Regulatory, Integrative and Comparative Physiology

Effects of freshwater and saltwater adaptation and dietary salt on fluid compartments, blood pressure, and venous capacitance in trout

Kenneth R. Olson, Todd M. Hoagland


Trout are of interest in defining the relationship between fluid and salt balance on cardiovascular function because they thrive in freshwater (FW; volume loading, salt depleting), saltwater (SW; volume depleting, salt loading), and FW while fed a high-salt diet (FW-HS; volume and salt loading). The effects of chronic (>2 wk) adaptation to these three protocols on blood volume (51Cr red cell space), extracellular fluid volume (99mTc-diethylene triaminepenta-acetic acid space), arterial (dorsal aortic; PDA) and venous (ductus Cuvier; Pven) blood pressure, mean circulatory filling pressure (zero-flow Pven), and vascular capacitance were examined in the present study on unanesthetized rainbow trout. Blood volume, extracellular fluid volume, PDA, Pven, and mean circulatory filling pressure progressively increased in the order SW < FW < FW-HS. Vascular capacitance in SW fish appeared to be continuous with the capacitance curve of FW fish and reflect a passive volume-dependent unloading of the venous system of FW fish. Vascular capacitance curves for FW-HS fish were displaced upward and parallel to those of FW fish, indicative of an active increase in unstressed blood volume without any change in vascular compliance. These studies are the first in any vertebrate to measure the relationship between fluid compartments and cardiovascular function during independent manipulation of volume and salt balance, and they show that volume, but not salt, balance is the primary determinant of blood pressure in trout. They also present a new paradigm with which to investigate the relative contributions of water and salt balance in cardiovascular homeostasis.

  • fish
  • unstressed blood volume
  • mean circulatory filling pressure
  • osmoregulation
  • ionoregulation

the pioneering studies of Goldblatt et al. (13), Guyton et al. (15), Dahl and Heine (8), and others clearly showed that salt and water balance were primary effectors of blood pressure in mammals. Although a direct effect of salt balance on blood pressure has been suggested (2), it is generally accepted that fluid volume is the primary effector of blood pressure and that salt balance is regulated to maintain fluid volume. This is not surprising, as water loss in the urine or through evaporation is a chronic problem for terrestrial vertebrates, and, because water is not actively transported, it can only be retained if osmotically coupled to an osmolyte, usually salt. However, the inexorable coupling of water and salt has been problematic in understanding the pathophysiology of blood pressure regulation (2, 19, 35).

The relative impact of volume regulation and salt balance on blood pressure regulation in fish may be more amenable to experimentation. Saltwater (SW) fish live in an environment with a osmolarity around 1,000 mosM, whereas environmental osmolarity for freshwater (FW) fish is usually <1 mosM. These environments impose high, but opposite, transepithelial osmotic and ionic gradients on fish, and it is likely that this places considerable strain on the cardiovascular system. Trout and other euryhaline fish show a remarkable ability to adapt to either FW or SW and yet maintain plasma osmolarity close to 300 mosM. However, it is not clear whether water or salt balance takes precedence during the transition between these environments or after long-term adaptation. In fact, arguments have been presented for the primacy of either salt (37, 39) or volume (10, 22, 25, 36) regulation by fish.

Takei et al. (37) proposed that body fluid regulation in fish is considerably different from that of mammals due to the almost complete absence of gravity in the former and the differences in salt and water availability in SW, FW, and terrestrial environments. According to their hypothesis, ion regulation supersedes water balance in fish. Thus sodium excretion in SW and sodium reabsorption in FW are the most tightly regulated parameters, while sodium and water are coupled in mammals. The obvious implication is that colonization of land necessitated a reprioritization of salt and water homeostasis.

