Ammonia excretion in rainbow trout (Oncorhynchus mykiss): evidence for Rh glycoprotein and H+-ATPase involvement

Excretory nitrogen metabolism and defence against ammonia toxicity in air-breathing fishes

With the development of air-breathing capabilities, some fishes can emerge from water, make excursions onto land or even burrow into mud during droughts. Air-breathing fishes have modified gill morphology and morphometry and accessory breathing organs, which would tend to reduce branchial ammonia excretion. As ammonia is toxic, air-breathing fishes, especially amphibious ones, are equipped with various strategies to ameliorate ammonia toxicity during emersion or ammonia exposure. These strategies can be categorized into (1) enhancement of ammonia excretion and reduction of ammonia entry, (2) conversion of ammonia to a less toxic product for accumulation and subsequent excretion, (3) reduction of ammonia production and avoidance of ammonia accumulation and (4) tolerance of ammonia at cellular and tissue levels. Active ammonia excretion, operating in conjunction with lowering of ambient pH and reduction in branchial and cutaneous NH₃ permeability, is theoretically the most effective strategy to maintain low internal ammonia concentrations. NH₃ volatilization involves the alkalization of certain epithelial surfaces and requires mechanisms to prevent NH₃ back flux. Urea synthesis is an energy-intensive process and hence uncommon among air-breathing teleosts. Aestivating African lungfishes detoxify ammonia to urea and the accumulated urea is excreted following arousal. Reduction in ammonia production is achieved in some air-breathing fishes through suppression of amino acid catabolism and proteolysis, or through partial amino acid catabolism leading to alanine formation. Others can slow down ammonia accumulation through increased glutamine synthesis in the liver and muscle. Yet, some others develop high tolerance of ammonia at cellular and tissue levels, including tissues in the brain. In summary, the responses of air-breathing fishes to ameliorate ammonia toxicity are many and varied, determined by the behaviour of the species and the nature of the environment in which it lives.

Rh proteins and NH4(+)-activated Na+-ATPase in the Magadi tilapia (Alcolapia grahami), a 100% ureotelic teleost fish

The small cichlid fish Alcolapia grahami lives in Lake Magadi, Kenya, one of the most extreme aquatic environments on Earth (pH ~10, carbonate alkalinity ~300 mequiv l(-1)). The Magadi tilapia is the only 100% ureotelic teleost; it normally excretes no ammonia. This is interpreted as an evolutionary adaptation to overcome the near impossibility of sustaining an NH3 diffusion gradient across the gills against the high external pH. In standard ammoniotelic teleosts, branchial ammonia excretion is facilitated by Rh glycoproteins, and cortisol plays a role in upregulating these carriers, together with other components of a transport metabolon, so as to actively excrete ammonia during high environmental ammonia (HEA) exposure. In Magadi tilapia, we show that at least three Rh proteins (Rhag, Rhbg and Rhcg2) are expressed at the mRNA level in various tissues, and are recognized in the gills by specific antibodies. During HEA exposure, plasma ammonia levels and urea excretion rates increase markedly, and mRNA expression for the branchial urea transporter mtUT is elevated. Plasma cortisol increases and branchial mRNAs for Rhbg, Rhcg2 and Na(+),K(+)-ATPase are all upregulated. Enzymatic activity of the latter is activated preferentially by NH4(+) (versus K(+)), suggesting it can function as an NH4(+)-transporter. Model calculations suggest that active ammonia excretion against the gradient may become possible through a combination of Rh protein and NH4(+)-activated Na(+)-ATPase function.

Rh vs pH: the role of Rhesus glycoproteins in renal ammonia excretion during metabolic acidosis in a freshwater teleost fish

Increased renal ammonia excretion in response to metabolic acidosis is thought to be a conserved response in vertebrates. We tested the hypothesis that Rhesus (Rh) glycoproteins in the kidney of the freshwater common carp Cyprinus carpio play a critical role in regulating renal ammonia excretion during chronic metabolic acidosis. Exposure to water pH 4.0 (72 h) resulted in a classic metabolic acidosis with reduced plasma pHa, [HCO3-], no change in PCO2, and large changes in renal function. Urine [NH4+] as well as [titratable acidity–HCO3-] rose significantly over the acid exposure, but the profound reduction (5-fold) in urine flow rates eliminated the expected elevations in renal ammonia excretion. Low urine flow rates may be a primary strategy to conserve ions, as urinary excretion of Na+, Cl- and Ca2+ were significantly lower during the acid exposure relative to the control period. Interestingly, renal Rhcg1 mRNA and protein levels were elevated in acid relative to control groups, along with mRNA levels of several ion transporters, including the Na+/H+ exchanger (NHE3), H+ATPase and Na+/K+ATPase (NKA). Immunofluorescence microscopy showed a strong apical Rhcg1 signal in distal tubules. Taken together, these data show that renal Rh glycoproteins and associated ion transporters are responsive to metabolic acidosis, but conservation of ions through reduced urine flow rates takes primacy over renal acid-base regulation in the freshwater C. carpio. We propose that an “acid/base-ion balance” compromise explains the variable renal responses to metabolic acidosis in freshwater teleosts.

Compensatory responses in common carp (Cyprinus carpio) under ammonia exposure: additional effects of feeding and exercise

Ammonia is an environmental pollutant that is toxic to all aquatic animals. The toxic effects of ammonia can be modulated by other physiological processes such as feeding and swimming. In this study, we wanted to examine these modulating effects in common carp (Cyprinus carpio). Fish were either fed (2% body weight) or starved (unfed for seven days prior to the sampling), and swimming at a sustainable, routine swimming speed or swum to exhaustion, while being exposed chronically (up to 28 days) to high environmental ammonia (HEA, 1 mg/L ~58.8 μmol/L as NH4Cl at pH 7.9). Swimming performance (critical swimming speed, Ucrit) and metabolic responses such as oxygen consumption rate (MO2), ammonia excretion rate (Jamm), ammonia quotient, liver and muscle energy budget (glycogen, lipid and protein), plasma ammonia and lactate, as well as plasma ion concentrations (Na(+), Cl(-), K(+) and Ca(2+)) were investigated in order to understand metabolic and iono-regulatory consequences of the experimental conditions. Cortisol plays an important role in stress and in both the regulation of energy and the ion homeostasis; therefore plasma cortisol was measured. Results show that during HEA, Jamm was elevated to a larger extent in fed fish and they were able to excrete much more efficiently than the starved fish. Consequently, the build-up of ammonia in plasma of HEA exposed fed fish was much slower. MO2 increased considerably in fed fish after exposure to HEA and was further intensified during exercise. During exposure to HEA, the level of cortisol in plasma augmented in both the feeding regimes, but the effect of HEA was more pronounced in starved fish. Energy stores dropped for both fed and the starved fish with the progression of the exposure period and further declined when swimming to exhaustion. Overall, fed fish were less affected by HEA than starved fish, and although exercise exacerbated the toxic effect in both feeding treatments, this was more pronounced in starved fish. This suggests that fish become more vulnerable to external ammonia during exercise, and feeding protects the fish against the adverse effects of high ammonia and exercise.

Ammonia transport across the skin of adult rainbow trout (Oncorhynchus mykiss) exposed to high environmental ammonia (HEA)

Recent molecular evidence points towards a capacity for ammonia transport across the skin of adult rainbow trout. A series of in vivo and in vitro experiments were conducted to understand the role of cutaneous ammonia excretion (J amm) under control conditions and after 12-h pre-exposure to high environmental ammonia (HEA; 2 mmol/l NH4HCO3). Divided chamber experiments with bladder-catheterized, rectally ligated fish under light anesthesia were performed to separate cutaneous J amm from branchial, renal, and intestinal J amm. Under control conditions, cutaneous J amm accounted for 4.5 % of total J amm in vivo. In fish pre-exposed to HEA, plasma total ammonia concentration increased 20-fold to approximately 1,000 μmol/l, branchial J amm increased 1.5- to 2.7-fold, and urinary J amm increased about 7-fold. Urinary J amm still accounted for less than 2 % of total J amm. Cutaneous J amm increased 4-fold yet amounted to only 5.7 % of total J amm in these fish. Genes (Rhcg1, Rhcg2, Rhbg, NHE-2, v-type H(+)-ATPase) known to be involved in ammonia excretion at the gills of trout were all expressed at the mRNA level in the skin, but their expression did not increase with HEA pre-exposure. In vitro analyses using [(14)C] methylamine (MA), an ammonia analog which is transported by Rh proteins, demonstrated that MA permeability in isolated skin sections was higher in HEA pre-exposed fish than in control fish. The addition of basolateral ammonia (1,000 μmol/l) to this system abolished this increase in permeability, suggesting ammonia competition with MA for Rh-mediated transport across the skin of HEA pre-exposed trout; this did not occur in skin sections from control trout. Moreover, in vitro J amm by the skin of fish which had been pre-exposed to HEA was also higher than in control fish in the absence of basolateral ammonia, pointing towards a possible cutaneous ammonia loading in response to HEA. In vitro MA permeability was reduced upon the addition of amiloride (10(-4) mol/l), but not phenamil (10(-5) mol/l) suggesting a role for a Na/H-exchanger (NHE) in cutaneous ammonia transport, as has been previously described in the skin of larval fish. Overall, it appears that under control conditions and in response to HEA pre-exposure, the skin makes only a very minor contribution to total J amm, but the observed increases in cutaneous J amm in vivo and in cutaneous J amm and MA permeability in vitro demonstrate the capacity for ammonia transport in the skin of adult trout. It remains unclear if this capacity may become significant under certain environmental challenges or if it is merely a remnant of cutaneous transport capacity from early life stages in these fish.

