INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CARDIAC HYPERTROPHY and its underlying etiology have been studied intensively in mammals, but until recently it has received little attention in fish. Several recent investigations have established that hypertrophy of the cardiac ventricle is associated with reproductive maturation in male, but not female, rainbow trout (Oncorhynchus mykiss) (2, 4, 12). These studies demonstrate that relative ventricle mass [RVM; (ventricle weight/body wt) × 100] is up to threefold greater in mature males with a large gonadosomatic index [GSI; (gonad weight/body wt) × 100] when compared with females or immature males. In mammals, the primary precursor for both pathological and physiological enlargement of the heart is an elevated hemodynamic workload (30). Studies in trout (10, 16) also suggest that cardiac work influences ventricular growth in fish. Because reproductive maturation and spawning activities increase functional demands on the heart of male salmonids (1, 28), it is plausible that hemodynamic overload promotes the dramatic ventricular hypertrophy observed in mature male trout.

Types of hemodynamic overloads leading to cardiac hypertrophy in mammals can be segregated qualitatively into pressure and volume overloads. Defined broadly, a pressure overload is an increase in resistance to ejected blood volume. Although increased vascular resistance is the etiologic factor most commonly cited to explain cardiac hypertrophy in mammals (30), the presence of the gill circuit between the heart and systemic vessels in fish imposes constraints on pressure development during systole. Dampening of the pressure pulse is accomplished by the elastic bulbus arteriosus, which is analogous to the ascending aorta in mammals. Adequate functioning of the bulbus depends on its distensibility (i.e., fractional change in volume per unit change in pressure) and compliance (i.e., change in volume per unit change in pressure) (9, 11). A reduction in bulbus distensibility or compliance relative to ventricular output could elevate cardiac afterload and lead to ventricular hypertrophy, as occurs with decreased aortic compliance in humans (14). Volume overload, wherein end diastolic volume is increased, can be invoked by an expanded blood volume (V_B). Increased V_B could elevate central venous pressure, enhance end diastolic volume, and result in a greater stroke volume (6, 31). Although V_B has not been compared previously between mature and immature fish, Franklin and Davie (12) demonstrated that maximum stroke volume increases with ventricular hypertrophy in mature male trout.

The purpose of this study was to ascertain whether ventricular hypertrophy in male trout is associated with an increased hemodynamic load. Using male trout with a broad range of RVMs (representing increasing degrees of hypertrophy), we measured blood pressure and pulse pressure dynamics in the ventral aorta, V_B with the use of Evans blue dye (EBD), and the pressure-volume dynamics of the bulbus arteriosus. We describe relationships between these hemodynamic variables and the relative mass of the ventricle. Our observations indicate that ventricular hypertrophy in reproductively mature male trout is associated with hypertension and hypervolemia, and the pressure-volume dynamics of the bulbus do not compensate for ventricular enlargement. Furthermore, our data demonstrate that higher blood pressure has important implications for estimations of blood volume.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In Vivo Measurements of Hemodynamics in Trout

Surgical procedures. Individual fish were netted and transferred to an oxygenated anesthetic bath containing 1% NaCl (to reduce osmotic stress), 5.0 mM sodium bicarbonate, and 0.02% MS-222 (Crescent Chemical, Phoenix, AZ). Upon loss of locomotory ability, fish were weighed and placed upright in a holder made from 8-in. polyvinyl chloride (PVC) pipe split lengthwise. Animals were kept moist throughout surgery, and constant retrograde irrigation of the gills was provided (8.6 l/min) by means of a recirculating, oxygenated solution (~7°C) containing 1% NaCl, 5.0 mM sodium bicarbonate, and 0.006% MS-222.