Our laboratory (10, 22, 25) has proposed that, although salt balance is important, the primary concern for both SW and FW fish is volume regulation, as it is in mammals. This argument is founded on the numerous fundamental similarities in piscine and mammalian cardiovascular systems (25) and can be supported by an alternative interpretation of the evidence presented by Takei et al. (37) regarding the physiology of fish in SW and FW, as follows. In hyperosmotic SW, fish passively lose water and gain salt (reviewed in Ref. 21). The hallmark response of SW fish is increased drinking of hypertonic medium to restore volume and a secondary excretion of salt by the gills to correct the salt balance. Thus SW fish incur an even greater salt load to replenish lost fluids. Conversely, in hypoosmotic FW, fish passively gain water and lose salt (21). Urine output is increased in FW fish, and, because the fish urinary tract is limited in its capacity to conserve salt, there is a net loss of salt in the urine, and this, as well as diffusive salt loss across the gills, is replenished by active branchial absorption (21). This increase in urine output, despite even greater salt loss, suggests to us that the hyponatremia is adjusted secondarily to the hypervolemia.

The effects of SW and FW on salt balance have been examined in considerable detail in many euryhaline fish (cf. Ref. 21); however, there are no comprehensive studies of the effects of these conditions on fluid compartments (blood and extracellular volumes) and the relationship between these fluid compartments and cardiovascular homeostasis (22). In the present studies, we added a third paradigm by feeding FW fish a high-salt diet (FW-HS), which is also well-tolerated by trout (33). Thus, with these three treatments, we produced fish that were potentially volume depleted and salt loaded (SW), volume loaded and salt depleted (FW), and volume loaded and salt loaded (FW-HS). This enabled us to independently manipulate salt and volume load noninvasively and under conditions in which the fish thrive. We then measured vascular and extravascular fluid volumes (ECFV), arterial and venous pressures (Pven), mean circulatory filling pressure (MCFP), and venous capacitance in unanesthetized fish. Stressed and unstressed blood volumes (USBV) and venous compliance were obtained from the capacitance curves and allowed further evaluation of venous function. To our knowledge, these are the first studies in any vertebrate to independently manipulate water and salt balance under simulated natural conditions and examine their effects on fluid compartments and cardiovascular homeostasis.



Rainbow trout (Oncorhynchus mykiss, mixed Kamloops strain; 0.4–0.7 kg) of both sexes were purchased from a hatchery in northern Michigan and kept in circulating 2,000-liter tanks at 12–14°C and under 12:12-h light-dark cycles. Fish were fed up to 48 h before experimentation. Because fluid compartments vary somewhat with different stocks of fish and with season, control fish were examined concurrently with each group of experimental fish. Experiments were approved by the Indiana University School of Medicine Institutional Animal Care and Use Committee.

SW adaptation.

A maximum of 20–25 trout were adapted to SW (1,000 mosM) at one time in a 500-liter Instant Ocean Culture System (model WM-500, Aquarium Systems, Eastlake, OH) using synthetic sea salts supplied by the manufacturer. The fish were initially adapted to 300 mosM SW for 3–5 days, and the osmolarity was gradually increased to full-strength SW over the next 2 wk. They were maintained at this salinity for a minimum of an additional 2–3 wk before experimentation. Temperature, pH, osmolarity, and total ammonia were measured daily and, with the exception of osmolarity, were not different from FW.

HS diet.

Rainbow trout (0.4–0.7 kg) of either sex were randomly sorted into two groups of 50 and placed in separate 750-liter flow-through tanks in aerated well water (12 ± 1°C). One group was fed a control diet (1% NaCl; FW), and the other group a HS diet containing 12% NaCl (Zeigler Brothers, Gardners, PA) at 3% total body weight/day, for a minimum of 3 wk before experimentation (FW-HS). This HS diet feeding regime has been shown by Salman and Eddy (32) to increase gill chloride cell numbers and Na+-K+-ATPase activity in FW rainbow trout. Fish were fed using a pair of automated vibrational feeders (AF7 Fish Feeder, 3–12 V, Sweeney Enterprises, Boerne, TX), with one feeder per treatment tank with the appropriate pelleted diets in each. These vibrational feeders were connected to a common rheostat (Aquatic Ecosystems, Apopca, FL), which sent signals to dispense the food every 4 h, 24 h/day. As fish were removed from these tanks, the feeders were recalibrated to continually deliver their respective diets at 3% of the total biomass in each tank per day. A 12:12-light-dark photoperiod was maintained.