An in vitro analysis of intestinal ammonia handling in fasted and fed freshwater rainbow trout (Oncorhynchus mykiss)

Ammonia transport and metabolism were investigated in the intestinal tract of freshwater rainbow trout which had been either fasted for 7 days, or fasted then fed a satiating meal of commercial trout pellets. In vivo, total ammonia concentrations (T amm) in the chyme were approximately 1 mmol L(-1) across the entire intestine at 24 h after the meal. Highest chyme pH and P NH3 values occurred in the posterior intestine. In vitro gut sac experiments examined ammonia handling with mucosal (Jmamm) and serosal (Js amm) fluxes under conditions of fasting and feeding, with either background (control ≤ 0.013 mmol L(-1)) or high luminal ammonia concentrations (HLA = 1 mmol L(-1)), the latter mimicking those seen in chyme in vivo. Feeding status (fasted or fed) appeared to influence ammonia handling by each individual section. The anterior intestine exhibited the greatest Jm amm and Js amm values under fasted control conditions, but these differences tended to disappear under typical post-feeding conditions when total endogenous ammonia production (Jt amm = Js amm - Jm amm, signs considered) was greatly elevated in all intestinal sections. Under fasted conditions, glutamate dehydrogenase (GDH) and glutaminase (GLN) activities were equal across all sections, but the ammonia-trapping enzyme glutamine synthetase (GS) exhibited highest activity in the posterior intestine, in contradiction to previous literature. Feeding clearly stimulated the total rate of endogenous ammonia production (Jt amm), even in the absence of a high luminal ammonia load. This was accompanied by an increase in GDH activity of the anterior intestine, which was also the site of the largest Jt amm. In all sections, during HLA exposure, either alone or in combination with feeding, there were much larger increases in endogenous Jt amm, most of which was effluxed to the serosal solution. This is interpreted as a response to avoid potential cytotoxicity due to overburdened detoxification mechanisms in the face of elevated mucosal ammonia. Thus T amm of the intestinal tissue remained relatively constant regardless of feeding status and exposure to HLA. Ammonia production by the gut may explain up to 18 % of whole-body ammonia excretion in vivo under fasting conditions, and 47 % after feeding, of which more than half originates from endogenous production rather than from absorption from the lumen.

Modulation of Rh glycoproteins, ammonia excretion and Na+ fluxes in three freshwater teleosts when exposed chronically to high environmental ammonia

We investigated relationships among branchial unidirectional Na(+) fluxes, ammonia excretion, urea excretion, plasma ammonia, plasma cortisol, and gill transporter expression and function in three freshwater fish differing in their sensitivity to high environmental ammonia (HEA). The highly ammonia-sensitive salmonid Oncorhynchus mykiss (rainbow trout), the less ammonia-sensitive cyprinid Cyprinus carpio (common carp) and the highly ammonia-resistant cyprinid Carassius auratus (goldfish) were exposed chronically (12-168 h) to 1 mmol l(-1) ammonia (as NH4HCO3; pH 7.9). During HEA exposure, carp and goldfish elevated ammonia excretion (JAmm) and Na(+) influx rates ( ) while trout experienced higher plasma ammonia (TAmm) and were only able to restore control rates of JAmm and . All three species exhibited increases in Na(+) efflux rate ( ). At the molecular level, there was evidence for activation of a 'Na(+)/NH4(+) exchange metabolon' probably in response to elevated plasma cortisol and TAmm, though surprisingly, some compensatory responses preceded molecular responses in all three species. Expression of Rhbg, Rhcg (Rhcg-a and Rhcg-b), H(+)-ATPase (V-type, B-subunit) and Na(+)/K(+)-ATPase (NKA) mRNA was upregulated in goldfish, Rhcg-a and NKA in carp, and Rhcg2, NHE-2 (Na(+)/H(+) exchanger) and H(+)-ATPase in trout. Branchial H(+)-ATPase activity was elevated in goldfish and trout, and NKA activity in goldfish and carp, but NKA did not appear to function preferentially as a Na(+)/NH4(+)-ATPase in any species. Goldfish alone increased urea excretion rate during HEA, in concert with elevated urea transporter mRNA expression in gills. Overall, goldfish showed more effective compensatory responses towards HEA than carp, while trout were least effective.

Sensitivity of ventilation and brain metabolism to ammonia exposure in rainbow trout, Oncorhynchus mykiss

Ammonia has been documented as a respiratory gas that stimulates ventilation, and is sensed by peripheral neuroepithelial cells (NECs) in the gills in ammoniotelic rainbow trout. However, the hyperventilatory response is abolished in trout chronically exposed (1+ months) to high environmental ammonia [HEA; 250 μmol l(-1) (NH4)2SO4]. This study investigates whether the brain is involved in the acute sensitivity of ventilation to ammonia, and whether changes in brain metabolism are related to the loss of hyperventilatory responses in trout chronically exposed to HEA ('HEA trout'). Hyperventilation (via increased ventilatory amplitude rather than rate) and increased total ammonia concentration ([TAmm]) in brain tissue were induced in parallel by acute HEA exposure in control trout in a concentration-series experiment [500, 750 and 1000 μmol l(-1) (NH4)2SO4], but these inductions were abolished in HEA trout. Ventilation was correlated more closely to [TAmm] in brain rather than to [TAmm] in plasma or cerebrospinal fluid. The close correlation of hyperventilation and increased brain [TAmm] also occurred in control trout acutely exposed to HEA in a time-series analysis [500 μmol l(-1) (NH4)2SO4; 15, 30, 45 and 60 min], as well as in a methionine sulfoxamine (MSOX) pre-injection experiment [to inhibit glutamine synthetase (GSase)]. These correlations consistently suggest that brain [TAmm] is involved in the hyperventilatory responses to ammonia in trout. The MSOX treatments, together with measurements of GSase activity, TAmm, glutamine and glutamate concentrations in brain tissue, were conducted in both the control and HEA trout. These experiments revealed that GSase plays an important role in transferring ammonia to glutamate to make glutamine in trout brain, thereby attenuating the elevation of brain [TAmm] following HEA exposure, and that glutamate concentration is reduced in HEA trout. The mRNAs for the ammonia channel proteins Rhbg, Rhcg1 and Rhcg2 were expressed in trout brain, and the expression of Rhbg and Rhcg2 increased in HEA trout, potentially as a mechanism to facilitate the efflux of ammonia. In summary, the brain appears to be involved in the sensitivity of ventilation to ammonia, and brain ammonia levels are regulated metabolically in trout.

Cutaneous nitrogen excretion in the African clawed frog Xenopus laevis: effects of high environmental ammonia (HEA)

Ammonia is a highly toxic molecule and often introduced in considerable amounts into aquatic environments due to anthropogenic activities. Many aquatic and semi-aquatic amphibians utilize, in addition to their kidneys, the skin for osmoregulation and nitrogen excretion. In the present study the effects of prolonged (7-21 days) exposure to high environmental ammonia (HEA, 1 mmol l(-1) NH4Cl) on cutaneous nitrogen excretion and gene expression of key-transporters involved in nitrogen excretion and acid-base regulation were investigated in the fully aquatic African clawed frog, Xenopus laevis. The study revealed that X. laevis excretes predominately ammonia of which approximately 50% is excreted via the skin. Both the ventral and dorsal skin were capable to generate a net ammonia efflux, which was significantly activated by 10 mmol l(-1) of the phosphodiesterase blocker theophylline. The obtained data further suggest that the ammonia efflux was promoted by an acidification of the unstirred boundary layer, likely generated by an apical localized V-ATPase, with NH3 being transported via cutaneous expressed ammonia transporters, Rhbg and Rhcg. Prolonged HEA exposure did significantly reduce the net-flux rates over the ventral skin with Vmax changing from 256 nmol cm(-2) h(-1) in control frogs to 196 nmol cm(-2) h(-1) in HEA exposed animals. Further, prolonged HEA exposure caused a decrease in mRNA expression levels of the ammonia transporter Rhbg, Na(+)/K(+)-ATPase (α-subunit) and V-ATPase (subunit H) in the ventral and dorsal skin and the kidney. In contrast, Rhcg expression levels were unaffected by HEA in skin tissues.

Immunohistochemical localization of urea and ammonia transporters in two confamilial fish species, the ureotelic gulf toadfish (Opsanus beta) and the ammoniotelic plainfin midshipman (Porichthys notatus)

This study aims to illustrate potential transport mechanisms behind the divergent approaches to nitrogen excretion seen in the ureotelic toadfish (Opsanus beta) and the ammoniotelic plainfin midshipman (Porichthys notatus). Specifically, we wish to confirm the expression of a urea transporter (UT), which is found in the gill of the toadfish and which is responsible for the unique "pulsing" nature of urea excretion and to localize the transporter within specific gill cells and at specific cellular locations. Additionally, the localization of ammonia transporters (Rhesus glycoproteins; Rhs) within the gill of both the toadfish and midshipman was explored. Toadfish UT (tUT) was found within Na(+)-K(+)-ATPase (NKA)-enriched cells, i.e., ionocytes (probably mitochondria-rich cells), especially along the basolateral membrane and potentially on the apical membrane. In contrast, midshipman UT (pnUT) immunoreactivity did not colocalize with NKA immunoreactivity and was not found along the filaments but instead within the lamellae. The cellular location of Rh proteins was also dissimilar between the two fish species. In toadfish gills, the Rh isoform Rhcg1 was expressed in both NKA-reactive cells and non-reactive cells, whereas Rhbg and Rhcg2 were only expressed in the latter. In contrast, Rhbg, Rhcg1 and Rhcg2 were expressed in both NKA-reactive and non-reactive cells of midshipman gills. In an additional transport epithelium, namely the intestine, the expression of both UTs and Rhs was similar between the two species, with only subtle differences being observed.

