Observations on Blood Viscosity in Striped Bass, Morone saxatilis (Walbaum) Associated with Fish Hatchery Conditions

S.L. Lebelo,¹* D.K. Saunders,¹ and T.G. Crawford²

Division of Biological Sciences, Box 4050, Emporia State University, Emporia, KS 66801. Correspondence and e-mail to D.K. Saunders: saunderd@emporia.edu.
Kansas Department of Wildlife and Parks, Milford Fish Hatchery, Junction City, KS 66441, USA.

*Present Address: Medical University of Southern Africa, Physiology Department, P.O.Box 130, MEDUNSA, 0204, South Africa.

This article is published in the Transactions of the Kansas
Academy of Science, vol. 104, no. 3/4, p. 183-194 (2001).

Table of Contents
Introduction Materials and methods
Results Discussion
Literature cited Related sites

**Table of Contents**
Introduction	Materials and methods
Results	Discussion
Literature cited	Related sites

ABSTRACT

Blood viscosity and blood parameters were studied in striped bass, Morone saxatilis (Walbaum) under various fish hatchery conditions. Twenty-seven adult striped bass weighing between 1040 g and 1800 g were divided into four groups: 1) healthy fish; 2) short-term hypoxia, fish exposed to oxygen concentrations of 4 mg/L for four hours; 3) simulated transport, exposed to crowding with oxygen concentrations of approximately 8.4 mg/L; and 4) diseased fish, infected with henneguya causing inflammation of the gill surface leading to hypoxemia in these animals. Plasma and apparent (whole) blood viscosity measurements were made using a Wells-Brookfield cone/plate viscometer at ten different shear rates. At packed cell volumes (PCV) of 30% and 40% and at high shear rates (75 s^-1 and 150 s^-1) the hypoxic group had a significantly higher apparent blood viscosity when compared to healthy fish, likely the result of a significant increase in the total plasma protein concentration in this group. The diseased group had a significantly higher PCV than all other groups resulting in a significant increase in blood viscosity. No significant difference was observed among the four groups in mean cell volume, mean cell hemoglobin, and mean cell hemoglobin concentration. The increase blood viscosity noted in the hypoxic and diseased groups could lead to a decreased blood flow and oxygen delivery to the tissues.

INTRODUCTION

Striped bass, Morone saxatilis (Walbaum) are often exposed to hatchery conditions that include exposure to hypoxia, disease, and the stresses of handling and transport. Hypoxia, handling, and exercise have been shown to alter substantially the packed cell volume (PCV) in various fish species (Casillas and Smith, 1977; Hughes and Kikuchi, 1984; Boutilier and others, 1988; Murad, Houston, and Samson, 1990; Cech and others, 1996; Soldatov, 1996; Wendelaar Bonga, 1997). The change in PCV can be brought about by the release of catecholamines during the primary stages of stress, which can mobilize red blood cells from the spleen (Wells and Weber, 1990) or induce red blood cell swelling as a result of fluid shift into the intracellular compartment (Nikinmaa and Huestis, 1984; Fuchs and Albers, 1988; Chiocchia and Motais, 1989).

Although the increased PCV might be adaptive for increased oxygen carrying capacity, such packed cell volumes could result in high blood viscosity, thus leading to increased difficulty in maintaining blood flow (Chien and others, 1971). An increase in PCV resulting in an increased blood viscosity has been well documented in a variety of fish species (Graham and Fletcher, 1983; Fletcher and Haedrich, 1987; Wells and Forster, 1989). In addition, the deformability of the red blood cells also plays a role in blood viscosity. Blood viscosity increases with decreasing red blood cell deformability (Chien and others, 1971). Fish, as well as reptiles, amphibians, and birds have nucleated red blood cells, which are thought to be less deformable than nonnucleated red blood cells (Usami and others, 1970; Chien, 1975; Gaehtgens, Schmidt, and Will, 1981; Nash and Egginton, 1993). Thus, the potential increase in PCV associated with some fish hatchery conditions and the decreased red blood cell deformability of nucleated red blood cells could result in an increased resistance to blood flow.

There have been few viscometric studies aimed at evaluating the possible hemorheological effects associated with various fish hatchery environments. The main objective of this investigation was to determine the blood viscosity and other related blood parameters in striped bass reared and maintained under various fish hatchery conditions.