Methods for cannulation of the dorsal aorta and ductus Cuvier have been described in detail (23). Trout were anesthetized in benzocaine (ethyl-p-aminobenzoate; 1:12,000 wt/vol) before surgery. The dorsal aorta was cannulated percutaneously through the roof of the buccal cavity with heat-tapered polyethylene tubing (PE-60); the gills were not irrigated during this brief (<1 min) procedure. Thereafter, gills were continuously irrigated with 4°C aerated water containing 1:24,000 wt/vol benzocaine during placement of the venous cannula and ventral aorta occluder. Only the dorsal aorta was cannulated for volume measurement studies.

The pericardial cavity was exposed with a midline ventral incision, and the right horn of the ductus Cuvier was nonocclusively cannulated with 5-cm-long, 0.51-mm-inner diameter silicone tubing (Dow Corning veterinary grade; Konigsberg Instruments, Pasadena, CA). The free end of the cannula was connected to 60 cm of PE-90. All cannulas were filled with heparinized saline (100 USP units/ml heparin in 9.0 g/l NaCl) and connected to Gould P23 pressure transducers. An inflatable vascular occluder was fitted around the distal bulbus-proximal ventral aorta, and the venous cannula and occluder tubing were exteriorized and secured to the fish with cyanoacrylate gel (super glue gel). Details on the construction of the occluder can be found in Zhang et al. (41). The ventral incision was closed with interrupted silk sutures and sealed with a gel cyanoacrylate glue. The fish were revived and placed in black plastic tubes and placed in individual 3-liter aerated aquaria. A four-way stopcock in the dorsal aortic cannula served as the site for blood infusion and withdrawal. Experiments were conducted 24 h after surgery.

Analog pressure signals were recorded with Hewlett-Packard 7853A patient monitors (Palo Alto, CA). Digitized signals of pressure and flow were collected at 0.1-s intervals, and 1-s averages were stored on computer. The pressure transducers were calibrated with a water manometer. Resting pressures were monitored for 1–2 h before experimentation to ensure stability; control parameters were recorded for a minimum of 5 min before blood volume manipulation.

Fluid volume.

Blood volume and ECFV were determined using previously published indicator dilution techniques (3), modified by substituting technetium-99m-diethylene triaminepenta-acetic acid (99mTc-DTPA) for 60Co-EDTA, as 60Co was no longer available. Briefly, red blood cells were collected from donor fish, washed in five volumes of saline (0.9% NaCl), centrifuged, and resuspended; the wash was repeated five times. They were then incubated on a rotary mixer for at least 4 h at 12°C with 7.4 × 105 Bq/ml (20 μCi/ml) 51Cr as sodium chromate (Amersham, Arlington Heights, IL). The cells were again washed five times and resuspended in saline. 99mTc-DTPA was purchased locally (Spectrum Pharmacy, Mishawaka, IN) and used to measure ECFV (6). Due to its short half-life (6 h), 99mTc-DTPA was added to the red cells to give an activity ratio of ∼5 μCi 99mTc-DTPA per 1 μCi of 51Cr. The hematocrit of the dual-labeled blood was adjusted to ∼33 with saline before injection.