Nitrogen metabolism of the intestine during digestion in a teleost fish, the plainfin midshipman (Porichthys notatus)

Digestion affects nitrogen metabolism in fish, as both exogenous and endogenous proteins and amino acids are catabolized, liberating ammonia in the process. Here we present a model of local detoxification of ammonia by the intestinal tissue of the plainfin midshipman (Porichthys notatus) during digestion, resulting in an increase in urea excretion of gastrointestinal origin. Corroborating evidence indicated whole-animal ammonia and urea excretion increased following feeding, and ammonia levels within the lumen of the midshipman intestine increased to high levels (1.8±0.4 μmol N g(-1)). We propose that this ammonia entered the enterocytes and was detoxified to urea via the ornithine-urea cycle (O-UC) enzymes, as evidenced by a 1.5- to 2.9-fold post-prandial increase in glutamine synthetase activity (0.14±0.05 and 0.28±0.02 μmol min(-1) g(-1) versus 0.41±0.03 μmol min(-1) g(-1)) and an 8.7-fold increase in carbamoyl phosphate synthetase III activity (0.3±1.2 versus 2.6±0.4 nmol min(-1) g(-1)). Furthermore, digestion increased urea production by isolated gastrointestinal tissue 1.7-fold, supporting our hypothesis that intestinal tissue synthesizes urea in response to feeding. We further propose that the intestinal urea may have been excreted into the intestinal lumen via an apical urea transporter as visualized using immunohistochemistry. A portion of the urea was then excreted to the environment along with the feces, resulting in the observed increase in urea excretion, while another portion may have been used by intestinal ureolytic bacteria. Overall, we propose that P. notatus produces urea within the enterocytes via a functional O-UC, which is then excreted into the intestinal lumen. Our model of intestinal nitrogen metabolism does not appear to be universal as we were unab le to activate the O-UC in the intestine of fed rainbow trout. However, literature values suggest that multiple fish species could follow this model.

Rhcg1 and Rhbg mediate ammonia excretion by ionocytes and keratinocytes in the skin of zebrafish larvae: H+-ATPase-linked active ammonia excretion by ionocytes

In zebrafish, Rhcg1 was found in apical membranes of skin ionocytes [H⁺-ATPase-rich (HR) cells], which are similar to α-type intercalated cells in mammalian collecting ducts. However, the cellular distribution and role of Rhbg in zebrafish larvae have not been well investigated. In addition, HR cells were hypothesized to excrete ammonia against concentration gradients. In this study, we attempted to compare the roles of Rhbg and Rhcg1 in ammonia excretion by larval skin and compare the capability of skin cells to excrete ammonia against concentration gradients. Using in situ hybridization and immunohistochemistry, Rhbg was localized to both apical and basolateral membranes of skin keratinocytes. A scanning ion-selective electrode technique (SIET) was applied to measure the NH₄⁺ flux at the apical surface of keratinocytes and HR cells. Knockdown of Rhbg with morpholino oligonucleotides suppressed ammonia excretion by keratinocytes and induced compensatory ammonia excretion by HR cells. To compare the capability of cells to excrete ammonia against gradients, NH₄⁺ flux of cells was determined in larvae exposed to serial concentrations of external NH₄⁺. Results showed that HR cells excreted NH₄⁺ against higher NH₄⁺ concentration than did keratinocytes. Knockdown of the expression of either Rhcg1 or H⁺ -ATPase in HR cells suppressed the capability of HR cells.

Close Association of Carbonic Anhydrase (CA2a and CA15a), Na(+)/H(+) Exchanger (Nhe3b), and Ammonia Transporter Rhcg1 in Zebrafish Ionocytes Responsible for Na(+) Uptake

Freshwater (FW) fishes actively absorb salt from their environment to tolerate low salinities. We previously reported that vacuolar-type H⁺-ATPase/mitochondrion-rich cells (H-MRCs) on the skin epithelium of zebrafish larvae (Danio rerio) are primary sites for Na⁺ uptake. In this study, in an attempt to clarify the mechanism for the Na⁺ uptake, we performed a systematic analysis of gene expression patterns of zebrafish carbonic anhydrase (CA) isoforms and found that, of 12 CA isoforms, CA2a and CA15a are highly expressed in H-MRCs at larval stages. The ca2a and ca15a mRNA expression were salinity-dependent; they were upregulated in 0.03 mM Na⁺ water whereas ca15a but not ca2a was down-regulated in 70 mM Na⁺ water. Immunohistochemistry demonstrated cytoplasmic distribution of CA2a and apical membrane localization of CA15a. Furthermore, cell surface immunofluorescence staining revealed external surface localization of CA15a. Depletion of either CA2a or CA15a expression by Morpholino antisense oligonucleotides resulted in a significant decrease in Na⁺ accumulation in H-MRCs. An in situ proximity ligation assay demonstrated a very close association of CA2a, CA15a, Na⁺/H⁺ exchanger 3b (Nhe3b), and Rhcg1 ammonia transporter in H-MRC. Our findings suggest that CA2a, CA15a, and Rhcg1 play a key role in Na⁺uptake under FW conditions by forming a transport metabolon with Nhe3b.

Diuretics with carbonic anhydrase inhibitory action: a patent and literature review (2005 - 2013)

INTRODUCTION

The benzothiadiazines and high ceiling diuretics (hydrochlorothiazide, hydroflumethiazide, quinethazone, metolazone, chlorthalidone, indapamide, furosemide and bumetanide) contain primary sulfamoyl moieties acting as zinc-binding groups in the metalloenzyme carbonic anhydrase (CA, EC 4.2.1.1). These drugs are widely used clinically and were recently shown to weakly inhibit isoforms CA I and II, but to possess stronger activity against isoforms involved in other important pathologies, for example, obesity, cancer, epilepsy and hypertension.

AREAS COVERED

The class of clinically used diuretics, with CA inhibitory properties, is the main topic of the review. A patent literature review covering the period from 2005 to 2013 is presented.

EXPERT OPINION

This section presents an overview of the patent literature in the sulfonamide diuretic field. Most of the patents deal with the combination of diuretic sulfonamide CA inhibitors with other agents useful in the management of cardiovascular diseases and obesity. Such combinations exert a better therapeutic activity compared to similar diuretics that do not inhibit CAs, raising the question of the polypharmacological and drug repositioning effects of these old drugs. These effects seem to be due to the potent inhibition of such drugs against CA isoforms present in kidneys and blood vessels, which explain both the blood pressure lowering effects as well as organ-protective activity of the drugs. An explanation of these data is provided by the fact that inhibition of the renal CAs leads to a large increase of the nitrite excretion in urine, suggesting that renal CAs are involved in nitrite reabsorption in humans. Important lessons for the drug design of sulfonamide CA inhibitors (CAIs) can be drawn from these data.

Marine, freshwater and aerially acclimated mangrove rivulus (Kryptolebias marmoratus) use different strategies for cutaneous ammonia excretion

Rhesus (Rh) glycoproteins are ammonia gas (NH(3)) channels known to be involved in ammonia transport in animals. Because of the different osmoregulatory and ionoregulatory challenges faced by teleost fishes in marine and freshwater (FW) environments, we hypothesized that ammonia excretion strategies would differ between environments. Also, we hypothesized that cutaneous NH(3) volatilization in air-acclimated fish is facilitated by base secretion. To test these hypotheses, we used the skin of the euryhaline amphibious mangrove rivulus (Kryptolebias marmoratus). The skin excretes ammonia and expresses Rh glycoproteins. Serosal-to-mucosal cutaneous ammonia flux was saturable (0-16 mmol/l ammonia, K(m) of 6.42 mmol/l). In FW, ammonia excretion increased in response to low mucosal pH but decreased with pharmacological inhibition of Na(+)/H(+) exchangers (NHE) and H(+) ATPase. Conversely, in brackish water (BW), lowering the mucosal pH significantly decreased ammonia excretion. Inhibitors of NHE also decreased ammonia excretion in BW fish. Immunofluorescence microscopy demonstrated that both the Rh isoform, Rhcg1, and NHE3 proteins colocalized in Na(+)/K(+) ATPase expressing mitochondrion-rich cells in the gills, kidney, and skin. We propose that the mechanisms of cutaneous ammonia excretion in FW K. marmoratus are consistent with the model for branchial ammonia excretion in FW teleost fish. NH(4)(+) excretion appeared to play a stronger role in BW. NH(4)(+) excretion in BW may be facilitated by apical NHE and/or diffuse through paracellular pathways. In aerially acclimated fish, inhibition of NHE and H(+) ATPase, but not the Cl(-)/HCO(3)(-) exchanger, significantly affected cutaneous surface pH, suggesting that direct base excretion is not critical for NH(3) volatilization. Overall, K. marmoratus use different strategies for excreting ammonia in three different environments, FW, BW, and air, and Rh glycoproteins and NHE are integral to all.