MATERIALS AND METHODS

Fish Maintanence

Twenty-seven adult striped bass (M. saxatilis), weighing between 1040-1800 g were obtained from Milford Fish Hatchery, Junction City, KS, USA. Fish were fed a diet of trout chow daily, equivalent to about 3% of their body weight. Average water temperature ranged from 23-28°C during experimental periods (June-August, 1997). Fish were divided into four groups: healthy fish, simulated transportation, hypoxia, and diseased groups. Transportation of fish was simulated by placing seven animals in a 400 L tank at 24°C with aeration. Although no measurements of oxygen content in the water were taken, this process should result in oxygen saturation of approximately 8.4 mg/L (Wetzel, 1983). The fish were maintained under these conditions for 24 hours after which blood was immediately taken. The simulated transport conditions in terms of density of fish, temperature, and oxygenation of the water are similar to those used to transport fish from the Milford Fish Hatchery to reservoirs throughout Kansas. For hypoxic simulation, fish were placed in a 1,000 L tank and water flow was shut down to allow dissolved oxygen levels to drop to sub-optimal production levels for 4 hours. At the end of the 4-hour period, dissolved oxygen levels had dropped approximately 50% to 4 mg/L. Dissolved oxygen levels were measured using a YSI dissolved oxygen meter. Blood samples were collected immediately at the end of the 4-hour period. At the Milford Fish Hatchery, summer oxygen levels are generally around 8 mg/L. The diseased group consisted of fish that had become infected with henneguya disease (proliferative gill disease or Hamberger gill disease). This group was obtained one week after the signs of the disease had first been noticed at the fish hatchery. Clinically, these fish develop numerous white cysts on the skin and gills. Cysts on the gills (both intralamellar and interlamellar) can cause extensive granulomatous inflammation and hyperplasia of the gill surface, leading to serious respiratory problems (Moeller, 1996). No attempt was made to assess the state of the disease by counting the number of cyst. Henneguya usually occurs in a small population of striped bass at the Milford Fish Hatchery. As such, seven of the diseased fish were randomly sampled from this population. Seven fish from the healthy population were randomly assigned to each of the remaining three groups, except in the hypoxic group where six fish were sampled (one died during experimental preparation).

Blood Sampling

Fish were anaesthetized at the fish hatchery by placing them in a receptacle containing 62.5 mg/L solution of tricaine methanesulphonate (MS 222). Blood samples were taken via cardiac puncture using a heparinized syringe, and 6-8 ml of blood from each fish was placed into a separately labeled Vacutainer containing heparin. Each Vacutainer was vortexed immediately to insure mixing of the blood and packed cell volume was then determined on these samples using the microcapillary hematocrit method. Once packed cell volume determinations were performed, the Vacutainers were placed on ice and transported back to Emporia State University for further analysis.

Viscosity Measurements

Blood samples from individual fish were placed into three Eppendorf tubes. Each tube was centrifuged at 1200xg for approximately 30 seconds. Plasma from the first tube was added to the second tube in order to create packed cell volumes above (tube 1) and below (tube 2) that determined from the original sample. The packed cell volume in the third tube represented the normal packed cell volume and was not altered. Packed cell volumes were determined for blood samples in each of the three Eppendorf tubes. All tubes were kept in an ice bath until viscosity measurements could be made. Viscosity measurements were made using a Wells-Brookfield cone/plate viscometer (Model DV-II+, Brookfield Engineering Lab, Stoughton, MA, USA), with a CP-40 cone using 0.5 ml of sample. The viscometer was calibrated with distilled water or standard oil of known viscosity before actual data recording. The temperature of the sample cup was kept at 24°C with an external water bath. Viscosity determinations were made over a range of ten different shear rates corresponding with the rotational speed of the cone, and the results were reported in centipoise (cP). The shear rates were 3.75 s^-1, 7.5 s^-1, 15 s^-1, 18.8 s^-1, 30 s^-1, 37.5 s^-1, 75 s^-1, 150 s^-1, 375 s^-1, and 750 s^-1. For some of the samples tested, we were unable to obtain viscosity measurements at shear rates above 150 s^-1 because of the limited range of the viscometer, so these data were not included. Once viscosity values were determined for the blood in each of the three Eppendorf tubes for a given fish, a linear regression equation was calculated for each shear rate on plots of log apparent viscosity versus packed cell volume. This process was performed for each individual fish. From the regression equation of each individual fish, apparent (whole) blood viscosity values were predicted for a range of packed cell volumes (PCVs from 10 to 70%) for each animal. The average R² value for these regressions was 0.96. Plots were made for the apparent viscosity versus packed cell volume using Lotus 1,2,3 spreadsheet software. Plasma viscosity was also determined on each fish at ten different shear rates at a constant temperature of 24°C. Relative blood viscosity values were calculated for each animal for each shear rate by dividing apparent blood viscosity by the plasma viscosity.

