Niacin (nicotinic acid) is an essential nutrient, but it exerts a hypolipidemic action when taken in large amounts (DiPalma and Thayer 1991, Drood et al. 1991, Grundy et al. 1981, Perry 1986). Although the physiological requirement is 17 mg/MJ, the dose necessary to achieve niacin's pharmacological effect is usually in the range of 2-4 g/d. The lipid-lowering action of niacin has also been clinically tested in conjunction with drugs, such as cholestyramine, colestipol, clofibrate and lovastatin (Blankenhorn et al. 1987, Coronary Drug Project Research Group 1975, Kane et al. 1981, Kuo et al. 1987). A noticeable synergistic effect has been observed in these studies.

The potential risk of long-term use of large doses of niacin has not been critically considered in the same way as with other lipid-lowering drugs. Niacin's only consistent deleterious effect in high doses has been the cutaneous flushing and/or itching, which is thought to be caused by prostaglandin-mediated vasodilation (Morrow et al. 1989). In isolated reports, the use of niacin in large doses has been found to be accompanied by cholestatic jaundice with delayed bromosulfthalein clearance, implying potential hepatoxicity (Sugerman and Clark 1974, Winter and Boyer 1973).

Niacin is excreted as methylated pyridones (Shibata and Matsuo 1989). Methylation is carried out by a simple methyl transfer reaction in which S-adenosylmethionine is the methyl donor. Niacin is water soluble; thus, it is not stored in the body beyond its tissue saturation level. Because niacin excretion is dependent upon methionine (Shibata and Matsuo 1989), an intake of niacin 200-400 times greater than the physiological need may affect the metabolism of this essential amino acid. Methionine synthesis requires 5-methyltetrahydrofolate as a methyl donor and vitamin B-12 as a cofactor, and its degradation to cysteine is catalyzed by cystathionine -synthase for which vitamin B-6 is the cofactor (Stipanuk 1986).

The present study was undertaken to investigate the effect of megadoses of niacin on methionine metabolism and on the biochemical status of the vitamins involved, with particular reference to vitamin B-6.

MATERIALS AND METHODS

Table 1. Composition of the semisynthetic diet1 [View Table]

Table 2. Effect of feeding niacin-supplemented diets for 3 wk on the food intake and growth rates of rats (Experiment 1)1 [View Table]

Experiment 1. The rats were randomly divided into three groups of six. The groups received diets containing 47, 400 or 4000 mg niacin/kg diet. All rats had free access to water and their respective diets for 3 wk.

Experiment 2. Three groups of six rats were fed diets containing: 47 or 4000 mg niacin/kg diet or 4000 mg /kg plus vitamin B-6 (10 mg pyridoxine-HCl/kg) for a period of 6 wk.

Table 3. Effect of feeding niacin-supplemented diets for 3 wk on plasma lipid concentrations of rats (Experiment 1)1 [View Table]

Table 4. Effect of feeding niacin-supplemented diets for 3 wk on the plasma and urinary levels of free methionine and cysteine of rats (Experiment 1)1 [View Table]

Table 5. Effect of feeding niacin-supplemented diets for 3 wk on the plasma levels of vitamins of rats (Experiment 1)1 [View Table]

Table 6. Effect of feeding rats diets supplemented with niacin (4000 mg/kg) either alone or in combination with vitamin B-6 (10 mg/kg) for 6 wk on the plasma and urinary levels of methionine and its metabolites in their free forms (Experiment 2)1 [View Table]

RESULTS

Rats fed niacin-supplemented diets (400 mg/kg) had significantly lower plasma concentrations of TG and LDL-cholesterol whereas HDL-cholesterol concentrations were higher than those of the control rats (Table 3). Plasma total cholesterol concentration was also lower in rats fed the niacin-supplemented diet than in controls, but this was significant only in rats fed the highest dose of niacin.

Plasma and urinary methionine levels were significantly higher in niacin-supplemented rats while cysteine levels were lower (Table 4). These differences were significant and dose related.

Plasma concentrations of vitamin B-12 and folic acid were the same in all groups (Table 5). Plasma PLP levels, however, were significantly lower in rats given the highest dose of niacin compared with rats in the other two groups.

In Experiment 2, rats treated with niacin (4000 mg/kg for 6 wk) had higher plasma and urinary levels of not only methionine but also homocysteine than controls (Table 6). As before, cysteine levels were lower in plasma and urine. However, when vitamin B-6 was administered along with the niacin, concentrations of these three substances no longer differed from those in unsupplemented rats.