Differential responses in ammonia excretion, sodium fluxes and gill permeability explain different sensitivities to acute high environmental ammonia in three freshwater teleosts

We examined the acute physiological responses to high environmental ammonia (HEA), particularly the linkages between branchial ammonia fluxes and unidirectional Na(+) fluxes, as well as urea excretion, cortisol, and indicators of gill permeability in three freshwater teleosts differing in their sensitivities to ammonia; the highly sensitive salmonid Oncorhynchus mykiss (rainbow trout), the less sensitive cyprinid Cyprinus carpio (common carp) and the highly resistant cyprinid Carassius auratus (goldfish). Fish were acutely exposed to two sub-lethal ammonia concentrations (as NH(4)HCO(3)) at pH 7.9: 1 mM for a period of 12 h, identical for all species, and 5 mM for the cyprinids and 1.4 mM for the trout for 3 h. Elevation of plasma cortisol at both levels of HEA was apparent in all species. At 1 mM, ammonia excretion (J(amm)) was inhibited to a greater extent in trout than cyprinids and concurrently a significantly higher plasma ammonia level was evident in trout. However J(amm) was reversed in all species at 5 or 1.4 mM. Goldfish showed a significant increase in urea excretion rate (J(urea)) during HEA exposure. In carp and trout, neither level of HEA elevated J(urea) but urea production was increased as evidenced by a considerable elevation of plasma urea. At 1mM HEA, Na(+) imbalance became progressively more severe in trout and carp due to a stimulation of unidirectional Na(+) efflux (J(out)(Na)) without a concomitant increase in unidirectional Na(+) influx (J(in)(Na)). Additionally, a transient reduction of J(in)(Na) was evident in trout. Goldfish showed an opposite trend for J(out)(Na) with reduced efflux rates and a positive Na(+) balance during the first few hours of HEA. However, after 12 h of exposure, both J(in)(Na) and J(out)(Na) were also increased in both carp and goldfish, whereas only J(out)(Na) was increased in trout, leading to a net Na(+) loss. Na(+) homeostasis was entirely disrupted in all three species when subjected to the 5 or 1.4 mM ammonia for 3 h: J(in)(Na) was significantly inhibited while considerable activation of J(out)(Na) was observed. Diffusive water efflux rates and net K(+) loss rates across the gills were enhanced during HEA only in trout, indicating an increment in gill transcellular permeability. Transepithelial potential was increased in all the species during ammonia exposure, but to the least extent in goldfish. Overall, for several different physiological systems, trout were most disturbed, and goldfish were least disturbed by HEA, helping to explain the differential ammonia tolerance of the three species.

Molecular and physiological responses to long-term sublethal ammonia exposure in Atlantic salmon (Salmo salar)

The objective of this study was to determine the underlying physiological and molecular responses to long-term sublethal ammonia exposure in Atlantic salmon (Salmo salar) parr. Previous studies have predominately focused on mechanisms during acute, short-term exposure. For that purpose Atlantic salmon parr were exposed to four ammonia concentrations between 4 and 1800 μmol l(-1) total ammonia nitrogen (TAN), and subjected to two feeding regimes for 15 weeks. Elevated environmental ammonia and full feeding strength caused an initial increase in plasma ammonia levels ([T(amm)]) after 22 days of exposure, which thereafter declined and remained similar to the control animals towards the end of the study. On the other hand, a progressive decrease in plasma urea levels was evident throughout the entire exposure period and depended on the concentration of environmental ammonia, with the largest decrease in urea levels observed at the highest ammonia concentrations (1700 and 1800 μmol l(-1) TAN). We hypothesized that the successful adaptation to long-term elevated ammonia levels would involve an increased capacity for carrier-facilitated branchial excretion. This hypothesis was strengthened by the first evidence of an up-regulation of branchial transcription of the genes encoding the Rhesus (Rh) glycoproteins, Rhcg1 and Rhcg2, urea transporter (UT) and aquaporin 3a (Aqp3a), during long-term exposure. Of the Rhesus glycoprotein (Rh) mRNAs, Rhcg1 was up-regulated at all tested ammonia levels, while Rhcg2 showed a concentration-sensitive increase. Increased transcription levels of V-type H(+)-ATPase (H(+)-ATPase) were observed at the highest ammonia concentrations (1700 and 1800 μmol l(-1) TAN) and coincided with an up-regulation of Rhcg2 at these concentrations. Transcription of UT and Aqp3a was increased after 15 weeks of exposure to low ammonia levels (470 and 480 μmol l(-1) TAN). A significant increase in brain glutamine (Gln) concentration was observed for full fed Atlantic salmon after 22 days and in fish with restricted feeding after 105 days of exposure to 1800 and 1700 μmol l(-1) TAN, respectively, without any concomitant decrease in brain glutamate (Glu) concentrations. These results suggest that Gln synthesis is an ammonia detoxifying strategy employed in the brain of Atlantic salmon parr during long-term sublethal ammonia exposure. Full feed strength had an additive effect on plasma [T(amm)], while the restricted feeding regime postponed the majority of the observed physiological and molecular responses. In conclusion, Atlantic salmon parr adapts to the long-term sublethal ammonia concentrations with increased branchial transcription levels of ammonia and urea transporting proteins and ammonia detoxification in the brain.

Expression pattern of potential biomarker genes related to growth, ion regulation and stress in response to ammonia exposure, food deprivation and exercise in common carp (Cyprinus carpio)

Waterborne ammonia has become a persistent pollutant of aquatic habitats. During certain periods (e.g. winter), food deprivation may occur simultaneously in natural water. Additionally, under such stressful circumstances, fish may be enforced to swim at a high speed in order to catch prey, avoid predators and so on. Consequently, fish need to cope with all these stressors by altering physiological processes which in turn are controlled by their genes. In this present study, toxicogenomic analyses using real time PCR was used to characterize expression patterns of potential biomarker genes controlling growth, ion regulation and stress responses in common carp subjected to elevated ammonia (1 mg/L; Flemish water quality guideline for surface water) following periods of feeding (2% body weight) and fasting (unfed for 7 days prior to sampling). Both feeding groups of fish were exposed to high environment ammonia (HEA) for 0 h (control), 3h, 12h, 1 day, 4 days, 10 days, 21 days and 28 days, and were sampled after performing swimming at different speeds (routine versus exhaustive). Results show that the activity and expression of Na(+)/K(+)-ATPase, an important branchial ion regulatory enzyme, was increased after 4-10 days of exposure. Effect of HEA was also evident on expression patterns of other ion-regulatory hormone and receptor genes; prolactin and cortisol receptor mRNA level(s) were down-regulated and up-regulated respectively after 4, 10 and 21 days. Starvation and exhaustive swimming, the additional challenges in present study significantly further enhanced the HEA effect on the expression of these two genes. mRNA transcript of growth regulating hormone and receptor genes such as Insulin-like growth factor I, growth hormone receptor, and the thyroid hormone receptor were reduced in response to HEA and the effect of ammonia was exacerbated in starved fish, with levels that were remarkably reduced compared to fed exposed fish. However, the expression of the growth hormone gene itself was up-regulated under the same conditions. Expression of somatolactin remained unaltered. Stress representative genes, cytochrome oxidase subunit 1 showed an up-regulation in response to HEA and starvation while the mRNA level of heat shock protein 70 was increased in response to all the three stressors. The expression kinetics of the studied genes could permit to develop a "molecular biomarker system" to identify the underlying physiological processes and impact of these stressors before effects at population level occur.

Waterborne copper exposure inhibits ammonia excretion and branchial carbonic anhydrase activity in euryhaline guppies acclimated to both fresh water and sea water

Inhibition of ammonia excretion (J(amm)) is a common response to Cu exposure in freshwater (FW) and seawater (SW) organisms. To determine the mechanism of this response, a euryhaline species of guppy (Poecilia vivipara) was exposed to 20 μg Cu/l in FW (0 ppt) and SW (25 ppt) for 96 h. In both salinities, Cu transiently inhibited ammonia excretion (J(amm)) followed by a full recovery by the end of the 96 h exposure. The activities of Na(+)/K(+)-ATPase, H(+)-ATPase, and carbonic anhydrase (CA) were examined in the gills at 12 and 96 h of Cu exposure. In both salinity acclimations, CA activity was significantly inhibited following 12h of Cu exposure in P. vivipara, marking the first in vivo evidence of Cu-induced inhibition of CA in fish. Moreover, the inhibition and recovery of this enzyme were correlated with the inhibition and recovery of J(amm) in both salinity acclimations. The blockade of CA potentially acts as a common mechanism of J(amm) inhibition in FW and SW. There were no significant effects on Na(+)/K(+)-ATPase or H(+)-ATPase activity at either time point or salinity. However, H(+)-ATPase activity was upregulated at 96 h relative to the 12h time point, potentially involving this enzyme in re-establishing J(amm).

Toxicity of silver nanoparticles to rainbow trout: a toxicogenomic approach

Silver (Ag) nanoparticles are used as antimicrobial adjuvant in various products such as clothes and medical devices where the release of nano-Ag could contaminate the environment and harm wildlife. The purpose of this study was to examine the sublethal effects of nano-Ag and dissolved Ag on Oncorhynchus mykiss rainbow trout. Hepatic Ag contents and changes in gene expression were monitored to provide insights on bioavailability and mode of action of both forms of silver. Fish were exposed to increasing concentrations (0.06, 0.6 and 6 μg L(-1)) of nano-Ag (20 nm) and silver nitrate (AgNO(3)) for 96 h at 15°C. A gene expression analysis was performed in the liver using a DNA microarray of 207 stress-related genes followed by a quantitative polymerase chain reaction on a selection of genes for validation. The biochemical markers consisted of the determination of labile zinc, metallothioneins, DNA strand breaks, lipid peroxidation (LPO) and vitellogenin-like proteins. The analysis of total Ag in the aquarium water revealed that nano-Ag was mostly aggregated, with 1% of the total Ag being dissolved. Nevertheless, hepatic Ag content was significantly increased in exposed fish. Indeed, dissolved Ag was significantly more bioavailable than nano-Ag only at the highest concentration with 38 ± 10 and 11 ± 3 ng Ag mg(-1) proteins for dissolved and nano-Ag respectively. Exposure to both forms of Ag led to significant changes in gene expression for 13% of tested gene targets. About 12% of genes responded specifically to nano-Ag, while 10% of total gene targets responded specifically to dissolved Ag. The levels of vitellogenin-like proteins and DNA strand breaks were significantly reduced by both forms of Ag, but DNA break levels were lower with nano-Ag and could not be explained by the presence of ionic Ag. Labile zinc and the oxidized fraction of metallothioneins were increased by both forms of Ag, but LPO was significantly induced by nano-Ag only. A discriminant function analysis revealed that the responses obtained by biochemical markers and a selection of ten target genes were able to discriminate completely (100%) the effects of both forms of Ag. Exposure to nano-Ag involved genes in inflammation and dissolved Ag involved oxidative stress and protein stability. Hence, the toxicity of Ag will differ depending on the presence of Ag nanoparticles and aggregates.