A Dramel Moto-Tool (model 3801; Emerson Electric, Racine, WA) fitted with a 3.3-mm-diameter stainless steel grinder was used to smooth the teeth on the tongue to prevent abrasion of the cannula. We cannulated the ventral aorta (VA), using slight variations of the procedures presented in Gamperl et al. (13). Cannulas (PE-50 Intramedic; Clay Adams, Parsippany, NJ) were cut to a constant length (0.8 m), and a small bubble was made 1.5 cm from the tip of each cannula to assist in securing it to the animal. We used a drilling motion to insert a beveled (60°), 12-mm-long, 23-gauge needle into the isthmus between the first and second gill arch. When resistance ceased, the needle was removed, and a heparinized cannula fitted with an indwelling stainless steel wire (18-gauge guitar string) was inserted immediately. After insertion, the wire was removed slowly, and the cannula was positioned to maximize blood flow. We cleared blood from the cannula by injecting ~0.5 ml of filtered (0.22 µm), sterile 1% NaCl containing 100 U/ml lithium heparin. The cannula was connected to a blood pressure analyzer (see Measurement of blood pressure) to facilitate final positioning of the cannula within the VA and to monitor blood pressure during the remainder of surgery. The cannula was secured to the tongue with ligatures (4-0 coated Vicryl and curved RB-1 needle; Ethicon, Somerville, NJ) applied immediately anterior and posterior to the bubble. A third ligature was applied ~1 cm posterior to the tip of the tongue to prevent cannula movement with respiratory movements or swimming. A 12-gauge steel needle was used to insert a 2-cm-long heat-flared sleeve (PE-205, Intramedic) through the bottom of the mandible at a position slightly rostral to the tip of the tongue and 0.5 cm lateral from the midline of the jaw. The cannula was disconnected from the blood pressure monitor, exteriorized through the sleeve, and cleared with 0.5 ml heparinized saline, and the end was tied. Total time under anesthesia for each fish was 10-20 min.

Fish were placed into holding tubes manufactured of PVC pipes (8 in. diameter, 30 in. long) with rigid plastic mesh (30 × 30 mm lattice) covering the ends and narrow slits cut into the tops. Tubes were immersed in concrete raceways and oriented so that fish faced upstream into flowing (3.7 cm/s), 14°C spring water. Cannulas were then exteriorized from the tube through the slits. Fish always recovered equilibrium and swimming ability within 10 min and were subsequently able to maintain position in the tubes with minimal exertion. Preliminary results indicated that 10 h after surgery was sufficient time for stabilization of blood pressure. We permitted fish to recover for 13-18 h before blood pressure measurements and 18-32 h before blood volume determinations.

Measurement of blood pressure. Cannulas were flushed with 0.3 ml of sterile, filtered (0.22 µm) 1% NaCl containing 100 U/ml lithium heparin and then connected via a 23-gauge needle hub to a blood pressure analyzer and transducer (DigiMed BPA 200a; Micro-Med, Louisville, KY). The DigiMed BPA 200a facilitated high-resolution recordings because minimal fluid flow (~0.02 nl/mmHg) across the transducer element was required for detection. The analyzer and chart recorder were calibrated daily with a water-filled manometer. Pressure tracings were recorded for 1 min and monitored for 5 min on a model L45 two-channel flatbed recorder (IITC, Woodland Hills, CA). Recordings were made at chart speeds of 2 and 10 mm/s (Fig. 1). Pressures were recorded for each animal at ~13 and ~22 h postsurgery.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Representative tracings of blood pressure (BP) in ventral aorta (VA) of unanesthetized, unrestrained male rainbow trout. Pictured are examples of BP tracings from an immature trout (top) with normal ventricle [relative ventricle mass (RVM) = 0.078] and a sexually mature trout (bottom) exhibiting ventricular hypertrophy (RVM = 0.199). Note change in time scale halfway through tracing.

Hemodynamic variables. Peak systolic blood pressure (SBP), end diastolic pressure (DBP), pulse pressure (PP), heart rate (HR), and absolute and relative duration of arterial systole were determined directly from chart recordings by means of a digital caliper (Ultra-Cal Mark III; Fowler, Newton, MA). The highest and lowest points on each pressure spike were used to determine SBP and DBP, respectively. Preliminary analyses indicated that deriving mean arterial blood pressure (MAP) with a standard mammalian calculation [MAP = DBP + 1/3(SBP  DBP)] overestimated MAP. Therefore, MAP was derived by digitizing pressure tracings on a Summagraphics digitizing tablet (Summagraphics, Houston, TX) and measuring area under the curve with SigmaScan software (Jandel Scientific, San Rafael, CA). Maximum rates of arterial pressure development (+dP/dt) and decay (dP/dt) were determined from tangents drawn to the rising and falling phases, respectively, of pressure tracings.