To account for excretion of radiolabeled indicators, fish were placed in individual containers in a cold room (12–14°C) after cannulation, and experiments were performed the following day. Well water was continuously circulated for FW fish until 30 min before injection of the radiolabeled blood, at which time flow was stopped. SW was changed every 4–6 h until 30 min before injection. Duplicate samples of the labeled blood were removed immediately before injection to serve as reference, and hematocrit was determined from additional duplicate samples. One-half milliliter was injected into each fish via the dorsal aorta cannula, which was flushed with 0.4 ml of saline (twice the volume of the cannula and stopcock). The experiment was terminated at 4 h, which is sufficient for complete mixing of both intravascular and extravascular indicators in trout (3). One milliliter of blood was drawn from the dorsal aorta cannula, and duplicate aliquots were saved for radioactivity analysis and hematocrit. The remaining sample was centrifuged, and the plasma was stored at −80°C for later analysis of electrolytes. One milliliter of Nembutal (50 mg/ml) was injected into the dorsal aorta, and, after 1 min of circulation, a hemostat was placed over the isthmus to clamp the ventral aorta to prevent further mixing. The carcass was then quick-frozen in an ethanol-water-dry ice slurry and stored at −60°C for subsequent tissue analysis. Ten milliliters of aquarium water were also collected and counted for radioactivity to correct for loss in the urine. 51Cr did not appear in the aquarium water, indicating that blood loss, if any, was minimal.

Approximately 0.5- to 1.5-g samples of 29 tissues were removed from each fish, weighed, and counted for radioactivity on a Beckman 4000 gamma counter. Tissues sampled were fins (pectoral, pelvic, dorsal, caudal), skeletal muscle (anterio-dorsal white, midventral white, red), skin, gill arches (I, II, III, IV), spleen, liver, gall bladder, esophagus, stomach, cecum, intestine (anterior, posterior), fat (visceral), swim bladder, kidney (anterior, posterior), eye, brain, skull, operculum, and urinary bladder. The samples were then dried at 90°C until weight stabilized and reweighed to determine water content.

Whole body and tissue spaces for 51Cr-red cells and 99mTc-DTPA were determined using standard indicator dilution relationships. Blood volume was calculated from the red cell space divided by 1 − hematocrit. Whole-animal 99mTc-DTPA space was corrected for urine loss based on the appearance of 99mTc in the ambient water.

Plasma electrolytes.

Blood was collected from 24 control and 21 HS diet trout and centrifuged, and the plasma was analyzed for Na+, K+, Cl, Ca2+, and Mg2+ by a local medical laboratory using ion-selective electrodes.

Vascular capacitance.

Vascular capacitance curves were constructed from measurements on conscious trout in vivo using the ventral aorta occlusion method to establish zero-flow cardiac output, as described previously (41). Fish were placed in individual 3-liter aerated aquaria, and FW or SW was continuously circulated from the respective holding tank. Dorsal aortic blood pressure (PDA) and Pven were measured in unanesthetized fish before and during occlusion of the ventral aorta. Inflation of the occluder (∼5 s duration) produced a rapid decrease in PDA and increase in Pven. At this point, Pven was assumed to be equivalent to MCFP. The occlusion was then released, and the blood pressure returned to normal within two to three beats. Vascular capacitance curves were obtained by measuring MCFP at 80, 90, 100, 110, and 120% of normovolemic blood volume. Heparinized whole blood from a donor fish was used for volume expansion, and all volume adjustments were made via the dorsal aorta cannula within 30 s of occlusion, and blood volume was restored within 30 s after occlusion. The order of volume withdrawal or addition was randomized, and the interval between each volume perturbation, during which the fish were normovolemic, was at least 15 min. Thus the fish were exposed to hyper- or hypovolemia for only a short portion of the experiment. Only one capacitance curve was obtained from each fish.