Measuring titratable alkalinity by single versus double endpoint titration: An evaluation in two cyprinodont species and implications for characterizing net H+ flux in aquatic organisms

In this study, Na(+) uptake and acid-base balance in the euryhaline pupfish Cyprinodon variegatus variegatus were characterized when fish were exposed to pH 4.5 freshwater (7mM Na(+)). Similar to the related cyprinodont, Fundulus heteroclitus, Na(+) uptake was significantly inhibited when exposed to low pH water. However, it initially appeared that C. v. variegatus increased apparent net acid excretion at low pH relative to circumneutral pH. This result is opposite to previous observations for F. heteroclitus under similar conditions where fish were observed to switch from apparent net H(+) excretion at circumneutral pH to apparent net H(+) uptake at low pH. Further investigation revealed disparate observations between these studies were the result of using double endpoint titrations to measure titratable alkalinity fluxes in the current study, while the earlier study utilized single endpoint titrations to measure these fluxes (i.e.,. Cyprinodon acid-base transport is qualitatively similar to Fundulus when characterized using single endpoint titrations). This led to a comparative investigation of these two methods. We hypothesized that either the single endpoint methodology was being influenced by a change in the buffer capacity of the water (e.g., mucus being released by the fish) at low pH, or the double endpoint methodology was not properly accounting for ammonia flux by the fish. A series of follow-up experiments indicated that buffer capacity of the water did not change significantly, that excretion of protein (a surrogate for mucus) was actually reduced at low pH, and that the double endpoint methodology does not properly account for NH(3) excretion by fish under low pH conditions. As a result, it overestimates net H(+) excretion during low pH exposure. After applying the maximum possible correction for this error (i.e., assuming that all ammonia is excreted as NH(3)), the double endpoint methodology indicates that net H(+) transport was reduced to effectively zero in both species at pH 4.5. However, significant differences between the double endpoint (no net H(+) transport at low pH) and single endpoint titrations (net H(+) uptake at low pH) remain to be explained.

Structure and function of ionocytes in the freshwater fish gill

Freshwater fishes lose ions to the external medium owing to the steep concentration gradients between the body fluids and the water. To maintain homeostasis, they use ionocytes to actively extract Na(+), Cl(-), and Ca(2+) from the dilute external medium and excrete acidic (H(+)) or basic (HCO(3)(-)) equivalents by specialized cells termed ionocytes that are responsible for transport of ions. Freshwater fishes have evolved diverse approaches to solving these similar ionic and acid-base problems. In the few well-studied species, there are clearly different patterns in the physiology and morphology for ionocytes in the gill. In this review, we describe the varying nomenclature of ionocytes that have been used in the past 80 years to allow direct comparison of ionocytes and their common functions in different species. We focus on the recent advancement in our understanding of the molecular mechanisms of ion and acid-base regulation as represented by ionocyte subtypes found in rainbow trout, killifish, tilapia and zebrafish gill.

Seven things fish know about ammonia and we don't

In this review we pose the following seven questions related to ammonia and fish that represent gaps in our knowledge. 1. How is ammonia excretion linked to sodium uptake in freshwater fish? 2. How much does branchial ammonia excretion in seawater teleosts depend on Rhesus (Rh) glycoprotein-mediated NH(3) diffusion? 3. How do fish maintain ammonia excretion rates if branchial surface area is reduced or compromised? 4. Why does high environmental ammonia change the transepithelial potential across the gills? 5. Does high environmental ammonia increase gill surface area in ammonia tolerant fish but decrease gill surface area in ammonia intolerant fish? 6. How does ammonia contribute to ventilatory control? 7. What do Rh proteins do when they are not transporting ammonia? Mini reviews on each topic, which are able to present only partial answers to each question at present, are followed by further questions and/or suggestions for research approaches targeted to uncover answers.

Characterization of rainbow trout gonad, brain and gill deep cDNA repertoires using a Roche 454-Titanium sequencing approach

Rainbow trout, Oncorhynchus mykiss, is an important aquaculture species worldwide and, in addition to being of commercial interest, it is also a research model organism of considerable scientific importance. Because of the lack of a whole genome sequence in that species, transcriptomic analyses of this species have often been hindered. Using next-generation sequencing (NGS) technologies, we sought to fill these informational gaps. Here, using Roche 454-Titanium technology, we provide new tissue-specific cDNA repertoires from several rainbow trout tissues. Non-normalized cDNA libraries were constructed from testis, ovary, brain and gill rainbow trout tissue samples, and these different libraries were sequenced in 10 separate half-runs of 454-Titanium. Overall, we produced a total of 3million quality sequences with an average size of 328bp, representing more than 1Gb of expressed sequence information. These sequences have been combined with all publicly available rainbow trout sequences, resulting in a total of 242,187 clusters of putative transcript groups and 22,373 singletons. To identify the predominantly expressed genes in different tissues of interest, we developed a Digital Differential Display (DDD) approach. This approach allowed us to characterize the genes that are predominantly expressed within each tissue of interest. Of these genes, some were already known to be tissue-specific, thereby validating our approach. Many others, however, were novel candidates, demonstrating the usefulness of our strategy and of such tissue-specific resources. This new sequence information, acquired using NGS 454-Titanium technology, deeply enriched our current knowledge of the expressed genes in rainbow trout through the identification of an increased number of tissue-specific sequences. This identification allowed a precise cDNA tissue repertoire to be characterized in several important rainbow trout tissues. The rainbow trout contig browser can be accessed at the following publicly available web site (http://www.sigenae.org/).

The interactive effects of ammonia exposure, nutritional status and exercise on metabolic and physiological responses in gold fish (Carassius auratus L.)

This study aimed to elucidate the physiological effects of high environmental ammonia (HEA) following periods of feeding (2% body weight) and starvation (unfed for 7 days prior to sampling) in gold fish (Carassius auratus). Both groups of fish were exposed to HEA (1 mg/L; Flemish water quality guideline) for 0 h (control), 3 h, 12 h, 1 day, 4 days, 10 days, 21 days and 28 days. Measurements of weight gain (%), oxygen consumption (MO2), ammonia excretion rate, ammonia quotient (AQ), critical swimming speeds (Ucrit), plasma and muscle ammonia accumulation, plasma lactate, liver and muscle glycogen, lipid and protein content were done at various time intervals during the experimental periods. Overall, ammonia excretion rates, plasma ammonia accumulation and AQ were significantly affected by food regime in ammonia free water. HEA, the additional challenge in the present study, significantly altered all the studied parameters among fed and starved groups in days-dependent manner. Results show that weight gain (%), MO2, Ucrit, protein content in liver and muscle, and glycogen content in muscle among starved fish under HEA were considerably reduced compared to control and fed fish. Additionally a remarkable increase in plasma ammonia level, muscle ammonia, lactate accumulation and AQ was seen. However in fed fish, MO2, ammonia excretion rate, AQ and lactate level augmented after exposure to HEA. These results indicate that starved fish appeared more sensitive to HEA than fed fish. Furthermore, as expected, the toxic effect of ammonia exposure in both feeding treatments was exacerbated when imposed to exhaustive swimming (swum at 3/4th Ucrit). Such effects were more pronounced in starved fish. This suggests that starvation can instigate fish more vulnerable to external ammonia during exercise. Therefore, it was evident from our study that feeding ameliorates ammonia handling and reduces its toxicity during both routine and exhaustive swimming. Moreover, recovery was observed for some physiological parameters (e.g. MO2, ammonia excretion, Ucrit, plasma ammonia) during the last exposure periods (21-28 days) while for others (e.g. growth, tissue glycogen and protein content, muscle ammonia) effects only became apparent at this time. In the future, these results need to be considered in ecological context as fish in ammonia polluted may experience different phenomenon (starvation and exercise) simultaneously.

Branchial and extra-branchial ammonia excretion in goldfish (Carassius auratus) following thermally induced gill remodeling

Under cold acclimated conditions, goldfish (Carassius auratus) express an interlamellar cell mass (ILCM) which limits diffusive ion loss but may also impede branchial ammonia excretion (J(amm)). In the present study, goldfish were subjected to a 2-week 5 or 25 °C acclimation in order to modulate the degree of ILCM gill coverage and determine potential effects on J(amm). 25 °C-fish displayed gill coverage which was significantly lower than the 5 °C-fish, though the ILCM was not completely absent in these fish. 5 °C-fish demonstrated J(amm) values approximately 60% lower than those of 25 °C-fish. The magnitude of anterior (branchial) J(amm) strongly correlated with gill coverage (r(2)=0.83), suggesting that the ILCM may impede branchial J(amm). Divided chamber experiments demonstrated that relative to the 25 °C-fish, 5 °C-fish relied more upon posterior routes of excretion. In response to high external ammonia (HEA; 1.5mM NH(4)HCO(3)) exposures, 25 °C-fish displayed ammonia uptake while 5 °C-fish maintained excretion against HEA, suggesting that the ILCM may act as a barrier preventing ammonia uptake. In summary, the ILCM appears to impede branchial J(amm), such that 5 °C-rely more on extra-branchial routes of excretion. We hypothesize that gill remodeling in these fish may be intimately tied to physiological adjustments on the whole-body scale.