Determination of blood volume and Evans blue kinetics. Within 30 min of final blood pressure measurements, 0.5 ml of blood was withdrawn via the VA cannula of each fish and used to determine hematocrit and to establish a plasma reference. Only animals with a hematocrit > 20% were included in the study. EBD (T-1824; 7.0 mg/ml in sterile, filtered 1% NaCl) was injected (1.0 ml/kg body wt) into each animal via the VA cannula, and the cannula was flushed with 0.25 ml sterile, filtered saline containing 10 U/ml lithium heparin. Blood samples (0.5 ml) were withdrawn every 15 min for 90 min. Each sample was placed into heparinized (20 µl of 10 U/ml lithium heparin) 1.5-ml polypropylene microcentrifuge tubes and centrifuged at 600 g for 1 min. Plasma was removed and frozen immediately in cryovials at the temperature of liquid nitrogen. The pellet containing blood cells was resuspended gently in a volume of sterile, filtered saline equal to the volume of plasma removed, and the resuspended cells were reinjected into the animal via the cannula. Hematocrit was determined again at the final sample.

To determine the concentration of EBD at each time point, plasma samples (and plasma controls) were diluted 1:10 with 1% NaCl, and the absorbance at 600 nm was measured spectrophotometrically. We converted each absorbance to a concentration of dye (mg/ml plasma) by means of a standard curve, plotted concentrations vs. time, and used least-squares regression analysis over the initial linear portion of the curve (15-60 min) to determine the theoretical concentration of dye at time 0. The following standard pharmacokinetic formula (25) was used to calculate plasma volume:

(1)

where V_P = plasma volume, D₀ = amount of dye injected (mg), and P_C = plasma concentration of dye (mg/ml). To calculate the volume of blood in each animal, we employed the following equation:

(2)

Finally, we expressed V_B as milliliters of blood per kilogram animal body weight.

EBD is known to leak from fish vasculature even when bound to plasma proteins (20, 21, 29). This raised the possibility that if hemodynamics were different in animals exhibiting ventricular hypertrophy, the loss of dye (and therefore volume estimates) would not be comparable among all animals. Preliminary analysis suggested that V_B was elevated up to 70% as RVM increased. We questioned the magnitude of this apparent increase, as it seemed likely that the elevated blood pressures observed as ventricles enlarged might have inflated our estimates of V_B. To evaluate this possibility, we used an open two-compartment pharmacokinetic model to calculate the kinetics of EBD in our animals (19, 25). This type of model was appropriate based on the biexponential decay in dye concentration over time, as demonstrated in Fig. 2. We used the following design in our two-compartment model: where compartment 1 represents the central compartment or blood volume, and compartment 2 represents the extravascular compartment. The rate constants represent the movement of EBD from compartment 1 to compartment 2 (k₁₂), from compartment 2 to compartment 1 (k₂₁), and elimination of drug from the central compartment (k₁₀). Using this model, we calculated the volume of distribution (V_C) of EBD in the central compartment as a function of body weight:

(3)

where A and B are the y intercepts for the distribution and elimination phases, respectively, of the plasma-level time curve (Fig. 2). We then used our calculated V_C to estimate clearance of EBD from the central compartment:

(4)

Our analysis indicated that elevated SBP was associated with greater clearance and a larger V_C for EBD (see RESULTS), suggesting that EBD was being removed or driven out of the vasculature at higher blood pressures. To more accurately estimate V_B and avoid inflation of our volume estimates caused by extravasation of EBD, we recalculated V_B based on our earliest sampling point (15 min). Our reasoning was that there would be little loss of EBD from the central compartment at an early sampling time (3).

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Representative semilog plot showing concentration of Evans blue dye (EBD) in trout plasma for each sampling time. Regression lines highlight biexponential nature of EBD disappearance in trout blood, indicating that an open, 2-compartment pharmacokinetic model is appropriate. Dotted line represents exchange of EBD between compartments (distribution phase), and solid line represents slow elimination of EBD from animal (elimination phase). A and B represent y-intercepts for each regression line and are used to calculate clearance of dye.

