Lability of Blood Volume in Snakes and its Relation to Activity and Hypertension

The importance of blood volume is well established in vertebrate physiology, and a suite of control mechanisms appears to maintain the volume of circulating blood within relatively narrow limits. Blood volumes have been measured in numerous species of reptiles with a variety of techniques, and typically represent 5–8 % of the body mass in restrained or anaesthetized animals (e.g., Bradshaw & Shoemaker, 1967; Thorson, 1968; Lillywhite & Smith, 1981; Pough & Lillywhite, 1984).

We have previously investigated the responses of blood volume in snakes subjected to experimental haemorrhage and found that the ability of snakes to restore blood volume through transcapillary shifts of extracellular fluid apparently exceeds that of other vertebrates studied (Lillywhite & Smith, 1981; Lillywhite & Pough, 1983; Smits & Lillywhite, 1984). These investigations suggest that the capacity for transferring fluid between the vascular and extravascular compartments is relatively great, and that blood volume within an individual snake might be quite labile. We infer that if the resistance to fluid movement across capillaries is indeed comparatively small then inducing transient increases in capillary hydrostatic pressure could significantly reduce blood volume by enhancing filtration of plasma. In this paper we report that locomotion and arterial hypertension both induce substantial reductions of circulating blood volume in two species of snakes.

We measured changes of blood volume in four rat snakes (Elaphe obsoleta) and three Pacific rattlesnakes (Crotalus viridis) obtained from commercial suppliers. Snakes were prepared for experimentation by cannulating the dorsal aorta just anterior to the vent with heparinized, saline-filled PE-10 or PE-50 occluding catheters. Snakes were anaesthetized by lowering their body temperatures in chipped ice and allowing several inhalations of halothane (Fluothane) vapour. Surgical procedures and exteriorization of the catheter followed those reported by Lillywhite & Seymour (1978). Snakes were allowed 36–48 h for post-operative recovery within transparent acrylic tubes provided with openings for ventilation and exit of the catheter.

Measurements of blood volume were obtained by labelling a snake’s own erythrocytes with the isotope ⁵¹Cr. Details of the labelling procedure were developed by Smits & Lillywhite (1984). Briefly, approximately 1 ml of a snake’s red blood cells were incubated with 10 μCi of isotopic sodium chromate at room temperature for 1 h. The red cells were then repeatedly washed with a plasma-saline solution to eliminate the unbound isotope. Finally, the red blood cells were diluted in saline to a volume originally withdrawn from the snake, and a known volume and specific activity of the labelled blood was infused into the aortic catheter. The labelled cells were allowed to mix with the circulating blood volume overnight.

Because ⁵¹Cr binds irreversibly to the haemoglobin within the red blood cells (Jones & Mollison, 1956), and because red cell longevity is apparently long in reptiles (Altland & Brace, 1962; Smeller, Bush & Seal, 1978), the ⁵¹Cr marker remained within the vascular space for the duration of the experiments. Therefore, instantaneous blood volumes of snakes were determined at specific times during the experiments by measuring the specific activity of gamma radiation in the blood of the snake. Blood samples (0·3 ml) withdrawn for blood volume determinations were partitioned into microcapillary tubes (0·05 ml) for haematocrit measurement (IEC Microcentrifuge) and into pre-weighed 13 × 100 mm tubes (0·25 ml) for gamma counting. The counting tubes were then reweighed and counted for 5 min each (approx. 1 % sigma) in a Packard Auto-Gamma 800C scintillation counter. Blood volume was obtained by dividing the total activity infused into a snake by the specific activity of a given blood sample, correcting for decay. The specific gravity of whole blood (1·05 gml^−l) was used to convert units of mass to volume.

Changes in blood volume induced by locomotor activity were measured the day following the labelling of the red cells. Aortic pressures were measured by means of a Gould-Statham P23 ID pressure transducer and recorded on chart paper with a Grass Model 7D Polygraph. A blood sample (0·3 ml) was withdrawn for initial measurements of blood volume (BV) and haematocrit (Hct) after a snake exhibited a steady heart rate (HR) and blood pressure (BP) for at least 30 min of inactive within the acrylic tube. The blood volume and haemodynamic values measured while the snake was at rest were considered control data, and the cardiovascular responses of each snake to experimental treatments were compared to its own controls.

To induce activity snakes were removed from the acrylic tube and allowed to move across the laboratory floor. Snakes were prodded to move at a steady pace, and both sedentary behaviour and frantic bursts of locomotion were discouraged. Snakes were returned to the acrylic tube following 15 min of activity; measurements of BV, Hct, HR and BP were made immediately and after a 10-min period of recovery. Levels of Hct were remeasured at 5 h, and in one Elaphe both Hct and BV were determined at hourly intervals for 7 h.

A second experiment was conducted on each snake to measure responses of blood volume to arterial hypertension. Control values of BV, Hct, HR and BP were determined as before. Blood pressure was then elevated by infusing 100 μg of norepinephrine per kg body mass (Calbiochem DL Arterenol, <50 μl saline carrier) into the dorsal aorta. Heart rate and blood pressure were measured continuously, and blood samples for BV and Hct determinations were taken when BP peaked and at 10 and 20 min thereafter. These measurements were repeated at hourly intervals for 4 h in one of the Elaphe.

The resting blood volumes (BV) of the two species of snake measured prior to the experiments were not significantly different (P>0·2; t-test) and represented 6·07 ± 0·40% and 6·41 ± 0·06% of the body mass (mean ± S.E.) of Elaphe and Crotalus, respectively (Table 1).