A similar pattern was observed with body weight gain and food intake (Table 7). Thus, the lower body weight gain and food intake seen in rats supplemented with niacin was not observed when vitamin B-6 was also given.

Table 7. Effect of feeding rats diets supplemented with niacin (4000 mg/kg) either alone or in combination with vitamin B-6 (10 mg/kg) for 6 wk on food intake, growth and plasma lipid concentrations (Experiment 2)1 [View Table]

As in the first experiment, rats supplemented with niacin had a lower plasma concentration of total cholesterol, TG, and LDL-cholesterol but a higher level of HDL-cholesterol (Table 7). Addition of vitamin B-6 to the diet containing niacin produced plasma lipid concentrations essentially the same as those in rats given niacin but without vitamin B-6.

DISCUSSION

The versatile action of niacin on lipoprotein metabolism, as well as its low cost, should make it a drug of choice in patients with dyslipidemia and/or coronary heart disease. Unfortunately, however, these beneficial effects are accompanied by some adverse reactions. The most prominent of these is an intense flushing (Coronary Drug Project Research Group 1975, Knopp et al. 1985), which occurs in most people with consumption of as little as 100 mg orally. This unpleasant side effect, however, is often diminished if niacin is taken with food or aspirin (Kaijser et al. 1979).

The present study is the first to provide evidence suggesting that megadoses of niacin induce hyperhomocysteinemia, a possible risk factor for arterial occlusive diseases. Thus, mild hyperhomocysteinemia has been linked to coronary heart disease (Ubbink et al. 1996), stroke (Coull et al. 1990), and peripheral vascular disease (Taylor et al, 1991). The possible mechanisms by which homocysteine may promote these disorders have been suggested to include oxidative modification of LDL-cholesterol (Olszewski and McCully 1993), vascular endothelial injury and enhanced binding of lipoprotein (a) to fibrin in atherosclerotic plaque (Harpel et al. 1996).

In normal plasma, the majority of homocysteine (~70%) is protein bound, and storage of plasma may cause redistribution of thiols resulting in an increase of protein-bound fraction at the expense of the free form (Ueland and Refsum 1989). Because the present study measured only the free thiols, which comprise only a small percentage of total plasma homocysteine, the results should be interpreted with caution.

Homocysteine is derived from methionine, which may be converted to cysteine via cystathionine in the transulfuration pathway by two PLP-dependent enzymes, cystathionine -synthase and cystathionase (Ubbink et al. 1993). Subsequently, it may be remethylated to methionine in reactions requiring vitamin B-12 and folic acid. Rats fed a niacin-supplemented diet showed elevation of not only free homocysteine but also free methionine levels in both plasma and urine, whereas free cysteine levels were decreased. It was of further interest that plasma PLP levels were markedly reduced in these rats. However, plasma concentrations of vitamin B-12 and folic acid were unaffected. These results suggest that the elevated free homocysteine levels in niacin-treated animals may be a consequence of vitamin B-6 deficiency. This hypothesis is consistent with the important observation that concuurent supplemetation of rats with vitamin B-6 (10 mg/kg diet) and niacin (4000 mg/kg diet) normalized the niacin effects on both the plasma and urinary excretory levels of the free forms of methionine, homocysteine and cysteine. These results are in agreement with others who demonstrated that hyperhomocysteinemia can be successfully treated with a modest daily use of vitamin B-6 (Brattstrom 1996).

The untoward effects of the use of niacin in large doses, such as hyperhomocysteinemia and hypocysteinemia, appear to be possible limiting factors in the use of niacin as a hypolipidemic agent. This is an important observation warranting further studies which should include measurements of total (free + protein-bound) homocysteine and cysteine concentrations in plasma. Niacin in large doses appears to interfere with PLP. The mechanism of interaction between high niacin and PLP cannot be determined at the present time. It is possible that niacin induces catabolic loss of vitamin B-6 or impairs its conversion to an operational coenzyme.

Perhaps the most important observations made in the present study are that the plasma and urinary concentrations of all sulfur amino acids were reversed to their levels in control rats when niacin and vitamin B-6 were given together. It is also noteworthy that the simultaneous administration of the two vitamins did not alter the hypolipidemic action of niacin. These results clearly suggest that a combination of these two vitamins may be a better choice of therapy for lowering lipid status than niacin alone. Clinical trials are warranted to determine whether these experimental results can be extrapolated to humans.