Experiments were conducted on rainbow trout to determine the impact of dietary salt on arterial blood pressure. After 4–6 wk, fish fed a salt-enriched diet exhibited a 37% elevation of dorsal aortic pressure (from 23.8 ± 1.2 to 32.6 ± 1.4 mmHg) and an 18% increase in ventral aortic pressure (from 33.0 ± 1.5 to 38.9 ± 1.3 mmHg). The hypertension presumably reflected the increase in cardiac output (from 31.0 ± 0.8 to 36.4 ± 2.2 ml·min−1·kg−1) because systemic and branchial resistances were statistically unaltered by salt feeding. The chronic hypertension was associated with a decrease in the pressor responses of the systemic vasculature to catecholamines and hypercapnia in the salt-fed fish. The reduction in α-adrenergic responsiveness of the systemic vasculature is consistent with desensitization or loss of functional α-adrenoceptors (α-ARs). In support of this idea, the salt-fed fish exhibited significantly decreased levels of α1D-AR mRNA in the dorsal aorta and the afferent (ABA) and efferent branchial arteries (EBA). In contrast, however, the results obtained from norepinephrine dose-response curves for EBA and ABA vascular rings in vitro did not provide evidence for loss of function of branchial artery α1-ARs in the salt-fed fish. Indeed, the EC50 for the EBA norepinephrine dose-response curve was significantly reduced (from 3.75 × 10−7 to 2.12 × 10−7 M) in the salt-fed fish, indicating an increase in the binding affinity of the α1-ARs.
- blood pressure
- systemic resistance
- salt feeding
- efferent branchial artery
- afferent branchial artery
although controversial, the results of epidemiological studies reveal a weak, but significant relationship, between dietary salt intake and arterial blood pressure (for reviews, see Refs. 1a and 11). The mechanisms underlying salt-induced hypertension in mammals, however, are poorly understood. Generally, the salt-mediated hypertension is related to extracellular fluid or plasma volume expansion and/or to increased total peripheral resistance (1a, 11, 23). Although it is widely accepted that high levels of dietary salt will promote primary hypertension only in those individuals with heightened sensitivity to salt, the underlying mechanisms for this phenomenon remain uncertain. Several factors are thought to be involved, including an increase in sympathoadrenal activity (10, 20), activation of the renin-angiotensin-system (9), and vascular endothelial dysfunction (2).
Total peripheral resistance is controlled by a variety of factors, but arguably the most important determinant is the extent of adrenergic vasomotor tone, which, in turn, is set predominantly by sympathetic stimulation of vascular α1-adrenoceptors (α1-ARs). On the basis of the accumulated evidence of pharmacological and molecular studies (29), it is now accepted that in mammals, α1-ARs comprise three subtypes: α1A, α1B, and α1D. It has been suggested that α1-ARs may be involved in the pathogenesis and maintenance of primary hypertension (8). For example, a high density of α1D-ARs was reported in the resistance vessels of hypertensive rats (22, 27). Furthermore, pharmacological studies of spontaneously hypertensive rats demonstrated that α1D-ARs may be involved in vascular hyperactivity leading to the hypertensive state (8) (for a review, see Ref. 21). More recently, the results of knockout studies revealed a functional role of α1D-ARs in the development of dietary (19) and central (4) salt-induced hypertension in mice. Recently, we provided evidence that α1A- and α1D-ARs are involved in blood pressure regulation in rainbow trout, Oncorhynchus mykiss (3).
Although previous studies have examined aspects of dietary salt loading on renal and branchial function in rainbow trout (15–18), there are no published data on the cardiovascular effects of salt loading in any fish species. Because freshwater fish are hyperosmotic with respect to their environment, they are faced with the continuous osmotic entry of water (largely across the gills) that must be excreted by the kidney. Because this large daily influx of water would be further increased by salt loading, we reasoned that the freshwater rainbow trout might represent a useful vertebrate model to study salt-induced hypertension. Thus, the first goal of this study was to determine whether elevated dietary salt would promote hypertension in rainbow trout with a fully operational kidney. The second goal was to determine whether the salt-induced hypertension was associated with any potentially compensatory adjustments to cardiovascular α1-ARs. Specifically, we hypothesized that long-term hypervolemic hypertension (if occurring in trout) would promote a decrease in the responsiveness of α1-ARs to sympathetic stimulation as a mechanism to limit further increases in blood pressure. This was tested by comparing the systemic and branchial cardiovascular responses of control and salt-fed fish to exogenous catecholamines or external hypercapnia [treatments known to stimulate the α1-ARs of the systemic vasculature (12)] and by assessing in selected blood vessels the levels of mRNA for α1A- and α1D-ARs and their responsiveness in vitro to adrenergic stimulation.