Fig. 1 illustrates the percentage change of blood volume in both species of snake after 15 min of locomotor activity. Both species demonstrated substantial reductions in BV to approximately 21 % below control volumes (mean ± S.E. were −20·8 ± 2·26 and − 21·8 ± 1·86% for Elaphe and Crotalus, respectively), indicating that at least one-fifth of the circulating fluid had left the vascular space. BV of both species remained at these levels during the 10 min of recovery following activity. Measurements of BV made later in recovery indicated that the restoration of control BV required longer than 1 h and varied due to changes in the activity of snakes while they were held inside the acrylic tubes (Fig. 2).

Arterial hypertension caused by the infusion of norepinephrine also reduced BV in both species of snake (Fig. 3). Decreases in BV occurred notably 10–20 min after arterial blood pressure peaked. The maximum BV reductions (at +20 min) in Elaphe and Crotalus were 8·6 ± 1·9 and 17·0 ± 5·7%, respectively. The return of BV to control level required 1–3 h but varied according to the activity of snakes, as in the other experiments (Fig. 4).

Haemodynamic responses to activity were proportionately similar in Elaphe and Crotalus (Fig. 5): HR increased dramatically (83 and 87%, respectively) while increases in BP were more modest (59 and 55 %, respectively). The significant increases in Hct in Elaphe (22%) and Crotalus (29%) are consistent with the estimates of BV reduction as measured by the ⁵¹Cr-labelled erythrocytes (Fig. 1). The infusion of norepinephrine caused proportionately greater hypertension in Elaphe than Crotalus (154 vs 121% increase in BP). Heart rates remained unchanged, with a slight bradycardia at the peak of arterial pressure in Elaphe. Increases in Hct in response to arterial hypertension were smaller than increases due to activity. Maximum increases in Hct above control values in Elaphe (7 %) and Crotalus (18 %) occurred 10–20 ms enter the pressure peak, and corresponded to the degree of BV reduction in each species (Fig. 3).

The steady-state distribution of fluid between the vascular and extravascular compartments is determined principally by the hydrostatic and osmotic forces associated with capillaries, and the functional capabilities of lymphatics. Present findings indicate that BV of active or hypertensive terrestrial snakes differs substantially from BV of snakes at rest, apparently due to an increase in the capillary filtration of plasma out of the vascular space. Muscular activity promotes the translocation of plasma more effectively than does hypertension per se, for if BP is elevated pharmacologically, the level of arterial pressure is higher but the fluid shift is less than that which is induced by locomotion (Figs 1–5). The state of hypertension promoted by norepinephrine reflects a generalized vasoconstriction in peripheral vasculature, whereas the increase in BP resulting from activity must accompany an increase in blood flow to muscle tissue. It is evident from mammalian studies that elevations of capillary hydrostatic pressure subsequent to a fall in precapillary resistance increase the movement of fluid out of skeletal muscle capillaries during and for varying periods following the activation of muscle in exercise (Haddy, Scott & Grega, 1976). The increased fluid efflux may be augmented by transient increases in tissue osmolality and capillary surface area (Kjellmer, 1964). Thus, skeletal muscle is likely to be a principal site of the transcapillary fluid fluxes observed during locomotion in snakes.

Exercise in mammals reduces plasma volume roughly 5–15 %, depending on the intensity and duration of muscular activity (Ekelund & Holmgren, 1964; Greenleaf et al. 1977; Novosadová, 1977; Thomas & Etheridge, 1982). By comparison, the 21 % reduction of BV demonstrated in snakes represents a translocation of plasma which exceeds that of mammals by factors of 2 to 7. This supports our inference that these reptiles possess a comparatively small resistance to fluid transfer between the vascular and extravascular spaces, and may also indicate a limited removal capacity of the lymphatic system.

The potential contribution of the spleen (releasing or sequestering erythrocytes) to our measurements is negligible for several reasons. First, the spleen of the snake is comparatively small (0·086 g wet mass/100g body mass in Elaphe obsoleta, N = 4; A. W. Smits, unpublished data) and could not possibly account for the changes in Hct we observed. Secondly, the increases in Hct during activity and hypertension were reversible, and Hct returned to control values in all snakes studied. Finally, we have never previously observed such dramatic increments in Hct in these snakes, even during experiments involving severe haemorrhage when one would expect spleen contraction.

Both species of snake exhibit similar haemodynamics and B V changes with activity, but the responses of each species to arterial hypertension are different. Elaphe develops proportionately greater hypertension and reflexogenic response (bradycardia) to norepinephrine and experiences less plasma filtration than does Crotalus. These findings are consistent with previous research indicating that rat snakes possess score potent autonomic adjustments of vascular tone and exhibit less blood pooling in dependent vasculature than do rattlesnakes when the circulation is disturbed been head-up tilting (H. B. Lillywhite, unpublished data).

Two obvious consequences of the decrease in plasma volume observed during exercise are increases in the oxygen capacity and viscosity of the blood. It is not known wHcther these changes confer any respiratory (e.g. oxygen delivery) or cardiovascular (e.g. flow changes) advantages or disadvantages to snakes during these dramatic reductions of plasma volume. Our results do indicate a need for caution while interpreting or comparing instantaneous measurements of blood volume in these vertebrates; differences between species may clearly be confounded by variability due to the activity or hypertensive states of animals. This may also apply to other vertebrates as well.

This research was supported by USPHS, NIH grant no. HL 24640. We are grateful to Kerry Young who obtained the Crotalus viridis under the authority of California Department of Fish and Game permit no. 82-8-1, and we thank Martin Feder for his comments on the manuscript.