Hematology

Hemoglobin concentration and cell counts were determined within 24 hours of sampling. Previous work by Korcock, Houston, and Gray (1988) noted no effect on hemoglobin concentration when using MS 222 as an anesthetic, heparin as an anticoagulant, and storage of blood at 0-2°C over a 24-hour period. Red blood cell counts (RBCC) were determined by diluting the blood in a standard red blood cell pipette (1:200) with 0.9% NaCl saline solution (Dacie and Lewis, 1984). The diluted blood was placed on a hemocytometer and cells were counted using the method previously described by Hesser (1960). White blood cell counts (WBCC) were determined by diluting the blood with Shaw's fluid (1:20) in a standard white blood cell pipette (Shaw, 1930) and the same procedure was followed as for the red blood cell count (Shaw, 1930; Hesser, 1960).

Hemoglobin concentration was measured using the cyanomethemoglobin method (Sigma Chemicals, St. Louis, MO). Mean cell hemoglobin concentration (MCHC), mean cell volume (MCV), and mean cell hemoglobin (MCH) were calculated using the following equations: MCHC = (Hb / PCV)·100, MCH = (Hb / RBCC)·10 and MCV = (PCV / RBCC)·100 (Wickham, Costa, and Elsner, 1990).

Once plasma viscosity had been determined, the plasma was frozen for later plasma protein analysis. Plasma protein concentrations were determined with the BioRad method using bovine serum albumin as a standard (Bradford, 1976). Plasma samples were read against a standard curve obtained from bovine serum albumin. All measurements were made in duplicate and averaged.

Statistical Analysis

Means and standard deviations were determined for all the hematological parameters. A one-way analysis of variance (ANOVA), with a Student-Newman-Keuls (SNK) multiple range test were used to compare means of apparent and plasma viscosity measurements of the various treatment groups for a same packed cell volume and shear rate. The same tests were used to compare means of MCHC, MCH, MCV, RBCC, and WBCC. Differences were considered significant at P < 0.05 (Zar, 1996).

RESULTS

Table 1 gives mean values for packed cell volume, hemoglobin concentration, red blood cell count, white blood cell count, derived indices (MCH, MCV, MCHC), and plasma protein concentration for each group. The diseased group had a significantly higher mean packed cell volume and hemoglobin value as related to those in healthy fish (P < 0.05). The white blood cell counts observed in the diseased, hypoxic, and transport groups were also significantly higher as compared to healthy fish (Table 1). The hypoxic group showed a significantly higher plasma protein concentration as compared to healthy animals, whereas the plasma protein concentration observed in the diseased and simulated transport groups were significantly lower (P < 0.05). No significant differences in RBCC, MCH, MCV, or MCHC occurred among groups.

Table 1. Hematological parameters, MCH, MCV, MCHC, and plasma protein concentration for all groups of striped bass.

Figure 1 shows mean values of apparent viscosity for whole blood vs. packed cell volume for the healthy, hypoxic, simulated transport, and diseased fish. Apparent blood viscosity increased with an increase in packed cell volume in all the groups at all shear rates. The shapes of the apparent viscosity curves were similar at low shear rates (3.75 s^-1 - 37.5 s^-1) and as such we show only data for blood viscosity at 15.0 s^-1 to represent low shear rates. Shear rates of 75 s^-1 and 150 s^-1 represent medium and higher shear rates respectively. At shear rates of 75 s^-1 and 150 s^-1, blood from the hypoxic animals had a significantly higher apparent viscosity as compared to controls (P < 0.05) at packed cell volumes of 30-70% (Fig. 1).

Figure 1. Influence of packed cell volume on apparent blood viscosity at three different shear rates (SR). Means (± SD) for control, hypoxic, simulated transport, and diseased groups are shown. Note, scale shown for apparent viscosity at 15 s^-1 (0-40 cP) differs from scale shown for apparent viscosity at 75 s^-1 and 150 s^-1 (0-25 cP). Click on the small image to see a full-sized version.

Plasma viscosity was not shear rate dependent between shear rates of 75 s^-1 and 750 s^-1. The plasma viscosity observed for the hypoxic group (1.76 cP ± 0.14) was significantly higher than that of the healthy animals (1.52 cP ± 0.11). No differences were noted in the plasma viscosity between healthy fish and the diseased group (1.37 cP ± 0.11) or the transport group (1.34 cP ± 0.18). Relative viscosity determined at shear rates of 15 s^-1, 75 s^-1, or 150 s^-1 was not significantly different among any of the groups at packed cell volumes of 30% or 40% (Table 2).