MATERIALS AND METHODS
Rainbow trout (Oncorhynchus mykiss), weighing ∼400–700 g, were obtained from Linwood Acres Trout Farm (Campellcroft, ON, Canada). Fish were transported to the University of Ottawa Aquatic Care Facility and were maintained in fiberglass holding tanks (1,275 liter) containing well-aerated, dechloraminated City of Ottawa tap water at 13°C. Fish were subjected to a constant 12:12-h light-dark photoperiod and fed five times a week with commercial trout pellets [composed of: 41.0% crude protein (minimum); 11.0% crude fat (minimum); 3.5% crude fiber (maximum); 1.0% calcium (actual); 0.85% phosphorus (actual); 0.45% sodium (actual); 6,800 IU/kg vitamin A (minimum); 2,100 IU/kg vitamin D (minimum); 80 IU/kg vitamin E (minimum); 200 IU/kg vitamin C (minimum); Martin Mills, Tavistock, ON, Canada]. Receptor characterization experiments were conducted between August and October. All experiments were approved by the University of Ottawa Animal Care Protocol Review Committee and conform to the guidelines established by the Canadian Council on Animal Care for the use of animals in research.
Normal and High-Salt Diets
Fish were separated into four groups, and each group was placed into individual tanks (1,275 liter). Two control groups were fed a commercial diet (see above), while the other two were fed a reconstituted diet containing 11% (wt/wt) NaCl (17). The salt diet was prepared by grinding the normal pellets, adding 11% (wt/wt) NaCl and appropriate water, reconstituting the mixture into pellets under pressure (Bizerba, Elpack, Rostrevor, SA, Australia) and then air drying. The amount of the high-salt diet fed to the trout was gradually increased over an 8-day period to reach the final concentration of 11%. The fish were then fed (1% body mass per day) a normal or full (11%) salt-diet for 4–6 wk.
Trout were anesthetized by immersion in an oxygenated benzocaine solution (0.1 g l−1) until voluntary activity ceased. Fish were then transferred to an operating table that enabled continuous irrigation of the gills with the same anesthetic solution. An indwelling cannula [Clay-Adams polyethylene (PE)-50 tubing] was implanted into the dorsal aorta (17a) for measurements of blood pressure. To permit drug injections, the caudal vein was cannulated (PE-50; Clay-Adams, Sparks, MD) in the anterior direction using standard surgical procedures (1). To measure cardiac output, a 3-S or 4-S ultrasonic flow probe (Transonic Systems, Ithaca, NY) was placed around the bulbus arteriosus. The bulbus was exposed by a small (∼1.5 cm) ventral midline incision and removal of the pericardium. Lubricating jelly (K-Y Personal Lubricant; Johnson & Johnson, Skillman, NJ) was used with the flow probe as an acoustic couplant. The caudal cannula and flow probe were anchored to the skin, and all incisions were closed using silk sutures. In a separate series of experiments, the bulbus was cannulated (12) to permit measurements of ventral aortic pressure, and estimates of branchial vascular resistance. In these experiments, the caudal vein was not cannulated, and drugs were injected into bulbus cannula. The tubing and cannulas were flushed with cold heparinized saline (100 IU/ml sodium heparin). Fish were revived by flushing the gills with aerated water until opercular movements became pronounced and regular. The fish were placed in individual opaque acrylic chambers supplied with aerated, flowing water (flow rate > 2.5 l/min), where they were allowed to recover for 24 h prior to beginning an experiment.
Dorsal aortic (PDA) and ventral aortic (PVA) blood pressures were measured using UFI model 1050BP (UFI, Morro Bay, CA) pressure transducers calibrated against a column of water. Cardiac output (Vb) was determined by connecting the ultrasonic flow probe to a small animal blood flow meter (T106, Transonic Systems). The analog signals (PDA and Vb) were converted to digital data and stored by interfacing with a data acquisition system (Biopac Systems, Santa Barbara, CA) using AcKnowledge data acquisition software (sampling rate set at 30 Hz) and a PC. Cardiac frequency (fH) was calculated from the pulsatile Vb recordings using an automatic rate function feature of the data acquisition software. Systemic resistance (RS) was calculated as mean PDA (mmHg)/mean Vb (ml·min−1·kg−1). Branchial vascular resistance (RG) was calculated as (mean PVA − mean PDA)/mean Vb.