A nose-to-nose comparison of the physiological and molecular responses of rainbow trout to high environmental ammonia in seawater versus freshwater

Steelhead rainbow trout acclimated to either freshwater (FW) or seawater (SW) were exposed to high environmental ammonia (HEA, 1000 μmol l(-1) NH(4)HCO(3), pH 7.8-8.0) for 24 h. SW trout restored ammonia excretion more rapidly (3-6 h versus 9-12 h in FW), despite higher production rates and lower plasma pH. Plasma total ammonia levels stabilized at comparable levels below the external HEA concentration, and blood acid-base disturbances were small at both salinities. The electrochemical gradients for NH(4)(+) entry (F(NH(4))(+)) were the same in the two salinities, but only because FW trout allowed their transepithelial potential to rise by ∼15 mV during HEA exposure. Elevation of plasma [cortisol] during HEA exposure was more prolonged in SW fish. Plasma [glucose] increased in SW, but decreased in FW trout. Plasma [urea-N] also decreased in FW, in concert with elevated urea transporter (UT) mRNA expression in the gills. Of 13 branchial transporters, baseline mRNA expression levels were higher for Rhcg1, NHE2, NKCC1a and UT, and lower for NBC1 and NKA-α1a in SW trout, whereas NKA-α1b, NHE3, CA2, H(+)-ATPase, Rhag, Rhbg and Rhcg2 did not differ. Of the Rh glycoprotein mRNAs responding to HEA, Rhcg2 was greatly upregulated in both FW and SW, Rhag decreased only in SW and Rhcg1 decreased only in FW. H(+)-ATPase mRNA increased in FW whereas NHE2 mRNA increased in SW; NHE3 did not respond, and V-type H(+)-ATPase activity declined in SW during HEA exposure. Branchial Na(+),K(+)-ATPase activity was much higher in SW gills, but could not be activated by NH(4)(+). Overall, the more effective response of SW trout was explained by differences in physical chemistry between SW and FW, which greatly reduced the plasma NH(3) tension gradient for NH(3) entry, as well as by the higher [Na(+)] in SW, which favoured Na(+)-coupled excretion mechanisms. At a molecular level, responses in SW trout showed subtle differences from those in FW trout, but were very different than in the SW pufferfish. Upregulation of Rhcg2 appears to play a key role in the response to HEA in both FW and SW trout, and NH(4)(+) does not appear to move through Na(+),K(+)-ATPase.

Revisiting the effects of crowding and feeding in the gulf toadfish, Opsanus beta: the role of Rhesus glycoproteins in nitrogen metabolism and excretion

Models of branchial transport in teleosts have been reshaped by the recent discovery of Rhesus (Rh) glycoproteins, a family of proteins that facilitate the movement of NH3 across cell membranes. This study examines the effects of crowding and feeding on ammonia excretion in gulf toadfish (Opsanus beta) within the context of Rh glycoproteins and the ammonia-fixing enzyme, glutamine synthetase (GS). Four Rh isoforms (Rhag, Rhbg, Rhcg1 and Rhcg2) were isolated from toadfish. Tissue distributions showed higher levels of mRNA expression in the gills and liver, moderate levels in the intestine and lower levels in the stomach. Crowding significantly lowered branchial Rh expression and ammonia excretion rates in fasted toadfish. A comparison of Rh expression in the digestive tract revealed relatively low levels of Rhcg1 and Rhcg2 in the stomach and high mRNA abundance of Rhbg, Rhcg1 and Rhcg2 in the intestine of fasted, crowded toadfish. We speculate that these trends may reduce secretion and enhance absorption, respectively, to minimize the amount of ammonia that is lost through gastrointestinal routes. By contrast, these patterns of expression were modified in response to an exogenous ammonia load via feeding. Post-prandial ammonia excretion rates were elevated twofold, paralleled by similar increases in branchial Rhcg1 mRNA, gastric Rhcg1 mRNA and mRNA of all intestinal Rh isoforms. These changes were interpreted as an attempt to increase post-prandial ammonia excretion rates into the environment owing to a gradient created by elevated circulating ammonia concentrations and acidification of the digestive tract. Overall, we provide evidence that toadfish modulate both the expression of Rh isoforms and urea synthesis pathways to tightly control and regulate nitrogen excretion.

Interactions between cortisol and Rhesus glycoprotein expression in ureogenic toadfish, Opsanus beta

In their native environment, gulf toadfish excrete equal quantities of ammonia and urea. However, upon exposure to stressful conditions in the laboratory (i.e. crowding, confinement or air exposure), toadfish decrease branchial ammonia excretion and become ureotelic. The objective of this study was to determine the influences of cortisol and ammonia on ammonia excretion relative to expression of Rhesus (Rh) glycoproteins and the ammonia-fixing enzyme, glutamine synthetase (GS). In vivo infusions and/or injections were used to manipulate corticosteroid activity and plasma ammonia concentrations in ureotelic toadfish. Metyrapone treatment to lower circulating cortisol levels resulted in a 3.5-fold elevation of ammonia excretion rates, enhanced mRNA expression of two of the toadfish Rh isoforms (Rhcg1 and Rhcg2), and decreased branchial and hepatic GS activity. Correspondingly, cortisol infusion decreased ammonia excretion 2.5-fold, a change that was accompanied by reduced branchial expression of all toadfish Rh isoforms (Rhag, Rhbg, Rhcg1 and Rhcg2) and a twofold increase in hepatic GS activity. In contrast, maintenance of high circulating ammonia levels by ammonia infusion enhanced ammonia excretion and Rh expression (Rhag, Rhbg and Rhcg2). Toadfish treated with cortisol showed an attenuated response to ammonia infusion with no change in Rh mRNA expression or GS activity. In summary, the evidence suggests that ammonia excretion in toadfish is modulated by cortisol-induced changes in both Rh glycoprotein expression and GS activity.

Rhcg1 and NHE3b are involved in ammonium-dependent sodium uptake by zebrafish larvae acclimated to low-sodium water

To investigate whether Na+ uptake by zebrafish is dependent on NH4+ excretion, a scanning ion-selective electrode technique was applied to measure Na+ and NH4+ gradients at the yolk-sac surface of zebrafish larvae. Low-Na+ acclimation induced an inward Na+ gradient (uptake), and a combination of low Na+ and high NH4+ induced a larger inward Na+ gradient. When measuring the ionic gradients, raising the external NH4+ level (5 mM) blocked NH4+ excretion and Na+ uptake; in contrast, raising the external Na+ level (10 mM) simultaneously enhanced Na+ uptake and NH4+ excretion. The addition of MOPS buffer (5 mM), which is known to block NH4+ excretion, also suppressed Na+ uptake. These results showed that Na+ uptake and NH4+ excretion by larval skin are associated when ambient Na+ level is low. Knockdown of Rhcg1 translation with morpholino-oligonucleotides decreased both NH4+ excretion and Na+ uptake by the skin and Na+ content of whole larvae. Knockdown of nhe3b translation or inhibitor (5-ethylisopropyl amiloride) treatment also decreased both the NH4+ excretion and Na+ uptake. This study provides loss-of-function evidence for the involvement of Rhcg1 and NHE3b in the ammonium-dependent Na+ uptake mechanism in zebrafish larvae subjected to low-Na+ water.

Ammonia excretion in the freshwater planarian Schmidtea mediterranea

In aquatic invertebrates metabolic nitrogenous waste is excreted predominately as ammonia. Very little is known, however, of the underlying mechanisms of ammonia excretion, particularly in freshwater species. Our results indicate that in the non-parasitic freshwater planarian Schmidtea mediterranea ammonia excretion depends on an acidification of the apical unstirred layer of the body surface and consequent ammonia trapping. Buffering of the environment to a pH of 7 or higher decreased excretion rate. Inhibitor experiments suggested further that the excretion mechanism involves the participation of the V-type H+-ATPase and carbonic anhydrase and possibly also the Na+/K+-ATPase and Na+/H+ exchangers (NHEs). Alkalinization (pH 8.5, 2 days) of the environment led to a 1.9-fold increase in body ammonia levels and to a down-regulation of V-ATPase (subunit A) and Rh-protein mRNA. Further, a two day exposure to non-lethal ammonia concentrations (1 mmol L-1) caused a doubling of body ammonia levels and led to an increase in Rh-protein and Na+/K+-ATPase (α-subunit) mRNA expression levels. In-situ hybridization studies indicated a strong mRNA expression of the Rh-protein in the epidermal epithelium. The ammonia excretion mechanism proposed for S. mediterranea reveals striking similarities to the current model suggested to function in gills of freshwater fish.

Digestion of a single meal affects gene expression of ion and ammonia transporters and glutamine synthetase activity in the gastrointestinal tract of freshwater rainbow trout

Experiments on freshwater rainbow trout, Oncorhynchus mykiss, demonstrated how digestion affected the transcriptional expression of gastrointestinal transporters following a single satiating meal (~3% body mass ration) after a 1-week fast. Quantitative real-time polymerase chain reaction was employed to measure the relative mRNA expression of three previously cloned and sequenced transporters [H(+)-K(+)-ATPase (HKA), Na(+)/HCO(3)(-) cotransporter (NBC), and the Rhesus glycoprotein (Rhbg1; an ammonia transporter)] over a 24-h time course following feeding. Plasma total ammonia increased about threefold from pre-feeding levels to 288 μmol l(-1), whereas total ammonia levels in chyme supernatant reached a sixfold higher value (1.8 mmol l(-1)) than plasma levels. Feeding did not appear to have a statistically significant effect on the relative mRNA expression of the gastric HKA or Rhbg1. However, the relative mRNA expression of gastric NBC was increased 24 h following the ingestion of a meal. Along the intestinal tract, feeding increased the relative mRNA expression of Rhbg1, but had no effect on the expression of NBC. Expression of the gastric HKA was undetectable in the intestinal tract of freshwater rainbow trout. Digestion increased the activity of glutamine synthetase in the posterior intestine at 12 and 24 h following feeding. This study is among the first to show that there are digestion-associated changes in gene expression and enzyme activity in the gastrointestinal tract of teleost fish illustrating the dynamic plasticity of this organ. These post-prandial changes occur over the relative short-term duration of digesting a single meal.