Vascular compliance and USBV were determined over the range of three blood volume intervals, 80–100, 90–110, and 100–120%, by regression analysis of the three consecutive pressure-volume data points within each interval. Because we did not know the actual blood volume when these experiments were conducted, we assumed a blood volume of 35 ml/kg body wt and added or withdrew 3.5 and 7 ml/kg body wt. The actual percent blood volume added or withdrawn was later determined from the volume measurements in FW, SW, and FW-HS fish (Table 1), and these values were used for Figs. 1 and 2. In the capacitance curves (Figs. 1C and 2C and Fig. 3, B and D), MCFP by convention was treated as the independent variable, and the slope of the resultant volume-pressure line (Δvolume/Δpressure) was, therefore, equal to vascular compliance. The intercept of this line with the blood volume axis at MCFP = 0 was assumed to be the percentage of the total blood volume in the unstressed compartment (30). This percentage, multiplied by the actual blood volume, equals the predicted USBV. The product of percent blood volume times actual blood volume in the three groups of fish permitted conversion of vascular compliance and USBV into actual volumes, i.e., ml·mmHg−1·kg−1 and ml/kg, respectively.

Fig. 1.

Effects of freshwater (FW; circles; n = 9) and saltwater (SW; triangles; n = 12) adaptation on dorsal aortic blood pressure (PDA; A), central venous pressure (Pven; B), and vascular capacitance (C). Blood volume was manipulated by infusion or withdrawal of blood in increments of 3.5 or 7 ml/kg body wt and expressed as a percentage of total initial blood volume (Table 1). Solid symbols represent values obtained during uninterrupted cardiac output; open symbols represent values obtained during zero-flow cardiac output. Dotted lines in C are regression lines for mean circulatory filling pressure (MCFP) between 90 and 100% blood volume. Their slope is equivalent to vascular compliance, and the blood volume intercept at zero MCFP is the unstressed volume. Values are means ± SE. All blood pressures in SW fish are significantly lower than their FW counterparts.

Fig. 2.

Effects of FW (circles; n = 6) and FW, high-salt diet (triangles; n = 6) on PDA (A), Pven (B), and vascular capacitance (C). Blood volume was manipulated by infusion or withdrawal of blood in increments of 3.5 or 7 ml/kg body wt and expressed as a percentage of total initial blood volume (Table 1). Solid symbols represent values obtained during uninterrupted cardiac output; open symbols represent values obtained during zero-flow cardiac output. Dotted lines in C are regression lines for MCFP between 90 and 100% blood volume. Their slope is equivalent to vascular compliance, and the blood volume intercept at zero MCFP is the unstressed volume. Values are means ± SE.

Fig. 3.

Relationship between actual blood volume (ml/kg body wt; B and D) and Pven or vascular capacitance (A and C) in FW (circles) and SW-adapted (triangles) trout (A and B) and FW (circles) and high-salt diet (inverted triangles) trout (C and D). Solid symbols, values obtained during uninterrupted cardiac output; open symbols, values obtained during zero-flow cardiac output. Dashed lines in B and D are regression lines for MCFP between the three middle blood volumes (equivalent to 90–120% blood volume in Figs. 1 and 2). Values are means ± SE; n = 9 (FW; A and B), 12 (SW; A and B), and 6 (FW and high-salt diet; C and D).

View this table:
Table 1.

Effects of freshwater/saltwater adaptation and high-salt diet on dorsal aorta and central venous blood pressure, mean circulatory filling pressure, blood volume, and extracellular fluid volume in unanesthetized trout


Comparisons were made with Student's t-test or repeated-measures ANOVA. Significance was assumed at P ≤ 0.05. Values were expressed as means ± SE, or means + SE in many figures.


FW-SW adaptation.

SW adaptation decreased PDA and Pven, MCFP, and whole body blood volume and ECFV (Table 1). SW adaptation also significantly decreased red cell space in pectoral and pelvic fins, skin, gill arches I and II, and gall bladder; red cell space was not significantly lower in other tissues (data not shown). ECFV was significantly lower in white (anterio-dorsal and midventral) skeletal muscle, skin, stomach, swim bladder, and urinary bladder, whereas it was significantly higher in spleen, liver, anterior and posterior kidney, and eye of SW-adapted fish. Total tissue water was lower in caudal fin, white (anterio-dorsal and midventral) skeletal muscle, gill arches I, II, and III, gall bladder, esophagus, stomach, swim bladder, and urinary bladder and significantly higher in cecum and anterior intestine.