Blood samples (0.5 ml, 24 h post-surgery) for Na+ and Cl− analyses were withdrawn prior to beginning any treatments.
Effects of External Hypercapnia
Trout respond to external hypercapnia with a marked increase in RS, owing to α-AR-mediated vasoconstriction (12). To assess the effects of the high-salt diet on this response, trout were exposed to acute hypercapnia. External hypercapnia was achieved by pumping mixtures of CO2 in air (3–10%) through a gas equilibration column to achieve a water Pco2 (PwCO2) of 6–7 mmHg. The mixtures were supplied by a Wosthoff gas-mixing pump (model M301 A/F). Once the PwCO2 had stabilized (within 20 min), the cardiovascular parameters were recorded. Pco2 was measured by pumping water across a CO2 electrode (Cameron Instruments) housed in a thermostatted cuvette. The CO2 electrode was connected to a blood gas meter (BGM 200; Cameron Instruments, Guelph, ON, Canada), and the analog output from the meter was interfaced with a data acquisition system (see above). The CO2 electrode was calibrated prior to each experiment using mixtures of 0.5% and 1.0% CO2 derived from the Wosthoff gas-mixing pump.
In a separate experimental series, fish were injected (0.6 ml/kg) via the caudal vein or bulbus cannula over a 30-s interval with a catecholamine cocktail containing equimolar concentrations (2.5 × 10−5·mol·l−1 in 150 mmol/l NaCl) of norepinephrine and epinephrine. Before (30 min) injecting the catecholamines, 150 mmol/l NaCl (0.6 ml/kg) was injected as a vehicle control. To ensure that the catecholamine dose was appropriate (i.e., able to distinguish potentially subtle effects of salt feeding on α1-AR responsiveness), several preliminary dose-response curves were constructed, demonstrating that the selected dose was eliciting an approximate 50% vasoconstrictory response. Additional preliminary experiments were performed to ensure that the catecholamine injection was not causing downregulation or desensitization of α1-ARs. Serial injections (every 15 min) of the catecholamine cocktail into fish showed that the increases in PDA and RS were unaffected until the 9th injection. The result of this experiment ruled out the possibility that a single catecholamine injection could downregulate α1-ARs and thereby affect the prazosin injection results (see below).
Because the resting blood pressure of the salt-fed fish was significantly higher than the control fish, an additional series of experiments was performed to assess whether the ability of the injected catecholamines to raise blood pressure in the salt-fed fish was constrained by the higher basal blood pressure. In this series, four fish were injected with 0.6 ml/kg of a high dose of an equimolar (2.5 × 10−4 mol/l) cocktail of norepinephrine and epinephrine. A second injection of the same dose was then administered before blood pressure was able to return to preinjection levels.
The α-AR antagonist prazosin (1 mg/ml dissolved in 1 part 150 mmol l−1 NaCl and 1 part methanol) was injected (1 ml/kg) into fish via the caudal vein or bulbus cannula over a 10-min period. The prazosin injection was performed 30 min after the catecholamine injection, at which time all cardiovascular variables had stabilized and returned to baseline levels.
Plasma Sodium and Chloride Assays
Plasma sodium concentration was measured by flame emission spectrophotometry (Varian, model SpectraAA 250 Plus). Plasma chloride concentration was determined by a mercuric thiocyanate spectrophotometric assay method (28).
Isolation of RNA
Using a separate group of similarly treated fish, total RNA was isolated from fresh tissues [including efferent branchial artery (EBA), afferent branchial artery (ABA), coeliacomesenteric artery (CMA), ventral aorta (VA), dorsal aorta (DA) and spleen)] using Trizol reagent (Gibco BRL, Burlington, ON, Canada), according to the protocol provided by the manufacturer. RNA concentrations and quality were verified using spectrophotometry (Eppendorf Biophotometer) and RNA gel electrophoresis. To aid the precipitation of RNA from the small tissue samples, 4 μl of linear acrylamide (2 mg/μl) (Ambion, Austin, TX) was added after the addition of isopropanol; the samples were then placed on dry ice for 30 min, at which point normal extraction proceeded. cDNA was synthesized from 5 μg total RNA using StrataScript reverse transcriptase (Stratagene, La Jolla, CA) and random hexamer primers.
