Nutrient Metabolism

Carnitine Deﬁciency and Supplementation Do Not Affect the Gene Expression of Carnitine Biosynthetic Enzymes in Rats1,2 Alan T. Davis3 and Thomas J. Monroe* Departments of Surgery, Michigan State University and Spectrum Health, Grand Rapids, MI and *Department of Molecular Biology, Spectrum Health, Grand Rapids, MI

ABSTRACT Starved male weanling rats supplemented with 20 mmol/L pivalate in their drinking water exhibit Downloaded from https://academic.oup.com/jn/article/135/4/761/4663769 by guest on 30 September 2021 significantly depressed concentrations of carnitine in tissues and plasma. In addition, pivalate supplementation has been linked with increased renal and hepatic trimethyllysine hydroxylase (TMLH) activity, whereas carnitine supplementation has been associated with significantly decreased hepatic ␥-butyrobetaine hydroxylase (BBH) activity. The purpose of this study was to determine whether pivalate or carnitine supplementation affects the activity and genetic expression of 2 enzymes of carnitine (Cn) biosynthesis, TMLH and BBH, expressed as mRNA abundance, relative to the abundance of ␤-actin mRNA. Male weanling rats were administered the control treatment (C; n ϭ 6), the pivalate treatment (P; n ϭ 7), or the pivalate treatment plus supplemental dietary carnitine (PϩCn; n ϭ 7). Rats in group P had elevated renal TMLH activity, relative to the other groups (P Ͻ 0.05). The groups did not differ in the abundance of renal or hepatic TMLH or BBH mRNA. A previously unreported finding was the quantifiable level of renal BBH mRNA, which was verified by direct sequencing of the BBH cDNA product amplified from kidney RNA. The groups did not differ in renal BBH mRNA abundance and renal BBH enzyme activity was not detected. Thus, the alterations in enzyme activities in the pivalate-treated rats are not regulated at the transcriptional level, and are apparently related to post-transcriptional effects on the enzymes themselves. J. Nutr. 135: 761–764, 2005.

KEY WORDS: ● carnitine ● carnitine biosynthesis ● trimethyllysine ● ␥-butyrobetaine

Carnitine is a naturally occurring compound in mammalian ciency of ␥-butyrobetaine metabolized to carnitine. In addi- energy metabolism. Its functions include the facilitation of tion, thyroxine was reported to significantly increase liver long-chain fatty acid oxidation, elimination of toxic metabo- carnitine concentration, as well as hepatic BBH activity lites of acyl CoA excess, modulation of the free CoA to acyl (11,12). Whether these alterations in carnitine concentration CoA ratio, storage of energy as acetylcarnitine, and the inter- are directly related to or enhanced by altered TMLH or BBH compartmental shuttling of energy substrates (1). The first activities is unknown. enzyme in the carnitine biosynthetic pathway, trimethyllysine A previous study from this laboratory, using the pivalate 4 hydroxylase (TMLH), hydroxylates trimethyllysine to 3-hy- model of secondary carnitine deficiency in rats (13), showed droxy-trimethyllysine, whereas the final enzyme in the path- that TMLH activity was greater in kidney, liver, and heart of ␥ ␥ way, -butyrobetaine hydroxylase (BBH), hydroxylates -bu- pivalate-treated rats compared with controls (14). In addition, tyrobetaine to carnitine. The biosynthesis of carnitine is BBH activity was depressed in rats fed a carnitine-supple- thought to be regulated by the availability of trimethyllysine mented diet relative to controls. It was unclear from these (2,3). In their review of previous studies (2–6), Vaz et al. (1) results, however, whether these effects were related to a direct noted that the capacity of the carnitine biosynthetic pathway ␥ effect upon the enzymes themselves, or to an alteration in the to generate carnitine from trimethyllysine and -butyrobe- expression of the enzymes. taine far exceeds the amount of carnitine utilized. The purpose of the current study was to determine whether It has been noted, however, that during starvation (7,8), pivalate alone or in combination with supplemental carnitine clofibrate administration (9), and lactation (10), there are marked effects upon carnitine distribution and/or the effi- alters the metabolism of trimethyllysine, via alterations in TMLH activity, expression of the TMLH mRNA, tissue concentration, and/or urinary excretion. In addition, a goal