PDA was consistently lower in SW-adapted fish when blood volume was increased or decreased by ∼20% from 100% (control) blood volume (Fig. 1A). Similarly, Pven was lower in the SW group when expressed as a percentage of control blood volume (Fig. 1B). Blood withdrawal did not significantly affect PDA in either FW or SW fish. Blood infusion did not affect PDA in SW fish, and only 20% volume expansion significantly increased PDA in FW fish. Pven increased around sixfold over the range of blood volume perturbation (Fig. 1B). During zero-flow conditions (ventral aortic occlusion), PDA decreased by ∼8 mmHg in both FW and SW trout and at all blood volumes (Fig. 1A); Pven consistently increased by 0.5–1.0 mmHg (Fig. 1B).

SW fish had significantly greater vascular compliance at 90–110% and 100–120% blood volume and greater unstressed volume at 80–100% blood volume than FW-adapted fish (Fig. 1A and Table 2). There were no significant differences between the two groups of fish in compliance at 80–100% blood volume or unstressed volume at 90–110% and 100–120% blood volume.

View this table:
Table 2.

Effects of freshwater/saltwater adaptation and high-salt diet on unstressed blood volume and vascular compliance in unanesthetized trout

However, when percent blood volume was converted to estimated actual blood volume using blood volume values from Table 1 (i.e., 100% blood volume equals 37.2 and 32.8 ml/kg body wt for FW and SW fish, respectively), the relationship between Pven and blood volume for FW and SW fish at either normal or zero-flow cardiac output appeared to form a continuous line (Fig. 3A), and, at any given blood volume, there was no difference between FW and SW fish. Similarly, when vascular capacitance was expressed in terms of actual blood volume (ml/kg body wt), the capacitance curves for SW and FW fish (Fig. 3B) were nearly superimposed, and, at any blood volume, the MCFP were similar.

HS diet.

Compared with FW trout, FW-HS trout had a significantly higher PDA, Pven, MCFP, and blood volume, whereas their ECFV were not significantly different (Fig. 2, A and B, Table 1). 51Cr-red blood cell spaces in FW-HS trout were significantly greater in liver, fat, anterior intestine, and cecum; all other tissue averages were numerically, but not significantly, greater as well. ECFV in FW-HS fish was greater in urinary bladder and anterior and posterior intestine, but lower in caudal fin, anterior kidney, and gallbladder. Total water was significantly greater in urinary bladder, all red and white muscle, and anterior intestine, and lower in posterior kidney in FW-HS fish (tissue values not shown). Plasma Na+ concentration was slightly, but significantly, greater in HS diet trout (148.8 ± 0.8 vs. 146.8 ± 0.6 mmol/l). There were no differences in plasma K+, Cl, Ca2+, or Mg2+ between HS and control diet trout.

Compared with FW control trout, FW-HS fish had consistently higher PDA and Pven when blood volume was increased or decreased from 80 to 120% (Fig. 2, A and B). Blood withdrawal did not affect PDA in either group, and only infusion of 7 ml/kg body wt blood increased PDA in FW fish. Pven increased around threefold over the range of blood volume perturbation (Fig. 2B). In both groups, zero-flow (ventral aortic occlusion) decreased PDA by ∼8 mmHg (Fig. 2A) and increased Pven by ∼1.0 mmHg (Fig. 2B).

USBV was lower in FW-HS fish when measured between 90 and 110% blood volume. There were no other differences in compliance or unstressed volume between these two groups (Fig. 2C and Table 2).