Ammonia sensing by neuroepithelial cells and ventilatory responses to ammonia in rainbow trout

Ammonia, the third respiratory gas in teleost fish, acts as an acute stimulant to ventilation in ammoniotelic rainbow trout. We investigated whether this sensitivity is maintained in trout chronically exposed (1+ months) to high environmental ammonia [HEA, 250 μmol l–1 (NH4)2SO4] in the water, and whether gill neuroepithelial cells (NECs) are involved in ammonia sensing. Hyperventilation was induced both by acute external (NH4)2SO4 exposure [250 or 500 μmol l–1 (NH4)2SO4] and by intra-arterial (NH4)2SO4 injection (580 μmol kg–1 of ammonia) in control trout, but these responses were abolished in chronic HEA animals. Hyperventilation in response to acute ammonia exposure persisted after bilateral removal of each of the four gill arch pairs separately or after combined removal of arches III and IV, but was delayed by removal of gill arch I, and eliminated by combined removal of arches I and II. NECs, identified by immunolabeling against 5-HT, were mainly organized in two lines along the filament epithelium in all four gill arches. In control trout, NECs were slightly smaller but more abundant on arches I and II than on arches III and IV. Chronic HEA exposure reduced the density of the NECs on all four arches, and their size on arches I and II only. Fura-2 fluorescence imaging was used to measure intracellular free calcium ion concentration ([Ca2+]i) responses in single NECs in short-term (24–48 h) culture in vitro. [Ca2+]i was elevated to a comparable extent by perfusion of 30 mmol l–1 KCl and 1 mmol l–1 NH4Cl, and these [Ca2+]i responses presented in two different forms, suggesting that ammonia may be sensed by multiple mechanisms. The [Ca2+]i responses to high ammonia were attenuated in NECs isolated from trout chronically exposed to HEA, especially in ones from gill arch I, but responses to high K+ were unchanged. We conclude that the hyperventilatory response to ammonia is lost after chronic waterborne HEA exposure, and that NECs, especially the ones located in gill arches I and II, are probably ammonia chemoreceptors that participate in ventilatory modulation in trout.

Ammonia excretion via Rhcg1 facilitates Na⁺ uptake in larval zebrafish, Danio rerio, in acidic water

The involvement of a Na(+)/H(+) exchanger (NHE) in mediating Na(+) uptake by freshwater fish is currently debated. Although supported indirectly by empirical molecular and pharmacological data, theoretically its operation should be constrained thermodynamically, owing to unfavorable chemical gradients. Recently, there has been an increasing focus on ammonia channels (Rh proteins) as potentially contributing to Na(+) uptake across the freshwater fish gill. In this study, we tested the hypothesis that Rhcg1, a specific apical isoform of Rh protein, is critically important in facilitating Na(+) uptake in zebrafish larvae via its interaction with NHE. Treating larvae (4 days postfertilization) with 5-(N-ethyl-N-isopropyl) amiloride (EIPA), an inhibitor of NHE, caused a significant reduction in Na(+) uptake in fish reared in acidic water (pH ∼ 4.0). A role for NHE in Na(+) uptake was further confirmed by translational knockdown of NHE3b, an isoform of NHE thought to be responsible for Na(+)/H(+) exchange in zebrafish larvae. Exposing the larvae reared in acidic water to 5 mM external ammonium sulfate or increasing the buffering capacity of the water with 10 mM HEPES caused concurrent reductions in ammonia excretion and Na(+) uptake. Furthermore, translational knockdown of Rhcg1 significantly reduced ammonia excretion and Na(+) uptake in larvae chronically (4 days) or acutely (24 h) exposed to acidic water. Unlike in sham-injected larvae, EIPA did not affect Na(+) uptake in fish experiencing Rhcg1 knockdown. Additionally, exposure of larvae to bafilomycin A1 (an inhibitor of H(+)-ATPase) significantly reduced Na(+) uptake in fish reared in acidic water. These observations suggest the existence of multiple mechanisms of Na(+) uptake in larval zebrafish in acidic water: one in which Na(+) uptake via NHE3b is linked to ammonia excretion via Rhcg1, and another facilitated by H(+)-ATPase.

Freshwater fish gill ion transport: August Krogh to morpholinos and microprobes

August Krogh proposed that freshwater fishes (and other freshwater animals) maintain body NaCl homoeostasis by extracting these ions from the environment via separate Na(+) /NH(4)(+) and Cl(-) /HCO(3)(-) exchangers in the gill epithelium. Subsequent data from other laboratories suggested that Na(+) uptake was more probably coupled to H(+) secretion via a vesicular proton pump (V-ATPase) electrically coupled to a Na(+) channel. However, despite uncertainty about electrochemical gradients, evidence has accrued that epithelial Na(+) /H(+) exchange indeed may be an alternative pathway for Na(+) uptake. The specific pathways for Na(+) uptake may be species and environment specific. An apical Cl(-) /HCO(3)(-) exchanger is generally accepted for most species (some species do not extract Cl(-) from freshwater), but the relative roles of anion exchanger-like (SLC4A1) vs. pendrin-like (SLC26Z4) exchangers are unknown, and also may be species specific. Most recently, data have supported the presence of an apical Na(+) + Cl(-) cotransporter (NCC-type), despite thermodynamic uncertainty. Ammonia extrusion may be via NH(3) diffusing through the paracellular junctions or NH(4) (+) substitution on both basolateral and apical ionic exchangers (Na(+) + K(+) -ATPase; Na(+) + K(+) + Cl(-) - cotransporter; and Na(+) /H(+) exchanger), but recent evidence suggests that Rhesus-glycoproteins mediate both basolateral and apical movement of ammonia.

Ion regulation in fish gills: recent progress in the cellular and molecular mechanisms

Fish encounter harsh ionic/osmotic gradients on their aquatic environments, and the mechanisms through which they maintain internal homeostasis are more challenging compared with those of terrestrial vertebrates. Gills are one of the major organs conducting the internal ionic and acid-base regulation, with specialized ionocytes as the major cells carrying out active transport of ions. Exploring the iono/osmoregulatory mechanisms in fish gills, extensive literature proposed several models, with many conflicting or unsolved issues. Recent studies emerged, shedding light on these issues with new opened windows on other aspects, on account of available advanced molecular/cellular physiological approaches and animal models. Respective types of ionocytes and ion transporters, and the relevant regulators for the mechanisms of NaCl secretion, Na(+) uptake/acid secretion/NH(4)(+) excretion, Ca(2+) uptake, and Cl(-) uptake/base secretion, were identified and functionally characterized. These new ideas broadened our understanding of the molecular/cellular mechanisms behind the functional modification/regulation of fish gill ion transport during acute and long-term acclimation to environmental challenges. Moreover, a model for the systematic and local carbohydrate energy supply to gill ionocytes during these acclimation processes was also proposed. These provide powerful platforms to precisely study transport pathways and functional regulation of specific ions, transporters, and ionocytes; however, very few model species were established so far, whereas more efforts are needed in other species.

Mechanism of sodium uptake in PNA negative MR cells from rainbow trout, Oncorhynchus mykiss as revealed by silver and copper inhibition

The rate of acid-stimulated and phenamil-sensitive sodium (Na(+)) uptake was measured in three different cell lineages: pavement cells (PVC), total mitochondrion-rich (MR) cell populations, and peanut lectin agglutinin-negative mitochondrion-rich cells (PNA(-) MR) isolated from the rainbow trout gill epithelium. Despite the presence of basal levels of Na(+) uptake in PVC, this transport was not enhanced by acidification, nor was it inhibited by independent treatment with bafilomycin (i.e., a V-type H(+)-ATPase inhibitor), phenamil (i.e., a specific inhibitor of ENaC), or Ag (a specific inhibitor of active Na(+) transport in fish). In contrast, Na(+) uptake in PNA(-) MR cells was increased by ~220% above basal levels following acidification of near 0.4 pH units in the presence of 1.0 mM external Na(+). Acid-stimulated Na(+) transport was entirely inhibited by both phenamil and bafilomycin. Silver (Ag) and copper (Cu), which are known to interfere with active Na(+) transport in fish, were also responsible for inhibiting acid stimulated Na(+) uptake in PNA(-) MR cells, but by themselves had no effect on basal Na(+) transport. Thus, we demonstrate that Ag specifically prevented acid-stimulated Na(+) uptake in PNA(-) MR cells in a dose-dependent manner. We also demonstrate rapid (<1 min) and significant inhibition of carbonic anhydrase (CA) by Ag in PNA(-) MR cells, but not in PVC. These data lend further support to the idea of a PNA(-) MR cell type as the primary site for Na(+) uptake in the freshwater (FW) gill phenotype of rainbow trout. Moreover, these findings provide support for the importance of intracellular protons in regulating the movement of Na(+) across the apical surface of the fish gill.