was to 1 Presented at Experimental Biology 03, April, 2003, San Diego, CA [Davis, determine whether pivalate alone or in combination with A. T. & Monroe, T. J. (2003) Expression of carnitine biosynthetic enzymes is supplemental carnitine alters the biosynthesis of carnitine, via unaltered in carnitine deficient and carnitine supplemented rats (Program addendum, abstract LB403)]. alterations in BBH activity and expression of the BBH 2 Supported by grants from the Blodgett Butterworth Health Care Foundation. mRNA, as well as tissue concentration and/or urinary excre- 3 To whom correspondence should be addressed. E-mail: [email protected]. tion of ␥-butyrobetaine and carnitine. Specifically, the work- 4 Abbreviations used: BBH, ␥-butyrobetaine hydroxylase; C, control rats; P, rats receiving pivalate; PϩCn, rats receiving pivalate and supplemental carnitine; ing hypothesis was that the activity and mRNA expression of TMLH, trimethyllysine hydroxylase. both enzymes would be increased in the pivalate-treated rats

761 762 DAVIS AND MONROE and decreased in the carnitine-supplemented rats relative to 100% Eluent C. From 40.9 to 41 min after injection, a linear gradient controls. between Eluents C and A resulted in a concentration of 100% Eluent A. The column was then reequilibrated for 15 min before the next injection. MATERIALS AND METHODS RNA isolation and PCR amplification. Expression of TMLH mRNA was determined using RT-PCR with fluorescence quantita- Animals and diets. Male weanling Sprague-Dawley rats (n ϭ 20; tion, using the set of primers described by Vaz et al. (23). Expression Charles River) were housed individually in polycarbonate cages in a of BBH mRNA was determined using RT-PCR with fluorescence room maintained at 21 Ϯ 2°C and 50 Ϯ 10% humidity with a 12-h quantitation, using a set of primers described by Galland et al. (24). light:dark cycle. The rats were housed at the West Michigan Regional The expression of ␤-actin was used as a control (25). TMLH, BBH, Laboratory, whose Animal Care and Use Committee approved the and ␤-actin cDNA were amplified from 600 ng total RNA extracted study. Rats were maintained in accordance with the NIH guidelines from rat kidney and liver using the Superscript 1-step RT-PCR system for the care and use of laboratory animals. The rats were randomly (Invitrogen). Amplification linearity of the 3 RNA species using the assigned to 1 of 3 groups. Control rats (group C, n ϭ 6) were freely same thermocycling profile and cycle number was determined empir- fed a nutritionally complete purified diet, AIN-76A [(15); Harlan ically before quantitative experiments. The RT-PCR amplification Teklad]. Analysis in this laboratory determined the carnitine con- product quantification was determined using PicoGreen (Molecular centration of this diet to be 1.2 nmol/g. Rats in Group C were Probes) and a fluorescence microplate reader (Cytofluor series 4000, Downloaded from https://academic.oup.com/jn/article/135/4/761/4663769 by guest on 30 September 2021 administered 20 mmol/L sodium bicarbonate in their drinking water, Perspective Biosystems). as described previously (13). Rats in the pivalate group (Group P, n Statistics. The values shown in the text and tables are means ϭ 7) were fed the same diet as the rats in Group C, and their water Ϯ SEM, except where indicated otherwise. Due to the wide range of contained 20 mmol/L sodium pivalate. Rats in the carnitine-supple- variability for the total carnitine concentrations and for the urine mented group (Group PϩCn, n ϭ 7) were fed the AIN-76A diet ␥-butyrobetaine excretion, all of these values were transformed by supplemented with 0.067 mmol carnitine/g diet. This concentration taking the natural logarithm of the original data before analysis. The of carnitine in the diet was used previously and produced 400 and data were analyzed using 1-way ANOVA and the Fisher’s Protected 200% increases in plasma and skeletal muscle total carnitine concen- Least Significant Difference (FPLSD) test. Differences were consid- tration, respectively (16). Rats in Group PϩCn were administered 20 ered significant at P Ͻ 0.05. All analyses were conducted with NCSS mmol/L sodium pivalate in their drinking water. The rats in all 3 2004 (Number Cruncher Statistical Systems). groups remained in their cages and were administered these treat- ments for 14 d, at which time the rats were transferred to individual metabolism cages for an additional 48-h period. During this time, RESULTS 24-h urinary excretions were collected (in 6 mol/L HCl), and 24-h food and fluid intake were recorded. Carnitine and carnitine precursor concentrations in tissue At the end of the study period, the rats were anesthetized with and urine. Provision of pivalate in the drinking water of the isoflurane and killed by decapitation. Blood was collected into hep- Group P rats significantly decreased total carnitine in plasma arinized tubes and plasma separated by centrifugation at 1500 ϫ g for and urine compared with controls (Table 1). Addition of 10 min. Plasma samples were frozen at Ϫ80°C until they were carnitine to the diet of Group PϩCn rats significantly and analyzed. Samples of liver, skeletal muscle, heart, and kidney were markedly increased plasma and urine total carnitine relative to obtained from each rat, and freeze-clamped in aluminum tongs cooled rats in the other 2 groups. Group PϩCn rats excreted signif- in liquid nitrogen. A smaller sample of liver and kidney from each rat icantly less trimethyllysine than either of the other 2 groups. was first submerged in RNAlater (Ambien) before freeze-clamping ϩ Ϫ Group P Cn rats had significantly higher concentrations of the tissue. The tissues remained frozen at 80°C until they were ␥-butyrobetaine in all tissues tested, in addition to signifi- analyzed. The volume of urine was measured in a graduated cylinder, ␥ then mixed thoroughly, filtered, and an aliquot saved for further cantly higher -butyrobetaine excretion in urine relative to analysis. Frozen blood, urine, liver, skeletal muscle, heart, and kidney were analyzed for carnitine and free trimethyllysine as described previously TABLE 1 (17). TMLH activity was determined by the method of Davis (18), whereas BBH was determined by the technique of Vaz et al. (19). ␥ ␥ Concentrations of total carnitine, trimethyllysine, and -Butyrobetaine assay. -Butyrobetaine samples were prepared ␥ using the technique of Janssens et al. (20), utilizing hexanoylcarni- -butyrobetaine in plasma, liver, skeletal muscle and urine tine as the internal standard. The derivatization and analysis of the of rats in the C, P, and PϩCn groups1 samples were conducted using the techniques of VanKempen and Odle (21) and Minkler et al. (22), as modified below. The HPLC CP PϩCn system consisted of 2 Beckman 110B pumps (Beckman Coulter), an Alcott Chromatography Model 718 autosampler, and a FD-300 flu- n 67 7 orescence detector (Groton Technology). A switching valve was used Total carnitine to switch between eluents A and B (Neptune Research). The fluo- Plasma, ␮mol/L 18.6 Ϯ 1.1b 3.6 Ϯ 0.2a 132.7 Ϯ 10.4c rescence detector was operated with an excitation wavelength of 259 Liver, nmol/g 89.3 Ϯ 17.4a 56.8 Ϯ 4.7a 780.8 Ϯ 159.9b nm, an emission wavelength of 394 nm, a lamp frequency of 60 Hz, Urine, nmol/24 h 191 Ϯ 12b 53 Ϯ 8a 278,894 Ϯ 23,808c and a response time of 1 s. The analytical column was a 100 ϫ 4.6 Trimethyllysine ␮ Plasma, ␮mol/L 2.0 Ϯ 0.7 2.2 Ϯ 0.4 1.0 Ϯ 0.2 mm i.d. Nucleosil C8 (3 m, 120 A) column purchased from Phe- Ϯ Ϯ Ϯ nomenex. Liver, nmol/g 1.0 0.2 1.9 0.2 1.8 0.3 Skeletal muscle, Three eluents were used: Eluent A contained acetonitrile:water nmol/g 19.6 Ϯ 2.6 21.5 Ϯ 4.2 13.5 Ϯ 1.7 (80:20), Eluent B was 100% water, and Eluent C contained 800 mL Urine, nmol/24 h 447 Ϯ 41b 471 Ϯ 64b 245 Ϯ 32a acetonitrile, 200 mL water, 8 mL triethylamine, and 6.4 mL phos- ␥-Butyrobetaine phoric acid. Initially, 100% Eluent A was pumped at a flow rate of 1.0 Plasma, ␮mol/L 0.7 Ϯ 0.2a 0.7 Ϯ 0.2a 11.5 Ϯ 1.2b mL/min. At 0.2 min after sample injection, Eluent A was replaced Liver, nmol/g 2.1 Ϯ 0.6a 1.4 Ϯ 0.3a 4.7 Ϯ 0.8b with 100% Eluent B. From 2.9 to 3 min after injection, a linear Skeletal muscle, gradient between Eluents B and C resulted in a concentration of 90% nmol/g 9.1 Ϯ 0.8a 14.6 Ϯ 1.7b 31.0 Ϯ 2.2c Eluent B:10% Eluent C. From 3 to 38 min after injection, a linear Urine, nmol/24 h 312 Ϯ 39a 285 Ϯ 42a 174,067 Ϯ 73,626b gradient between Eluents B and C resulted in a concentration of 48% Eluent B:52% Eluent C. From 38 to 38.1 min after injection, a linear 1 Values are means Ϯ SEM. Means in a row with superscripts gradient between Eluents B and C resulted in a concentration of without a common letter differ, P Ͻ 0.05. EFFECT OF CARNITINE DEFICIENCY OR EXCESS IN RATS 763 rats in the other 2 groups. Group P rats had a significantly TABLE 2 higher concentration of ␥-butyrobetaine in skeletal muscle than the control rats. TMLH and BBH mRNA abundance, relative to ␤-actin mRNA, Renal and hepatic TMLH and BBH activities and mRNA in liver and kidney of rats in the C, P, and PϩCn groups1 expression. Rats in Group P had significantly higher renal TMLH activity relative to rats in the other 2 groups, which did CPPϩCn not differ from one another (Fig. 1). Hepatic BBH activity in ϩ n 677 Group P Cn was significantly lower than in the other 2 Liver groups, which did not differ. The expression of renal and BBH/␤-actin mRNA 0.81 Ϯ 0.35 0.77 Ϯ 0.21 1.09 Ϯ 0.34 hepatic TMLH and BBH mRNA did not differ among the TMLH/␤-actin mRNA 0.16 Ϯ 0.05 0.20 Ϯ 0.05 0.14 Ϯ 0.12 groups (Table 2; Fig. 2). Hepatic TMLH was not measured, Kidney and no renal BBH activity was detected. BBH/␤-actin mRNA 0.12 Ϯ 0.04 0.10 Ϯ 0.02 0.09 Ϯ 0.03 TMLH/␤-actin mRNA 0.21 Ϯ 0.05 0.21 Ϯ 0.03 0.24 Ϯ 0.04

DISCUSSION 1 Values are means Ϯ SEM. Downloaded from https://academic.oup.com/jn/article/135/4/761/4663769 by guest on 30 September 2021 There has been renewed interest in carnitine biosynthesis over the past several years, primarily spurred by the characterization of several of the carnitine biosynthetic enzymes and supplemented rats had significantly decreased hepatic BBH their cDNA sequences (19,23,24,26–28). Nevertheless, as re- activity. ported by Vaz and Wanders in their review of carnitine bio- In the current study, renal TMLH activity was significantly ϩ synthesis (1), most of the additional information tends to increased in group P rats, relative to rats in groups C or P Cn support the hypotheses of Davis and Hoppel (2) and Rebouche (Fig. 1). However, hepatic BBH activity did not differ between et al. (3) from the mid-1980s, who both hypothesized that rats in Group C and Group P. This was somewhat surprising in carnitine biosynthesis was regulated by trimethyllysine avail- light of the carnitine deficiency caused by pivalate adminis- ϩ ability. However, it was noted that alterations in the enzyme tration. Conversely, rats in group P Cn had renal TMLH activities of TMLH and BBH occur with various alterations in activity that did not differ from that in rats in groups C, yet diet and physiologic state. In addition, Galland et al. (12) they had significantly lower hepatic BBH activity than rats in showed alterations in both enzyme activity and expression for Group C. These differences in activity, however, were not BBH after thyroxine administration to rats. Thus, there is an reflected in the hepatic or renal expression of mRNA for either interest in determining whether these changes in enzyme of the 2 hydroxylases (Table 2). activity are related to changes in enzyme expression, and One unexpected finding was the expression of BBH mRNA whether either of these changes reflect true physiologic alter- in kidney. It was demonstrated previously that rat kidney does ations in carnitine concentration. not exhibit renal BBH activity, a finding that was repeated in In this study, 2 models known to affect carnitine metabo- the present study (data not shown). Galland et al. (24), using lism were used. One model involved the addition of sodium adult male Wistar rats, found BBH mRNA only in liver, testis, ␤ pivalate to the drinking water, which was shown previously to and epididymis, using -actin as a control. Galland et al. (27) cause secondary carnitine deficiency in rats (13). The second noted that antibodies to BBH cross-reacted with proteins in model utilized carnitine supplementation of the diet to rats the rat kidney. These proteins were 40 and 44 kDa, whereas given sodium pivalate in the drinking water. These modifica- the protein that they purified from rat liver was 43 kDa. tions were specifically chosen in a previous study (14) to Vaz et al. (26) independently identified the cDNA encod- determine whether TMLH and BBH activities would be af- ing BBH using Wistar rats. However, noting the previous work fected. In that previous study, it was shown that pivalate- concerning the tissue distribution of rat BBH activity, the treated rats had an almost 100% increase in TMLH activity in researchers demonstrated the presence of mRNA expression of the kidney compared with controls, whereas the carnitine- the enzyme only in rat liver. Concerned about the specificity of the results, the identity of the mRNA in the current study