When expressed in terms of actual blood volume, Pven in FW fish at both normal and zero-flow conditions appeared higher than the respective Pven in HS diet fish at the same blood volume (Fig. 3C). When expressed in terms of actual blood volume, the capacitance curve for the fish fed the HS diet was displaced upward and to the left (Fig. 3D). The unstressed volume of fish fed the HS diet was ∼25% greater than that of the control fish, whereas the compliance was identical.


Fluid and salt balance were independently perturbed in these studies by placing trout in SW (volume depleting and salt loading), FW (volume loading and salt depleting), or feeding FW trout a HS diet (volume loading and salt loading). Compared with FW fish, SW fish had a decreased blood and ECFV and lower arterial, Pven, and MCFP, whereas these variables increased when FW fish were fed a HS diet. The effect of these experiments on the cardiovascular system appeared to be more consistent with the potential perturbation of fluid balance than salt balance, because blood pressure continually increased as fish became more volume loaded (SW < FW < FW-HS), whereas blood pressure of salt-loaded fish was either lower (SW) or higher (FW-HS) than blood pressure of salt-depleted (FW) fish.

These studies confirm previous observations that trout PDA falls when trout are adapted from FW to SW (24) and increases when FW fish are fed a HS diet (5, 9). There is accumulating evidence that both FW-to-SW transfer and HS diet are associated with a salt overload. After transfer from FW to SW or dilute SW (∼2/3 SW) plasma ions or osmolarity increase dramatically (1, 4, 7, 14, 38). After long-term adaptation (days to weeks), they may return to pretransfer levels (4, 7, 12, 20, 38) or remain elevated (14, 17), whereas the osmoregulatory organs retain features characteristic of salt secretion (21). FW-HS trout also experience an elevation in plasma electrolytes for up to 7 h after feeding (34). Although the electrolytes may (5) or may not (18, 28; this study) return to prefeeding levels within 24–48 h, there are physiological responses, such as net sodium efflux (34), increased drinking (28), and changes in gill morphology and enzymes (27, 32), that are consistent with salt-loaded SW trout. However, unlike SW trout, urinary water loss is increased in FW-HS trout (31), consistent with a concomitant volume overload, which was also reflected in the increased blood volume of FW-HS trout (Table 1). Thus salt load does not correlate with blood pressure perturbations observed in these experiments.

Correlation of blood volume with aortic pressure, Pven, and vascular capacitance (Table 1, Fig. 3) suggests that fluid balance has a significant impact on cardiovascular function. It is not clear if the physiological responses (drinking, transbranchial salt flux, or renal excretion) are not quite sufficient to restore blood pressures, or if blood pressures are intentionally offset to help maintain fluid balance (e.g., pressure diuresis in hypertensive FW-HS fish and decreased golmerular filtration and increased tubular reabsorption in SW fish). In the stenohaline, osmo-conforming hagfish, Eptatretus cirrhatus, PDA initially increased within 1 h when fish were transferred from SW to 90% SW and decreased after transfer from SW to 110% SW (11). Pressure in the supraintestinal vein mirrored that of the dorsal aorta, whereas posterior cardinal vein pressure fell in both experimental groups. The changes in PDA and supraintestinal vein pressure were consistent with a passive response to volume loading and volume depletion, as predicted from changes in plasma osmolarity (11). The fall in posterior cardinal vein pressure upon transfer to 90% was interpreted as an attempt to lower blood pressure during the volume expansion, and it suggests that there are attempts, albeit not completely successful, to regulate pressure in response to the volume perturbation. Unfortunately, neither 90% SW nor 110% SW were well tolerated by hagfish for 24 h (11). Thus, although it was not possible to evaluate long-term responses, their (11) results suggest that the hagfish, like the trout, does not have the capacity to fully restore blood pressure when volume status is perturbed.