Real-Time PCR Analysis of A1A-AR And A1D-AR mRNA Expression
Real-time PCR analysis was performed using a Stratagene MX-4000 real-time QPCR machine and DNA amplification was followed using SYBR Green (Molecular Probes, Eugene, OR), according to the manufacturer's protocol. α1-ARs (α1A-AR and α1D-AR) and β-actin (internal control) were assayed. To avoid genomic DNA amplification, primers were designed to span a hypothetical intron. On the basis of previous studies of α1-ARs genomic structure, there is a conserved intron region between transmembrane domain (TMD) 6 and TMD 7. The following primers were designed and synthesized (Invitrogen) to yield 170-bp amplicons: α1A-AR, forward: 5′-CTGCGCCTCCTCAAGTTCTC-3′; reverse, 5′-CCCCAGCCAGAAGGTGATCT-3′; α1D-AR, forward: 5′-CTGTCCGTGCGCTTGATGAA-3′; reverse, 5′-AACCCAGCCAGAAGATGACC-3′; β-actin, forward: 5′-CGTCCCAGGCATCAGGGAGT-3′; and reverse: 5′-TCTCCATGTCGTCCCAGTTG -3′.
Initially, the PCR products were cloned to ensure the specificity of the primers. Forty cycles of a two-step PCR protocol were used: 1 × 15 min at 95°C, 40 × (30 s at 95°C, 30 s at 58°C, and 30 s at 72°C), and 1 × 10 min at 72°C. A no template control (cDNA was replaced by water) for each master mix, and a no RT control was included in all QPCR runs. Standard curves for each pair of primers were established by using receptor-specific plasmid DNA as template. The slopes of the standard curves for α1A-AR and α1D-AR were −3.083 and −3.015, respectively, which corresponded to amplification efficiencies of 111 and 114.6%, respectively. Analysis of dissociation (melting) curves provided evidence for a single amplicon being produced in all cases; this was verified by gel electrophoresis. Real-time PCR data were analyzed using the Δ-ΔCt method (14) with β-actin as the reference gene.
Measurement of Plasma Catecholamine Levels
Blood samples (0.6 ml) were withdrawn from the caudal vein cannula before injecting catecholamines, and 2 and 15 min postinjection. Samples were centrifuged immediately (10,000 g for 30 s), and the plasma was removed and flash frozen in liquid N2. Samples were maintained at −80°C before determination of catecholamine levels. Plasma catecholamine levels were measured on alumina-extracted samples by HPLC (Varian, Palo Alto, CA) with electrochemical detection (model 400EC detector; E.G., & G. Parc, Princeton, NJ), according to the basic method of Woodward (25a). 3,4-Dihydroxybenzylamine was used as an internal reference standard in all analyses.
Adrenergic Responsiveness of Afferent and Efferent Branchial Arteries in Vitro
Segments (3–5 mm long) of EBA and ABA were isolated in Ottawa, placed in HEPES buffer (in mM): 145 NaCl, 3 KCl, 0.57 MgSO4·7H2O, 2 CaCl2·2H2O, 5 glucose, 3 HEPES acid, and 7 HEPES Na+ salt, at pH 7.8 at 4°C and transported within 24 h to University of Notre Dame, where experiments were conducted over 2 days. Arteries were mounted on 280-μm-diameter stainless-steel wire hooks and suspended in 5-ml water-jacketed smooth muscle baths filled with HEPES-buffered trout saline (14°C) and gassed with room air. The bottom hooks were stationary; the upper ones were connected to Grass model FT03C force-displacement transducers (Grass Instruments, West Warwick, RI). Tension was measured on a Grass Model 7E or 7F polygraph (Grass Instruments). Data were archived on a computer at 1 Hz using Softwire software (Measurement Computing, Middleboro, MA). The chart recorders and software were calibrated before each experiment. With this system, a change in tension equivalent to 5 mg could be detected. Baseline (resting) tension of 500 mg was applied to the vessel rings for 0.5–1 h before experimentation. The vessel rings were then contracted with 80 mmol/l KCl, washed twice, and resting tension reestablished. This process was repeated, and resting tension was reestablished for a minimum for 30 min before further experimentation. These two “pretreatment” contractions have been established by this laboratory to be necessary to produce optimal responses thereafter (K. R. Olson, unpublished observations). The cumulative dose-response characteristics of norepinephrine were examined in otherwise unstimulated EBA and ABA vessel rings from control trout and fish fed a high-salt diet. Vessel rings were exposed to norepinephrine concentrations beginning at 0.1 nmol/l and then incrementally increased to 0.2, 0.5, 1, 2, 5 nmol/l, and so forth until no further contractions were achieved with increasing concentration. Vessel rings were then washed twice with fresh HEPES-buffered trout saline and washed a third time 30 min later. Baseline tension was then restored, and the vessel rings were undisturbed for a minimum of 2 h following the final norepinephrine exposure. The vessel rings were then treated with either 3 μM BMY-7378 (BMY, an α1D-selective antagonist; see Ref. 7) or 0.1 μM RS-17053 (RS, an α1A-selective antagonist; see Ref. 6), creating a total of four treatment groups for both the EBA and ABA: 1) control-BMY; 2) high-salt diet-BMY; 3) control-RS; and 4) high-salt diet-RS. Approximately 20 min after antagonist exposure, norepinephrine dose-response curves were repeated. As an additional control, one group of EBAs (n = 8) collected from trout housed at the University of Notre Dame were subjected to consecutive cumulative norepinephrine dose-response experiments in the absence of any antagonists. Relaxation was calculated as the percentage of force reduction relative to force produced by the second KCl contraction. This will produce negative values if the resting (baseline) tension is decreased by a drug. EC50s were determined for each vessel using a Table Curve (Jandel Scientific, San Rafael, CA).