After the collection of blood samples, animals were killed by a sharp blow to the head. Body weight and fork length were recorded, and the testes and ventricle were isolated, rinsed briefly with 1% NaCl, blotted dry, and weighed. The GSI was employed as an indicator of reproductive maturation (12). We dissected the gill isthmus of several animals to confirm that we had placed the cannula correctly in the VA. Our designation of ventricles as hypertrophied was based on the definition proposed by Clark and Rodnick (4), whereby a ventricle is considered hypertrophied if the RVM exceeds by 30% or more the mean RVM of immature male and female trout in the population from which the study animals were drawn. For the population used in this study, an RVM > 0.11% was considered hypertrophied.

Pressure-Volume Relationships of the Bulbus Arteriosus

Experimental procedures. Animals were anesthetized with 0.02% MS-222 and killed by a sharp blow to the head. We recorded body weight and fork length and then exposed the heart and bulbus arteriosus with a single ventral incision. After determining the in vivo length of the bulbus, we excised the intact ventricle and bulbus, being careful to include at least 0.5 cm of the VA. The ventricle was cut away, leaving a small portion still attached to the bulbus, and the ventricle and bulbus were rinsed with 1% NaCl at 15°C, blotted dry, and weighed. A 14-gauge needle was inserted through the ventricular-bulbar valve and secured with 3-0 surgical silk. The needle hub was attached to a three-way plastic stopcock, which was connected to a calibrated pressure transducer (Digimed BPA 200a; Micro-Med) and a calibrated syringe pump (ISCO, Lincoln, NE). Air was evacuated from the preparation by flushing the bulbus lumen with 1% NaCl followed by tying the ventral aorta with 3-0 silk to prevent leakage. The bulbus surface was kept moist throughout the procedure by frequent external applications of 1% NaCl. The bulbus was inflated with 1% NaCl (1 ml/min) to a maximum lumen pressure of 80 mmHg then deflated. Tracings were captured on a Model L45 2-channel flatbed recorder (IITC, Woodland Hills, CA) set to record at a rate of 1 mm/s. Each bulbus was subjected to three preliminary inflation-deflation cycles, followed by four cycles that were recorded for analysis. At the end of the experiment, the bulbus minus the VA and ventricle was weighed.

Data analysis. Pressure tracings were transcribed with a digital caliper. Volumes were calculated by using the time component on the chart recorder and factoring the syringe pump rate (1 ml/min). Plots of bulbus volume vs. pressure were used subsequently to describe the hemodynamic characteristics of the bulbus. Based on our tracings, we identified the range of greatest compliance (i.e., the portion with minimal slope) for the pressure-volume curve for each animal and calculated compliance and distensibility over this range by means of the following formulas

(5)

(6)

where V and P refer to the change in volume and pressure, respectively, and V refers to the initial volume at the beginning of this compliant range. We used this range to determine the average pressure at which bulbi were most compliant, as well as the scope of the range. Finally, we estimated maximal lumen volume as the volume of the bulbus arteriosus at a lumen pressure of 80 mmHg, which represented the point at which P/V rose exponentially for all animals.

Statistical Analyses

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Reproductive Maturation Induces Ventricular Hypertrophy

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Relationship between sexual maturity and RVM [(ventricle weight/body wt) × 100] in male rainbow trout (n = 16; y = 0.027x + 0.091; r2 = 0.91; P < 0.001). Mass of testes relative to body weight [gonadosomatic index (GSI)] serves as an index of sexual maturation.

Ventricular Hypertrophy is Associated with Altered Hemodynamics

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Relationships between RVM and blood pressure in VA of unrestrained, unanesthetized rainbow trout (n = 15). Peak systolic pressure (A; y = 172.4x + 24.6; r2 = 0.59), mean arterial pressure (MAP) (C; y = 102.1x + 23.9; r2 = 0.47), and pulse pressure (PP) (D; y = 98.1x + 2.6; r2 = 0.7) were correlated significantly with RVM (* P < 0.01; ** P < 0.001). MAP was determined by measuring area under the curve from digitized pressure tracings.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Relationships between RVM and PP dynamics as measured in the VA of unrestrained, unanesthetized rainbow trout (n = 15). We determined maximum rates of arterial pressure development (+dP/dt) (A; y = 444.4x  2.9; r2 = 0.75) and decay (dP/dt) (B; y = 371.5x  23.3; r2 = 0.80) from pressure tracings by drawing tangents to rapid pressure development and decay phases of PP then determining slopes. Pressure tracings were also used to determine heart rate (HR) (C; y = 117.9x + 60.8; r2 = 0.60) and absolute (D) and relative (E) duration of arterial systole. (y = 58.0x + 31.3; r2 = 0.62). * P < 0.01; ** P < 0.001.

Blood Volume and Clearance of EBD are Increased with Ventricular Hypertrophy

1

1

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Clearance (A) and volume of distribution (B) of EBD in trout as a function of blood pressure (* P < 0.01). Both kinetic variables are expressed as a function of body weight. Animals were given bolus dose of EBD, and kinetics were calculated with an open 2-compartment pharmacokinetic model. Systolic blood pressure was correlated with greater clearance (y = 0.08x  1.91; r2 = 0.77) and elevated volume of distribution (y = 0.04x + 0.89; r2 = 0.60) of EBD in trout.

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Relationship between blood volume (VB) and RVM in unrestrained, unanesthetized rainbow trout (n = 15; y = 82.6x + 32.8; r2 = 0.64; P < 0.01). EBD was given intravenously in a single bolus dose, and VB was determined by calculating plasma volume from concentration of EBD at 15 min postinjection and correcting for hematocrit.

Bulbus Pressure-Volume Characteristics are Unchanged with Ventricular Hypertrophy

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Representative tracings of lumen pressure of bulbi arteriosi at a range of lumen volumes. Each tracing shown represents 1 animal and is the average of 4 inflation cycles. Tracings were obtained from bulbi arteriosi taken from trout with a wide range of RVMs (0.08-0.18%).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Despite the essential role hemodynamic load plays in inducing cardiac growth in mammals (30), few studies have examined whether increasing workload promotes growth of the fish heart. The aim of our study was to investigate whether the well-documented ventricular enlargement of mature rainbow trout is associated with elevated blood pressure and/or blood volume. Our approach was to measure hemodynamic variables in reproductively immature and mature male rainbow trout and subsequently describe the hemodynamic load on the heart.

Trout Exhibiting Ventricular Hypertrophy are Hypertensive and Hypervolemic

In contrast to studies in mammals, only two studies in fish have suggested a correlation between increased cardiac workload and ventricle growth. First, Houlihan et al. (16) demonstrated that increased workload and power output in an in vitro trout heart preparation induced greater fractional rates of protein synthesis in the ventricle. An increase in protein synthesis within hours of hemodynamic overload is considered a marker of cardiac growth in mammals (30). Second, a recent report demonstrated that parasitic infestations of the bulbus arteriosus in Sheepshead minnows (Cyprinodon mariegatus) resulted in fibrosis and narrowing of the lumen diameter of the bulbus, leading to ventricular hypertrophy in infected animals (5). This observation stresses how the functional status of the bulbus can influence ventricular function and growth. Additional support for the effects of increased workload on cardiac growth comes from morphometric studies. In mammals, increased end diastolic volume and wall stress induces myocyte elongation and thickening, whereas increased afterload and strain results predominantly in increased myocyte diameter (15). We reported recently (4) that ventricular hypertrophy in reproductively mature trout results from cardiomyocyte hypertrophy, specifically increased cross-sectional area (+83%) and length (+31%). The fact that myocytes grew predominantly by thickening suggests that increased afterload may promote ventricular hypertrophy in mature trout, although it does not exclude increased preload as a factor influencing ventricle growth.

We acknowledge that factors other than hemodynamics may be important in the development of ventricular hypertrophy in trout. Blood concentrations of androgens are higher in reproductively mature male trout (24), and androgen supplementation in young male and female trout is associated with ventricular hypertrophy (8, 27). Despite the anabolic effect androgens exert on numerous tissues, a direct link between elevated androgens and cardiomyocyte hypertrophy has not been established in fish. We suggest that elevated androgens in mature male trout may influence ventricular growth indirectly through volume expansion or increasing blood pressure.

Hemodynamics and the Role of the Bulbus Arteriosus in Ventricular Hypertrophy

Blood Volume Estimates and Evans Blue Distribution Kinetics are Dependent on Blood Pressure

Perspectives