QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Blair, P. V. Articles by Harris, R. A. Search for Related Content PubMed PubMed Citation Articles by Blair, P. V. Articles by Harris, R. A.

---

Article

Dietary Thiamin Level Influences Levels of Its Diphosphate Form and Thiamin-Dependent Enzymic Activities of Rat Liver

Paul V. Blair*⁴, Rumi Kobayashi*, Hardy M. Edwards, III,, Neil F. Shay**, David H. Baker and Robert A. Harris*

^* Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, IN 46202–5122 and Division of Nutritional Sciences, Department of Animal Sciences and ^** Department of Food Science and Human Nutrition, University of Illinois, Urbana, IL 61801

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
This study was prompted by our incomplete understanding of the mechanism responsible for the clinical benefits of pharmacological doses of thiamin in some patients with maple syrup urine disease (MSUD) and the question of whether thiamin diphosphate (TDP), a potent inhibitor of the activity of the protein kinase that phosphorylates and inactivates the isolated branched-chain -ketoacid dehydrogenase (BCKDH) complex, affects the activity state of the complex. Rats were fed a chemically-defined diet containing graded levels of thiamin (0, 0.275, 0.55, 5.5, and 55 mg thiamin/kg diet). Maximal weight gain was attained over a 3-wk period only in rats fed diets with 5.5 and 55 mg thiamin/kg. Feeding rats the thiamin-free diet for just 2 d caused loss of nearly half of the TDP from liver mitochondria. Three more days caused over 70% loss, an additional 3 wk, over 90%. Starvation for 2 d had no effect, suggesting a mechanism for conservation of TDP in this nutritional state. Mitochondrial TDP was higher in rats fed pharmacological amounts of thiamin (55 mg thiamin/kg diet) than in rats fed adequate thiamin for maximal growth. Varying dietary thiamin had marked but opposite effects on the activities of -ketoglutarate dehydrogenase (-KGDH) and BCKDH. Thiamin deficiency decreased -KGDH activity, increased BCKDH activity, and increased the proportion of BCKDH in the active, dephosphorylated, state. Excess dietary thiamin had the opposite effects. TDP appears to be more tightly associated with -KGDH than BCKDH in thiamin-deficient rats, perhaps denoting retention of -KGDH activity at the expense of BCKDH activity. Thus, thiamin deficiency and excess cause large changes in mitochondrial TDP levels that have a major influence on the activities of the keto acid dehydrogenase complexes.

KEY WORDS: • rats • thiamin • liver • mitochondria • thiamin-dependent enzymes

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Thiamin in the diphosphate form (TDP)⁵ is an important coenzyme for transketolation and for three decarboxylation-dehydrogenation enzyme complexes, namely pyruvate dehydrogenase, -ketoglutarate dehydrogenase (-KGDH), and branched-chain -ketoacid dehydrogenase (BCKDH). Thiamin is transported across cell membranes by three mechanisms—two saturable components that are responsive to low and high physiological concentrations of thiamin (Bettendorff 1995 , Bettendorff & Wims 1994 ) and one nonsaturable (diffusive) component (Casirola et al. 1988 , Yamamoto et al. 1981 ). A defect in one saturable-specific component, (Pekovich et al. 1998 ) presumably a membrane specific carrier protein, results in thiamin-responsive megaloblastic anemia (TRMA), a syndrome that can be partially corrected by the administration of pharmacological levels of thiamin (or a lipophilic form, benzoyloxymethyl-thiamin), which is probably transported by the diffusive component (Rindi et al. 1994 , Valerio et al. 1998 ). In another deficiency abnormality, the catalytic function of the BCKDH complex is defective, which manifests as maple syrup urine disease (MSUD). The branched-chain amino acids (BCAA) leucine, isoleucine, and valine and their derived keto acids (BCKA) accumulate in the blood of afflicted patients (Menkes et al. 1954 , Westall et al. 1957 ). In the white blood cells of MSUD patients transamination is unaffected, whereas decarboxylation of the BCKA is blocked (Dancis et al. 1960 ). A phenotype of MSUD patients who were responsive to pharmacological doses of dietary thiamin, coupled with restricted intake of BCAA, was reported (Scriver et al. 1971 and 1985 ). In general, thiamin-responsive patients are heterogenous, and consequently the administration of pharmacological thiamin for relief of the symptoms of the MSUD syndrome has had limited success (Chuang and Shih 1995 ).

The first sign of thiamin deficiency is the depletion of tissue thiamin stores, of which TDP comprises over 80% (Rindi & de Giuseppe 1961 ). Before enzymic activity in those reactions that require TDP as a cofactor is curtailed, thiamin is drastically depleted (Warnock et al. 1978 ). There is a rapid decrease in total TDP in rat tissues after only 10 d of feeding a thiamin-deficient diet. Depletion of TDP is greatest and fastest in the liver, followed by the heart, with only a minimal drop occurring in the brain (Dreyfuss 1961 ). It is probable that, in addition to adequate, though not excessive, levels of essential nutrients and energy components in the sustaining diet, the nutritional state of animals (or patients) before initiation of thiamin treatment, is a key factor in the determination of deficiency characteristics (Baker 1986 ).

The mechanism by which pharmacological thiamin doses exert a beneficial effect on MSUD patients is not understood. In the current report we have addressed this problem by using graded levels of thiamin, including a pharmacological level of 55 mg/kg diet, blended into a chemically defined diet to evaluate thiamin status in reference to the activity of the BCKDH complex. In a recent publication (Rains et al. 1997 ), a chemically defined diet for weanling rats was used to specify the minimum thiamin requirement to promote maximal weight gain over 3 wk. This diet easily accommodated the addition of graded levels of thiamin and was used as the basal diet for studying thiamin deficiency and a pharmacological level of dietary thiamin.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Animals and diets.

All animal regimens were in accordance with Indiana University School of Medicine animal care recommendations. Male Sprague-Dawley rats (Harlan, Indianapolis, IN) weighing 95 ± 10 g were confined individually in stainless steel (wire mesh bottom) cages in a controlled (23°C; 12 h light and dark cycle) environment. Each replication commenced with nine rats randomly assigned to treatments. Four replications were conducted over a period of 5 wk. Rat 1 received no treatment and was killed at Time zero. Rats 2 and 3 were fed a non-purified diet(Teklad Rodent Diet (W) No. 8604; Harlan Industries, Indianapolis, IN)—one for 5 d before being killed, the other for 26 d before being killed. Rats 4–9 were acclimated for 5 d on a thiamin-free chemically-defined diet. One was then killed and the remaining five were fed the chemically-defined diet (Rains et al. 1997 ) with thiamin added at a level of 0, 0.275, 0.55, 5.5, or 55 mg thiamin/kg diet. Water was freely available, and droppings through the wire bottom were washed away routinely. Weights of rats were measured before acclimation started (Time zero); upon completion of acclimation (5 d); and twice each week for 3 wk, with the last weight taken just before they were killed.

In the experiment that tested the effects of 5 d acclimation to the thiamin-free chemically-defined diet, four individually caged rats were fed the defined diet and four were fed the non-purified reference diet. After weighing, rats were killed at the end of 5 d.

To compare the results of starvation to the results of feeding the thiamin-free chemically-defined and non-purified reference diets, 3–5 rats, individually caged and weighed, were placed in each group. The starved and non-purified reference diet-fed rats were 3 d older than those fed the thiamin-free diet. After 48 h of treatment, the rats were weighed and killed. All rats were fed the reference diet before starvation or dietary treatments were started.

Animals were killed with a guillotine. Blood (1–3 mL) was collected in 30 mL beakers lined with heparin, then immediately chilled to 0°C in an ice bath. The blood was transferred to 1.5 mL tubes for centrifugation at ~11,000 x g in an Eppendorf microcentrifuge for 2 min. The supernatant fluid (plasma) was decanted into new tubes. Both plasma and cells were quickly frozen at -85°C.

When bleeding was over, the abdominal cavity of each rat was opened and the liver excised. Approximately 1/2 of each liver was quickly frozen in a clamping device immersed in liquid nitrogen just before and after clamping the liver sample. After freezing, the liver half was weighed and then stored at -85°C. The remaining liver half was weighed before placing in chilled (0°C) mitochondrial isolation medium (~30 mL).

Isolation of liver mitochondria.

The chilled livers were rinsed twice with 250 mmol sucrose/L -1 mmol EGTA/L (0°C, pH 7.4) before mincing with scissors (30 s). Before homogenizing and centrifuging by a described procedure (Blair 1977 ), the volume was brought to ~40 mL. A 1-mL sample of the homogenate was withdrawn and quickly frozen (-85°C). The final mitochondrial pellet was suspended in a small volume of sucrose-EGTA.

Protein concentrations of blood plasma, liver homogenates, and liver mitochondria were estimated by a biuret method (Gornall et al. 1949 ) modified to treat the peculiarity of each sample after solubilization with sodium deoxycholate. For example, liver homogenate samples were routinely centrifuged after solubilization and heating (100°C for 20 s) to insure turbidity did not interfere with the specific spectrophotometric reading. Crystalline bovine serum albumin served as the protein standard. After determining oxygen consumption rates stimulated by rat liver mitochondria (RLM), the RLM were frozen (-85°C). All samples were stored at -85°C until the measurable parameters were completed.

Mitchondrial oxygen consumption.

Rat liver mitochondria were added to an isoosmotic reaction solution in a 2 mL water-jacketed cell of a Gilson Oxygraph maintained at 30°C. Oxidizable substrates used were succinate (plus rotenone) and glutamate (plus malate) each at 10 mmol/L. The addition of 1 mg of RLM protein started state-2 respiration (Chance and Williams 1955 ). State-3 respiration was initiated with 600 nmol of adenosine diphosphate (ADP). When ADP-stimulated respiration decreased to approximately the same rate as that seen in state-2, it was assumed the added ADP was consumed [converted to adenosine triphosphate (ATP)], which in turn was the start of state-4 respiration. The respiratory control ratio (state-3 respiration rate ÷ state-4 respiration rate) was used to evaluate RLM integrity.

Thiamin determinations.

Total thiamin, corresponding primarily to TDP, with much lesser amounts of thiamin, thiamin monophosphate, and thiamin triphosphate was determined by a fluorescence assay (Edwin 1979 ) modified to suit our particular experiments. Samples of blood plasma, liver homogenate, or liver mitochondria (1 mg protein in 1 mL) were extracted with HCl acid, and the precipitate was removed by centrifugation in an Eppendorf microcentrifuge (1.5 mL tube, 2 min). The supernatant fluid (1 mL) was neutralized with NaOH prior to diluting in an alcoholic-aqueous solution. The reduced fluorescence spectrum of the solution was run from 390 nm to 490 nm in a Varian (Palo Alto, CA) SF-330 Spectrofluorometer with the excitation wavelength set at 368 nm. The reduced form of thiamin has little or no fluorescence. The thiamin and thiamin phosphates (total thiamin) in solution were oxidized to thiochrome with K₃Fe(CN)₆. After thiochrome formation, the hydrophobic thiamin can be separated from the thiamin phosphates by extraction into isobutanol. The phosphate derivatives remain in the alcoholic-aqueous phase. The ferricyanide color must be cleared with H₂O₂ before the oxidized spectrum can be run. The thiochrome is highly fluorescent. Total thiamin concentration is estimated by the difference in fluorescence of the reduced versus the oxidized form at peak fluorescence, which is about 430 nm. Concentrations were obtained by comparison to standard concentrations of purified thiamin treated and scanned in the same manner. The fluorescence was linear within reasonable limits, at least over the levels of thiamin used in our assays. Thiamin and the thiamin phosphates yield the same quantity of fluorescence under the conditions of our assay.

Activity of enzymes.

Cell-free extracts for the assay of liver BCKDH and -KGDH were prepared essentially as described in previous reports (Aftring et al. 1986 , Shimomura et al. 1990 ). Briefly, one portion of frozen liver was powdered in liquid nitrogen, weighed, and homogenized in an extraction buffer (Shimomura et al. 1990 ) containing Triton-X100, protease inhibitors, -chloroisocaproate (BCKDH kinase inhibitor), potassium fluoride (BCKDH phosphatase inhibitor), and TDP (0.5 mmol/L). A second portion of the liver was homogenized with this same extraction buffer, except that TDP was omitted. Insoluble material was removed from both extracts by centrifugation. Polyethylene glycol (final concentration 9%) was then added to the supernatant fluid to precipitate the BCKDH and -KGDH complexes. The polyethylene glycol precipitates from the extracts prepared with TDP were taken up in a suspending buffer containing 0.5 mmol TDP/L, as described previously (Shimomura et al. 1990 ). The precipitates prepared without TDP were taken up in the same suspending buffer without TDP.

The extracted BCKDH and -KGDH complexes were assayed spectrophotometrically in a cocktail (Paxton et al 1986 , Shimomura et al. 1987 ) with -ketoisovalerate and -ketoglutarate as substrates, respectively. The assays were carried out with and without TDP (0.5 mmol/L) in the assay cocktail. One unit of enzyme activity catalyzed the reduction of 1 µmol of NAD⁺/min at 30°C.

Statistical analysis.

ANOVA (using a multiple crossed classification procedure with body weight, pmol thiamin/mg protein, and mU enzymatic activity/g wet liver as measured variables) was applied to all data, and single degree-of-freedom comparisons were made among treatment means for each diet and measured variable.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Effect of graded levels of dietary thiamin on growth and liver total thiamin levels.

Rats fed the thiamin-free purified diet for 5 d (acclimation period) gained about the same amount of weight as those fed the non-purified reference diet (Table 1). During the 21-d period of feeding the experimental diet, body weights responded to thiamin dosing, particularly after 7 d of feeding. In fact, between d 7 and 21, rats fed either 0 or 0.275 mg thiamin/kg diet lost weight. Weight change at Day 21 of the feeding period increased as a function of dietary thiamin level between 0 and 5.5 mg thiamin/kg diet (P < 0.05). Dietary thiamin levels of 5.5 and 55 mg/kg produced weight gains that did not differ (P > 0.10), and the weight gains of rats fed these diets were not different (P > 0.10) from those of rats fed the non-purified reference diet containing 10.8 mg thiamin/kg diet. Changes in liver weight paralleled the changes in body weight, although liver weight as a percentage of body weight was lower (P < 0.05) in rats receiving thiamin-deficient diets (0.0 to 0.55 mg thiamin/kg diet, respectively) compared to those fed diets containing 5.5 or 55 mg thiamin/kg diet (data not shown).

(Table 2)

(Table 3)

The quality of the mitochondria isolated from the livers of all rats was good, as judged by oxygen consumption studies. The respiratory control ratio (state-3 respiration ÷ state-4 respiration) was in most cases >5, using succinate as oxidizable substrate. The ratio was invariably higher when glutamate was used as a substrate. Glutamate (source of -ketoglutarate) oxidation rates of RLM were not affected by dietary thiamin concentration (data not shown).

Effects of graded levels of dietary thiamin on activity and activity state of the BCKDH complex.

To determine the actual activity of the enzyme, BCKDH activity was measured before activation with a phosphatase, and to give the maximum total activity, after activation with a phosphatase (Table 4). The first response of BCKDH to thiamin deficiency was a lowering of activity as seen after 5 d of acclimation to the thiamin-free diet (Table 4) . However, in rats fed a thiamin-deficient diet during the following 21 d, BCKDH activity was restored and actually exceeded the activity found in rats fed >0.55 mg thiamin/kg diet. The livers of rats fed chemically-defined diets containing 0 and 0.275 mg thiamin/kg diet had BCKDH activities that were double (P < 0.05) those of rats fed 5.5 or 55 mg thiamin/kg diet, in spite of having only 10% of the total thiamin content (Table 2) . A greater (P < 0.05) proportion of BCKDH was in the active (nonphosphorylated) form in rats fed 0 and 0.275 mg thiamin/kg diet versus rats fed 5.5 and 55 mg thiamin/kg diet (65 vs. 33%). In rats fed a 0.55 mg thiamin/kg diet there was an intermediate level of total activity, even though the active form was less prevalent than in rats fed the lower thiamin levels (26 vs. 65%). As expected from previous work (Gillim et al. 1983 , Harris et al. 1985 ), the BCKDH was mostly in the active (nonphosphorylated) form in rats fed the non-purified reference diet. A lower activity state of the complex was expected in the rats fed the purified diet because of its lower protein (amino acid) concentration (16 vs. 24 g/100 g for the non-purified reference diet).

View this table: [in this window] [in a new window]   Table 4. Effect of graded levels of dietary thiamin, fed to rats, on the total liver branched-chain -ketoacid dehydrogenase (BCKDH) activity and the portion of the BCKDH complex present in the active stated

Effects of graded levels of dietary thiamin on latency of BCKDH and -KGDH complex activities.

BCKDH and -KGDH complex activities were measured in liver extracts prepared and assayed in the presence and absence of TDP (Table 5). The portion of the enzyme activity that was latent, i.e. the portion requiring exogenous TDP (apoenzyme) for expression of activity, was then calculated as previously described by Pekovich et al. (1996) . As expected, the activity of BCKDH was lower (P < 0.05) in the absence (apoenzyme) than in the presence of added TDP (holoenzyme), when rats were fed either the chemically-defined diet or the non-purified reference diet. However, the proportion of liver holoenzyme was much less in rats fed the chemically-defined diet (10–31%) than in rats fed the non-purified reference diet (>50%), even though the TDP level in liver mitochondria of the rats fed the chemically-defined diet with high thiamin was comparable to that of rats fed the non-purified reference diet.

View this table: [in this window] [in a new window]   Table 5. Effect of graded levels of thiamin, fed to rats, on the activities of hepatic branched-chain -ketoacid dehydrogenase (BCKDH) and -ketoglutarate dehydrogenase (-KGDH) complexese

The activity of the liver -KGDH complex of rats fed thiamin-deficient diets (0, 0.275, and 0.55 mg thiamin/kg diet) was lower (P < 0.05) than in rats fed diets containing 5.5 mg thiamin/kg diet (Table 5) . When the -KGDH complex was extracted and assayed without added TDP, a substantial amount of activity was found in livers of thiamin-deficient rats. When dietary thiamin exceeded the National Research Council recommended level, almost all of the -KGDH activity was manifested without adding TDP.

Additional evidence for firm binding of TDP to the -KGDH was seen during the adjustment to the chemically-defined diet. The 5 d of acclimation to the chemically-defined thiamin-free diet decreased the liver thiamin from ~200 pmol/mg protein to 60 pmol/mg protein (Table 2) , but the -KGDH remained 90% saturated with TDP when it was extracted and assayed without adding TDP (1,230 mU/g wet liver compared to 1,111 mU/g).

During the 5 d of acclimation to the chemically-defined thiamin-free diet the livers of rats lost 70% of their thiamin (Table 2) , 30% of their total BCKDH activity (Table 4) , and 40% of their -KGDH activity (Table 5) . The amount of nonphosphorylated (active) BCKDH was 23%, and the amount of BCKDH with TDP bound was 5%, whereas the amount of -KGDH with TDP bound was 90%. This clearly signifies that TDP is more firmly bound to -KGDH than to BCKDH, and -KGDH activity is preferentially conserved when TDP becomes limiting in mitochondria.

Only when dietary thiamin is severely limiting (0 mg thiamin/kg diet) for 26 d does liver -KGDH lose most (89%) of its activity (Table 5) , but high latent activity, or apoenzyme, (requires added TDP) remains. In rats fed the other two deficient diets (0.275 and 0.55 mg/kg) more activity (holoenzyme) was retained, with a loss in latency or apoenzyme (Table 5) . In rats fed adequate dietary thiamin (5.5 or 55 mg/kg), almost full -KGDH activity (over 90%) was manifested without added TDP. Although the liver BCKDH of rats fed the thiamin-free diet retained ~12% of its total activity, additional thiamin in the diet (5.5 mg/kg) did not restore latent activity. In fact, even with high thiamin in the purified diet (55 mg/kg) and in the liver (233 pmol/mg protein), 69% of the BCKDH was inactive (phosphorylated) or latent (needed added TDP).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The current National Research Council (1995) recommendation for thiamin in diets of growing rats is 4.0 mg/kg. Recently, the validity of these recommendations or requirements was challenged by Rains et al. (1997) , who suggested the minimum thiamin requirement for maximal weight gain of weanling rats was 0.55 mg thiamin/kg diet. Our experiments carried out with the chemically-defined diet of Rains et al. (1997) indicated that 0.55 mg thiamin/kg diet was inadequate to sustain normal growth for 21 d after 5 d of acclimation with no thiamin in the diet. It should be pointed out that during acclimation the liver thiamin declined from >200 pmol/mg protein to 60 pmol/mg protein (Table 2) , a level apparently sufficient to support weight gain for 7 d but clearly not for 21 d (Table 1) . However, by 9–14 d, thiamin in the liver reached a point where it was probably not sufficient to support maximal weight gain of rats fed diets with thiamin levels of 0.275 and 0.55 mg/kg. The point to be emphasized is that 0.55 mg thiamin/kg diet will not maintain an adequate thiamin concentration in the rat liver for an extended time, i.e., thiamin depletion occurs.

The protocol used by Rains et al. (1997) involved a 5-d pretest period, during which time 5 mg thiamin/kg diet was fed. In the subsequent 21 d, graded doses of thiamin were fed and weight gain and efficiency of diet utilization were maximized at a dietary thiamin level of 0.55 mg/kg. The data herein together with the study of Mercer et al. (1986) involved thiamin dose-titration using rats that had been depleted or reduced in thiamin stores (Table 2) . It appears clear from both our study and the earlier study of Mercer et al. (1986) that 0.55 mg thiamin/kg diet is inadequate to maximize weight gain and maintain thiamin (mainly TDP) stores in the liver of thiamin-depleted weanling rats.

Our data herein indicated that weight gains were considerably greater for rats fed 5.5 mg thiamin/kg diet than for those fed 0.55 mg/kg. Thus, the minimal thiamin requirement for maximal weight gain of thiamin-deprived weanling rats probably lies between 0.55 and 5.5 mg thiamin/kg diet. The surfeit level (55 mg thiamin/kg diet) of thiamin tested resulted in (liver) total thiamin levels in the tissues that were significantly greater (P < 0.05) than those of rats fed 5.5 mg/kg. We do not consider this grounds for suggesting that the dietary thiamin requirement exceeds 5.5 mg/kg. Maximal tissue stores of thiamin may be beneficial for patients with MSUD, but our data indicate it is unlikely that they are necessary for normal animals (or humans). This result may help in understanding the relief of symptoms of the MSUD syndrome in certain patients termed thiamin-responsive (Elsas and Danner 1982 , Fernhoff et al. 1985 , >Scriver et al. 1981 and 1985 ).

Although thiamin-responsive MSUD patients result from a mutation in the DNA that is the template for the E2 (acyl transferase) subunit of BCKDH (Danner et al. 1985 , Fisher et al. 1991 , Herring et al. 1992 ), the BCAA levels in blood are decreased by thiamin inclusion in the nutrient supply. Such MSUD patients may have as little as 3% of normal BCKDH activity, and maximal response to thiamin therapy may require 3 wk. Perhaps this is an indication of long-term thiamin-inducted expression of mutated BCKDH or its component proteins in compensation for the very low catalytic activity of the altered enzyme complex, though this induction of expression is not strongly supported by our data on the activity of the nonmutated BCKDH in liver of thiamin-deficient rats.

It is interesting that the thiamin concentration was not reduced in plasma, liver, or liver mitochondria of rats during starvation for 2 d. In fact, thiamin in liver homogenates was actually increased, perhaps because of a reduction in liver size (nearly 50%) with retention of intracellular thiamin phosphates. In contrast, rats fed the thiamin-free diet for only 2 d had a liver thiamin concentration ~30% of normal and a mitochondrial level ~50% of normal, while body weight gains were normal. Part of the difference in liver thiamin levels can be explained by a need for less thiamin as the body loses mass during starvation versus a need for more thiamin as the body continues to grow while fed a thiamin-free diet. The data also suggest, however, that food deprivation signals the liver of rats to restrict the loss of essential nutrients, such as thiamin.

The normal level of thiamin in the plasma of rats is 40–50 pmol/mg protein. In 5 d, the acclimation period, this level was reduced in rats fed the thiamin-free diet to virtually the minimal concentration (20 to 30 pmol/mg protein). Moreover, plasma thiamin remained at this level for the next 21 d when dietary thiamin levels < 0.55 mg/kg were fed. It takes a few days longer to minimize the liver thiamin, and it was never truly minimized during the 21-d trial in rats fed the 0.55 mg thiamin/kg diet. When the chemically-defined diet contained 5.5 mg thiamin/kg diet, tissue thiamins were replenished and supported maximal weight gain rates even though the plasma and liver concentrations were not yet maximized. This observation is in partial agreement with the study by Mercer et al. (1986) on the thiamin requirement for thiamin-depleted rats, which concluded that the minimal thiamin requirement for optimal growth should be 4.0 mg thiamin hydrochloride/kg diet.

Rats fed thiamin-deficient diets (with only 2.8% BCAA) had double the liver BCKDH activity of rats fed the same diet containing adequate (5.5 mg/kg) or excess thiamin (Table 4) . Almost all of the activity was latent (in the apoenzyme form or phosphorylated), requiring both TDP and dephosphorylation of the enzyme complex to recover full activity. The enhanced activity in livers of thiamin-deficient rats was not caused by increased BCAA in the blood plasma. Amino acid analysis revealed that BCAA in the plasma (108–125 µmol/L) were not different in rats fed the varying levels of thiamin in the chemically-defined diet (Blair, Liechty, & Harris, unpublished observations). The increase in BCKDH activity may be an induction response to the presence of BCAA (though not in excess) and corresponding BCKA, but very little free TDP is available to function in decarboxylation of the corresponding BCKA, which may accumulate. This notion is strengthened by the observation that rats with more than adequate TDP in the liver, fed high thiamin diets, had much less total BCKDH activity with an even lower proportion in the active (dephosphorylated) form, perhaps because the BCKA are kept in low concentration through oxidation. It was also apparent that high TDP concentrations in the liver mitochondria alone did not severely inhibit the BCKDH kinase reaction, as it did with the isolated complex (Lau et al. 1982 ).

It is tempting to speculate that the phosphorylated, inactive form of BCKDH has a much lower affinity for TDP than does the active form. When liver TDP is high and the active form is low, there is little BCKDH activity (<10%) until dephosphorylation is accomplished and TDP is added. TDP alone restores only ~25% of the BCKDH activity. On the other hand, when liver TDP is high and the active form is high (>80% of the total) there is ~50% of the total BCKDH activity in the absence of added TDP. Rats fed the non-purified reference diet containing 24% protein (of which 4.5% is BCAA) and 10.84 mg thiamin/kg diet had a liver concentration of 187 pmol TDP/mg protein, ~50% of total liver BCKDH activity in the absence of added TDP, and 82% of activity before activation with phosphatase. However, rats fed the chemically-defined diet containing 16% amino acids (of which 2.8% were BCAA) and 5.5 mg thiamin/kg diet had a liver concentration of 159 pmol TDP/mg protein, 7% of liver BCKDH activity in the absence of added TDP, and 26% of activity before activation with phosphatase. This chemically-defined diet may be considered to be a low-protein diet (Gilliam et al. 1983 , Harris et al. 1985 , Wohlheuter and Harper 1970 ), but it is adequate to promote maximal growth of weanling rats for 3 wk, indicating that protein synthesis is not restricted (Table 1) .