Data are presented as means ± SE. All statistical tests were performed using SigmaStat (ver. 3) software (SPSS, Chicago, IL). The effects of treatment were assessed by paired or unpaired Student's t-tests, where appropriate. A P value of <0.05 was regarded as significantly different and is shown in figures with an asterisk.
Effects of High-Salt Diet on Basal Cardiovascular Variables and Plasma Catecholamines
The PDA of salt-fed fish (32.6 ± 1.4 mmHg) was ∼37% higher than in the control fish (23.8 ± 1.2 mmHg; Table 1). In a separate experimental series, PVA in salt-fed fish (38.9 ± 1.3 mmHg) was determined to be 18% greater than in control fish (33.0 ± 1.3 mmHg). There were no significant differences in RS (P = 0.07), RG (P = 0.54) or fH (P = 0.66) between the two groups of fish (Table 1); Vb, however, was significantly increased from 31.0 ± 0.8 to 36.4 ± 2.2 (P = 0.034). The concentration of total catecholamines (norepinephrine plus epinephrine) in plasma did not differ between the control (5.5 ± 0.7 nmol/l) and salt-fed (4.1 ± 0.6 nmol/l) fish (Table 1).
Plasma Na+ and Cl− Levels
Plasma Na+ and Cl− concentrations were measured in control fish and fish fed a high-salt diet. Although there were trends for elevated levels of Na+ (139.1 ± 7.1 vs. 124.6 ± 5.3 mmol/l) and Cl− (154.5 ± 10.6 vs. 141.6 ± 4.0 mmol/l) in the fish fed a high-salt diet (Table 1), the differences were not statistically significant (P = 0.124 and 0.247, respectively).
Real-Time PCR Analysis
α1D-AR mRNA expression was examined in six tissues, including the EBA, ABA, CMA, VA, DA, and spleen. Figure 1 shows that in DA, ABA, and EBA, the expression of the α1D-AR gene was significantly lower in fish fed the high-salt diet compared with control fish (51% decrease in DA, 50% in ABA, and 35% in EBA). Although similar trends existed for α1D-AR mRNA expression in the VA, the differences were not statistically significant. Similarly, an apparent increase in α1D-AR mRNA in spleen of the high-salt-fed fish was not statistically significant.
α1A-AR mRNA expression also was examined in the DA, CMA, and EBA, in which levels remained unchanged after salt feeding (data not shown). The ABA was not analyzed because a previous study (X. Chen, S. F. Perry, S. Aris-Brosou, C. Selva, and T. W. Moon, unpublished data) demonstrated an absence of α1A-AR mRNA expression in this tissue.
Fish (n = 6) were exposed to acute increases in ambient Pco2 to achieve (within 20 min) a stable level of hypercapnia of 6.3 ± 0.4 mmHg (0.84 ± 0.05 kPa). In the control fish, hypercapnia was associated with absolute increases in PDA and RS of 11.0 ± 2.6 mmHg and 0.38 ± 0.14 mmHg·ml−1·min−1·kg−1, respectively (Fig. 2). The changes in PDA (3.9 ± 0.7 mmHg) and RS (0.11 ± 0.04 mmHg ml−1·min−1·kg−1) were significantly lower (by ∼70%) in fish fed the high-salt diet (Fig. 2). The changes in Vb were highly variable and thus there were no effects of hypercapnia (regardless of diet) on this variable (data not shown). fH was decreased to a similar extent in the control and salt-fed groups (8.8 ± 2.6 min−1 and 7.9 ± 7.2 min−1, respectively).