The consequences of reversible gill remodelling on ammonia excretion in goldfish (Carassius auratus)

Goldfish acclimated to cold water (e.g. 7°C) experience a marked reduction in functional lamellar surface area owing to the proliferation of an interlamellar cell mass (ILCM), a phenomenon termed gill remodelling. The goal of the present study was to assess the consequences of the reduced functional surface area on the capacity of goldfish to excrete ammonia. Despite the expected impact of ambient temperature on functional surface area, fish acclimated to 7°C and 25°C exhibited similar rates of ammonia excretion (J(net,amm)); the Q₁₀ values for fed and starved fish were 1.07 and 1.20, respectively. To control for possible temperature-related differences in rates of endogenous ammonia production, J(net,amm) was determined at the two acclimation temperatures after loading fish with 1.12 μmol g₋₁ of NH₄Cl. In the 3 h post-injection period, J(net,amm) was elevated to a greater extent in the 25°C fish. To estimate the potential contribution of increased ventilation and cardiac output to ammonia clearance in the warmer fish, the ammonia loading experiment was repeated on the 7°C fish immediately after they were exercised to exhaustion. The rate of excretion of ammonia was significantly increased in the exercised 7°C fish (presumably experiencing increased ventilation and cardiac output for at least some of the measurement period) suggesting that differences in external and internal convection may at least partially explain the enhanced capacity of the 25°C fish to clear the ammonia load. To more specifically assess the contribution of the different functional surface areas on the differing rates of ammonia clearance at the two acclimation temperatures, the 7°C fish were exposed for 7 days to hypoxia (P(O₂)=10 mmHg=1.33 kPa), a treatment known to cause the disappearance of the ILCM. The results demonstrated that the hypoxia-associated loss of the ILCM was accompanied by a significant increase in the rate of ammonia clearance in the 7°C fish when returned to normoxic conditions. To determine whether compensatory changes in the ammonia transporting proteins might be contributing to sustaining J(net,amm) under conditions of reduced functional lamellar surface area, the relative expression and branchial distribution of four Rh proteins were assessed by western blotting and immunocytochemistry. Although the relative expression of the Rh proteins was unaffected by acclimation temperature, there did appear to be a change in the spatial distribution of Rhag, Rhbg and Rhcg1. Specifically, these three Rh proteins (and to a lesser extent Rhcg2) appeared to localize in cells on the outer edge of the ILCM that were enriched with Na(+)/K(+)-ATPase. Thus, we suggest that despite the impediment to ammonia excretion imposed by the ILCM, goldfish acclimated to 7°C are able to sustain normal rates of excretion owing to the redistribution of ammonia transporting cells.

Ammonia and urea excretion in the Pacific hagfish Eptatretus stoutii: Evidence for the involvement of Rh and UT proteins

The nature of ammonia and urea excretion was examined in the Pacific hagfish (Eptatretus stoutii), which, under resting conditions, excreted similar quantities of nitrogen as either ammonia or urea. In the presence of high external ammonia (HEA) concentrations, ammonia was taken up at high rates and then excreted at similarly high rates upon return to normal water. However, although elevated by HEA, plasma ammonia levels were maintained at approximately 1-4 μmolNg⁻¹, reflecting time-dependent decreases in the rates of ammonia uptake, the possible conversion of ammonia to urea, and the potential active excretion of ammonia against a gradient. Internal injections of NH₄Cl caused marked increases in the rate of ammonia excretion and a delayed increase in urea excretion that may have resulted from increasing urea levels in the plasma. Conversely, when the rate of urea excretion was reduced in the presence of 0.1 mM phloretin, ammonia excretion was significantly elevated. Rates of urea excretion were initially increased by approximately 1000-fold following internal urea injections while the presence of high external urea levels (5-100 mM final concentration) resulted in associated linear increases in plasma urea levels. Using hagfish skin mounted in Ussing chambers, the rate of diffusion of ammonia across the skin exceeded that of urea by approximately four times when equivalent gradients were imposed. Based on western blotting and immunocytochemistry, hagfish gill appears to possess Rh proteins (Rhag, Rhbg and Rhcg1) and urea transporter proteins. Despite the tolerance of hagfish to high levels of ammonia and urea, it is suggested that the presence of ammonia and urea transporter proteins may be required during the period of time hagfish spend in burrows or while feeding, when conditions of high ammonia and/or urea might be encountered.

Ammonia production, excretion, toxicity, and defense in fish: a review

Many fishes are ammonotelic but some species can detoxify ammonia to glutamine or urea. Certain fish species can accumulate high levels of ammonia in the brain or defense against ammonia toxicity by enhancing the effectiveness of ammonia excretion through active NH4+transport, manipulation of ambient pH, or reduction in ammonia permeability through the branchial and cutaneous epithelia. Recent reports on ammonia toxicity in mammalian brain reveal the importance of permeation of ammonia through the blood-brain barrier and passages of ammonia and water through transporters in the plasmalemma of brain cells. Additionally, brain ammonia toxicity could be related to the passage of glutamine through the mitochondrial membranes into the mitochondrial matrix. On the other hand, recent reports on ammonia excretion in fish confirm the involvement of Rhesus glycoproteins in the branchial and cutaneous epithelia. Therefore, this review focuses on both the earlier literature and the up-to-date information on the problems and mechanisms concerning the permeation of ammonia, as NH(3), NH4+ or proton-neutral nitrogenous compounds, across mitochondrial membranes, the blood-brain barrier, the plasmalemma of neurons, and the branchial and cutaneous epithelia of fish. It also addresses how certain fishes with high ammonia tolerance defend against ammonia toxicity through the regulation of the permeation of ammonia and related nitrogenous compounds through various types of membranes. It is hoped that this review would revive the interests in investigations on the passage of ammonia through the mitochondrial membranes and the blood-brain barrier of ammonotelic fishes and fishes with high brain ammonia tolerance, respectively.

Inhibition of glutamine synthetase during ammonia exposure in rainbow trout indicates a high reserve capacity to prevent brain ammonia toxicity

Glutamine synthetase (GSase), the enzyme that catalyses the conversion of glutamate and ammonia to glutamine, is present at high levels in vertebrate brain tissue and is thought to protect the brain from elevated ammonia concentrations. We tested the hypothesis that high brain GSase activity is critical in preventing accumulation of brain ammonia and glutamate during ammonia loading in the ammonia-intolerant rainbow trout. Trout pre-injected with saline or the GSase inhibitor methionine sulfoximine (MSOX, 6 mg kg(-1)), were exposed to 0, 670 or 1000 micromol l(-1) NH(4)Cl in the water for 24 and 96 h. Brain ammonia levels were 3- to 6-fold higher in ammonia-exposed fish relative to control fish and MSOX treatment did not alter this. Brain GSase activity was unaffected by ammonia exposure, while MSOX inhibited GSase activity by approximately 75%. Brain glutamate levels were lower and glutamine levels were higher in fish exposed to ammonia relative to controls. While MSOX treatment had little impact on brain glutamate, glutamine levels were significantly reduced by 96 h. With ammonia treatment, significant changes in the concentration of multiple other brain amino acids occurred and these changes were mostly reversed or eliminated with MSOX. Overall the changes in amino acid levels suggest that multiple enzymatic pathways can supply glutamate for the production of glutamine via GSase during ammonia exposure and that alternative transaminase pathways can be recruited for ammonia detoxification. Plasma cortisol levels increased 7- to 15-fold at 24 h in response to ammonia and MSOX did not exacerbate this stress response. These findings indicate that rainbow trout possess a relatively large reserve capacity for ammonia detoxification and for preventing glutamate accumulation during hyperammonaemic conditions.

Rh glycoprotein expression is modulated in pufferfish (Takifugu rubripes) during high environmental ammonia exposure

Rhesus (Rh) protein involvement in ammonia transport processes in freshwater fish has received considerable attention; however, parallel investigations in seawater species are scant. We exposed pufferfish to high environmental ammonia (HEA; 1 and 5 mmol l–1 NH4HCO3) and evaluated the patterns of ammonia excretion and gill Rh mRNA and protein expression. Gill H+-ATPase, NHE1, NHE2, NHE3, Na+/K+-ATPase (NKA), Na+/K+/2Cl– co-transporter (NKCC1) mRNA, H+-ATPase activity, NKA protein and activity, were also quantified. Activation of NKA by NH4+ was demonstrated in vitro. The downregulation of Rhbg mRNA and simultaneous upregulations of Rhcg1, H+-ATPase, NHE3, NKA, NKCC1 mRNA, H+-ATPase activity, and NKA protein and activity levels suggested that during HEA, ammonia excretion was mediated mainly by mitochondria-rich cells (MRCs) driven by NKA with basolateral NH4+ entry via NKA and/or NKCC1, and apical NH3 extrusion via Rhcg1. Reprotonation of NH3 by NHE3 and/or H+-ATPase would minimise back flux through the Rh channels. Downregulated Rhbg and Rhag mRNA observed in the gill during HEA suggests a coordinated protective response to minimise the influx of external ammonia via the pavement cells and pillar cells, respectively, while routing ammonia excretion through the MRCs. Exposure to hypercapnia (1% CO2 in air) resulted in downregulated gill and erythrocyte Rhag mRNA. Surprisingly, Rhag, Rhbg, Rhcg1 and Rhcg2 proteins responded to both hypercapnia and HEA with changes in their apparent molecular masses. A dual NH3/CO2 transport function of the pufferfish Rh proteins is therefore suggested. The results support and extend an earlier proposed model of pufferfish gill ammonia excretion that was based on immunolocalisation of the Rh proteins. Passive processes and/or Rhbg and Rhcg2 in the pavement cells may maintain basal levels of plasma ammonia but elevated levels may require active excretion via NKA and Rhcg1 in the MRCs.