Apparent Paradox of Low-Fat “Healthy” Diets Increasing Plasma Levels of Oxidized Low-Density Lipoprotein and Lipoprotein(a)
There is a wealth of evidence to suggest that low-fat diets, particularly those rich in fruits and vegetables, are “healthy.” In this issue of Arteriosclerosis, Thrombosis, and Vascular Biology, Silaste et al1 report what appears to be a paradox. Feeding a diet low in total fat and saturated fat to 37 healthy women volunteers, even when supplemented with vegetables, berries, and fruit, caused an increase in the plasma levels of oxidized low-density lipoproteins (OxLDL) and lipoprotein(a) [Lp(a)]. The measurement of OxLDL (normalized to apolipoprotein B [apoB]) was made with an antibody (EO6) that specifically recognizes oxidized phospholipids. As previously reported by Tsimikas et al,2 the epitope recognized by EO6 was mainly associated with Lp(a) as indicated by the virtually perfect concordance between Lp(a) levels and EO6. Both of the test diets (low-fat, low-vegetable and low-fat, high-vegetable), produced small but significant decreases in HDL cholesterol and triglyceride levels. Only the low-fat, high-vegetable diet caused a small but significant decrease in total cholesterol and LDL cholesterol levels. Both diets caused a significant increase in plasma α-carotene, β-cryptoxanthin, and vitamin C levels, and both caused a small but significant decrease in plasma lycopene levels. The low-fat, low-vegetable diet caused a small but significant decrease in luetin–zeaxanthin whereas the low-fat, high-vegetable diet increased lutein–zeaxanthin as well as β-carotene levels. Interestingly, there was no change in autoantibody titers to OxLDL with either diet.
See page 498
The authors suggest that these “healthy” diets may have induced a reverse transport of oxidized phospholipids and Lp(a) out of the artery wall and other tissues accounting for the increase in plasma levels. This is certainly a possibility. However, there are other possibilities. Although the mean plasma values for OxLDL-EO6 increased by 87%±51% on the low-fat, low-vegetable diet (P<0.01) and increased by 77%±46% on the low-fat, high-vegetable diet (P<0.01) the median increases of plasma OxLDL were only 27% and 19%, respectively. The authors note in the Methods section that the distribution of OxLDL-EO6 and Lp(a) “… were not normally distributed, and non-parametric tests were used in the statistical analyses of these parameters.” The author’s statistical treatment seems perfectly appropriate. A review of Figure 5 indicates that 15 of the 37 healthy women placed on the low-fat, low-vegetable diet actually had a decrease in OxLDL-EO6 levels whereas one woman had a 200% increase, one a 450% increase, and one a nearly 2000-fold increase with this diet. Similar considerations apply to the Lp(a) levels. The mean Lp(a) levels increased by 11%±4% on the low-fat, high-vegetable diet with a median increase of 9%. Review of Figure 5 indicates that 14 of the 37 women actually had a decrease in Lp(a) levels on the low-fat, high-vegetable diet. Thus, the response to the diets was not homogenous. Despite these observations, the statistical analysis appears valid, and clearly a majority of individuals showed an increase in OxLDL-EO6 and Lp(a) levels on these diets.
Another explanation for these seemingly paradoxical responses may relate to the findings of Hedrick et al.3 They found that short-term feeding (up to 7 days) of an atherogenic diet to LDL receptor null mice caused a dramatic decrease in plasma paraoxonase (PON) activity and mass without a change in hepatic mRNA levels. The decreased PON activity and mass were temporally related to an increase in plasma and HDL lipid hydroperoxides and to a decrease in native apolipoprotein A-I (apoA-I) and apolipoprotein A-II (apoA-II). As the native apoA-I protein disappeared from the circulation, higher molecular weight forms of apoA-I appeared, some of which contained epitopes recognized by the EO6 antibody. After the mice consumed an atherogenic diet for 7 days, the apoA-I-EO6 level reached a maximum. Switching to a chow diet for 3 days resulted in a dramatic increase in native apoA-I to baseline levels, with virtual disappearance of the high molecular weight forms of apoA-I, including the form recognized by EO6. IgG and IgM antibodies were found to be associated with apoA-I in control animals and were dramatically decreased after the 7-day atherogenic diet, returning to near or above baseline levels after switching to the chow diet for 3 days. The authors concluded that in LDL receptor null mice, there are always detectable levels of oxidized lipids in HDL and that these are associated with complexed autoantibodies. In response to the atherogenic diet, there was an increase in the oxidized lipid per HDL particle, which in turn may have led to an increased number of antibodies per particle. The authors speculated that when some threshold was exceeded, this could have led to enhanced plasma clearance of the immune complexes.
Lp(a) levels increased with both diets in the studies by Silaste et al,1 whereas the levels of autoantibodies to the EO6 epitope did not change (in contrast to the situation in acute coronary syndromes in which the levels of autoantibodies to the EO6 epitope, OxLDL-EO6, and Lp(a) all increased2). It is possible that the apparent increase in plasma levels of OxLDL-EO6 in the studies by Silaste et al1 actually reflected a decrease in the rate of formation of Lp(a)-EO6 epitopes. As a result, the number of particles containing a threshold density of Lp(a)-EO6 epitopes required for rapid removal from the circulation via autoantibodies was less and the clearance of Lp(a) was reduced. Thus, the apparent increase in plasma OxLDL-EO6 might actually represent an increase in the number of Lp(a) particles in the plasma, each containing a decreased EO6 density but with an increase in the number of Lp(a) particles sufficient to raise the total plasma level of EO6 epitopes. Because plasma apoB levels remained virtually unchanged by the diets, and because Lp(a) accounts for only a small fraction of apoB, the ratio of EO6 to apoB would increase. This scenario could explain the apparent paradox of increasing oxidation epitopes and increasing levels of Lp(a) resulting from diets that raise plasma antioxidant levels, which are diets known to be associated with decreased atherosclerosis.
Whatever the explanation, the findings by Silaste et al1 are sure to provide the basis for further exciting and potentially important studies.
This work was supported in part by USPHS grant HL 30568.
Silaste M-L, Rantala M, Alfthan G, Aro, A, Witztum JL, Kesaniemi YA, Horkko S. Changes in dietary fat intake alter plasma levels of oxidized low-density lipoprotein and lipoprotein(a). Arterioscler Thromb Vasc Biol. 2004; 24: 498–503.
Hedrick CC, Hassan K, Hough GP, Yoo JH, Simzar S, Quinto CR, Kim SM, Dooley A, Langi S, Navab m, Witztum JL, Fogelmen AM. Short-term feeding of atherogenic diet to mice results in reduction of HDL and paraoxonase that may be mediated by an immune mechanism. Arterioscler Thromb Vasc Biol. 2000; 20: 1946–1952.