Catecholamine and Prazosin Injections In Vivo
The potential role of α1-ARs in salt-induced high BP was assessed by injecting α1-AR agonists and an antagonist into cannulated trout in vivo. The natural catecholamines, epinephrine and norepinephrine, were used as nonspecific α1-AR agonists to determine whether the chronic hypertension associated with salt feeding had affected the capacity of endogenous α1-ARs to promote vasoconstriction. The nonspecific α1-AR antagonist prazosin was used as a tool to determine whether α-adrenergic vascular tone was affected by chronic hypertension.
The injection of the natural catecholamine mixture significantly increased PDA, PVA, and RS in both groups of fish (Fig. 3) without affecting RG (P = 0.93). However, the salt-fed fish displayed markedly smaller changes in PDA (24 vs. 68%) and RS (28 vs. 55%) after catecholamine injection when compared with the control group. Vb and fH were unaffected by catecholamine injection (data not shown). To determine the potential effect of elevated blood pressure itself on influencing the pressor responses to catecholamine injections, fish (n = 4) were injected with a higher dose of catecholamines to raise PDA from 22.7 ± 2.1 to 82.7 ± 1.9 mmHg. Blood pressure was then allowed to fall back to 56.1 ± 0.2 mmHg, at which point the fish were given a second injection of the catecholamine cocktail; blood pressure was increased to 93.8 ± 3.1 mmHg, a value that was statistically greater (P = 0.023) than the maximal blood pressure attained after the first injection.
The injection of prazosin caused similar decreases in PDA and RS in both groups of fish (Fig. 4) and similar increases in RG; Vb, and fH were unaffected (data not shown).
Adrenergic Responsiveness of Afferent and Efferent Branchial Arteries In Vitro
The norepinephrine dose response curves for ABA and EBA are depicted in Fig. 5, and the EC50 values are summarized in Table 2. For the EBA, the EC50 was decreased significantly from 3.75 ± 0.83 mol/l in the control fish to 2.12 ± 0.32 × 10−7 mol/l in the salt-fed fish. The EC50 was significantly increased in both groups by roughly two orders of magnitude in the presence of the α1D-AR antagonist BMY-7378 (Table 2). The effects of the α1A-AR antagonist RS-17053 on increasing EC50 (nine- and fourfold for the control and salt-fed fish, respectively), while statistically significant, were markedly lower than for BMY-7378. The EC50 for the ABA were similar in the normal and salt-fed fish (Table 2). Unlike for the EBA, the two receptor antagonists caused similar increases in the EC50 values.
A first goal of the present study was to develop a piscine model for salt-induced hypertension. The results clearly demonstrated that feeding rainbow trout a diet supplemented with 11% NaCl for 4–6 wk caused a significant rise in arterial blood pressure. To our knowledge, this is the first study to demonstrate salt-induced hypertension in a nonmammalian vertebrate. The second goal of this research was to determine whether the salt-induced hypertension was associated with any potentially compensatory adjustments to cardiovascular α1-ARs. The results demonstrated a reduction in the pressor responses of the systemic vasculature mediated by exogenously administered catecholamines or external hypercapnia, indicating loss of function of systemic vessel α-ΑRs. At the gill, there was a reduction of α1D-AR mRNA in EBA and ABA in the salt-fed fish, yet these vessels exhibited either heightened responsiveness (EBA) or no change in responsiveness (ABA) to adrenergic stimulation. Thus, unlike in mammals in which salt-induced hypertension is, in part, believed to reflect a heightened sensitivity of α1-ARs to sympathetic stimulation (for a review, see Ref. 19), the responses of trout cardiovascular α-ΑRs appears to be more complex and tissue dependent.