While ECFV of SW fish was significantly decreased, it was not significantly increased in FW-HS fish (Table 1). The lack of a significant increase in ECFV may be an artifact of the high variability and small sample size, or it may have physiological significance. Trout capillaries appear to be quite permeable to plasma protein (26), and, lacking an appreciable offsetting contribution from plasma oncotic pressure, blood volume would become quite susceptible to blood pressure, unless something limited interstitial compliance. This may be accomplished by the skin. The stratum compactum of fish skin is composed mostly of alternating sheets of parallel collagen fibers that are wrapped in alternating left and right helices around the fish (16). This forms a cylinder wall around the body that “…can strongly resist internal pressures, yet is capable of extensions in any direction except those of the fibers themselves” (16). By resisting distension, this collagen corset would also limit expansion of the ECFV.

The effects of the three protocols on vascular capacitance were also consistent with changes in fluid volume, and, from a semiquantitative perspective, the venous system was either the most affected by alterations in fluid balance, or the most responsive to it. In SW (relative to FW) trout PDA, blood volume and ECFV were all decreased by 12%, whereas Pven was down by 40% and MCFP was down by 36%. In FW-HS fish (relative to FW), PDA was increased by 12%, blood volume increased by 24%, whereas Pven and MCFP were increased by 35 and 33%, respectively. Closer examination of the capacitance curves shows that different mechanisms appear to be involved in volume depletion and volume expansion. When corrected for actual blood volume, vascular capacitance of SW fish appears to overlay that of FW fish, albeit at lower blood volumes and Pven (Fig. 3, A and B). This suggests that the apparent increase in compliance in SW fish (Fig. 1C, Table 2) is an artifact resulting from normalizing the capacitance curve to an assumed blood volume (i.e., 100%) rather than actual blood volume (bespeaking to the danger of this practice). It also suggests that there is no active venous response of SW fish to the hypovolemia, as they merely appear to be operating off of the lower end of the capacitance curve of FW fish. The SW capacitance curve is the portion of the curve where compliance passively increases due to deflation of the veins (40), and Pven are proportionally decreased. However, the capacitance curve for FW-HS fish (Fig. 3D) is clearly shifted upward and parallel relative to the FW curve. This means the USBV (the theoretical blood volume at MCFP = 0) is increased without a change in venous compliance, an indication of an active reduction in venous tone (29). One mechanism whereby this could be accomplished is via decreased sympathetic venous constriction, and this has some support. It has been shown that the sympathetic nervous system exerts tonic stimulation of trout veins (41) and that trout fed a HS diet experience a chronic decrease in vascular responsiveness to catecholamines (5). Other factors, such as cardiac natriuretic peptides, also appear to contribute to this response (K. Johnson and K. Olson, unpublished observations), and it is likely that other neuroendocrine effectors may also be involved. Because this active change in vascular capacitance was found in FW-HS, but not SW trout, it appears to be driven by volume expansion, not salt loading. Interestingly, the hagfish also appears to respond to volume expansion better than volume depletion (11).


Much of our current knowledge of the effects of salt and water balance on blood pressure has been derived from studies on mammals. However, like all terrestrial vertebrates, mammals continually experience water loss due to evaporation and obligatory urine output and, therefore, osmotically couple salt and water balance for long-term regulation of fluid volume and blood pressure. In this study, we examined an alternative vertebrate model, the rainbow trout. Because these euryhaline fish thrive in FW, SW, or in FW while fed a HS diet, they provide us with a unique opportunity to independently, and physiologically, regulate salt and water balance. As fish are also the progenitors of the vertebrate cardiovascular system, this virtually untapped model offers a variety of unique experimental opportunities with which to further examine extant cardiovascular regulatory mechanisms, or provide evidence for novel ones.


This work was supported in part by National Science Foundation Grant IBN-0235223.


The authors express gratitude to Dr. D. Duff and D. Hambleton for technical assistance.

Present address of T. M. Hoagland: Department of Anatomy and Neurobiology, Boston University School of Medicine, Boston, MA 02118.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract