The functions of thiamin have been well characterized in both humans and laboratory animals. Present mainly as the diphosphate form (TDP),⁵ thiamin serves as a cofactor for several enzymes in energy metabolism, including pyruvate dehydrogenase and -ketoglutarate dehydrogenase. In the brain, the triphosphate form appears to play a role in nerve membrane function. Hence, a deficiency of thiamin rapidly results in anorexia, impairment of carbohydrate metabolism, and eventually central nervous system dysfunction (Haas 1988).

The occurrence of classic thiamin deficiency syndromes such as beriberi are uncommon in industrialized countries with the exception of alcohol and opiate abusers (Iber et al. 1982). However, mild-to-moderate deficiencies have been reported in populations such as the elderly and athletes (Iber et al. 1982, Van Dam 1978), but the health risks and functional complications of marginal thiamin deficiency are unclear (Carney and Barry 1985, Van der Beek et al. 1988). Although few human or animal studies have measured biochemical and behavioral alterations during a preclinical thiamin deficiency, those which examined the effects of rats consuming <50% of the current NRC recommendation (NRC 1995) for thiamin (3.1 µg/g) were unsuccessful in achieving marginal deficiency symptoms such as reduced growth or decreased erythrocyte transketolase activity (Tagliaferro and Levitsky 1982). The lack of biochemical or physical changes at 50% of the NRC thiamin requirement suggests that the current NRC requirement is overestimated for the rat.

New methodologies and strategies for estimating nutrient requirements in animals have recently been established. In particular, the use of a chemically defined diet, in which crystalline amino acids supply dietary nitrogen, has been shown to estimate nutrient requirements more accurately without the interference of dietary ingredients contributing to the total nutrient content of the diet (Hirakawa et al. 1984). Rate and efficiency of weight gain and total daily food intake are similar in studies using chemically defined or semipurified diets (Hirakawa et al. 1984). In the present study, the thiamin requirement of rats was estimated using titrated doses of thiamin as a component of a chemically defined diet.

MATERIAL AND METHODS

Table 1. Composition of experimental diets [View Table]

RESULTS

Thiamin levels used in trial 1 indicated that the minimum thiamin requirement for maximum growth was close to 0.5 mg/kg. Therefore, we repeated the study using lower dietary levels of thiamin. Food intake during the 3-wk period was significantly lower (P < 0.05) in the 0- and 0.25-mg thiamin/kg groups (9.9 and 13.4 g/d, respectively) than for rats fed 0.5, 0.75, 1.0, 4.0 and 5.0 mg thiamin/kg (16.1 g/d mean; Table 2). Using a broken-line least-squares analysis, weight gain reached a plateau (6.8 g/d) at a thiamin concentration of 0.55 ± 0.07 mg thiamin/kg (Fig. 2). We observed increased spillage of the thiamin-free group which was caused by a few individual rats within the group.

Table 2. Weight gain and food intake of rats fed graded doses of dietary thiamin from thiamin mononitrate (trial 2)1 [View Table]

Hepatic transketolase activity was measured in trial 2 in the groups receiving 0, 0.25, 0.50, 1.0 and 5.0 mg thiamin/kg to compare the effects of thiamin status on weight gain vs. activity of a thiamin-dependent enzyme (Fig. 3). Transketolase activity did not plateau in the dose ranges examined, although significant differences (P < 0.05) were apparent between the thiamin-deficient group and the other groups. Groups fed 0.25, 0.50 and 0.75 mg thiamin/kg had significantly lower (P < 0.05) TK activities than the group provided 5.0 mg thiamin/kg. Using the two-slope broken-line model to determine an objective inflection point, relative TK activity reached an inflection point at 0.53 mg thiamin/kg (Fig. 3). The activity coefficient, expressed as the TPP-effect, was higher (P < 0.05) in the thiamin-deficient group compared with the other groups (data not shown).

DISCUSSION

A study by Mercer et al. (1986) reported that thiamin levels below 1.25 mg thiamin/kg are too low for optimal growth of male rats averaging ~200 g body weight. Graded doses of thiamin hydrochloride (0, 0.3, 0.6, 1.5, 6.0, 30.0 and 100.0 mg /kg) were fed to thiamin-deficient Sprague-Dawley rats as a component of a purified diet containing cornstarch, vitamin-free casein and corn oil. Weight gain responded linearly up to 1.5 mg thiamin/kg, but a saturation kinetics model predicted the thiamin requirement to be 4.0 mg thiamin hydrochloride/kg. The saturation kinetics model for determination of nutritional requirements is based on a sigmoidal weight gain response from titrated levels of a particular nutrient (Mercer et al. 1986 and 1989). Although there is considerable debate concerning the correct statistical method to determine a nutrient requirement (Baker 1986, Mercer et al. 1989, Robbins et al. 1979), our data clearly showed an inflection point in weight gain at a concentration of 0.55 mg thiamin/kg diet.

The minimum thiamin requirement of young pigs is estimated at 1.5 mg/kg (NRC 1988) and that of young chicks is estimated at 1.8 mg/kg (NRC 1994). It seems unreasonable that rats growing relatively more slowly than pigs or chicks and practicing coprophagy would require a higher level of dietary thiamin than chicks or pigs. Our rats fed 0.25 mg thiamin/kg gained weight (3.5 g/d) at a rate that was 50% of maximal growth. In contrast, the thiamin-depleted rats used by Mercer et al. (1986) lost body weight when fed 0.30 mg thiamin hydrochloride/kg. The fact that Mercer et al. (1986) depleted body thiamin stores for a 2-wk period prior to the dose-response study suggests that their value of 4.0 mg thiamin hydrochloride/kg represents the thiamin requirement for repletion of thiamin status in rats weighing ~200 g.

Transketolase and the TPP-effect were measured in trial 2 to compare the effects of thiamin status on weight gain and activity of a thiamin-dependent enzyme. Transketolase is the thiamin-dependent enzyme whose activity is most sensitive to thiamin status and can be detected in high quantities in both erythrocytes and liver (Trebukhina et al. 1983). We chose to measure hepatic TK due to reports of erythrocyte TK variation between younger and more mature blood cells (Harris and Kellenmeyer 1970, Trebukhina et al. 1983). Hepatic TK has been shown to be representative of thiamin status (Harris and Kellenmeyer 1970). Transketolase activities were indicative of thiamin status and increased progressively in rats fed diets containing from 0 to 5.0 mg thiamin/kg. Values did not plateau as did the weight gain data, and a thiamin requirement estimate based on enzyme activity is not apparent from the current data. However, maximal TK activity is probably not necessary for maximum growth (Mercer et al. 1984) and is not the rate-limiting enzyme in the pentose phosphate shunt ( McCandless et al. 1976, Page et al. 1989). Nevertheless, it is of interest to note that using the two-slope broken-line model, relative TK activity reached an inflection point at 0.53 mg thiamin/kg, a level that parallels the thiamin concentration determined to be the requirement of maximum growth.

In a typical diet, thiamin found in dietary ingredients including whole grains, enriched cereals, meat and poultry products can contribute to the total thiamin content of a diet. In our laboratory, male Sprague-Dawley rats consuming 0.25 mg added thiamin/kg in a purified diet containing cornstarch, sucrose, egg white and soy oil did not exhibit reduced body weight gain compared with controls (4.0 mg thiamin/kg) in contrast to the present crystalline amino acid diet (unpublished results). However, the thiamin status of a typical diet can also be compromised by compounds inherently found or added during food processing. These polyphenolic and sulfite compounds (thiaminases) and antithiamin factors have the potential to reduce thiamin bioavailabilty. High temperatures can also interfere with the available thiamin in a diet due to its susceptibility to form maillard products with other food constituents. However, identification of the minimum thiamin requirement of rats consuming a crystalline amino acid diet will enable future investigators the opportunity to explore the physiological alterations of compromised thiamin status without the complications associated with severe thiamin deficiency.