A Piscine Model for Salt-Induced Hypertension
Salman and Eddy (17) demonstrated that rainbow trout fed a salt-enriched diet exhibited an increased glomerular filtration rate (GFR). As a possible explanation, these authors proposed that salt feeding was promoting an elevation of blood pressure. The possibility of increased renal blood pressure contributing to elevated GFR is certainly feasible given that in the present study, dorsal aortic blood pressure was increased by ∼37%. The intent of this study was to assess potential adrenergic adjustments associated with prolonged salt-induced hypertension rather than to specifically delineate the underlying mechanism of the hypertension. Nevertheless, it would appear that the increase in blood pressure with salt feeding was the result of concurrent increases in cardiac output and systemic resistance (branchial vascular resistance was unaffected). In mammals with fully functional kidneys, the relationship between dietary salt load and blood pressure remains controversial (11). However, there is general agreement that when occurring in mammals, salt-induced hypertension reflects an increase in systemic resistance (1a). This phenomenon, while also apparent in the present study, was unrelated to increased systemic α-adrenergic vascular tone (see below). The prolonged increase in cardiac output that was demonstrated in this study presumably reflected a volume expansion of the vascular compartment, owing to the movement of water into the NaCl-enriched plasma. Indeed, in a related but separate study, trout fed a similar salt-enriched diet exhibited an increase in blood volume from 33.2 ± 1.7 to 41.1 ± 1.2 ml/kg (K. R. Olson, unpublished data). As in mammals, water could be derived from the interstitial fluids (at least transiently) and by drinking. In freshwater (FW) fish, drinking is probably not a major route of water entry, although there is one report of an increase in drinking in trout fed a high-salt diet (15). Unlike in mammals, FW fish possess an additional route for fluid entry into the blood and that is transepithelial entry of water at the gill. Indeed, the passive movement of water across the gill constitutes a massive daily influx that is generally matched by the production of copious volumes of dilute urine. The influx of water across the gills would be further increased, owing to an increased osmotic gradient associated with salt loading. Although a previous study showed that urine flow rate increased by ∼33% in trout fed a high-salt diet (17), this diuresis would not appear to be sufficient to restore vascular fluid volume and hence cardiac output. Thus, unlike in mammalian models that generally require experimentally induced renal dysfunction, salt-induced hypertension can occur in fish possessing a fully functional kidney.
Although renal salt loss increases in salt-fed fish (17), it nevertheless constitutes only a small fraction of whole body salt excretion. By far, the most important site of salt balance in FW fish is the gill (5) where, under normal conditions, there is a slight net gain of salt to balance losses at the kidney. In salt-fed fish (depending on the severity of the salt load), the gills become an important site for net Na+ excretion owing to a decrease in Na+ uptake and an increase in Na+ efflux (15, 18). The mechanisms underlying the changes in unidirectional fluxes at the gill are unknown, although they are accompanied by an increased number of branchial mitochondria-rich (MR) cells (chloride cells) and Na+/K+-ATPase activity (16). These changes, which are typically observed when euryhaline fish are transferred from FW to saltwater (SW), suggest that salt loading, regardless of its origin (e.g., SW or dietary), may be causing similar alterations in the gill MR cells, whereby they become more important in salt extrusion (13).
Consequences of Prolonged Hypertension on Cardiovascular Function and Adrenergic Responses
Resting vasomotor tone is high in rainbow trout and largely reflects the combined actions of basal sympathetic nerve activity and circulating catecholamines on α1-ARs within the systemic vasculature (24, 26). Thus, intravascular injections of α-AR antagonists can evoke large reductions in systemic resistance and blood pressure (see Fig. 4). In the face of prolonged hypervolemic hypertension, we reasoned that trout might reduce α-adrenergic systemic vascular tone to lower resistance and hence also lower blood pressure. This hypothesis was refuted, however, because α-AR blockade caused identical reductions in systemic resistance and blood pressure in all fish regardless of diet and basal blood pressure. Interestingly, branchial resistance was significantly increased after α-AR blockade in both groups of fish. Because activation of branchial α-ARs is known to increase gill resistance, the increase in RG following their inhibition was unexpected but may have reflected, in part, the massive fall in cardiac output (∼33%; data not shown) associated with prazosin treatment.
Two other protocols were employed to assess the impact of prolonged hypertension on the α-adrenergic responsiveness of the systemic vasculature; intra-arterial injection of catecholamines and exposure to environmental hypercapnia. The latter treatment is known to cause a pronounced increase in systemic resistance and blood pressure, owing to the combined actions of increased sympathetic nerve activity and circulating catecholamine levels (12). In teleost fish, systemic resistance is regulated by the opposing influences of α1-AR-mediated vasoconstriction and β-AR-mediated vasodilation (11a). Because of the greater impact of α1-AR-mediated vasoconstriction, the net effect of increased sympathetic nerve activity or circulating catecholamine levels is an elevation of systemic resistance (24, 25). The novel finding of the present study is that the fish experiencing salt-induced hypertension exhibited attenuated vasopressor responses to both catecholamine injection and hypercapnia. The most parsimonious explanation for these results is that peripheral α1-ARs were either desensitized or reduced in number in the hypertensive fish. However, without additional experimentation, we cannot exclude the possibility of heightened sensitivity of systemic β-ARs. Although we were unable to directly quantify α1-AR numbers in the present study, we did quantify the levels of α1-AR mRNA in selected blood vessels. The results demonstrated a marked reduction of α1D-AR mRNA in the dorsal aorta, ABA, and EBA, as well as an increase in α1A-AR mRNA in the ventral aorta. Thus, while impossible to measure, it is tempting to speculate that the α1D-AR mRNA levels were also decreased in the resistance vessels of the systemic circulation and that reduced levels of transcript would be accompanied by a reduction in the numbers of functional α1D-ARs.