Table 2. Relative viscosities (± SD) determined for packed cell volumes of 30% and 40% at three different shear rates for all groups of striped bass.

DISCUSSION

Hematological Variables

In this study, we observed that hematological and hemorheological parameters of adult striped bass were altered by different conditions that occur in the fish hatchery environment. Hematological values (PCV, Hb, RBCC, and WBCC) observed for the healthy striped bass in this study were similar to those previously reported by Tisa, Strange, and Peterson (1983); Hunn, Greer, and Grady (1992); and Cech and others, (1996), for unstressed striped bass.

There was a significant increase in packed cell volume and hemoglobin concentration in the diseased group relative to that of the healthy fish. The proliferative gill disease (henneguya) causes inflammation and hyperplasia of the gills, and ultimately impairs oxygen uptake resulting in hypoxemia. The resulting chronic hypoxemia could in turn cause a release of red blood cells from the spleen or an increase in red blood cell production. Although we did notice an increase in red blood cells and mean cell hemoglobin in the diseased group, these values were not significantly different from that of the healthy fish. This is likely the result of a small sample size and large variation in red blood cell counts. It therefore is difficult to resolve if the increase in PCV and hemoglobin concentration is the result of spleenic contraction, enhanced erythrocyte production, or red blood cell swelling.

A significant increase in the concentration of white blood cells also was seen in the diseased group relative to healthy fish. This increase presumably was the result of a pathological response of the fish, similar to that observed in other teleosts (Pages and others, 1995). The values for plasma protein concentrations in diseased fish were similar to those reported by Barham, Smit, and Schoonbee (1980) for bacterially infected rainbow trout Oncorhynchus mykiss (Walbaum). Plasma proteins such as fibrinogen, prothrombin, albumin, and alpha, beta, and gamma globulins have been shown to be lowered during pathological conditions (Barham, Smit, and Schoonbee, 1980). Pages and others, (1995) observed the same trend in gilthead seabream Sparus aurata, and suggested that reduction could result from increased protein catabolism during stress or possibly from globulin degradation associated with immunological responses. Interestingly, the concentration of white blood cells were elevated significantly in striped bass placed in simulated transport conditions and those subjected to short-term hypoxia relative to healthy fish.

The hypoxic group failed to show any significant difference in packed cell volume, hemoglobin concentration, red blood cell counts, or any of the calculated hematological indicies. This was unexpected as numerous studies have shown an increased packed cell volume associated with the stress of hypoxia (Soivio, Westmand, and Nyholm, 1974; Scott and Rogers, 1981; Hughes and Kikuchi, 1984; Wells and Weber, 1991). The increase in packed cell volume associated with the above mentioned studies has been suggested to be the result of cell swelling, release of erythrocytes from storage pools such as the spleen, or to result from fluid shifts out of the vascular space. It may have been that the level of hypoxia in our study (4 mg/L) and the rate and duration of hypoxic exposure (a decrease in oxygen concentrations by about 50% over a 4-hour period) were not sufficiently severe to induce substantial changes in hematological parameters. Scott and Rogers (1981) subjected channel catfish to 1.5 mg/L oxygen concentrations for 24, 48, and 72 hours. After 24 hours, only hemoglobin concentration showed a significant increase above control animals. Additionally, Boutilier and others, (1988) working with rainbow trout (Salmo gairdneri), showed little change in hematocrit, hemoglobin concentration, and MCHC in trout exposed for 24 hours to inspired oxygen partial pressures of 120 torr and 90 torr as compared to trout maintained in water with an oxygen partial pressure of 150 torr. Drastic changes were seen in the above parameters when the trout were exposed to inspired oxygen partial pressures of 50 and 30 torr over a 24 hour period.

Plasma protein concentration was significantly elevated in the hypoxic group relative to healthy fish. If such change was the result of fluid shifts, then one would expect corresponding changes in packed cell volume. This was not the case (Table 1). Bouck (1966) working with hypoxic stressed rock bass, Ambloplites rupestris, showed that changes in plasma protein were in part due to entry of proteins from the tissues into the blood. It was suggested that anaerobic enzymes such as lactate dehydrogenase might become elevated in blood plasma, increasing plasma protein concentrations, before adaptations occur in the hematological parameters. Additionally, some fluid shifts may have occurred in the hypoxic animals in our study, but due to the large variation in red blood cell counts and packed cell volumes coupled with the small sample size, such changes would be difficult to detect.