Before thiamin deficiency can increase BCKDH activity in rats, weight gain must be curtailed, liver thiamin concentration minimized, and perhaps weight loss must start. Consuming the thiamin-free diet for 5 d dramatically reduces the active form (nonphosphorylated) of BCKDH and causes a concomitant, almost absolute, requirement for exogenous TDP; very little TDP (~5%) is bound to the enzyme complex (Tables 4 and 5) . However, at some point during the next 21 d, in rats fed the thiamin-free diet (or 0.275 mg thiamin/kg diet) this response was reversed to the extent that total BCKDH activity was more than doubled with >60% of the enzyme complex in the active form (Table 4) . It may be that low (inadequate) liver TDP coupled to adequate BCAA in the diet induces the expression of BCKDH because the amino acids are not used for protein synthesis, but are used for energy via transamination to the corresponding ketoacids that subsequently are oxidized. But if TDP is deficient, the BCKA may induce the synthesis of BCKDH to oxidize the BCKA. However, more BCKDH will not solve the problem, whereas more TDP would.

Our experiments with weanling rats indicate a delicate balance between thiamin deficiency and adequacy. Cycling between the two could help relieve the symptoms of MSUD by lowering BCAA and BCKA in the blood by using excess amino acids for energy. Our experiments do not provide evidence that additional thiamin in the liver of rats leads to greater BCKDH activity and subsequent lowering of BCAA in the blood plasma. It may be that increased TDP in the liver of patients with thiamin-responsive MSUD enhances the decarboxylation activity of the BCKDH complex and the removal of the BCKA, but not the acyl transferase, which is the mutated subunit in such patients (Danner et al. 1985 ). Thus, the ketoacid derivatives may fully saturate the active (dephosphorylated) complexes because the mutated subunit may exhibit as little as 3% of normal activity. This saturation could enhance the dephosphorylation of the mutated BCKDH to the extent that the ketoacid derivatives are reduced to an acceptable level by the catalytic action brought about by an increase in active faulty transferase.

Rats fed a chemically-defined diet containing inadequate thiamin (0, 0.275, and 0.55 mg thiamin/kg diet) to promote growth had low liver -KGDH activity and a low amount of TDP bound to the enzyme complex. Even at low mitochondrial TDP concentrations (16–21 pmol TDP/mg protein) substantial TDP was bound to the -KGDH, but over half of the enzyme activity was exhibited after the addition of TDP to the assay. In agreement with Pekovich et al. (1996) , -KGDH activity is resistant to thiamin depletion, and by inference the enzyme complex is resistant to mitochondrial proteases even when the prosthetic group (TDP) is not bound. Thus, TDP binding is not required to stabilize the -KGDH against degradative enzymes, although other coenzymes of the complex may be needed.

When nondeficient levels of thiamin were included in the chemically-defined diet fed to rats, -KGDH activity was >90% of maximal without adding TDP during extraction or assay. This indicates a high affinity or firm binding of the coenzyme to the protein. In contrast, at the same dietary thiamin levels, BCKDH activity was low before adding TDP and dephosphorylating the complex, a result which illustrates differential binding affinity of TDP to the two similar enzyme complexes. Certainly, in a condition where mitochondrial TDP is limited, -KGDH would be highly favored over BCKDH to bind TDP to the enzyme complex. Such a hierarchy could be very important for continued energy generation by a cell or tissue at times of thiamin insufficiency. This differential affinity of TDP for the two decarboxylation-dehydrogenation complexes is more sharply illustrated when comparisons are made after 5 d of acclimation to the chemically-defined thiamin-free diet. The mitochondrial TDP concentration was 65 pmol/mg protein (<30% of normal). The BCKDH activity (57 mU/g wet liver) in the absence of added phosphatase and TDP was 5% of its possible activity (1129 mU/g wet liver) in the presence of phosphatase (to dephosphorylate) and TDP, whereas -KGDH activity in the absence of added TDP was 90% of its possible activity (Tables 4 and 5) .

In summary, the minimal amount of thiamin in a chemically-defined diet required to promote maximal rates of gain in body weight for weanling rats for 21 d after 5 d of acclimation to a thiamin-free diet exceeds 0.55 mg thiamin/kg diet. Dietary thiamin deficiency reduces the plasma, liver, and liver mitochondrial levels of TDP, and excess thiamin significantly increases the liver and liver mitochondrial levels of TDP. Acclimation to the chemically-defined thiamin-free diet does not curtail growth, but reductions in thiamin compounds in the plasma and liver are significant. Starvation does not deplete plasma and liver thiamin. In thiamin deficiency, the tighter binding of TDP by -KGDH allows retention of -KGDH activity, perhaps at the expense of BCKDH activity. Excess dietary thiamin does not induce greater expression of BCKDH in the liver. In fact, dietary thiamin deficiency may induce this greater expression of BCKDH. Large alterations in mitochondrial TDP levels do not affect the stability of the BCKDH complex nor the activity of its kinase, nor is TDP binding required to stabilize -KGDH against degradative enzymes.

	FOOTNOTES

 
⁴ To whom correspondence should be addressed.

