H 7.4. The net acid excretion tended to hover around zero. The acidification of the g i l l boundary layer is dependent on the water buffer capacity. Seawater tends to be poorly buffered and thus the gi l l boundary layer w i l l be acidified. Experimentally, the boundary layer acidification can be eliminated by the addition of buffer to the water. In the mudskipper, the elimination of boundary layer acidification by buffering with 5 m M H E P E S p H 8.0 results in no change in JAmm- The H E P E S p H 8.0 did, however, stimulate a large JACID flux presumably by providing a large sink for protons. This effect on JACID has also been observed in freshwater trout (Salama et al. 1999). The addition of H E P E S buffer at p H 7.0 also had no effect on net ammonia flux rates suggesting NH3 trapping is not a necessary component of J A M M -Substitution of NH/ on some ion transport proteins Na+X-ATPase The gi l l M R cells are associated with high levels of N a + , K + - A T P a s e immunoreactivity and in vitro assays of (ouabain-sensitive) enzyme activity are also very high. The levels of N a + , K + - A T P a s e activity are 3-4 times higher than another mudskipper species (Boleophthalmus boddaeti; Randall et al. 1999) and double the values obtained for seawater adapted coho salmon (using the same assay conditions; Figure 4.5). The levels of immunoreactivity in the other gi l l cell types ( P V C , erythrocytes, pillar cells, F R cells) are below the level of detection although the absence of immunoreactivity is not a good criteria for concluding its absence. There are limits to the sensitivity of the technique in addition to the possibility that other isoforms of the a subunit are present in these cell types. Indeed, N a + , K + - A T P a s e is likely present in all cell types performing cellular housekeeping roles. In the G u l f toadfish (Opsanus beta) Claiborne et al. (1982) and Evans et al. (1989), were able to demonstrate that inhibition of N a + , K + - A T P a s e by ouabain significantly reduced JAmm in 167 this marine fish. However, it has been argued that under in vivo conditions (plasma [K ] ~ 5 m M and [ N H 4 + ] -0.1 m M ) K + w i l l out compete N H 4 + for transport binding sites on the N a + , K + -ATPase thus making the physiological significance of N H 4 + substitution questionable (Wall 1996). However, it should be noted that plasma [K + ] may over estimate [K + ] within the tubular system of the g i l l M R cell which is associated with high N a + , K + - A T P a s e activities, and likely have lower [ K + ] . Yet, in P.schlosseri kept under conditions of low external ammonia ouabain is without effect on TAmm (T. Kok , personal communication). Presumably under these conditions, NH3 and N H 4 + diffusion are capable of maintaining ammonia excretion (across the basolateral membrane of the epithelium at least). However, during exposure to high environmental ammonia when N H 4 + and NH3 gradients are directed inward (2mM) J A m m is sensitive to ouabain (Figure 5.1.5). This inhibition of JAmm is associated with a significant increase in plasma NFL; + levels (Table 5.1). Thus in the absence of favourable diffusion gradients N H 4 + substitution becomes physiologically significant. Mallery (1983) was also able to show that the sodium pump has a potentially higher affinity for N H 4 + than K + . Although J A m m was inhibited by apically applied ouabain, the effect was likely basolateral. Ouabain is a specific inhibitor of N a + , K + - A T P a s e activity and even though apical N a + , K + - A T P a s e has been described in some cell types (salivary gland; Just and Walz 1994), our immunogold observations indicate a distribution restricted to the basolateral tubular system (Figure 5.3.2). The basolateral effect of apically applied ouabain is believed to be due to entry of ouabain via the shallow tight junction found between M R chloride and accessory cells (Phillpot 1980). 168 Alternatively, the effect of inhibition of N a + , K + - A T P a s e by ouabain may not be a direct reflection of basolateral N H 4 + movement but an indirect effect of build up of intracellular N a + on an apical N a + / N H 4 + exchanger. ApicalNa+ /Ft (NH/) exchange The inhibition of the N H E by amiloride reduces JAMM by 50% yet the elimination of boundary layer acidification is without effect on JAMM- It therefore seems likely that an important component of ammonia elimination is achieved by direct N H 4 + transport by the N H E rather than a boundary layer p H effect by N a + / H + exchange. In renal microvillus membrane preparations N H 4 + has been shown to interact with N H E by Kinsella and Aronson (1981). The concentration of amiloride used in this study does not distinguish which N H E isoform is contributing to JAMM-There are at least five N H E isoforms ( N H E 1-4, P) although only the N H E 2 and N H E 3 are found apically. Immunoreactivity of both these isoforms is present in the apical crypt of gi l l M R cells. However, in the mammalian kidney medulla thick ascending limb ( M T A L ) , although both isoforms are also present, it is the N H E - 3 isoform that plays a greater role in the ammonium efflux of this segment (Paillard 1998). Recently, Claiborne et al. (1999) reported the presence of PNHE- l ike and NHE2- l i ke isoforms in gi l l of the sculpin using R T - P C R . In the marine fish Opsanus beta, Evans et al. (1989) found no evidence for apical N a + / N H 4 + exchange. Instead, they found that total ammonia excretion could be accounted for by non-ionic N H 3 (57%), paracellular ionic N H / (21%) diffusion and basolateral N a + , K + - A T P a s e (22%). However, their data could be interpreted as indicating that the N a + , K + - A T P a s e and N H E mechanisms are operating in series. In the same study they could not find evidence to support a role for N K C C in transbranchial ammonia transport. Goldstein et al. (1980, 1982) found that N H 4 + diffusion and N a + / N H 4 + exchange are important components of ammonia excretion in marine teleost fishes 169 noting the importance of paracellular leak and the inward N a gradient to drive N H E exchange in marine species. In seawater adapted rainbow trout challenged with I m M N H 4 + a base load results, consistent with N a + / N H 4 + exchange (Wilson and Taylor 1992). Randall et al. (1999) found increased [Na + ] p i in P.schlosseri challenged with high external ammonia consistent with the proposed mechanism of N a + / N H 4 + exchange. Carbonic anhydrase Carbonic anhydrase immunoreactivity has been observed in the apical region of gi l l M R cells and the sensitivity of JAmm to C A inhibition indicates that intracellular CO2 hydration may be important in providing H + for NH3 protonation to maintain PNH3 gradients across the basolateral membrane and providing N H 4 + for apical N a + / N H 4 + exchange. From the lack of ouabain-sensitive J A M M under conditions of low external ammonia it would appear that the facilitated movement of N H 4 + across the basolateral membrane is not important. The distribution of C A in the branchial M R cells is similar to that reported by Lacy (1983) in M R cells from the opercular epithelium of seawater adapted kill if ish. vH*-ATPase The role of the vH + -ATPase in ammonia excretion is not clearly defined in light of the absence of a p H dependent boundary layer effect and the in vivo experiments using K N O 3 which had a delayed effect on JAMM- Also , although K N O 3 reduced J A CID the effect was not different from the K C I control. However, immunoreactivity of the vH + -ATPase could be demonstrated in apical crypts of M R cells and NO3" and bafilomycin sensitive ATPase could be demonstrated in gil l homogenates using an in vitro assay. 170 CFTR It was initially thought that the C F T R could be used as a marker for M R cells involved in C l " secretion; a function normally ascribed to this cell type in sea water teleost fishes (chloride cells; Singer et al. 1998). In the initial investigation of the fine stucture of the branchial epithelium few chloride cell-accessory cell complexes were found so it was thought that few of the M R cells were actually involved in C l " elimination. However, CFTR- l ike immunoreactivity was associated with the apical crypts in practically every branchial M R cell. So contrary to our morphological observations of few accessory cells, it maybe that all M R cells are involved in Cl" elimination. This would be the conclusion i f the function of the C F T R - l i k e protein were limited to C l " conductance; however, the CFTR- l ike protein is also involved in the regulation of other apical channels as well as the conductance of other anions, notably HCO3" (Poulsen et al. 1994). It is this latter function that becomes quite intriguing. Assuming that C O 2 hydration catalyzed by intracellular C A provides H + ions for NH3 protonation to produce N H 4 + for N H E then an HCO3" is left behind. HCO3" must exit the cell either across the basolateral or apical membrane however, basolateral exit would result in a base load during ammonia excretion. Apica l exit facilitated by the C F T R , would be acid-base neutral and account for the lack of boundary layer p H effects. It also explains why JACID is^around zero and is not affected by the inhibition of J A M M with acetazolamide or amiloride. Salinity experiment In teleost fishes in general, there is a positive correlation between environmental salinity and N H 4 + permeability (Evans and Cameron 1986). This increased permeability is related to the formation of leaky tight junctions between chloride and accessory cells ( C C - A C ) that provides a route for the paracellular efflux o f N a + . C C - A C numbers show a positive correlation with salinity 171 because they are needed for N a C l elimination (see Laurent 1984). The euryhaline sculpin has been shown to reduce JAMM and a compensatory increase in J u r e a upon acute and chronic exposure to low salinity (Wright et al. 1995). In the mudskipper there is a trend in the data that follows this relationship but it is not significant. Curiously, plasma [ N H 4 + ] and P N H 3 show, a positive relationship with salinity. It is unclear why the fish at higher salinity would have higher plasma ammonia levels. In some other mudskipper species there is the opposite correlation with salinity: higher JAMM at lower salinity while JUREA is lower (Iwata et al. 1981; Gordon et al. 1965). It also is unclear why this relationship exists. None of the other plasma variables measured or gi l l ATPase levels were affected by environmental salinity. Unfortunately, I could not look at morphological changes in the gi l l (ie increased A C s or leaky tight junction formation) because of the poor quality of the tissue fixation. From my results it would appear that the paracellular route for NFL} + is not as important in the mudskipper as it is in other marine fishes (Evans et al. 1989). Randall et al. (1999) found that branchial T E P is unresponsive to increases in environmental ammonia indicating a very low N H 4 + permeability. The fact that this mudskipper lives in an estuarine environment and faces daily changes in salinity with the changing tides may explain why no differences were seen at the different salinity regimes. Also relatively few C C - A C type complexes were seen in the gil l suggesting the paracellular path may not be important for JAMM- In addition, i f the burrow in which this animal lives in is filled with 2 m M N H 4 + seawater then a high N H 4 + (paracellular) permeability would be disadvantageous. However, N a + efflux via a paracellular route is still likely as the equilibrium potential of N a + is close to the measured T E P consistent with other 172 euryhaline marine fishes (+10 m V ; see Chapter 4, and Potts 1984). It is possible that the barrier function of the shallow tight junctions shows selectivity for cations. Gill Morphology The abundance of M R cells within the lamellar epithelium is highly unusual for a marine fish. M R cells are generally restricted to the filament epithelium (reviews by Laurent 1984; Pisam and Rambourg 1991). This is also generally true for freshwater fish; however, under ion-poor conditions, numerous M R cells can be found covering their lamellae (Avella et al. 1987; Laurent and Perry 1991). In the freshwater aquatic air-breathing fish Arapiama gigas, the gills have reduced lamellae and numerous M R cells in the filament epithelium (Hulbert et al. 1978). This fish is a bimodal breather and the gil l is the site of CO2 excretion into water while the lung-like accessory respiratory organ (ARO) operates for O2 uptake (Randall et al. 1978). In this case, the gi l l presumably is still functioning in ion and acid-base regulation and ammonia excretion through the filament and reduced lamellar epithelium. The gills of the more aquatic mudskipper Boleophthalmus boddarti do not have the features of P. schlosseri and do not differ so greatly from other teleost fishes in both gross and fine structure (Hughes and Munshi 1979; Al-Kadhomiy and Hughes, 1988; L o w et al. 1990). Squamous epithelial cells cover the lamellae and M R cells are generally restricted to the filament epithelium. They do not have the abundance of lamellar M R cells or filament rich cells of P. schlosseri. From the illustrations of the gills of mudskippers Periophthalmus vulgaris and Periophthamus chrysospilos it appears as though the lamellae of these species have M R cells, although not in the same densities as P. schlosseri, and they also lack interlamellar fusions (Schottle 1931; Low et al. 1990). 