An additional explanation for the inability to show a significant change in packed cell volumes between the hypoxic group and healthy fish may have been the use of MS 222 as an anesthetic and heart puncture as a means for obtaining blood. Soivio and others, (1977) found an approximate 43% increase in hematocrit in rainbow trout after 15 minutes of anesthesia induced with MS 222, presumably due to cell swelling. In addition, Lowe-Jinde and Niimi (1983) also found an increase in hematocrit in rainbow trout after 5, 10, and 20 minutes of MS 222 treatment. Further, the use of cardiac puncture may also lead to cell swelling. Railo and others, (1985) showed an approximate 16% increase in hematocrit in fish sampled via cardiac puncture as compared to fish sampled using a dorsal aortic cannula. As such, cell swelling may have been induced in all groups. This would result in difficulty showing differences between groups regarding packed cell volume, mean cell volume, and mean cell hemoglobin concentration if swelling had occurred in the red blood cells of the healthy fish. However, extreme care was used to ensure that all sampling and processing of blood samples were similar to allow for reliable comparisons.

Hemorheology

No significant differences were noted in the apparent viscosity of the simulated transport group or the diseased group relative to that of healthy fish at the same packed cell volume at shear rates above 3.75 s^-1. However, our viscosity measurements demonstrated that hypoxia resulted in a significant increase in apparent blood viscosity at higher shear rates (shear rate = 30 s^-1, 37.5 s^-1, 75 s^-1, 150 s^-1), at packed cell volumes of 30-70%, relative to healthy fish. Chien and others, (1971) indicated that at higher shear rates, the main contributing factors for determining whole blood viscosity are red blood cell deformability and their orientation during flow. Therefore, the increased apparent (whole) blood viscosity observed in this study for the hypoxic group could be due to a decreased deformability of red blood cells and/or an increased plasma viscosity. However, the relative viscosity, which adjusts the apparent viscosity for the contribution of the plasma viscosity, was not significantly different in the hypoxic group relative to that of the healthy striped bass at any of the various shear rates tested (Table 2). This would suggest that no decreased deformability of the red blood cells had occurred in hypoxic animals. Previous studies actually have shown increased cell deformability with hypoxic exposure in carp (Hughes and Abers, 1988) and in rainbow trout (Hughes and Kikuchi, 1984; Chiocchia and Motais, 1989). Thus, it is likely that the increased apparent blood viscosity in the hypoxic group may be the result of their significantly higher plasma viscosity when compared to the plasma viscosity of healthy striped bass. The high plasma viscosity could result from the increased plasma protein concentration observed in the hypoxic group. Increased plasma protein concentration has been correlated with an increased plasma viscosity and could lead to increased apparent (whole) blood viscosity, particularly at lower shear rates (Chien and others, 1971).

The effect of increased packed cell volume on apparent blood viscosity, as occurred in the disease group, was determined by comparing the apparent blood viscosity of the diseased group at its original mean packed cell volume of 40% to healthy striped bass at their mean packed cell volume of 30%. At shear rates of 15 s^-1, 75 s^-1, and 150 s^-1, blood viscosity in the diseased group at a packed cell volume of 40% was 39%, 43%, and 48% greater, respectively, compared to the blood viscosity of healthy fish with packed cell volumes of 30%. Thus, the increased oxygen carrying capacity in the disease group might be offset somewhat by the increased viscosity associated with its higher packed cell volume.

This study showed that the effect of moderate, short-term hypoxia increased apparent blood viscosity in hypoxic fish. Increased apparent blood viscosity could lead to an increased resistance to blood flow, potentially resulting in a decreased blood flow in the systemic circulation and further impairment of oxygen delivery to the already oxygen-deprived tissues. Likewise, the increased packed cell volume observed in the diseased group results in an increased blood viscosity, that in turn could hamper the oxygen transporting ability, further minimizing the availability of oxygen to tissues in fish infected with henneguya. Simulated transportation of this species of fish in well-aerated water medium did not significantly alter hemorheological parameters of these animals.

Return to beginning.

ACKNOWLEDGEMENTS

The authors thank the Milford Fish Hatchery staff for providing fish and assistance, as well as Roger Ferguson, Jeff Witters, and Kham Noam Nang for their help with this project. We also would like to thank David Edds, Laurie Robbins, and Roger Fedde for their thoughtful review of this manuscript. Financial assistance for this project was provided by the Emporia State University Research and Creativity Committee.