¹ Presented in abstract form in part at American Society of Biochemistry and Molecular Biology Meetings, Washington, DC [Blair, P. V., Kobayashi, R., Harris, R. A., Baker, D. H., Edwards, H. M., III & Shay, N. F. (1998) Mitochondrial thiamin diphosphate level and the activities of thiamin-dependent enzymes. FASEB J. 12: A1415 (abs)].

Also presented in abstract form in part at American Society of Cell Biology Meeting, Washington, DC [Blair, P. V., Kobayashi, R., Baker, D. H. & Harris, R. A. (1997) Thiamin diphosphate levels in mitochondria isolated from livers of rats fed diets deficient in thiamin. Molec. Biol. Cell 8: 445A (abs)].

² Supported in part by awards from the Grace M. Showalter Trust, PHS DK19259 (rah), International Life Sciences Institute, North America, and the National Institutes of Health (AG13586).

³ The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.

⁵ Abbreviations used: -KGDH, -ketoglutarate dehydrogenase; ADP, adinosine diphosphate; ATP, adinosine triphosphate; BCAA, branched-chain amino acid; BCKA, branched-chain -ketoacid; BCKDH, branched-chain -ketoacid dehydrogenase; MSUD, maple syrup urine disease; NRC, National Research Council; RLM, rat liver mitochondria; TDP, thiamin diphosphate; TRMA, thiamin-responsive megaloblastic anemia.

Manuscript received October 13, 1998. Initial review completed November 4, 1998. Revision accepted December 1, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Aftring R. P., Black K. P., Buse M. G. Leucine and isoleucine activate skeletal muscle branched-chain -ketoacid dehydrogenase in vivo. Am. J. Physiol. 1986;250:E599-E604[Abstract/Free Full Text]

2. Baker D. H. Problems and pitfalls in animal experiments designed to establish dietary requirements for essential nutrients. J. Nutr. 1986;116:2339-2349

3. Bettendorff L., Wins P. Mechanism of thiamin transport in neuroblastoma cells inhibition of a high affinity carrier by sodium channel activators and dependence of thiamin uptake on membrane potential and intracellular ATP. J. Biol. Chem. 1994;269:14379-14385[Abstract/Free Full Text]

4. Bettendorff L. Thiamine homeostasis in neuroblastoma cells. Neurochem. Int. 1995;26:295-302[Medline]

5. Blair P. V. Induction of mitochondrial contraction and concomitant inhibition of succinate oxidation by magnesium ions. Arch. Biochem. Biophys. 1977;181:550-568[Medline]

6. Casirola D., Ferrari G., Gastaldi G., Patrini C., Rindi G. Transport of thiamin by brush-border membrane vesicles from rat small intestine. J. Physiol. 1988;398:329-339[Abstract/Free Full Text]

7. Chance B., Williams G. R. Respiratory enzymes in oxidative phosphorylation III: The steady state. J. Biol. Chem. 1955;217:409-427[Free Full Text]

8. Chuang D. T., Shih G. R. Disorders of branched-chain amino acid and keto acid metabolisms. Scriver C. R. Beaudet A. L. Sly W. S. Valle D. eds. The Metabolic and Molecular Basis of Inherited Diseases 1995;Vol. 1:1239-1277 McGraw-Hill New York, NY.

9. Dancis J., Kutzke J., Rokkones T. Metabolism of white blood cells in maple syrup urine disease. Biochim. Biophys. Acta 1960;43:342-343

10. Danner D. J., Armstrong N., Heffelfinger S. C., Sewell E. T., Priest J. H., Elsas L. J. Absence of branched chain acyltransferase as a cause of maple syrup urine disease. J. Clin. Invest. 1985;75:858-860

11. Dreyfuss P. M. The quantitative histochemical distribution of thiamine in deficient rat brain. J. Neurochem. 1961;8:139-145[Medline]

12. Edwin E. E. An improved procedure for the determination of thiamine. Methods in Enzymology 1979;62:51-54[Medline]

13. Elsas L. J., Danner D. J. The role of thiamin in maple syrup urine disease. Ann. NY Acad. Sci. 1982;378:404-427[Medline]

14. Fernhoff P. M., Lubitz D., Danner D. J., Dembure P. P., Schwartz H. P., Hillman R., Bier D. M., Elsas L. J. Thiamine response in maple syrup urine disease. Pediatr. Res. 1985;19:1011-1016[Medline]

15. Fisher C. R., Fisher C. W., Chuang D. T., Cox R. P. Occurence of a Tyr393Asn (Y393N) mutation in the El gene of the branched-chain -keto acid dehydrogenase complex in maple syrup urine disease patients from a Menonite population. Am. J. Hum. Genet. 1991;49:429-434[Medline]

16. Gillim S. E., Paxton R., Cook G. A., Harris R. A. Activity state of the branched-chain -ketoacid dehydrogenase complex in heart, liver, and kidney of normal, fasted, diabetic, and protein starved rats. Biochem. Biophys. Res. Commun. 1983;111:74-81[Medline]

17. Gornall G. A., Bardawill C. J., David M. M. Determination of serum proteins by means of the biuret reaction. J. Biol. Chem. 1949;177:751-766[Free Full Text]

18. Harris R. A., Powell S. M., Paxton R., Gillim S. E., Nagae H. Physiological covalent regulation of rat liver branched-chain -ketoacid dehydrogenase. Arch. Biochem. Biophys. 1985;243:542-555[Medline]

19. Herring W. J., McKean M., Dracopoli N., Danner D. J. Branched chain acyltransferase absence due to an Ala-based genomic deletion allele and an exon skipping in a compound heterozygote proband expressing maple syrup urine disease. Biochim. Biophys. Acta 1992;1138:236-242[Medline]

20. Lau K. S., Fatania H. R., Randle P. J. Regulation of the branched-chain 2-oxoacid dehydrogenase kinase reaction. FEBS Lett 1982;144:57-62[Medline]

21. Menkes J. H., Hurst P. L., Craig J. M. A new syndrome: Progressive familial infantile cerebral dysfunction associated with an unusual urinary substance. Pediatrics 1954;14:462-466[Abstract/Free Full Text]

22. Mercer L. P., Dodds S. J., Gustafson J. M. The determination of nutritional requirements: A modeling approach. Nutr. Rept. Int. 1986;34:337-350

23. National Research Council Nutrient Requirements of Laboratory Animals 4^th rev. ed. 1995 National Academy of Sciences Washington, D.C.

24. Paxton R., Scislowski P.W.D., Davis E. J., Harris R. A. Role of branched-chain 2-oxo acid dehydrogenase and pyruvate dehydrogenase in 2-oxobutyrate metabolism. Biochem. J. 1986;234:295-303[Medline]

25. Pekovich S. R., Martin P. R., Singleton C. K. Thiamin pyrophosphate-requiring enzymes are altered during pyrothiamin-induced thiamin deficiency in cultured human lymphoblasts. J. Nutr. 1996;126:1791-1798

26. Pekovich S. R., Poggi U., Martin R. R., Singleton C. K. Sensitivity to thiamin deficiency in cultured human cells is dependent on cell type and is enhanced in cells from thiamin-responsive megaloblastic anemia patients. J. Nutr. Biochem. 1998;9:215-222

27. Rains T. M., Emmert J. L., Baker D. H., Shay N. F. Minimum thiamin requirement of weanling Sprague-Dawley outbred rats. J. Nutr. 1997;127:167-170[Abstract/Free Full Text]

28. Rindi G., de Giuseppi L. A new chromotrophic method for the determination of thiamin and its mono-, di-, and tri-phosphates in animal tissues. Biochem. J. 1961;78:602-606

29. Rindi G., Patrini C., Laforcuza U., Mandel H., Berant M., Viana M. B., Poggi V., Zarra A.W.F. Further studies on erythrocyte thiamin transport and phosphorylation in seven patients with thiamin-responsive megaloblastic anemia. J. Inher. Metab. Dis. 1994;17:667-677[Medline]

30. Scriver C. R., MacKenzie S., Chow C. L., Delvin E. Thiamin responsive maple syrup urine disease. Lancet 1971;1:310-312[Medline]

31. Scriver C. R., Clow C. L., George H. So-called thiamin-responsive maple syrup urine disease: 15-year follow-up of the original patient. J. Pediatr. 1985;107:763-765[Medline]

32. Shimomura Y, Paxton R., Ozawa T., Harris R. A. Purification of branched-chain -ketoacid dehydrogenase complex from rat liver. Anal. Biochem. 1987;163:74-78[Medline]

33. Shimomura Y., Suzuki T., Saitoh S., Tasaki Y., Harris R. A., Suzuki M. Activation of branched-chain -ketoacid dehydrogenase complex by exercise: effect of high-fat diet intake. J. Appl. Physiol. 1990;68:161-165[Abstract/Free Full Text]

34. Valerio G., Franzese A., Poggi V., Tenore A. Long-term follow-up of diabetes in two patients with thiamin-responsive megaloblastic anemia syndrome. Diabetes Care 1998;21:38-41[Abstract]

35. Warnock L. G., Prudhomme C. R., Wagner C. The determination of thiamin pyrophosphate in blood and other tissues, and its correlation with erythrocyte transketolase activity. J. Nutr. 1978;108:421-427

36. Westall R. G., Dancis J., Miller S. Maple syrup urine disease. Am. J. Dis. Child. 1957;94:571-572

37. Wohlheuter R. M., Harper A. E. Co-induction of rat liver branched-chain -ketoacid dehydrogenase activities. J. Biol. Chem. 1970;245:2391-2401[Abstract/Free Full Text]

38. Yamamoto S., Koyama S., Kawasaki T. Properties of the active thiamin transport system in Ehrlich ascites tumor cells. J. Biochem. (Tokyo) 1981;89:809-816[Abstract/Free Full Text]

This article has been cited by other articles:

				L. Balk, P.-A. Hagerroth, G. Akerman, M. Hanson, U. Tjarnlund, T. Hansson, G. T. Hallgrimsson, Y. Zebuhr, D. Broman, T. Morner, et al. Wild birds of declining European species are dying from a thiamine deficiency syndrome PNAS, July 21, 2009; 106(29): 12001 - 12006. [Abstract] [Full Text] [PDF]

				D. T. Chuang, J. L. Chuang, and R. M. Wynn Lessons from Genetic Disorders of Branched-Chain Amino Acid Metabolism J. Nutr., January 1, 2006; 136(1): 243S - 249S. [Abstract] [Full Text] [PDF]

				J. L. Chuang, R. M. Wynn, C. C. Moss, J.-l. Song, J. Li, N. Awad, H. Mandel, and D. T. Chuang Structural and Biochemical Basis for Novel Mutations in Homozygous Israeli Maple Syrup Urine Disease Patients: A PROPOSED MECHANISM FOR THE THIAMIN-RESPONSIVE PHENOTYPE J. Biol. Chem., April 23, 2004; 279(17): 17792 - 17800. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Blair, P. V. Articles by Harris, R. A. Search for Related Content PubMed PubMed Citation Articles by Blair, P. V. Articles by Harris, R. A.