The evaluation and use of the pig (Sus domesticus) as a model for determining the effects of dietary lipids on lipid metabolism and thrombosis in human beings : a thesis prepared in partial fulfilment of the requirements for the degree of Doctor of Philosophy at Massey University
The domestic pig (Sus domesticus) has been used extensively as a model in atherothrombosis research because this species has many physiological similarities to human beings. In the series of studies presented here, the domestic pig was used and evaluated as a model for determining the effects of dietary lipids on lipid metabolism and thrombosis in human beings. In the first study, serum specimens were collected from five 8-week-old Large White pigs in the fasting and postprandial states to determine whether cholesterol fractions can be estimated without use of the ultracentrifuge. Sequential ultracentrifugation was used to determine cholesterol in VLDL (VLDLfuge), LDL (LDLfuge) and HDL (HDLfuge) fractions. VLDLfuge was compared with VLDL cholesterol (VLDL-C) concentration estimated as serum triglyceride concentration divided by four (VLDLtrig). HDLfuge was compared with cholesterol remaining in the supernatant after precipitation of apolipoprotein B-containing lipoproteins with Mn2+ and heparin (HDLppt). LDLfuge was compared with the concentration determined from the Friedewald formula (total cholesterol less HDLppt less VLDLtrig). After correcting the centrifuged fractions for recovery of total cholesterol, the mean difference between the LDLfriede and LDLfuge of fasting samples was <5% and the mean difference between fasting HDLppt and HDLfuge was <8%. Fasting VLDLtrig was more than twice VLDLfuge, after correction for cholesterol recovery, possibly because of very low recoveries in the ultracentrifuge or because of an incorrect divisor of total serum triglyceride. This study showed that whereas HDL cholesterol (HDL-C) and LDL cholesterol (LDL-C) can be reliably estimated in these specimens by simple methods, VLDL-C estimation requires further investigation. A secondary finding was that feeding significantly reduced the serum concentrations of total cholesterol and LDL-C but raised HDL-C. Using the simple methods for determining serum HDL-C and LDL-C concentrations established in the first study, fasting serum cholesterol concentrations were then determined in eight-week-old pigs (6 pigs per group) fed diets containing either fish oil, milkfat, coconut oil, olive oil or cornstarch at an inclusion rate of 4% (w/w) for 3 weeks in the second study. Serum total cholesterol concentration was significantly higher in pigs fed coconut oil than in pigs fed cornstarch or fish oil (p < 0.05). Pigs fed coconut oil, olive oil and milkfat had a significantly higher serum HDL-C concentration than those fed cornstarch or fish oil. There were no significant differences in serum LDL-C concentrations between groups. The serum triglyceride concentrations were higher in pigs receiving coconut oil and milkfat. This study showed that dietary fats that have a hypercholesterolaemic effect in humans tended to raise serum HDL-C rather than LDL-C concentrations in pigs. These findings suggest differences in lipoprotein metabolism between human beings and growing pigs and caution is warranted when making inferences about human lipoprotein metabolism from porcine studies. A balloon angioplasty model of arterial injury in pigs for the assessment of the thrombogenicity of dietary fats was evaluated in the third study. Eight-week-old pigs (8 pigs per group) were fed diets containing milkfat, fish oil or cornstarch at an inclusion rate of 10% w/w for 12 weeks. The group receiving the fish oil diet was withdrawn from the study after 6 weeks because the diet was unpalatable. At the end of the feeding period, angioplasty was performed on the left and right external iliac arteries simultaneously in each pig. One artery was randomly assigned to receive an angioplasty catheter with a 10-mm balloon while the contralateral artery was distended to 12 mm. One hour later, the damaged arterial segments were harvested and the size of the thrombi that formed in each artery were estimated with a technique using autologous, radiolabelled platelets and by a morphometric technique. The thrombi that formed in this study were platelet-rich, typical of those found at the site of atheroma rupture in human beings. Five of 32 arterial segments sustained deep injury, indicated by the presence of tears through the internal elastic lamina. Thrombi could also be seen at the site of balloon injury in arteries that did not sustain deep injury, a finding that contrasts with earlier angioplasty studies in pigs, which have shown that thrombi will only form at sites of deep injury. The presence of deep injury, and the length of the tear through the internal elastic lamina, appeared to influence the size of the thrombus that formed at the angioplasty site. Given that the length of the tear was difficult to control with balloon angioplasty, and that thrombi can be identified in the absence of tears, it may be more desirable to avoid deep injury in this model. When the arteries that sustained deep injury were excluded and thrombus size was evaluated by the radiolabelled platelet method, there was some evidence (p = 0.1) to suggest that the pigs fed the milkfat diet may have had a greater thrombotic tendency than those fed the cornstarch diet. This model, using autologous radiolabelled platelets for estimating thrombus size, shows promise as a method of evaluating the effects of dietary lipids on thrombosis. The activity of the coagulation cascade was assessed by determining the activated partial thromboplastin time (APTT), prothrombin time (TT), thrombin time (TT) and factor VII activity and plasma fibrinogen concentrations during the study evaluating the balloon angioplasty model of arterial injury in pigs (described above). These variables were determined at the beginning of the study and 1, 2, 4, 6, 8, 10 and 12 weeks later. There was some evidence (p = 0.07) to suggest that the group receiving the milkfat diet had a longer APTT than the group receiving the cornstarch diet. This suggests that the milkfat diet may induce less activity within the intrinsic clotting cascade and/or the common clotting cascade than the carbohydrate diet. This finding therefore, does not explain the tendency for thrombus to be larger at the site arterial injury induced by angioplasty in the pigs receiving milkfat than those receiving cornstarch. In the final study, the effect of dietary lipids on blood concentrations of markers of thrombosis, platelet indices, white and red blood cell counts and platelet reactivity in pigs was evaluated. The experimental animals and diets were the same as those presented in the second study. Variables evaluated included plasma TAT and fibrinogen concentrations, platelet numbers, mean platelet volume, plateletcrit, white blood cell (WBC) and red blood cell counts and platelet reactivity, assessed by platelet aggregation and by filtragometry. The inclusion of fish oil in the diet of pigs has affected some variables in a way that could be interpreted as 'prothrombotic' compared to the other dietary treatments. The fish oil group had the highest concentrations of TAT and fibrinogen, the highest platelet and WBC count and the greatest plateletcrit. In contrast, platelet function, as assessed by aggregometry, was lowest in the group receiving fish oil. Olive oil appeared to be relatively 'antithrombotic' when compared to the other dietary treatments. The pigs receiving the olive oil had the lowest, or equal lowest, TAT and fibrinogen concentrations, and plateletcrit and WBC counts. Platelet function, as assessed by aggregometry, was the second lowest. This is noteworthy because the consumption of a Mediterranean-style diet, which is rich in olive oil, has been previously been shown to be protective against coronary artery disease. These findings highlight the difficulty associated with using indicators of thrombosis to predict final thrombus size. The establishment of a thrombus end-point in an appropriate animal model is required in order to evaluate the overall thrombogenic potential of dietary fats. Nevertheless, evaluating the effect of dietary lipids on markers of thrombosis, such as those investigated in this study, may provide important insights into mechanisms that lead to thrombosis.
Content removed due to copyright restrictions:
Allan, F. J., Johnson, R. N., McNutt, P. V., James, K. A. C., Thompson, K. G., & Manktelow, B. W. (2000). Determination of fasting and postprandial lipoprotein cholesterol concentrations in pigs: A comparison of methods. Nutrition Research, 20(11), 1623-1631