Because resting blood pressure was already increased in the salt-fed fish, it was conceivable that the ability of catecholamines (endogenous or exogenous) to raise blood pressure further would be constrained by a maximum attainable pressure (i.e., a ceiling that could not be surpassed). To ensure that this was not the case, blood pressure was measured in fish injected with very high doses of catecholamines. The maximum blood pressure achieved in this experiment was 83 mmHg or ∼43 mmHg higher than levels attained using the lower dose of catecholamines. To further eliminate the higher resting blood pressure as a potential confounding factor, catecholamines were injected into these same fish, while blood pressure was still elevated; in this case, the average maximum blood pressure reached was 94 mmHg. Clearly, the higher resting blood pressure in the salt-fed fish was not constraining their capacity to increase pressure further.
Although the vasopressor responses to exogenous catecholamines and hypercapnia were diminished in the hypertensive salt-fed fish, there was no reduction in basal vasomotor tone. At first glance, these results are in apparent conflict with our contention that the attenuated responses in the salt-fed fish reflect a loss of functional α1-ARs. The simplest explanation is that vasomotor tone was being maintained in the face of reduced α1-AR numbers by spatial recruitment of previously inactive nerve fibers. A second explanation is that only a subset of α1-ARs were affected by salt loading. Conceivably, there could have been a reduction in the numbers of α1-ARs not in contact with sympathetic nerve fibers and no change in α1-AR numbers at postganglionic sympathetic synapses. If so, this could explain the reduced vasopressor responses during conditions of elevated circulating catecholamine levels (catecholamine injection and hypercapnia) without there being any effect on resting tone.
Effects of Salt-Induced Hypertension on Branchial Blood Vessel Responsiveness
On the basis of mRNA measurements, the EBA contains both α1A- and α1D-AR subtypes, whereas the ABA contains only the α1D-AR subtype (X. Chen, S. F. Perry, S. Aris-Brosou, C. Selva, and T. W. Moon, unpublished data). The predominant α-AR subtype in the EBA is the α1D-AR (ratio of α1D/α1A mRNA is ∼5). Thus, adrenergic responsiveness in both vessels may be largely under the control of the α1D-AR.
Despite lower levels of α1D-AR mRNA in EBAs isolated from the salt-fed fish, these vessels demonstrated a heightened adrenergic responsiveness based on the significant reduction in the EC50 for norepinephrine. Assuming that the reduced levels of α1D-AR mRNA reflect fewer α1D-receptors, it would appear that the remaining α1D-ARs possess an increased binding affinity for norepinephrine. Although the ABA from salt-fed fish also expressed decreased amounts of α1D-AR mRNA, adrenergic responsiveness was unaffected. It is unclear as to why the EBA and ABA responded differently.
In mammals, BMY-7378 and RS-17053 are selective for α1D- and α1A-AR antagonists, respectively (6, 7). Whether similar specificity exists in trout is unknown. Certainly, the much greater effect of BMY-7378 compared with RS-17053 on increasing the EC50 for norepinephrine in the EBA is consistent with the predominance of α1D-AR mRNA in this tissue. In the ABA, which lacks α1A-AR mRNA, both BMY-7378 and RS-17053 increased the EC50 for norepinephrine to similar levels, suggesting that RS-17053 is capable of inhibiting the α1D-AR.
The underlying causes of essential hypertension, while often linked to increased dietary intake of salt, remain unclear. Mammalian models of salt-induced hypertension are usually confounded by the need to concurrently impair kidney function. Thus, the finding in the present study of salt-induced hypertension in otherwise normal fish is significant. Continued use of the trout hypertension model may shed light on the causes of salt-mediated hypertension, as well as the consequences of chronically elevated blood pressure on physiological systems.
This work was supported by NSERC of Canada grants to S. F. Perry and T. W. Moon and NSF grants to K. R. Olson Xi Chen was supported by an Ontario Graduate Scholarship.
We appreciate the heroic efforts of Ms. Branka Vulesevic in performing surgery.
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.
- Copyright © 2007 the American Physiological Society