173 The study o f Yadav et al. (1990) deserves special mention as their observations are possibly on the same species {P. schlosseri). Although they also noticed interlamellar fusions, they are reported only at the tip of the filaments while I and L o w and co-workers (1990) observed interlamellar fusions along the length of the filament. Yadav et al. (1990) also make no mention of M R cells in the lamellae. From their observations o f lamellar vascular papillae and blood sinuses they suggest that the gi l l is still an important site for gas exchange while from my observation I cannot agree. It seems likely that they are looking at another species (Periophthalmodon septemradiatus or Periophthalmus minutus) as suggested by Clayton (1993). In P. schlosseri, despite the abundance o f lamellar M R cells, accessory cells and shared apical crypts were seldom observed. Laurent (1984) observed that N a C l elimination in marine teleosts depends on the intimate relationship between the chloride (mitochondria-rich) cell (CC) and accessory cell ( A C ) . The A C sends cytoplasmic projections into the neighbouring chloride cell which surface within the apical crypt of the C C . The tight junctions between C C and A C are short, consisting of 1-2 strands. In the classical N a C l elimination model (Silva et al. 1977; Sardet et al. 1979), C l " moves through the C C and N a + through the leaky paracellular channels (review by Karnaky 1998 and Chapter 4 Introduction). C l " moves into the C C across the basolateral membrane via a N K C C and out of the cell across the apical membrane via a CFTR- l ike C l" channel. A number of other changes in M R cell fine structure are associated with seawater adaptation, but it seems that the C C - A C relationship is the most diagnostic (Laurent 1984). For instance, apical crypts have been reported in Pimephales promelas, a freshwater species, following chronic exposure to acidic condition (pH 5; Leino and McCormick 1984). In these fish A C were reported to be absent and the crypts were explained to be involved in acid-base regulation at low p H . Also in the freshwater tilapia, apical crypts are a common feature 174 associated with M R cells (Perry et al. 1992; van der Heijden et al. 1997). On the basis of fine structure it seems, therefore, that only a minority of M R cells are involved in N a C l elimination in the gills of the mudskipper P. schlosseri, while the majority of M R cells, which lack A C , may be specialized for something other than N a C l elimination. However, the immunolocalization of CFTR- l ike channel to the apical crypt of most would indicate that these cells are also capable of C l " secretion. In P. schlosseri, M R cells are isolated from one another by neighbouring cells rich in filaments. To my knowledge, such an arrangement has not been observed in the branchial epithelium of any other fish, although a similar cell type has been observed in the opercular membrane of the kil l if ish (Fundulus heteroclitus; Lacy 1983). It is possible that the M R cells with their complex tubular system require some form of external reinforcement that may be provided by these cells. It appears that these filament-rich cells are not contractile in nature since the predominant filament type is intermediate and not microfilament. The unusual canaliculi-like structures that are found within these cells may provide a route for the movement of interstitial fluid. Gas exchange organ The gi l l lamellae of P. schlosseri are clearly not structurally designed for gas exchange. The interlamellar fusions while allowing the prevention of desiccation of the epithelium during air exposure also effectively extend the diffusion distance from the blood space to the apex of the lamellae. The movement of water through the interlamellar spaces is difficult to imagine. Thus, with the exception of the marginal channel, much of the blood is effectively shunted through the lamellae. Also , the high densities of M R cells within the tissue suggest that it is an oxygen-demanding organ rather than a site for oxygen uptake into the blood. 175 The inner-opercular lining with its increased surface area and intraepithelial capillaries makes a better gas exchange site than the lamellae. There is also a similar system of intraepithelial capillaries in areas of the leading edge of g i l l filaments and on the gi l l arch. The oxygenated blood from both these areas returns to the heart via the venous circulation (Schottle 1931; review by Laurent 1984) and contributes oxygenated blood to the returning systemic de-oxygenated blood resulting in a higher P 0 2 of blood perfusing the gills. The gills probably consume oxygen from the blood and perhaps the air, the end result is an arterial P 0 2 that has been observed to be quite low, between 25-40 torr (Kok et al. 1998; Ishimatsu et al. 1999). Summary In summary, the elimination of ammonia by P. schlosseri involves a N H E and C A and is not dependent on boundary layer p H effects. N a + , K + - A T P a s e plays a role in ammonia elimination only against a gradient. Although a vH + -ATPase is present its role in ammonia elimination is not certain. These animals are able to eliminate significant quantities of ammonia when out of water. The gills of P. schlosseri express all o f the above proteins and show some unique features that can be used to explain the animal's ability to actively eliminate ammonia. It also is clear that gas exchange across the gi l l lamellae would be severely handicapped by the long diffusion distances present. The observation of intra-epithelial capillaries in the inner operculum and to a lesser extent in the gil l filament would present more practical sites for gas exchange. 176 FIGURE 5.4 Illustration o f the mudskipper g i l l M R cell and its proposed role in active N F l / elimination. Ammonia enters the M R cells from the blood either by NH3 diffusion or active transport by Na + ,K + (NH4 + ) -ATPase and/or N a + : K + ( N H 4 + ) : 2C1" cotransporter. Intracellular carbonic anhydrase ( C A ) provides for NH3 protonation by catalyzing CO2 hydration. The N H 4 + formed is moved across the apical membrane in exchange for N a + by a N H E - l i k e carrier. The accumulated HCO3" exits the cell apically via an CFTR- l ike anion channel. 177 Mudskipper Model water | 4 S tdbf CFTR NKA v-ATPase NHE NKCC 6. GENERAL DISCUSSION The cellular and subcellular distributions of ion transport proteins can be determined using electrophysiological, biochemical, and immunological techniques. While the first two techniques have the advantage of providing evidence for the functional presence of a protein, the techniques have their limits in application (a point that is well illustrated with the fish gill). The gi l l epithelium is heterocellular, thin (1-15 pm), and arranged in three dimensions on filaments and lamellae. Direct study of the gi l l using these techniques has proven technically challenging. To date there is but one published report of whole cell recordings from fish g i l l cells estimating membrane potential differences (Clarke and Potts 1998) and a large number of transepithelial potential measurements (reviewed by Potts 1984). Biochemical studies have made use of differential cell isolation techniques ( M R cells from other cells, P V C s ) for characterizing cell types (e.g. Wong and Chan 1999) but have not had the benefit o f having protocols for isolating the apical plasma membrane fraction from gi l l epithelial tissue. There have not been any apical membrane markers identified in fish gi l l to make such an isolation possible and it also is unclear i f the techniques for isolating basolateral membrane also do not include apical membrane (Flik and Verbost 1994; Henry et al. 1997; Crockett 1999). Thus much of our understanding of the gil l 's contribution to ion regulation is constructed from data obtained from whole animal experiments, in vitro whole gi l l preparations, isolated epithelial cells, and surrogate models (e.g. L i et al. 1998; Marshall and Bryson 1998; Evans et al. 1999; Wong and Chan 1999). Immunological detection introduces a level of resolution that has helped in defining the function of branchial cell types. In this thesis, I have been able to demonstrate the usefulness of the immunological approach to ion transport protein localization. From my observation of freshwater fish species, 179 the vH + -ATPase and epithelial N a + channel have been shown to be co-localized. This lends credence to the vH + -ATPase driven N a + uptake model (Avella and Bornancin 1989; L i n and Randall 1995; Fenwick et al. 1999). There is not a clear distinction between the contributions of the M R cell versus the P V C that would allow me to generalize among teleost fishes. The involvement of the M R cell in N a + uptake shows species specificity. The freshwater M R cell does however seem to have a clear role in C l " uptake as an apical A E is localized to this cell in both the tilapia and coho salmon. Thus initial reference to this cell type as a chloride cell seems apt. The M R cell apical A E is lost in sea water adapted coho salmon but present in the stenohaline turbot. The role o f the N H E in freshwater N a + uptake is not clear, but the antibody I used in this study may prove useful as a marker for M R accessory cells that are found in marine and euryhaline fishes. In marine fishes, it is possible to localize the various ion transport proteins involved in active C l " excretion (Na + ,K + -ATPase , N K C C and C F T R ; Marshall and Bryson 1998). I was not able to find evidence to support the proposed role of branchial P V C s in active C l " excretion as suggested by the results of Avel la , Ehrenfeld and co-workers (1997, 1999; Duranton et al. 1997) studying seabass P V C primary cultures. Immunological localization techniques, of course, have limitations and I would not argue solely on the immunological evidence that these ion transport processes are taking place. So while the technique can demonstrate the presence of a protein, it cannot demonstrate i f it is functional. However, there is generally a priori evidence for the presence of many of these transporters in the g i l l o f freshwater and seawater fishes (Lin and Randall 1995; Goss et al. 1995; Evans et al. 1999). Thus the immunodetection data corroborates the physiological studies and provides additional evidence about ion transport protein cellular and subcellular distributions. In 180' the case of the mudskipper, basic morphological and physiological data was generated to support the immunolocalization data and construct a model for ion exchange in this species. In the mudskipper a convincing argument can be made that the unusual branchial M R cells are involved in the active N H 4 + excretion observed when the animal is out of water or in its water filled burrow which has ammonia concentrations around 2 m M ( Y - K Ip, personal communication). Assembled within the branchial M R cells are ion transport proteins that contribute to net ammonia flux. Lining the M R cell apical crypt are two immunoreactive N H E -like isoforms (NHE-2 and N H E - 3 ) . N H E has been found in vitro to transport N H 4 + ions (Kinsella and Aronson 1981), which in the case of the mudskipper makes use of inward N a + gradients to drive N H 4 efflux (Randall et al. 1999). A n apical CFTR- l ike anion channel may account for acid-base neutrality by facilitating the efflux of HCO3". Intracellular carbonic anhydrase ( C A ) catalyzes the CO2 hydration reaction to supply H + ions for NH3 protonation to provide N H 4 + for the N H E . Basolateral N a + , K + - A T P a s e contributes to ammonia excretion when the animal is faced with inward gradients of ammonia. The role of basolateral N K C C has yet to be established. The vfT-ATPase, although present in the M R cell, makes an uncertain contribution to ammonia excretion. Western analysis and control tissue staining using sera or IgG from the host animals (rabbit or mouse) were used to assess antibody specificity. In general, staining patterns seen also with the rabbit or mouse control were considered non-specific and not used; however, there are some difficulties with the Western blots. There are cases where antibodies useful for immunohistochemistry have failed to work on immunoblots; recognizing no bands ( N K C C ) or too many bands (NHE-2 , NHE-3 ) . Because of this, the data does have to be interpreted with at least some caution. However, crossreactivity may not have been observed for technical reasons 181 and the additional bands may represent breakdown products (lower M W ) or aggregates (higher M W ) of the protein o f interest. The presentation of proteins for immunodetection on tissue sections and immobilized on membrane supports is different. In tissue sections the protein is in its near native state, however proteins separated by S D S - P A G E and transferred to a support membrane have lost their native conformation and thus in the process may have lost epitopes previously recognized. The other antibodies used generally immunoreacted with bands in the correct molecular weight range (a-subunit N a + , K + - A T P a s e , C F T R , p-subunit E N a C , A-subunit vH + -ATPase) . Another point that needs to be addressed is negative results. Generally, the absence of immunoreactivity with tissue sections can be explained either by the absence of that protein with possibly a different isoform present within the tissue (i.e. the antibody is not specific for the protein) or again by technical problems (i.e. the antibody is unable to recognize the protein (epitope) in its present conformation). Without any sequence information on the protein in question, it not possible to distinguish the cause of this common outcome. The trout gi l l apical A E illustrates this point well because immunoreactivity within the branchial epithelium was not observed, however, there is much indirect evidence that the trout g i l l epithelium has an apical Cl" /HCO3" exchanger (correlation of C l " fluxes with M R cell morphometric changes, Goss et al. 1995; intracellular ion changes, Morgan et al. 1994, Morgan and Potts 1995). Apical immunoreactivity was observed in the freshwater tilapia and coho salmon. A non band 3-like A E isoform may be present in the trout M R cell similar to the mammalian collecting duct p-cell which expresses an apical A E 2 (Alper et al. 1997). There are again a number of technical reasons why crossreactivity was not observed ranging from masking of the epitope, poor antibody-antigen avidity (complex stability) or insufficient antigen. The absence of 182 immunodelectability of the basolateral N a + , K + - A T P a s e in P V C s also illustrates this latter point. N a + , K + - A T P a s e has a cellular housekeeping role contributing to the maintenance of the membrane potential and intracellular N a + and K + concentrations and thus is expressed in all vertebrate cells. Immunolocalization techniques used in this thesis and by others (Witters et al. 1996, Ura et al. 1996; Deng et al. 1999) have failed to detect the P V C basolateral N a + , K + -ATPase. Hootman and Phillpot (1979) were also unable to demonstrate the presence of N a + , K + -ATPase in branchial cells other than the M R cells using H-ouabain autoradiography. In the freshwater fish P V C s involved in N a + uptake, the basolateral N a + , K + - A T P a s e , would be expected to participate in the entry step of N a + into the blood (Richards and Fromm 1970; Pay an et al. 1975). It is possible that a different a-subunit isoform exists in the P V C that is not recognized by the monoclonal antibody a5 . Isoform variability and different a p dimer combinations impart different pump characteristics for specialized cell or tissue functions (Blanco and Mercer 1998). The usefulness of using non-homologous antibodies to proteins in such distantly related animals as mammals and teleost fishes is a legitimate concern. In addition, many of these ion transport proteins belong to large multigene families whose members may have divergent functions. It must be remembered however, that some of these proteins are also highly conserved and have functions essential for basic cellular homeostasis. One such example is the vH + -ATPase that is found in all l iving creatures and has a highly conserved A-subunit (archaebacteria to humans; e.g. Finbow and Harrisonl997). A peptide antibody was used to immunolocalize the A -subunit and a Blastp search (www.cbr.nrc.ca) of the peptide amino acid sequence (Sudhof et al. 1989) reveals high identity with other vertebrate and non-vertebrate A-subunit sequences (humans to Drosophilla melanogaster 100-73% identity, 16 hits). N o non- vH + -ATPase A -subunit sequences were found in the search. Thus it would seem that this sequence is also present 183 in the fish vH + -ATPase A-subunit. The peptide from P-subunit of the E N a C used to generate polyclonal antibodies produced similar results with the Blastp search. The P-subunit of the E N a C is the most highly conserved of the three E N a C subunits (a,p,y; Garty and Palmer 1997). The N a + , K + - A T P a s e a-subunit antibody is considered pan-specific recognizing the a-subunit in animals ranging from crayfish to humans (eg Barradas et al. 1999; Sabolic et al. 1998). Future directions While there is a general dearth of information provided by whole cell recordings, patch clamp techniques and the use o f ionophores stand to provide valuable information on transporter properties and intracellular ion concentrations. L i et al. (1998) have measured intracellular ion concentrations and p H in isolated cells and I know of at least three groups which are preparing to attempt some electrophysiology using patch clamping techniques ( C M . Wood, Hamilton Can.; R. Handy, Plymouth U K ; S. Thomas C N R S Fr.). The development of protocols for the isolation of apical membranes would allow for the purification of fish ion transporters for the development of homologous antibodies. I had attempted to do so by using a sucrose gradient and my antibodies as probes for membrane fractions but my method lacked resolution and I lacked the resolve. The use of surrogate models (primary culture and skin preparations) while currently uncertain, may yet provide results consistent with in vivo observations (Burgess et al. 1998; Wood et al. 1998; Gilmour et al. 1998). It is now possible to have M R cells in primary gil l culture using a double seeding technique (Kelly et al. 1999). With this advance it may be possible to demonstrate N a C l uptake as seen in vivo. Molecular techniques stand to contribute greatly to our understanding of ion regulation in fishes. The use of differential display has already lead to the identification of an upregulated 184 inwardly rectifying K + channel (Kir) in M R cells following seawater transfer (Suzuki et al. 1998). Cloning of the fish genes w i l l in turn lead to the generation o f homologous fish antibodies. However, with 25 000 known species of teleost fishes there are years of work ahead. The recent work on the vH + -ATPase B-subunit illustrates this point as two isoforms were found in the eel (Pelster et al. 1999) while only a single isoform in the rainbow trout (Perry et al. 1999). Also teleost fishes generally have more functional members of multigene families (paralogues) lacking mammalian counterparts (orthologue) (Wittbrodt et al. 1998). Wittbrodt et al. (1998) suggested that the complex genomic architechture of fishes has allowed them to speciate quickly in response to changing selective regimes. The mudskipper represents an interesting creature to study. The story of its ability to tolerate high environmental ammonia levels is far from over. The possible role of the C F T R in HCO3" efflux is worth following up. In the duodenum and airways of mammals the C F T R anion channel has been found to have a direct role in HCO3" efflux for alkalization of the lumen (Hogan et al. 1998; Lee et al. 1998). The inability of high environmental ammonia levels to elevate plasma ammonia levels in the mudskipper suggests that the apical membrane permeability to N H 3 is likely very low. During terrestrial excursions when ammonia accumulates in the surface film of water an impermeable membrane is also required to prevent a NH3 back flux of the excreted N H 4 + . There are examples of membranes impermeable to N H 3 such as the kidney medullary thick ascending limb of Henle (Kikeri et al. 1989). Interestingly, Kiker i et al. (1989) were also able to demonstrate that an apical N H E facilitates N H 4 + efflux but not influx across the apical membrane. Membrane permeability can be decreased by decreasing the fluidity of the membrane by altering the lipid composition either decreasing the amount of unsaturated 185 hydrocarbon chains, or increasing the amount of cholesterol (see Zeidel 1996). Future studies focusing on the properties of the apical membrane should provide most interesting results. Summary Conclusions 1. ) In freshwater teleost fishes, I have been able to show that there is a clear link between the vH + -ATPase and E N a C , offering strong support to the vH + -ATPase driven model of N a + uptake via an E N a C . I have also been able to demonstrate the presence of an apical band 3-like A E in M R chloride cells in tilapia and coho salmon supporting the idea of this cell type's role in active C l " uptake. 2. ) In marine teleost fishes, I have been able to demonstrate the presence of ion transport proteins involved in active C l " elimination in branchial M R chloride cells (CFTR-l ike , N K C C , N a + , K + - A T P a s e ) . However, I was unable to support the idea that P V C s are also involved in active C l " elimination. Branchial M R cells may also be involved in acid-base regulation by means o f N H E and A E proteins. Incidentally, the antibody 597 against N H E - 2 also appears to be good marker for the accessory cell type. 3. ) The mudskipper represents an unusual fish that spends much of its time out of water and is able to actively eliminate ammonia. In a series of studies I have been able to show that the branchial epithelium M R cells are likely the sites of active NFL} + elimination. N H 4 + elimination being mediate by an apical N H E - l i k e protein. 186 REFERENCES Al-Kadhomiy, N . K . and G . M . Hughes 1988 Histological study of different regions of the skin and gills in the mudskipper, Boleophthalmus boddarti with respect to their respiratory function. J.mar.biol.Ass.U.K., tfS:413-422. Alper, S.L., J. Natale, S. Gluck, H .F . Lodish, and D . Brown 1989 Subtypes of intercalated cells in rat kidney collecting duct defined by antibodies against erythroid band 3 and renal vacuolar H + -ATPase . Proc.Natl .Acad.Sci .USA, S6":5429-5433. Alper, S.L. 1991 The band 3-related anion exchanger (AE) gene family. Ann.Rev.Physiol. , 53:549-564. Alper, S.L. , A . Stuart-Tilley, D . Biemesderfer, B . E . Shmukler, and D . Brown 1997 Immunolocalization of A E 2 anion exchanger in rat kidney. Am.J.Physiol . , 273:F601-F614 Altendorf, K . , K . U . Bindseil , E.J . Bowman, A . Siebers, and A . Zeeck 1993 Inhibitory effect of modified bafilomycins and concanamycins on P- and V-type adenosinetriphosphatases. Biochemistry, 32:3902-3906. Anderson, M . P . , R .J . Gregory, S. Thompson, D . W . Souza, S. Paul, R . C . Mull igan, A . E . Smith, and M . J . Welsh 1991 Demonstration that C F T R is a chloride channel by alteration of its anion selectivity. Science, 253:202-205. Appel , C , S. Gloor, G . Schmalzing, M . Schachner, and R.P. Bernhardt 1996 Expression of a Na ,K-ATPase p 3 subunit during development of the zebrafish central nervous system. J.Neurosci.Res., 46:551-564. Appel l , K . C . and P.S. L o w 1981 Partial structural characterization of the cytoplasmic domain of the erythrocyte membrane protein, Band 3. J.Biol.Chem., 255:11104-11 111. Avel la , M . , A . Masoni, M . Bornancin, and N . Mayer-Gostan 1987 G i l l morphology and sodium influx in the rainbow trout (Salmo gairdneri) acclimated to artificial freshwater environments. J.exp.Zool., 247:159-169. Avel la , M . and M . Bornancin 1989 A new analysis of ammonia and sodium transport through the gills of the freshwater rainbow trout (Salmo gairdneri). J.exp.Biol., 742:155-175. Avel la , M . and M . Bornancin 1990 Ion fluxes in the gills of freshwater and seawater salmonid fish. In: Animal Nutrition and Transport Processes. 2. Transport, Respiration and Excretion: Comparative and Environmental Aspects. J.P. Truchot and B . Lahlou, eds. Comp Physiol., BasefKarger, pp. 1-13. Avel la , M . and J. Ehrenfeld 1997 Fish g i l l respiratory cells in culture: A new model for Cl" secreting epithelia. J.Membrane B i o l . , 756:87-97. 187 Avella , M . , P. Part, and J. Ehrenfeld 1999 Regulation of C l " secretion in seawater fish (Dicentrarchus labrax) g i l l respiratory cells in primary culture. J.Physiol., 5/6:353-363. Barradas, C , S. Dunel-Erb, and J. Lignon 1997 Transepithelial potential difference of a single g i l l filament isolated from the crayfish Astacus leptodactylus Esch.: A new method. Arch.Physiol.Biochem., 705:38-44. Barradas, C , J . M . Wilson, and S. Dunel-Erb 1999 N a + / K + - A T P a s e activity and immunocytochemical labeling in podobranchial filament and lamina of freshwater crayfish Astacus leptodactylus Eschscholtz: evidence for the existence of a sodium transport in the filaments. Tissue and Ce l l , (In Press) Bear, C . E . , C . L i , N . Kartner, R.J . Bridges, T.J. Jensen, M . Ramjeesingh, and J.R. Riordan 1992 Purification and functional reconstitution of the cystic fibrosis transmembrane conductance regulator (CFTR) . Ce l l , 65:809-818. Benos, D.J . 1982 Amiloride: a molecular probe of sodium transport in tissues and cells. Am.J.Physiol . , 242:C131-C145 Benos, D.J . , M . S . Awayda, I.I. Ismailov, and J.P. Johnson 1995 Structure and function of amiloride-sensitive N a + channels. J.Membrane B i o l . , 745:1-18. Bianchini, L . and J. Pouyssegur 1994 Molecular structure and regulation of vertebrate NaVtf" exchangers. J.exp.Biol., 796:337-345. Biemesderfer, D . , P . A . Rutherford, T. Nagy, J .H. Pizzonia, A . K . Abu-Alfa , and P.S. Aronson 1997 Monoclonal antibodies for high-resolution localization of N H E 3 in adult and neonatal rat kidney. Am.J.Physiol. , 275:F289-F299 Bijman, J., A . M . Hoogeveen, B . J . Scholte, M . Kansen, A . W . M . van der Kamp, and H.R. de Jonge 1990 Ion transport regulation in cystic fibrosis epithelia. In: Epithelial secretion of water and electrolytes. J .A. Young and P . Y . Wong, eds. Springer-Verlag, Berlin, Heidelberg, pp. 361-369. Bindon, S.D., K . M . Gilmour, J.C. Fenwick, and S.F. Perry 1994 The effects of branchial chloride cell proliferation on respiratory function in the rainbow trout Oncorhynchus my kiss. J.exp.Biol., 797:47-63. Blackston, C.R. , W. Holl imon, and J .B. Claiborne 1997 Isoforms of the N a + / H + antiporter (NHE) in g i l l m R N A of the marine long-horned sculpin (Myoxocephalus octodecimspinosus). Bull.Mt.Desert Isl .Biol.Lab., 22 Blanchard, A . , D . Eladari, F. Leviel , M . Tsimaratos, M . Paillard, and R . - A . Podevin 1998 N H / as a substrate for apical and basolateral N a + - H + exchangers of thick ascending limbs of rat kidney: evidence from isolated membranes. J.Physiol., 506.5:689-698. Blanco, G . and R . W . Mercer 1998 Isozymes of the Na-K-ATPase : heterogeneity in structure, diversity in function. Am.J.Physiol. , 275:F633-F650 188 Blattler, D.P. , F . Garner, K . van Slyke, and A . Bradley 1972 Quantitative electrophoresis in polyacrilamide gels o f 2-40%. ^Chromatography, 64:147-155. Bookstein, C , Y . X i e , K . Rabenau, M . W . Musch, R . L . McSwine, M . C . Rao, and E . B . Chang 1997 Tissue distribution of N a + / H + exchanger isoforms N H E 2 and N H E 4 in rat intestine and kidney. AmJ .Phys io l . , 27J:C1496-C1505 Borgese, F. , C . Sardet, M . Cappadoro, J. Pouyssegur, and R. Motais 1992 Cloning and expression a cAMP-activated N a + / H + exchanger; evidence that th cytoplasmic domain mediates hormonal regulation. Proc.Natl .Acad.Sci.USA, 59:6768-6769. Bornancin, M . , G . de Renzis, and R. Naon 1980 CF-HCO3"- ATPase in gills of the rainbow trout: evidence for its microsomal localization. AmJ.Phys io l . , 235:R251-R259 Bowman, E.J . , A . Siebers, and K . Altendorf 1988 Bafilomycins: A class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells. Proc.Natl .Acad.Sci .USA, 55:7972-7976. Bradbury, N . A . 1999 Intracellular C F T R : Localization and function. Physiol.Rev., 7P:S175-S191 Bradford, M . M . 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal.Biochem., 72:248-254. Breton, S., P.J.S. Smith, B . L u i , and D . Brown 1996 Acidification of the male reproductive tract by a proton pumping ( H + )-ATPase. Nature Medicine, 2:470-472. Brosius III, F .C . , S.L. Alper, A . M . Garcia, and H.F . Lodish 1989 The major kidney band 3 gene transcript predicts an amino-terminal truncated band 3 polypeptide. J.Biol.Chem., 264:7784-7787. Brosius III, F .C . , R . L . Pisoni, X . Cao, G . Deshmukh, D . Yannoukakos, A . Stuart-Tilley, C. Haller, and S.L. Alper 1997 A E anion exchanger m R N A and protein expression n vascular smooth muscle cells, aorta, and renal microvessels. AmJ .Phys io l . , 273:F1039-F1047 Brown, D . , S. Gluck, and HartwigJ 1987a Structure of the novel membrane-coating material in proton-secreting epithelial cells and identifcation as an Ff +-ATPase. J .Cell B i o l . , 705:1637-1648. Brown, D . , P. Weyer, and L . Orci 1987b Nonclathrin-coated vesicles are involved in endocytosis in kidney collecting duct intercalated cells. Anat .Rec, 275:237-242. Brown, D . , S. Hirsch, and S. Gluck 1988 Localization of a proton-pumping ATPase in rat kidney. J.Clin.Invest, 52:2114-2126. Brown, D . , E.J . Sorscher, D . A . Ausiello, and D.J . Benos 1989 Immunocytochemical localization of N a + channels in rat kidney medulla. AmJ.Phys io l . , 256:F366-F369 189 Brown, D . , I. Sabolic, and S. Gluck 1992 Polarized targeting of V-ATPase in kidney epithelial cells. J.exp.Biol., 772:231-243. Brown, D . , J. Lydon, M . McLaughl in , A . Stuart-Tilley, R. Tyszkowski, and S.L. Alper 1996 Antigen retrieval in cryostat tissue sections and cultured cells treatment with sodium dodecyl sulfate (SDS). Histochem.Cell B i o l . , 705:261-267. Burgess, D .W. , W.S . Marshall, and C M . Wood 1998 Ionic transport by the opercular epithelium of freshwater acclimated tilapia (Oreochromis niloticus) and kil l i f ish (Fundulus heteroclitus). Comp.Biochem.Physiol., 72/^4:155-164. Cabantchik, Z.I . and A . Rothstein 1972 The nature of the membrane sites controlling anion permeability of human red blood cells as determined by studies with disulfonic stilbene derivatives. J.Membrane B i o l . , 70:311-330. Cameron, B . A . , S.F. Perry, C. Wu , K . K o , and B . L . Tufts 1996 Bicarbonate permeability and immunological evidence for an anion exchanger-like protein in the red blood cells of the sea lamprey, Petromyzon marinus. J.Comp.Physiol., 766:197-204. Cameron, J .N. a n d N . Heisler 1983 Studies of ammonia in the rainbow trout: Physico-chemcal parameters, acid-base behaviour and respiratory clearance. J.exp.Biol., 705:107-125. Cameron, J .N. 1986 Responses to reversed NH3 and NH4"1" gradients in a teleost (Ictalurus punctatus), and elasmobranch (Raja erinacea) and a crustacean (Callinectes sapidus): evidence for N H 4 + / H + exchange in the teleost and elasmobranch. J.exp.Zool., 239:183 Canessa, C M . , L . Schild, G . Buel l , B . Thorens, I. Gautschi, J.-D. Horisberger, and B . C . Rossier 1994 Amiloride-sensitive epithelial N a + channel is made of three homologous subunits. Nature, 367:463-467. Chambrey, R., J . - M . Achard, P . L . St.John, D.R. Abrahamson, and D . G . Warnock 1997 Evidence for an amiloride-insensitive N a + / H + exchanger in rat renal cortical tubules. Am.J.Physiol . , 273:C1064-C 1074 Chew, S.F. and Y . K . Ip 1987 Ammoniagenesis in mudkippers Boleophthalmus boddaerti and Periophthamodon schlosseri. Comp.Biochem.Physiol., 575:941-948. Chretien, M . and M . Pisam 1986 Ce l l renewal and differentiation in the gi l l epithelium of fresh-or salt-water-adapted euryhaline fish as revealed by [ 3H]-thymidine radioautography. B i o l . C e l l , 56:137-150. Claiborne, J .B. , D . H . Evans, and L . Goldstein 1982 Fish branchial N a + / N H 4 + exchange is via basolateral N a + - K + activated ATPase. J.exp.Biol., 96 :431-434. Claiborne, J .B. and D . H . Evans 1988 Ammonia and acid-base balance during high ammonia exposure in a marine teleost (Myoxocephalus octodecimspinosus). J.exp.Biol., 740:89-105. 190 Claiborne, J .B. , E . Perry, S. Bellows, and J. Campbell 1997 Mechanisms of acid-base excretion across the gills of a marine fish. J.exp.Zool., 279:509-520. Claiborne, J .B. 1998 Acid-base regulation. In: The physiology of fishes. D . H . Evans, ed. C R C Press, Boca Raton, pp. 177-198. Claiborne, J .B. , C .R . Blackston, K . P . Choe, D . C . Dawson, S.P. Harris, L . A . MacKenzie , and A. I . Morrison-Shetlar 1999 A mechanism for branchial acid excretion in marine fish: Identification of multiple N a + / H + antiporter (NHE) isoforms in gills of two seawater teleosts. J.exp.Biol., 202:315-324. Clarke, A . P . and W . T . W . Potts 1998a Isolated filament potentials and branchial ion fluxes in the European flounder (Platichthys flesus L . ) . Evidence for proton pump mediated sodium uptake. J.Zool. , 246:433-442. Clarke, A . P . and W . T . W . Potts 1998b Sodium, net acid, and ammonia fluxes in freshwater-adapted European flounder (Platichthys flesus L . ) . Pharacological inhibition and effects on g i l l ventilation volume. J.Zool., 246:427-432. Clarke, L . L . and M . C . Harline 1998 Dual role of C F T R in cAMP-stimulated H C 0 3 " secretion across murine duodenum. AmJ.Phys io l . , 274:G718-G726 Clayton, D . A . 1993 Mudskippers. Oceanogr.Mar.Biol.Annu.Rev., 57:507-577. Cohen, L . H . , A . Mueller, and P.R. Steinmetz 1978 Inhibition of bicarbonate exit step in urinary acidifcation by a disulfonic stilbene. J.Clin.Invest, 67:981-986. Cohn, J .A. 1994 C F T R localization: Implications for Cel l and Tissue Pathphysiology. In: Cystic fibrosis: current topics. J .A. Dodge, D.J .H. Brock and J .H. Widdicombe, eds. Wiley, Chichester, U K , pp. 173-191. Conley, D . M . and J. Mallatt 1987 Histochamical localization of N a + - K + A T P a s e and carbonic anhydrase activity in gills of 17 fish species. Can.J.Zool., 66:2398-2405. Curran, R . C . and J. Gregory 1977 The unmasking of antigens in paraffin sections of tissue by trypsin. Experientia, 55:1400-1401. Cutler, C P . , I .L. Sanders, N . Hazon, and G . Cramb 1995a Primary sequence, tissue specificity and m R N A expression of the N a + , K + - A T P a s e (3i subunit in the European eel (Anguilla anguilla). Fish Physiol.Biochem., 74:423-429. Cutler, C P . , I .L. Sanders, N . Hazon, and G . Cramb 1995b Primary sequence, tissue specificity and m R N A expression of the N a + , K + - A T P a s e a i subunit in the European eel (Anguilla anguilla). Comp.Biochem.Physiol., 7775:567-573. Cutler, C P . , I .L. Sanders, and G . Cramb 1997 Expression of N a + , K + - A T P a s e beta suubunit isoforms in the European eel (Anguilla anguilla). Fish Physiol.Biochem., 77:371-376. 191 D'Souza, S., A . Garcia-Cabado, F. Y u , K . Teter, G . Lukacs, K . Skorecki, H.-P. Moore, J. Orlowski, and S. Grinstein 1998 The epithelial sodium-hydrogen antiporter N a + / H + exchanger 3 accumulates and is functional in recycling endosomes. J.Biol.Chem., 273:2035-2043. Dang, Z . - C . , R . A . Lock, G . Fl ik , and S.E. Wendelaar Bonga 1999 Responses of N a + / K + - A T P a s e , metallothioneins and Cortisol receptor in the gi l l epithelial cells of copper exposed fish. Comp.Biochem.Physiol., 124A :S 11 (Abstract) de Renzis, G . and J. Maetz 1973 Studies on the mechanism of chloride absorption by the goldfish gi l l . Relation with acid-base regulation. J.exp.Biol., 59:339-358. Degnan, K . J . , K . J . Karnaky, and J .A. Zadunaisky 1977 Active chloride transport in the in vitro opercular skin o f a teleost (Fundulus heteroclitus), a gill-l ike epithelium rich in chloride cells. J.Physiol., 277:155-191. Degnan, K . J . and J .A. Zadunaisky 1980 Passive sodium movements across the opercular epithelium: The paracellular shunt pathway and ionic conductance. J.Membrane B io l . , 55:175-185. Doyle, W . L . and D . Gorecki 1961 The so-called chloride cell o f the fish gi l l . Physiol.Zool., 34:81-85. Drenckhahn, D . , M . Oelmann, P. Schaaf, M . Wagner, and S. Wagner 1987 Band 3 is the basolateral anion exchanger of dark epithelial cells of turtle urinary bladder. Am.J.Physiol . , 252:C570-C574 Drose, S. and K . Altendorf 1997 Bafilomycins and concanamycins as inhibitors of V-ATPases and P-ATPase. J.exp.Biol., 200 : l -8 . Dschida, W.J .A . and B . J . Bowman 1995 The vacuolar ATPase: sulfite stabilization and the mechanism of nitrate inactivation. J.Biol.Chem., 270:1557-1563. Dubinsky, W.P . and L . B . Mont i 1986 Resolution of apical from basolateral membrane of shark rectal gland. Am.J.Physiol. , 257 :C721-C726 Dunel-Erb, S. and P. Laurent 1980 Ultrastructure of the marine teleost g i l l epithelium: S E M and T E M study of the chloride cell apical membrane. J.Morph., 7(55:175-186. Duranton, C , M . Tauc, M . Avel la , and P. Poujeol 1997 Chloride channels in primary cultures of seawater fish (Dicentrarchus labrax). Am.J.Physiol. , C874-C882 Ehrenfeld, J., I. Laccoste, F . Garcia-Romeu, and B . Harvey 1990 Interdependence o f N a + and H + transport in frog skin. In: Animal Nutrition and Transport Processes. 2. Transport, Respiration and Excretion: Comparative and Environmental Aspects. J.P. Truchot and B . Lahlou, eds. Comparative Physiology, Basel,Karger, pp. 152-170. 192 Ehrenfeld, J. and U . K le in 1996 A primary cation transport by a V-type ATPase of low specificity. J.exp.Biol., 799:1327-1334. Epstein, F . H . , P. Silva, and G . Kormanik 1980 Role of Na-K-ATPase in chloride cell function. AmJ .Phys io l . , 255:R246-R250 Eriksson, O., N . Mayer-Gostan, and P J . Wistrand 1985 The use of isolated fish opercular epithelium as a model tissue for studying intrinsic activities of loop diuretics. Acta Physiol.Scand., 725:55-66. Evans, D . H . 1979 Fish. In: Comparative physiology of osmoregulation in animals. G . M . O . Maloiy , ed. Academic Press, Orlando, pp. 305-390. Evans, D . H . and J .N. Cameron 1986 G i l l ammonia transport. J.exp.Zool., 259:17-23. Evans, D . H . , K . J . More, and S.L. Robbins 1989 Modes of ammonia transport across the g i l l epithelium of the marine teleost fish Opsanus beta. J.exp.Biol., 744:339-356. Evans, D . H . 1993 Osmotic and ionic regulation. In: The Physiology of Fishes. D . H . Evans, ed. C R C Press, Inc, Boca Raton, pp. 315-341. Evans, D . H . , P . M . Piermarini, and W . T . W . Potts 1999 Ionic transport in the fish gi l l epithelium. J.exp.Zool., 255:641-652. Feng, Y . and M . Forgac 1994 Inhibition o f vacuolar H + - A T P a s e by disulfide bond formation between cystein 254 and cysteine 532 in subunit A . J.Biol.Chem., 269:13224-13230. Fenwick, J . C , S.E. Wendelaar Bonga, and G . Fl ik (1999) In vivo bafilomycin-sensitive N a + uptake in young freshwater fish. J. exp. B i o l . , 202: 3659-3666. Finbow, M . and M . A . Harrison 1997 The vaculoar H + -ATPase : a universal proton pump of eukaryotes. Biochem J . , 524:697-712. Fl ik , G . and P . M . Verbost 1994 Ca transport across plasma membranes. In: Biochemistry and molecular biology of fishes. P .W. Hochachka and T.P. Mommsen, eds. Elsevier Science, Amsterdam, pp. 625-637. Fl ik , G . , P . M . Verbost, and B . Wendelaar 1995 Calcium transport processes in fishes. In: Fish Physiology. C M . Wood and T .W. Shuttleworth, eds. Academic Press, Boca Raton, pp. 317-342. Forgac, M . 1989 Structure and function of vacuolar class of ATP-dr iven proton pumps. Physiol.Rev., 69:765-796. Forgac, M . 1999 The vacuolar H + -ATPase of clathrin-coated vesicles is reversibly inhibited by S-nitrosoglutathione. J.Biol.Chem., 274:1301-1305. 193 Foskett, J .K. and C . Scheffey 1982 The chloride cell: definitive identification as the salt-secretory cell in teleosts. Science, 275:164-166. Fuller, C M . , M . S . Awayda, M . P . Arrate, A . L . Bradford, R . G . Morris, C M . Canessa, B . C . Rossier, and D.J . Benos 1995 Cloning of a bovine renal epithelial N a + channel subunit. AmJ .Phys io l . , 269:C641-C654 Garty, H . and L . G . Palmer 1997 Epithelial sodium channels: function, structure, and regulation. Ph.ysiol.Rev., 77:359-396. Garvin, J .L. , M . B . Burg, and M . A . Knepper 1985 Ammonium replaces potassium in supporting sodium transport by the Na-K-ATPase of renal proximal straight tubules. AmJ.Phys io l . , 249:F785-F788 George, A . A . , R . W . Lehrich, S.G. Al ler , and J .N. Forrest 1998 Partial cloning and apical membrane localization of an anion exchanger (AE-2) in shark (Squalus acanthias) rectal gland tubules: preliminary studies. Bull.Mt.Desert Isl .Biol.Lab., 37:32 Gilmour, K . M . , P. Prunet, M . Pisam, D . G . McDonald, and C M . Wood 1998 Permeability and morphology of a cultured branchial epithelium form the rainbow trout during prolonged apical exposure to fresh water. J.exp.Zool., 281:531-545. Gluck, S. 1992 V-ATPases of the plasma membrane. J.exp.Biol., 772:29-37. Gluck, S. and R. Nelson 1992 The role of the V-ATPase in renal epithelial H +transport. J.exp.Biol., 772:205-218. Gluck, S.L., R . D . Nelson, B .S . Lee, Z .Q. Wang, X . L . Guo, J .Y . Fu, and K . Zhang 1992 Biochemistry of the renal V-ATPase . J.exp.Biol., 772:219-229. Goldstein, L . , J .B. Claiborne, and D . E . Evans 1982 Ammonia excretion by the gills of two marine teleost fish: The importance of N F L * permeance. J.exp.Zool., 279:395-397. Goldstein, L . , J .B. Claiborne, and D . E . Evans 1982 Ammonia excretion by the gills of two marine teleost fish: The importance of N F L * permeance. J.exp.Zool., 279:395-397. Goss, G . , S. Perry, and P. Laurent 1995 Ultrastructural and morphometric studies on ion and acid-base transport processes in freshwater fish. In: Cellular and Molecular Approaches to Fish Ionic Regulation. AnonymousAcademic Press, San Diego, pp. 257-284. Goss, G . G . and C M . Wood 1990 N a + and C l " uptake kinetics, diffusive effluxes and scidic equivalent fluxes across the gills of rainbow trout I. Response to environmental hyperoxia. J.exp.Biol., 752:521-547. Goss, G . G . , P. Laurent, and S.F. Perry 1992 Evidence for a morphological component in acid-base regulation during environmental hypercapnia in the brown bullhead (Ictalurus nebulosus). Ce l l Tissue Res., 265:539-552. 194 Goss, G . G . , P. Laurent, and S.F. Perry 1994 G i l l morphology during hypercapnia in brown bullhead (Ictalurus nebulosus): role of chloride cells and pavement cells in acid-base regulation. J .F i shBio l . , 45:705-718. Goss, G . G . , S.F. Perry, J .N. Fryer, and P. Laurent 1998 G i l l morphology and acid-base regulation in freshwater fishes. Comp.Biochem.Physiol., 779/4:107-115. Graham, J .B. 1976 Respiratory adaptations of marine air-breathing fishes. In: Respiration of amphibious vertebrates. G . M . Hughes, ed. Academic Press, London, pp. 165-187. Graham, J .B. 1997 Air-breathing fishes. Academic Press, San Diego, pp. -299 Greenaway, P. and T. Nakamura 1991 Nitrogenous excretion in two terrestrial crabs (Gecarcoidea natalis and Geograpsus grayi). Physiol.Zool., 64:767-786. Gregory, R . B . 1977 Synthesis and total ecretion of waste nitrogen by fish of the Periophthalmusi (mudskipper) and Scartelaos families. Comp.Biochem.Physiol., 574:33-36. Handy, R .D . 1989 The ionic composition of rainbow trout body mucus. Comp.Biochem.Physiol., 93,4:571-575. Harvey, B . J . 1992 Energization of sodium absorption by the H + - A T P A s e pump in mitochondria-rich cells of frog skin. J.exp.Biol., 772:289-309. Harvey, W.R. 1992 Physiology of V-ATPases. J.exp.Biol., 772:1-17. Harvey, W.R. and H . Weiczorek 1997 Animal plasma membrane energization by chemiosmotic I f V-ATPases . J.exp.Biol., 200:203-216. He, X . , C M . Tse, M . Donowitz, S.L. Alper, S.E. Gabriel, and B . J . Baum 1997 Polarized distribution of key membrane transport proteins in the rat submandibular gland. Pflugers Arch. , 455:260-268. Heisler, N . 1993 Acid-Base Regulation. In: The Physiology of Fishes. D . H . Evans, ed. C R C Press Inc, Boca Raton, pp. 343-378. Hemken, P., X . L . Guo, Z .Q . Wang, K . Zhang, and S. Gluck 1992 Immunological evidence that vaculoar H * ATPases with heterogeneous forms of M r = 31,000 subunit have different membrane distributions in mammalian kidney. J.Biol.Chem., 267:9948-9957. Hirsch, S., A . Strauss, K . Masood, S. Lee, V . Sukhatme, and S. Gluck 1988 Isolation and sequence o f a c D N A alone encoding the 31-kDa subunit o f bovine kidney vacuolar H + -ATPase. Proc.Natl .Acad.Sci .USA, 55:3004-3008. Hong, L . N . A comparative study of the effects of environmental p H on two mudskippers. 1-94. 1997. School o f Biological Sciences, National Univerisity of Singapore. ( G E N E R I C ) Ref Type: Thesis/Dissertation 195 Hoogerwerf, W . A . , S.C. Tsao, O. Devuyst, S.A. Levine, C H . Yun , J.W. Y i p , M . E . Cohen, P .D. Wilson, A . J . Lazenby, C M . Tse, and M . Donowitz 1996 N H E 2 and N H E 3 are human and rabbit intestinal brush-border proteins. AmJ.Phys io l . , 270.G29-G41 Hootman, S.R. and C . W . Philpott 1979 Ultracytochemical localization of Na + ,K + -activated ATPase in chloride cells from the gills o f a euryhaline teleost. Anat .Rec , 793:99-130. Hootman, S.R. and C . W . Philpott 1980 Accessory cells in the teleost branchial epithelium. AmJ.Phys io l . , 235:R199-R206 Hughes, G . M . and E.R. Weibel 1972 Similarity of supporting tissue in fish gills and the mammalian reticuloendothelium. Journal of Ultrastrucure Research, 39:106-114. Hughes, G . M . and J.S.D. Munshi 1979 Fine structure of the gills of some Indian air-breathing fishes. J.Morph., 760:169-194. Hughes, G . M . 1984 General anatomy of the gills. In: Fish Physiology. W.S. Hoar and D.J . Randall, eds. Academic Press, New York, pp. 1-72. Hulbert, W . C , T .W. Moon , and P .W. Hochochka 1978 The erythrinid g i l l : correlations of structure, function, and metabolism. Can.J Physiol.Pharmacol., 56:814-819. Ip, Y . K . , C . Y . Lee, S.F. Chew, W.P . Low, and K . W . Peng 1993 Differences in the responses of two mudskippers to terrestrial exposure. Zool .Sci . , 70:512-519. Ishimatsu, A . , N . M . Aguilar, K . Ogawa, Y . Hishida, S. Oikawa, T. Kanda, and K . K . Huat 1999 Blood gas levels and cardiovascular functions during varying environmental conditions in a mudskipper fish Periophthalmodon schlosseri. J.exp.Biol., 202:1753-1762. Ismailov, I.L, B . K . Berdiev, A . L . Bradford, M . S . Awayda, C M . Fuller, and D.J . Benos 1996 Associated proteins and renal epithelial N a + channel function. J.Membrane B io l . , 749:123-132. Iwata, K . , I. Kakuta, M . Ikeda, S. Kimoto, and N . Wada 1981 Nitrogn metabolism in the mudskipper, Periophthalmus cantonensis : a role of free amino acids in detoxification of ammonia produced during its terrestrial life. Comp.Biochem.Physiol., 65,4:589-596. Iwata, K . 1988 Nitrogen metabolism in the mudskipper, Periophthalmus cantonesis: Changes in free amino acids and related compounds in various tissues under conditions of ammonia loading, with special reference to its high ammonia tolerance. Comp.Biochem.Physiol., 97^:499-508. Just, F. and B . Walz 1994 Immunocytochemical localization of N a + / K + - A T P a s e and V - H * -ATPase in the salivary glands of the cockroach, Periplaneta americana. Ce l l Tissue Res., 275:161-170. Justensen, N . P . B . , T. Dall-Larsen, and L . Klungsoyr 1993 Proton transport and chloride cells in the gi l l of rainbow trout (Oncorhynchus mykiss). Can.J.Fish.Aquat.Sci., 50:988-995. 196 Jiirss, K . and R. Bastrop 1995 The function of mitochondria-rich cells (chloride cells) in teleost gills. Rev.FishBiol .Fish . , 5:235-255. Karnaky, K . J . , L . B . Kinter, W . B . Kinter, and C E . Sterling 1976 Teleost chloride cell II: Autoradiographic localization o f g i l l Na ,K-ATPase in kil l if ish Fundulus heteroclitus adapted to low and high salinity environments. J .Cell B i o l . , 70:157-177. Karnaky, K . J . 1986 Structure and function of the chloride cell of Fundulus heteroclitus and other teleosts. Amer.Zool. , 2(5:209-224. Karnaky, K . J . 1998 Osmotic and ionic regulation. In: The physiology of fishes. D . H . Evans, ed. C R C Press, Boca Raton, pp. 157-176. Kawakami, K . , S. Noguchi, M . Noda, H . Takahashi, T. Ohta, M . Kawamura, H . Nojima, K . Nagano, T. Hirose, S. Inayama, H . Hayashida, T. Miyata, and S. Numa 1985 Primary structure of the a-subunit of Torpedo californica (Na + + K + ) ATPase deduced from c D N A sequence. Nature, 3/6:733-736. Kerstetter, T . H . and L . B . Kirschner 1972 Active chloride transport by the gills of rainbow trout (Salmo gairdneri). J.exp.Biol., 56:263-272. Kerstetter, T . H . and L . B . Kirschner 1974 HCO3"-dependent ATPase activity in the gills of rainbow trout (Salmo gairdneri). Comp.Biochem.Physiol., 48B:581 -589. Keys, A . 1931 Chloride and water secretion and absoption by the gills o f the eel. Z.vgl.Physiol., /5:364-388. Keys, A . and E . N . Willmer 1932 'Chloride secreting cells" in the gills o f fishes, with special reference to the common eel. J.Physiol., 76:368-378. Kiker i , D . , A . S U N , M . L . Zeidel, and S.C. Hebert 1989 Ce l l membranes impermeable to N H 3 . Nature, 339:478-480. K i m , J., W.J . Welch, J .K. Cannon, C C Tisher, and K . M . Madsen 1992 Immunocytochemical response of type A and type B intercalated cells to increased sodium chloride delivery. Am.J.Physiol . , 262:F288-F302 Kinne, R., E . Kinne-Saffran, H . Schutz, and B . Scholermann 1986 Ammonium transport in medullary thick ascending limb of rabbit kidney: Involvement of the N a + , K + , C f -cotransporter. J.Membrane B i o l . , 94:279-284. Kinsella, J .L . and P.S. Aronson 1981 Interaction of N H 4 + a n d L i + with the renal microvillus membrane N a + - H + exchanger. Am.J.Physiol. , 241 :C220-C226 Kisen, G . , C. Gallais, B . Auperin, H . Klungland, 0 . Sandra, P. Prunet, and O. Andersen 1994 Northern blot analysis of the N a + , K + - A T P a s e a-subunit in salmonids. Comp.Biochem.Physiol., / 07£ :225-259 . 197 Klanke, C . A . , Y . R . Su, D.F . Callen, Z . Wang, P. Meneton, N . Baird, R. Kandasamy, J. Orlowski, B . E . Otterud, M . Leppert, G . E . Shull, and A . G . Menon 1995 Molecular cloning and physical and genetic mapping of a novel human N a + / H + exchanger ( N H E 5 / S L C 9 A 5 ) to chromosome 16q22.1. Genomics, 25:615-622. Kle in , U . , M . Timme, W . Zeiske, and J. Ehrenfeld 1997 The H + pump in frog skin (Rana esculentd): Identification and localization of a V-ATPase . J.Membrane B i o l . , 757:117-126. Kleyman, T.R. and E.J . Cragoe Jr. 1988 Amiloride and its analogues as tools in the study of ion transport. J.Membrane B i o l . , 705:1-21. Kleyman, T.R. , P.R. Smith, and D.J . Benos 1994 Characterization and localization of epithelial N a + channels in toad urinary bladder. AmJ .Phys io l . , 266:Cl 105-C1111 Klungsoyr, L . 1987 Magnesium ion activated ATPase and proton transport in vesicles from the gi l l o f rainbow trout (Salmo gairdnerii). Comp.Biochem.Physiol., 88BA125-1134. Kohn, O.F., P.P. Mitchel l , and P.R. Steinmetz 1990 Characteristics of apical C1"HC0 3 " exchanger of bicarbonate-secreting cells in turtle bladder. AmJ .Phys io l . , 255:F9-F14 Kok, W . K . , C B . L i m , T J . Lam, and Y . K . Ip 1998 The mudskipper Periophthalmodon schlosseri respires more efficiently on land than in water and vice versa from Boleophthalmus boddaerti. J.exp.Zool., 250:86-90. Krogh, A . 1937 Osmotic regulation in fresh water fishes by active absorption of chloride ions. Z.vgl.Physiol. , 24:656-666. Krogh, A . 1939 Osmotic regulation in aquatic animals. Cambridge University Press, London, pp. 1 Kudrycki , K . E . and G . E . Shull 1989 Primary structure of the rat kidney band 3 anion exchange protein deduced from a c D N A . J.Biol.Chem., 264:8185-8192. Kurtz, I. and R.S . Balaban 1986 Ammonium as a substrate for N a + - K + - A T P a s e in rabbit proximal tubules. AmJ.Phys io l . , 250:F497-F502 Lacy, E .R . 1983 Histochemical and biochemical studies of carbonic anhydrase activity in the opercular epithelium of the eurhaline teleost, Fundulus heteroclitus. Am.J.Anat., 166:19-39. Laemmli, U . K . 1970 Cleavage of structural proteins during the assembly of the head bacteriophage T4. Nature, 227:680-685. Larsen, E . H . 1991 Chloride transport by high-resistance heterocellular epithelia. Physiol.Rev., 77:235-283. 198 Larsen, E . H . , N . J . Willumsen, and B . C . Christoffersen 1992 Role of proton pump of mitochondria-rich cells for active transport of chloride ions in toad skin epithelium. J.Physiol., 450:203-216. Laurent, P. 1984 G i l l internal morphology. In: Fish Physiology. W . Hoar and D.J . Randall, eds. Academic Press, N e w York, pp. 73-183. Laurent, P., H . Hobe, and S. Dunel-Erb 1985 The role of environmental sodium chloride relative to calcium in gi l l morphology i f freshwater salmonid fish. Cel l Tissue Res., 240:675-692. Laurent, P. and S.F. Perry 1990 Effects of Cortisol on gil l chloride cell morphology and ionic uptake in the freshwater trout, Salmo gairdneri. Cel l Tissue Res., 259:429-442. Laurent, P. and S.F. Perry 1991 Environmental effects on fish gi l l morphology. Physiol.Zool., 64:4-25. Laurent, P., G . G . Goss, and S.F. Perry 1994 Proton pumps in fish g i l l pavement cells? Arch.Int.Physiol.Biochim.Biophys., 102:11-19. Lee, M . C . , C M . Penland, J .H. Widdicombe, and J.J. Wine 1998 Evidence that Calu-3 human airway cells secret bicarbonate. Am.J.Physiol. , 274:L450-L453 Lee, T . H . , J .C. Tsai, M . J . Fang, M . J . Y u , and P.P. Hwang 1998 Isoform expression of N a + - K + -ATPase a-subunit in gills of the teleost Oreochromis mossambicus. Am.J.Physiol. , 275:R926-R932 Lehrich, R . W . , C . Marino, H . Henschel, C. Kelmenson, M . Ratner, and J .N. Forrest 1995 Immunolocalization of D F T R with the rodent C F T R antibody R3195 in the rectal gland of the dogfish shark, Squalus acanthias. Bull.Mt.Desert Isl .Biol.Lab., 34:28-31. Leino, R . L . and J .H. McCormick 1984 Morphological and morphometrical changes in chloride cells of the gills of Pimephales promelas after chronic exposure to acid water. Cel l Tissue Res., 236:121-128. Levine, S.A., M . H . Montrose, C M . Tse, and M . Donowitz 1993 Kinetics and regulation of three cloned mammalian N a + / H + exchangers stably expressed in a fibroblast cell line. J.Biol.Chem., 265:25527-25535. L i , J., J. Eygensteyn, R . A . Lock, S.E. Wendelaar Bonga, and G . F l i k 1997 N a + and C a 2 + homeostatic mechanisms in isolated chloride cells of the teleost Oreochromis mossambicus analyzed by confocal laser scanning microscopy. J.exp.Biol., 200:1499-1508. Lignot, J .H. , M . Charmantier-Daures, and G . Charmantier 1999 Immunolocalization of N a + , K + -ATPase in the orgnas o f the branchial cavity of the European lobster Homarus gammarus (Crustacea, Decapoda). Ce l l Tissue Res., 296:417-426.(Abstracf) 199 L i n , H . and D.J . Randall 1990 The effect of varying water p H on the acidification of expired water in rainbow trout. J.exp.Biol., 745:149-160. L i n , H . and D . Randall 1991 Evidence for the presence of an electrogenic proton pump on the trout g i l l epithelium. J.exp.Biol., 161:119-134. L i n , H . and D.J . Randall 1993 H + -ATPase activity in crude homogenates of fish gi l l tissue: inhibitor sensitivity and environmental and hormonal regulation. J.exp.Biol., 750:163-174. L i n , FL, D . C . Pfeiffer, A . W . V o g l , J. Pan, and D.J . Randall 1994 Immunolocalization of proton-ATPase in the g i l l epithelia of rainbow trout. J.exp.Biol., 795:169-183. L i n , H . and D . Randall 1995 Proton Pumps in Fish Gi l l s . In: Cellular and Molecular Approaches to Fish Ionic Regulation. AnonymousAcademic Press, San Diego, pp. 229-255. Low, W.P. , D.J .W. Lane, and Y . K . Ip 1988 A comparative study of terrestial adaptations of the gills in three mudskippers- Periophthalmus chrysospilos, Boleophthalmus boddaerti, and Periophthalmodon schlosseri. B io l .Bu l l . , 775:434-438. Low, W.P. , Y . K . Ip, and D.J .W. Lane 1990 A comparative study of th gi l l morphometry in the mudskippers- Periophthalmus chrysospilos, Boleophthalmus boddaerti and Periophthalmodon schlosseri. Zool.Sci. , 7:29-38. Lowry, O .H. , N . J . Rosebrough, A . L . Fair, and R J . Randall 1951 Protein measurement with the Fol in phenol reagent. J.Biol.Chem., 793:265-275. Lonnerholm, G . and P.J. Wistrand 1984 Carbonic anhydrase in the human kidney: A histochemical and immunocytochemical study. Kidney Int., 25:886-898. Lytle, C , J.C. X u , D . Biemesderfer, M . Haas, and B . Forbush III 1992 The N a - K - C l cotransport protien o f shark rectal gland. I Development of monoclonal antibodies, immunoaffinity purification and partial biochemical characterization. J.Biol.Chem., 2(57:25438-25443. Lytle, C , J .C. X u , D . Biemesderfer, and B . Forbush III 1995 Distribution and diversity of N a - K -C l cotransporter proteins: a study with monoclonal antibodies. AmJ .Phys io l . , 269:C1496-C1505 Madsen, K . M . and C C . Tisher 1985 Structure-function relationships in H+-sectreting epithelia. Federation P roc , 44:2704-2709. Madsen, K . M . , J .W. Verlander, J. K i m , and C C Tisher 1991 Morphological adaptation of the collecting duct to acid-base disturbances. Kidney Int., 40 Suppl. 33:S57-S63 Maetz, J. and R. Garica 1964 The mechanism of sodium and chloride uptake by the gills of a fresh-water fish, Carassius auratus. II Evidence for N H / / N a + and HCO3" / C l" exchanges. J.Gen.Physiol., 47:1209-1227. 2 0 0 Mallery, C H . 1983 A carrier enzyme basis for ammnium excretion in teleost g i l l . N H 4 -stimulated Na-dependent ATPase activity in Opsanus beta. Comp.Biochem.Physiol., 74,4:889-897. Marshall, T. and K . M . Will iams 1984 Artifacts associated with 2-mercaptoethanol upon high resolution two-dimensional electrophoresis. Analytical Biochemistry, 739:502-505. Marshall, W.S. , S.E. Bryson, A . Midelfart, and W.F. Hamilton 1995 L o w conductance anion channel activated by c A M P in teleost C l " secreting cells. Am.J.Physiol . , 26S:R963-R969 Marshall, W.S. , S.E. Bryson, P. Darling, C. Whitten, M . Patrick, M . P . Wilkie , C M . Wood, and J. Buckland-Nicks 1997 N a C l transport and ultrastructure of opecular epithelium from a freshwater-adapted euryhaline teleost, Fundulus heteroclitus. J.exp.Zool., 277:23-37. Marshall, W.S . and S.E. Bryson 1998 Transport mechanisms o f seawater teleost chloride cells: A n inclusive model of a multifunctional cell. Comp.Biochem.Physiol., 7/9/1:97-106. Marshall, W.S. , T.R. Emberley, T .D . Singer, S.E. Bryson, and S.D. McCormick 1999 Time course of salinity adaptation in a strongly euryhaline estuarine teleost, Fundulus heteroclitus: a multivariable approach. J.exp.Biol., 202:1535-1544. Martinez-Anso, E . , J .E. Castillo, J. Diez, J.F. Medina, and J. Prieto 1994 Immunochemical detection of chloride/bicarbonate anion exchangers in human liver. Hepatology, 19:1400-1406. McCormick, S.D. 1990 Fluorescent labelling of N a + , K + - A T P a s e in intact cells by use of a fluorescent derivative of ouabain: Salinity and chloride cells. Cel l Tissue Res., 260:529-533. McCormick, S.D. 1993 Methods for nonlethal gi l l biopsy and measurement of N a + , K + - A T P a s e activity. Can.J.Fish.Aquat.Sci., 50:656-658. McCormick, S.D. 1995 Hormonal control of gi l l N a + , K + - A T P a s e and chloride cell function. In: Fish Physiology. C M . Wood and T.W. Shuttleworth, eds. Academic Press, Boca Raton, pp. 285-315. McDonald, D . G . and C M . Wood 1981 Branchial and renal acid and ion fluxes in the rainbow trout, Salmo gairdneri, at low environmental p H . J.exp.Biol., 93:101-118. McDonald, D . G . , Y . Tang, and R . G . Boutilier 1989 A c i d and ion transfer across the gills of fish: mechanisms and regulation. Can.J.Zool., 67:3046-3054. McDonald, D . G . , V . Cavdek, and R. El l i s 1991 G i l l design in freshwater fishes: Interrelationships among gas exchange, ion regulation, and acid-base regulation. Physiol.Zool. , 64:103-123. 201 McDonald, F.J. , P . M . Snyder, P .B . McCray Jr, and M . J . Welsh 1994 Cloning, expression, and tissue distribution of a human arniloride-sensitive N a + channel. AmJ.Phys io l . , 266:L728-L734 McDonald, F.J. , M . P . Price, P . M . Snyder, and M . J . Welsh 1995 Cloning and expression of the P~and y-subunits of the human epithelial sodium channel. A m J . P h y s i o l , 265:C1 163 McLean, I .W. and P . K . Nakane 1974 Periodate-lysine-paraformaldehyde fixative. A new fixative for immunoelectron mircoscopy. J.Histochem.Cytochem., 22:1077-1083. Medina, J.F., A . A c i n , and J. Prieto 1997 Molecular cloning and characterization of the human A E 2 anion exchanger (SLC4A2) gene. Genomics, 59:74-85. Morgan, L P . , W . T . W . Potts, and K . Oates 1994 Intracellular ion concetrations in branchial epithelial cells of brown trout (Salmo trutta L. ) determined by x-ray microanalysis. J.exp.Biol., 794:139-151. Morgan, I.P. and W . T . W . Potts 1995 The effects of thiocynate on the intracellular ion concentrations of branchial epithelial cells of brown trout. J.exp.Biol., 795:1229-1232. Morgan, J.D. and G . K . Iwama 1999 Energy cost of N a C l transport in isolated gills o f cutthroat trout. AmJ.Phys io l . , 277:R631-R639 M o r i i , H . , K . Nishikata, and O. Tamura 1978 Nitrogen excretion of mudskipper fish Periophthalmus cantonensis and Boleophthalmus pectinirostris in water and on land. Comp.Biochem.Physiol., 604:189-193. Moriyama, Y . and N . Nelson 1987 The purified ATPase from chromaffin granule membrane is an anion-dependent proton pump. J.Biol.Chem., 26"2:9175-9180. Mount, D . B . , E . Delpire, G . Gamba, A . E . Ha l l , E . Poch, R.S. Hoover Jr., and S.C. Hebert 1998 The electroneutral cation-chloride cotransporters. J.exp.Biol., 207:2091-2102. Nagami, G.T. 1988 Luminal secretion o f ammonia in the mouse proximal tubule perfused in vitro. J.Clin.Invest., 57:159-164. Nelson, I.D., W . T . W . Potts, and H . Huddart 1997 The use of amiloride to uncouple branchial sodium and proton fluxes in the brown trout, Salmo trutta. J.Comp.Physiol., 767:123-128. Nelson, N . and W.R. Harvey 1999 Vacuolar and plasma membrane proton-adenosinetriphosphatases. Physiol.Rev., 79:361-385. Newstead, J.D. 1967 Fine structure of the respiratory lamellae of teleostean gills. Z.Zellforsch, 79:396-428. 202 Nielsen, S., A . B . Maunsbach, C . A . Ecelbarger, and M . A . Knepper 1998 Ultrastructural localization of Na-K-2C1 cotransporter in thick ascending limb and macula densa of rat kidney. Am.J.Physiol. , 275:F885-F893 O'Grady, S . M . , H . C . Palfrey, and M . Field 1987 Characteristics and functions of N a - K - C l cotransport in epithelial tissues. Am.J.Physiol. , 255":C177-C 192 Oluwatosin, Y . E . and P . M . Kane 1997 Mutations in the CYS4 gene provided evidence for regulation of the yeast vacuolar H + -ATPase by oxidation and reduction in vivo. J.Biol.Chem., 272:28149-28157. Onken, H . , K . Graszynski, and W. Zeiske 1991 Na+-independent, electrogenic C l " uptake across the posterior gills of the Chinese crab (Eriocheir sinensis): Voltage-clamp and microelectrode studies. J.Comp.Physiol., 767:293-301. Onken, H . and M . Putzenlechner 1995 A V-ATPase drives active, electrogenic andNa + -independent C l " absorption across the gills o f Eriocheir sinensis. J.exp.Biol., 198:161-114. Paillard, M . 1998 H + and HCO3" transporters in the medullary thick ascending limb of the kidney: molecular mechanisms, function and regulation. Kidney Int., 53:S-36-S-41 Pan, J., D . C . Pfeiffer, C L . Reimer, B . J . Crawford, A . W . V o g l , and N . Auersperg 1993 Lack of inter-species reactivity etween antigens and antibodies is overcome by protease treatment of Western blots. Electrophoresis, 74:892-898. Payan, P. and A . J . Matty 1975 The characteristics of ammonia excretion by a perfused head of trout (Salmo gairdneri): Effect of temperature and C02-free ringer. J.Comp.Physiol., 96:167-184. Part, P. and C M . Wood 1996 N a / H exchange in cultured epithelial cells from fish gills. J.Comp.Physiol.B, 766:37-45. Pelster, B . and H . Niederstatter 1999 A c i d release in cultured gas gland cells. Comp.Biochem.Physiol., 724,4 :S37(Abstract) Peng, K . W . , S.F. Chew, C B . L i m , S.S.L. Kuah, W . K . K o k , and Y . K . Ip 1998 The mudskipper Periophthalmodon schlosseri and Boleophthalmus boddaerti can tolerate environmental NH3 concentrations of 446 and 36 u M , respectiviely. Fish Physiol.Biochem., 79:59-69. Peranen, J., M . Rikkonen, and L . Kaariainen 1993 A method for exposing hidden antigenic sites in paraformaldehyde-fixed cultured cells, applied to initially unreactive antibodies. J.Histochem.Cytochem., 47:447-454. Pereira, C M . , J . M . Wilson, and A . Oliva-Teles 1999 Utilization of dietary carbonhydrate and fat by juvenile turbot (Scophthalmus maximus): The effect of rearing temperature. Comp.Biochem.Physiol., 724A:S\40(Abstract) 203 Perry, S.F., M . S . Haswell, D.J . Randall, and A . P . Farrell 1981 Branchial ionic uptake and acid-base regulation in the rainbow trout, Salmo gairdneri. J.exp.Biol., 92:289-303. Perry, S.F. and D.J . Randall 1981 Effects of amiloride and SITS on branchial ion fluxes in rainbow trout, Salmo gairdneri. J.exp.Biol., 275:225-228. Perry, S.F., G . G . Goss, and P. Laurent 1992 The interrelationships between gi l l chloride cell morphology and ionic uptake in four freshwater teleosts. Can.J.Zool., 70:1775-1786. Perry, S.F. and J .N . Fryer 1997 Proton pumps in the fish gi l l and kidney. Fish Physiol.Biochem., 77:363-369. Perry, S.F. 1997 The Chloride Cel l : Structure and function in the gills of freshwater fishes. Ann.Rev.Physiol. , 59:325-347. Perry, S.F., M . L . Beyers, and D . A . Johnson 1999 Cloning and molecular characterization of the trout (Oncorhynchus mykiss) vacuolar H + -ATPase B subunit. J.exp.Biol., (In Press) Philpott, C . W . 1980 Tubular system membranes of teleost chloride cells: osmotic response and transport sites. AmJ.Phys io l . , 255:R171-R184 Piermarini, P . M . and D . H . Evans 1999 Expression of sodlium, potassium-ATPase in the gills and rectal gland o f a euryhaline elasmobranch (Dasyatis sabina). The Society for Experimental Biology Annual Meeting, Animal and Cell Biology: 3 0(Abstract) Piiper, J. and P. Scheid 1984 Model analysis of gas transfer in fish gills. In: Fish Physiology. W.S. Hoar and D.J . Randall, eds. Academic Press, New York, pp. 229-262. Pisam, M . , C . Sardet, and J. Maetz 1980 Polysaccharidic material in chloride cell of teleostean gi l l : modifications according to salinity. AmJ.Phys io l . , 235:R213-R218 Pisam, M . 1981 Membranous systems in the "chloride cell" of teleostean fish g i l l ; their modifications in response to the salinity of the environment. Anat .Rec , 200:401-414. Pisam, M . , A . Caroff, and A . Rambourg 1987 Two types o f chloride cells in the gil l epithelium of a freshwater-adapted euryhaline fish: Lebistes reticulatus; their modifications during adaptation to seawater. Am.J.Anat., 779:40-50. Pisam, M . , P. Prunet, G . Boeuf, and A . Rambourg 1988 Ultrastructural features o f chloride cells in the g i l l epithelium of the Atlantic salmon. Salmo salar, and their modifications during smoltification. Am.J.Anat., 755:235-244. Pisam, M . , P. Prunet, and A . Rambourg 1989 Accessory cells in the gi l l epithelium of freshwater rainbow trout Salmo gairdneri. Am.J.Anat., 184:311-320. Pisam, M . , G . Boeuf, P. Prunet, and A . Rambourg 1990 Ultrastructural features o f mitochondria-rich cells in stenohaline freshwater and seawater fishes. Am.J.Anat. , 757:21-31. 204 Pisam, M . and A . Rambourg 1991 Mitochondria-rich cells in the gil l epithelium of teleost fishes: A n ultrastructural approach. Inter.Rev.Cytol., 730:191-232. Potts, W . T . W . 1994 Kinetics of sodium uptake in freshwater animals: a comparison of ion-exchange and proton pump hypotheses. AmJ.Phys io l . , 2<5<5:R315-R320 Potts, W . T . W . 1984 Transepithelial potentials in fish gills. In: Fish Physiology. W.S. Hoar and D.J . Randall, eds. Academic Press, SanDiego, pp. 105-128. Poulsen, J .H. , H . Fischer, B . Illek, and T.E . Machen 1994 Bicarbonate conductance and p H regulation capability o f cystic fibrosis transmembrane conductance regulator. Proc.Natl .Acad.Sci .USA, 91:5340-5344. Powell, M . D . , D.J . Speare, and G . M . Wright 1994 Comparative ultrastructural morphology of lamellar epithelial, chloride, and mucous cell glycocalyx of the rainbow trout (Oncorhynchus mykiss). J.Fish B i o l . , 44:725-730. Pressley, T . A . 1992 Phylogenetic conservation of isoform-specific regions within a-subunit of Na,K-ATPase . AmJ .Phys io l . , 26~2:C743-C751 Puopolo, K . , C . Kumamoto, I. Adachi , R. Magner, and M . Forgac 1992 Differential expression of the " B " subunit of the vacuolar H + -ATPase in bovine tissues. J.Biol.Chem., 267 :3696-3706. Rahim, S . M . , J.P. Delaunoy, and P. Laurent 1988 Identification and immunocytochemical localization of two different carbonic anhydrase isoenzymes in teleostean fish erythrocytes and gi l l epithelia. Histochem., 59:451-459. Randall, D . , H . L i n , and P . A . Wright 1991 G i l l water flow and the chemistry of the boundary layer. Physiol.Zool., (54:26-38. Randall, D J . , A . P . Farrell, and T. Herning 1978 Carbon dioxide excretion in the pirarucu (Arapaima gigas), an obligate air-breathing fish. Can.J.Zool., 5(5:970-973. Randall, D.J . and P .A . Wright 1987 Ammonia distribution and excretion in fish. Fish Physiol.Biochem., 3:107-120. Randall, D.J . , C M . Wood, T.P. Mommsen, S.F. Perry, H . Bergman, G . M . O . Maloiy , and P . A . Wright 1989 Urea excretion as a strategy for survival in a fish l iving in a very alkaline environment. Nature, 337:165-166. Randall, D J . , J . M . Wilson, K . W . Peng, W . K . Kok, S.S.L. Kuah, S.F. Chew, T J . Lam, and Y . K . Ip 1999 The mudskipper, Periophthalmodon schloesseri, actively transports N H 4 + against a concentration gradient. AmJ.Phys io l . , (In Press) Reynolds, E . 1963 The use of lead citrate at high p H as an electron-opaque stain in electron microscopy. J .Cell B i o l . , 77:208-212. 205 Richards, B . D . and P.O. Fromm 1970 Sodium uptake by isolated-perfused gills of rainbow trout (Salmo gairdneri). Comp.Biochem.Physiol., 55:303-310. Riestenpatt, S., G . Petrausch, and D . Siebers 1995 C l " influx across posterior gills of the Chinese crab (Eriocheir sinensis): potential energization by a V-type H + -ATPase . Comp.Biochem.Physiol., 7704:235-241. Riordan, J.R., B . Forbush III, and J.W. Hanrahan 1994 The molecular basis of chloride transport in shark rectal gland. J.exp.Biol., 796:45-418. Sabolic, I., C M . Herak-Kramberger, S. Breton, and D . Brown 1999 Na/K-ATPase in intercalated cells along the rat nephron revealed by antigen retrieval. Journal of the American Society of Nephrology, 70:913-922.(Abstract) Saha, N . and B . K . Ratha 1990 Alterations in excretion pattern of ammonia and urea in freshwater air-breathing teleost, Heteropneustes fossilis (Boch) during hyperammonia stress. IndJ .Exp.Biol . , 25:597-599. Salama, A . , I.J. Morgan, and C M . Wood 1999 The linkage between N a + uptake and ammonia excretion in rainbow trout: Kinetic analysis, the effects of (NH4)2S04 and NH4HCO3 infusion and the influence of gi l l boundary layer p H . J.exp.Biol., 202:697-709. Sandbacka, M . , H . Li l ius , M . O . K . Enkvist, and B . Isomaa 1998 Rainbow trout gi l l epithelial cells in primary culture communicate through gap junctions as demonstrated by dye-coupling. Fish Physiol.Biochem., 79:287-292. Sardet, C , M . Pisam, and J. Maetz 1977 The surface epithelium of teleostean fish gills. Cellular and junctional adaptations of the chloride cell in relation to salt adaptation. J.Cell B io l . , 50:96-117. Scala, C , G . Cenacchi, C. Ferrari, G . Pasquinelli, P. Preda, and G . C Manara 1992 A new acrylic resin formulation: A useful tool for histological, ultrastructural, and immunocytochemical investigations. J.Histochem.Cytochem., 40:1799-1804. Scharschmidt, B .F . , E . B . Keefe, N . M . Blankenship, and R . K . Ockner 1979 Validation of a recording spectrophotometeric method for measurement of membrane associated M g -and NaK-ATPase activity. J .Lab.Clin.Med., 93:790-799. Schonrock, C , S.D. Morley, Y . Okawara, K . Lederis, and D. Richter 1991 Sodium and potassium ATPase of the teleost fish Catostomus commersoni- sequence, protein structure and evolutionary conservation of the alpha subunit. Biol.Chem.Ffoppe-Seyler., 572:279-286. Schottle, E . 1931 Morphologie und physiologie der atmung bei wasser-, schlamm-, und landlebenden Gobiiformes. Z.Wiss.Zool. , 740:1-114. 206 Schuster, V . L . , S . M . Bonsib, and M . L . Jennings 1986 Two types of collecting duct mitochondria-rich (intercalated) cells: lectin and band 3 cytochemistry. AmJ.Phys io l . , 251 :C347-C355 Schwiebert, E . M . , D.J . Benos, M . E . Egan, M . J . Stutts, and W . B . Guggino 1999 C F T R is a conductance regulator as well as a chloride channel. Physiol.Rev., 79:S145-S166 Scoazec, J . -Y. , A . - F . Bringuier, J.F. Medina, E . Martinez-Anzo, D . Veissiere, G. Feldmann, and C. Housset 1997 The plasma membrane polarity of human biliary epithelial cells: in situ immunohistochemical analysis and functional implications. Journal o f Hepatology, 26:543-553. Shephard, K . L . 1992 Studies of the fish-gill microclimate: interactions between gi l l tissue, mucus and water-quality. Env.Biol .Fish. , 34:409-420. Silva, P., R. Solomon, K'.. Spokes, and F . H . Epstein 1977 Ouabain inhibition o f g i l l N a - K -ATPase: Relationship to active chloride transport. J.exp.Zool., 799:419-426. Simpson, I .A. and O. Sonne 1982 A simple, rapid, and sensitive method for measuring protein concentration in subcellular membrane fractions prepared by sucrose density ultracentrifugation. Analytical Biochemistry, 779:424-427. Singer, T .D. , S.J. Tucker, W.S. Marshall, and C.F. Higgins 1998 A divergent C F T R homologue: highly regulated salt transport in the euryhaline teleost F. herteroclitus. AmJ .Phys io l . , 274:C715-C723 Skou, J.C. and M . Esmann 1992 The Na,K-ATPase . J.Bioenerg.Biomembr., 24:249-261. Smith, H . W . 1929 The excretion of ammonia and urea by the gills o f fish. J.Biol.Chem., 81:121-742. Smith, J.J. and M . J . Welsh 1992 c A M P stimulates bicarbonate secretion across normal, but not cystic fibrosis airway epithelia. J.Clin.Invest, 59:1148-1153. Smith, P.R. and D.J . Benos 1991 Epithelial N a + channels. Ann.Rev.Physiol. , 53:509-530. Smith, P.R., A . L . Bradford, V . Dantzer, D.J. Benos, and E . Skadhauge 1993 Immunocytochemical localization of arniloride-sensitive sodium channels in the lower intestine of the hen. Ce l l Tissue Res., 272:129-136. Sorscher, E.J . , M . A . Accavitti , D . Keeton, E . Steadman, R . A . Frizzell , and D.J . Benos 1988 Antibodies against purified epithelial sodium channel protein from bovine renal papilla. AmJ .Phys io l . , 255:C835-C843 Speeg, K . V . and J.W. Campbell 1968 Formation and volatilization o f ammonia gasby terrestial snails. AmJ .Phys io l . , 274:1392-1402. 207 Stetson, D . L . and P.R. Steinmetz 1985 a and P types of carbonic anhydrase-rich cells in turtle bladder. Am.J.Physiol . , 249:F553-F565 Stuart-Tilley, A . , C . Sardet, J. Pouyssegur, M . A . Schwartz, D . Brown, and S.L. Alper 1994 Immunolocalization o f anion exchanger A E 2 and cation exchanger NHE-1 in distinct adjacent cells of gastric mucosa. Am.J.Physiol. , 266:C559-C568 Sullivan, G . V . , J .N. Fryer, and S.F. Perry 1995 Immunolocalization of proton pumps ( H + -ATPase) in pavement cells of rainbow trout g i l l . J.exp.Biol., 795:2619-2629. Sullivan, G . V . , J .N. Fryer, and S.F. Perry 1996 Localization of m R N A for the proton pump ( H + -ATPase) and C17HC0 3 " exchanger in the rainbow trout g i l l . Can.J.Zool., 74:2095-2103. Sun, A . M . , Y . L i u , L . D . Dworkin, C M . Tse, M . Donowitz, and K . P . Y i p 1997 N a + / H + exchanger isoform 2 (NHE2) is expressed in the apical membrane of the medullary thick ascending limb. Journal of Membrane Biology, 7(50:85-90. Suzuki, Y . , M . Itakura, M . Kashwagi, N . Nakamura, T. Matsuki, H . Sakuta, N . Naito, K . Takano, T. Fujita, and S. Hirose 1999 Identification by differential display of a hypertonicity-inducible inward rectifier potassium channel highly expressed in chloride cells. J.Biol.Chem., 274:11376-11382. Sudhof, T . C , V . A . Fried, D . K . Stone, and P . A . Johnston 1989 Human endomembrane H + pump strongly resembles the ATP-synthetase of Archaebacteria. Proc.Natl .Acad.Sci .USA, 56:6067-6071. Tago, K . , V . L . Schuster, and J.B. Stokes 1986 Stimulation of chloride transport by H C O 3 C O 2 in rabbit cortical collecting tubule. Am.J.Physiol. , 257:F49-F56 Takeyasu, K . , M . M . Tamkun, K . J . Renaud, and D . M . Fambrough 1988 Ouabain-sensitive (Na + + K + ) -ATPase activity expressed in mouse L cells by transfection with D N A encoding the a-subunit of an avian sodium pump. J.Biol.Chem., 263:4347-4354. Tamura, S.O., H . M o r i i , and M . Yuzuriha 1976 Respiration of the amphibious fishes Periophthalmus cantonensis and Boleophthalmus chinensis in water and on land. J.exp.Biol., 65:97-107. Taylor, C.R. , S.-R. Shi, and R.J . Cote 1996 Antigen retrieval for immunohistochemistry. Status and need for greater standardization. Applied Immunohistochemistry, 4:144-166. Teixeira da Silva Jr., J . C , R . D . Perrone, C . A . Johns, and N . E . Madias 1991 Rat kidney band 3 m R N A modulation in chronic respiratory acidosis. Am.J.Physiol. , 260:F204-F209 Thurston, R . V . , R . C . Russo, and G.R. Phillips 1983 Acute toxicity of ammonia to fathead minnow. Trans.Am.Fish.Soc, 772:705-711. Tokuyasu, K . T . and S.J. Singer 1976 Improved procedures for immunoferritin labeling of ultrathin frozen sections. J.Cell B i o l . , 77:894-906. 208 Towbin, H . , T. Staehelin, and J. Gordon 1979 Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc.Natl .Acad.Sci .USA, 76:4350-4354. Towle, D . W . and T. Ffolleland 1987 Ammonium ion substitutes for K + in ATP-dependent N a + transport by basolateral membrane vesicles. AmJ.Phys io l . , 252:R479-R489 Tse, C . M . , S .A. Levine, C . H . Yun , S.R. Brant, J. Pouyssegur, M . H . Montrose, and M . Donowitz 1993 Functional characteristics of a cloned epithelial N a + / H + exchanger (NHE3): resistance to amiloride and inhibition by protein kinase C . Proc.Natl .Acad.Sci.USA, 90:9110-9114. Ura, K . , K . Soyano, N . Omoto, S. Adachi, and K . Yamauchi 1996 Localization of Na-K-ATPase in tissues of rabbit and teleosts using an antiserum directed against a partial sequence ofthe a-subunit. Zool .Sci . , 73:219-227. Van Der Heijden, A . J . H . , P . M . Verbost, J. Eygensteyn, J. L i , S.E. Wendelaar Bonga, and G . F l ik 1997 Mitochondria-rich cells in gills of tilapia (Oreochromis mossambicus) adapted to fresh water or sea water: Quantification by confocal laser scanning microscopy. J.exp.Biol., 200:55-64. van Hi l le , B . , H . Richener, D . B . Evans, J.R. Green, and G . Bilbe 1993 Identification of two subunit A isoforms of the vacuolar H + -ATPase in human osteoclastoma. J.Biol.Chem., 265:7075-7080. van Hi l le , B . , H . Richener, P. Schmid, I. Puettner, J.R. Green, and G . Bilbe 1994 Heterogeneity of vacuolar H + -ATPase : differential expression of wo human subunit B isoforms. Biochem.J., 303:191-198. Verdouw, H . , C . J .A . van Echeld, and E . M . J . Dekkers 1978 Ammonia determination based on indophenol formation with sodium salicylate. Water Res., 72:399-402. Verlander, J.W., K . M . Madsen, P.S. Low, D.P. Al len , and C C . Tisher 1988 Immunocytochemical localization of band 3 protein in the rat collecting duct. AmJ .Phys io l . , 255:F115-F125 Vonnegut Jr., K . (1969). Slaughter house five or; The children's crusade, a duty dance with death. (New York: Delacorte Press). Wagner, S., R. Vogel , R. Lietzke, R. Koob, and D . Drenckhahn 1987 Immunochemical characterization of a band 3-like anion exchanger in collecting duct of human kidney. AmJ .Phys io l . , 253:F213-F221 Wal l , S . M . and L . M . Koger 1994 N H 4 + transport mediated by N a + - K + - A T P a s e in rat inner medullary collecting duct. AmJ.Phys io l . , 267:F660-F670 Wal l , S . M . 1996 Ammonium transport and the role of the Na,K-ATPase . Mineral and Electrolyte Metabolism, 22:311-317. 209 Walsh, P.J. 1998 Nitrogen excretion and metabolism. In: Physiology of fishes. D . H . Evans, ed. C R C Press, Boca Raton, pp. 199-214. Wang, Z . -Q. and S. Gluck 1990 Isolation and properties of bovine kidney brush border vacuolar H + -ATPase . J.Biol.Chem., 2(55:21957-21965. Wendelaar Bonga, S.E. and C . J . M . van der Meij 1989 Degeneration and death, by apoptosis and necrosis, o f the pavement and chloride cells in the gills o f the teleost Oreochromis mossambicus. Ce l l Tissue Res., 255:235-243. Werner, G . and H . Hagenmaier 1984 Bafilomycins, a new group of macrolide antibiotics. Production, isolation, chemical structure and biological activity. Journal of Antibiotics, 37:110-117. Werner, M . , R. von Wasielewski, and P. Komminoth 1996 Antigen retrieval, signal amplification and intensification in immunohistochemistry. Histochem.Cell B io l . , 705:253-260. Wieser, W . 1972 Oxygen consumption and ammonia excretion in Ligia beaudiana M . - E . Comp.Biochem.Physiol., 43^:869-876. Wilson, J . M . , D.J . Randall, A . W . Vog l , and G . K . Iwama 1997 Immunolocalization of proton-ATPase in the gills of the elasmobranch, Squalus acanthias. J.exp.Zool., 275:78-86. Wilson, R . W . and E . W . Taylor 1992 Transbranchial ammonia gradients and acid-base responses to high external ammonia concentration in rainbow trout (Oncorhynchus mykiss) acclimated to different salinities. J.exp.Biol., 7(5(5:95-112. Wilson, R . W . , P . M . Wright, S. Munger, and C M . Wood 1994 Ammonia excretion in freshwater rainbow trout (Oncorhynchus mykiss) and the importance of gil l boundary layer acidification: lack of evidence for N a + / N H 4 + exchange. J.exp.Biol., 797:37-58. Wilson, R .W. , K . M . Gilmour, R.P. Henry, and C M . Wood 1996 Intestinal base excretion in the seawater-adapted rainbow trout: A role in acid-base regulation? J.exp.Biol., 799:2331-2343. Wittbrodt, J., A . Meyer, and M . Schartl 1998 More genes in fish? BioEssays, 20:511-515. Witters, H . , P. Berckmans, and C . Vangenechten 1996 Immunolocalization of N a + , K + - A T P a s e in the g i l l epithelium of rainbow trout, Oncorhynchus mykiss. Ce l l Tissue Res., 253:461-468. Wong, C . K . C . and D . K . O . Chan 1999 Isolation of viable cell types from the gi l l epithelium of Japanese eel Anguillajaponica . AmJ .Phys io l . , 27(5:R363-R372 Wood, C M . and G . G . Goss 1990 Kinetic analysis of relationships between ion exchange and acid-base regulation at the gills of freshwater fish. In: Animal Nutrition and Transport 210 Processes. 2. Transport, Respiration and Excretion: Comparative and Environmental Aspects. J.P. Truchot and B . Lahlou, eds. Comp Physiol., Basel,Karger, pp. 119-136. Wood, C M . 1991 Branchial ion and acid-base transfer in freshwater teleost fish: Environmental hyperoxia as a probe. Physiol.Zool. , 64:68-102. Wood, C M . , K . M . Gilmour, and P. Part 1998 Passive and active transport properties of a gi l l model, the cultured branchial epithelium of the freshwater rainbow trout (Oncorhynchus mykiss). Comp.Biochem.Physiol., 779/4:87-96. Wright, P. A . and C M . Wood 1985 A n analysis of branchial ammonia excretion in the freshwater rainbow trout: Effects of environmental p H change and sodium uptake blockage. J.exp.Biol., 774:329-353. Wright, P .A . , D.J . Randall, and S.F. Perry II 1989 Fish gi l l water boundary layer: a site of linkage between carbon dioxide and ammonia excretion. J.Comp.Physiol., 755:627-635. Wright, P .A . , P. Part, and C M . Wood 1995 Ammonia and urea excretion in the tidepool sculpin (Oligocottus maculosus):sites o f excretion, effects of reduced salinity and mechanisms of urea transport. Fish Physiol.Biochem., 74:111-123. Wright, S.H. 1991 The inferface of animal and aqueous environment: strategies and constraints on the maintenance of solute balance. In: Biochemistry and molecular biology of fishes. P.W. Hochachka and T.P. Mommsen, eds. Elsevier Science, pp. 165-180. X u , J . C , C . Lytle, T.T. Zhu, J .A. Payne, J. Benz, and I.I.I. Forbush 1994 Molecular cloning and functional expression of the bumetanide-sensitive N a - K - C l cotransporter. Proc.Natl .Acad.Sci .USA, 97:2201-2205. Yadav, A . N . , M . S . Prasad, and B . R . Singh 1990 Gross structure of the respiratory organs and dimensions of the gi l l in the mud-skipper, Periophthalmus schlosseri (Bleeker). J.Fish B i o l . , 37:383-392. Zadunaisky, J .A. , S. Cardona, L . A u , D . M . Roberts, E. Fisher, B . Lowenstein, E.J . Cragoe Jr., and K . R . Spring 1995 Chloride transport activation by plasma osmolarity during rapid adaptation to high salinity of Fundulus heteroclitus. J.Membrane B i o l . , 143:207-217. Zadunaisky, J .A. 1996 Chloride cells and osmoregulation. Kidney Int., 49:1563-1567. Zaugg, W.S. 1982 A simplified preparation for adenosine triphospate determination in gi l l tissue. Can.J.Fish.Aquat.Sci., 59 :215-217. Zeidel, M . L . 1996 L o w permeabilities of apical membranes of barrier epithelia: what makes watertight membranes watertight? Am.J.Physiol. , 27/:F243-F245 Ziegler, A . 1997 Immunocytochemical localization of N a + , K + - A T P a s e in the calcium-transporting seternal epithelium of the terrestial isopod Porcellio scaber L . (Crustacea). J. Histochem. Cytochem., 45:437-446. 211 CO CD r d -r-» u - d • d OJ T3 J 3 "o _g CD CO -4-» CO CD )-< CD r d -r-> CJ-H O -r-» o Cl td r d -r-» CD o - d CD -r-» CD en CD • r H "d O - O O +-« co • r H r j X Q W CM a a a o o IS o <u CM co o r d co s CD co CD r ^ o CO i f e CD +-> co CD ^1 O l - d co o co CD r ^ r H N O O N O O oo cu C J a s-CU O N O O <u |<+H o 1 3 C O I T ) O N N O O N O O O O Ci ICQ I on ex O N O N ca CD | H - H r H CD r d o co O G O CD co H 0 0 | o o O N ' r H r H CD -a CO CD CD I S a o >-.| CD a -a r H ! O N <3 •*-H <u r H r H CD *d co CD a 2H r-H^I C l o CD C l H 3 r * o IrC co f >*l CD d 13 o CD O N I "a r H r H CD T 3 co CD s CD i-H CQ co O N o r H <D I u oo oo r H h5 > CD > '8 0 0 Cl O N M O rC| CO o -a CO CD > bO , CD Cl co |oo r H CD co C l O N O O tu , u I* o ICQ r - l CU C J s-S |£ CQ loo CD PH W CO co O a CD CQ CO O Cl CD ICQ N -r-> o Cl \& o C l o Q N o C l o Q o o o o a d o o • rH a l-d O CO Ci H O o rJ o ' r H CO CD C l CD 0 0 o I'm '1: s o CD ctj a 0 0 I C O o CD co 3 O a rO •8 r O r H '53 r H r D •8 r O CD co Q r O •s r H ,rO r H a W > OX) ••c I l3l l ^ 1 "I1 |5; co I ccj 1^1 CD ^ H C H CD OH lw 'I CO CD C l •> o rO r D d CO cd § CD CCJ CD CO cd OH H < > CD co cd H < I f > U cd rO a d rCJ . C H CD C H u cd s co. rC| C ^ C H c3 C l <u r H CD c > o r D o 00 00 I CD -r-> o C l o CD +-» o &i c l o «3 CD o r H C H d O CO <2 co - d 9 rO "o l l CD O H CD co d o £ 1 3 +-> C H CD CH g 'cd o I'd o • i-H a CO _cd "E, o -f-» . CD r H d rCj cl CD a 0 0 cd g cd a o T 3 '•8 r H CD CO d o W < CD • r H C H CD C H < u "cd • 1—( rCj o CD CD ll CD CD g > O l - d N O < H CD C o >/o CD C l o H U u r^ —I *-l flj ' 1 £ Q Z " . U u cd s ICS W |co W CO |W CO W W CO r D * d o CN W < CD CO cd PH H .< "cd IO co r-rH CD "J-> < •a u r^ ^ r D 0 0 o co ' r H CD > c PH cd" "d ^ S o Cl o • i-H td a CD CO O W *rH O CO CO 'co CD d CD 0 0 o § « *d <N d > ; f e a " c r d a CD £•* ^ U H U £ 2 +J 0 'C O ft >. CD O CD .tJ .£ m S Q 2 d •+-> CD ^ < 2 '1 rD cd I O CfH O CO r H CD > ' d CO d o co r H CD > cd r H o 42 ^ .tJ w ^ >^  C O HJH CO O CD d -M O g cy3 -2 *o .a W .2 'to " cd CQ 5 = 1 r d r - J Q C O 212 . oo CO CO u o is CD 0 ) OH 3 O VH H cd O O U CO DO ca J 3 oo « o Q o E-> H^ U ^ Si CO oo g cd bO cd 00 CU -3 O I + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 4^ + + + + + + + + + + + + + co cd H < cd u cd U Os <N W o 00 m K ; 213 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0089799/manifest

Comment

Related Items