FIGURE 2 Gel electrophoresis of hepatic and renal TMLH and BBH mRNA expression products in rats in the C, P, and PϩCn groups. FIGURE 1 Renal TMLH and hepatic BBH activities in rats in the C, Kidney: Lanes 1–10: Group C (lanes 1–3), Group P (lanes 4–7), and P, and PϩCn groups. Data are means Ϯ SEM, n ϭ 6 or 7. Means for an Group PϩCn (lanes 8–10). Liver: Lanes 11–20: Group C (lanes 11–13), enzyme without a common different letter differ, P Ͻ 0.05. Group P (lanes 14–16), and Group PϩCn (lanes 17–20). 764 DAVIS AND MONROE was verified by direct sequencing of the BBH cDNA product 12. Galland, S., Georges, B., Le Bourgne, F., Conductier, G., Viana Dias, J. & Demarquoy, J. (2002) Thyroid hormone controls carnitine status through mod- amplified from kidney RNA. Representative gels from rat ifications of ␥-butyrobetaine hydroxylase activity and gene expression. Cell. Mol. kidney and rat liver are shown in Figure 2. It should be noted Life Sci. 59: 540–545. that all 20 kidneys tested showed a positive result for the BBH 13. Bianchi, P. B. & Davis, A. T. (1991) Sodium pivalate treatment reduces mRNA. In addition, this band was not expressed in the ab- tissue carnitines and enhances ketosis in rats. J. Nutr. 121: 2029–2036. 14. Davis, A. T. (1999) Alterations in carnitine biosynthetic enzyme activ- sence of the primers for BBH. ities in carnitine deficient and carnitine supplemented rats. Experimental Biology In conclusion, the differences in enzyme activities seen in ’99, Washington, DC (Program addendum, abstract LB208). this study were not associated with altered expression of either 15. American Institute of Nutrition (1987) Second report of the ad hoc committee on standards for nutritional studies. J. Nutr. 110: 1726. TMLH or BBH. Thus, the alterations in enzyme activities are 16. Bianchi, P. B., Lehotay, D. C. & Davis, A. T. (1996) Carnitine supple- not regulated at the transcriptional level and are apparently mentation ameliorates the steatosis and ketosis induced by pivalate. J. Nutr. 126: related to direct effects on the 2 enzymes. Finally, although 2873–2879. renal BBH activity is not detectable in rats, we report for the 17. Davis, A. T. (1990) Tissue trimethyllysine biosynthesis and carnitine content in pregnant and lactating rats fed a lysine-limiting diet. J. Nutr. 120: first time the expression of BBH mRNA in rat kidney. 846–856. 18. Davis, A. T. (1987) Assay of trimethyllysine hydroxylase by high per-

formance liquid chromatography. J. Chromatogr. 422: 253–256. Downloaded from https://academic.oup.com/jn/article/135/4/761/4663769 by guest on 30 September 2021 ACKNOWLEDGMENTS 19. Vaz, F. M., van Gool, S., Ofman, R., Ijlst, L. & Wanders, R.J.A. (1998) ␥ We are grateful to Laura VanWyk and Mona Wojtas for their Carnitine biosynthesis: identification of the cDNA encoding human -butyrobetaine hydroxylase. Biochem. Biophys. Res. Commun. 250: 506–510. excellent technical assistance. We also thank Sigma Tau Pharmaceu- 20. Janssens, G.P.J., De Rycke, H., Hesta, M. & De Wilde, R.O.M. (1999) ticals, Incorporated, for providing the L-carnitine. Analysis of carnitine, betaine, ␥-butyrobetaine, and separate short-chain acylcar- nitines in pigeon plasma, crop milk and tissues by HPLC coupled with UV- detection. Biotechnol. Tech. 12: 231–234. LITERATURE CITED 21. van Kempen, T.A.T.G. & Odle, J. (1992) Quantification of carnitine 1. Vaz, F. M. & Wanders, R.J.A. (2002) Carnitine biosynthesis in mam- esters by high-performance liquid chromatography. Effect of feeding medium- mals. Biochem. J. 361: 417–429. chain triglycerides on the plasma carnitine ester profile. J. Chromatogr. 584: 2. Davis, A. T. & Hoppel, C. L. (1986) Effect of starvation on the dispo- 157–165. sition of free and peptide-linked trimethyllysine in the rat. J. Nutr. 116: 760–767. 22. Minkler, P. E., Brass, E. P., Hiatt, W. R., Ingalls, S. T. & Hoppel, C. L. 3. Rebouche, C. J., Lehman, L. J. & Olson, L. (1986) ⑀-N-Trimethyllysine (1995) Quantification of carnitine, acetylcarnitine, and total carnitine in tissues availability regulates the rate of carnitine biosynthesis in the growing rat. J. Nutr. by high-performance liquid chromatography: the effect of exercise on carnitine 116: 751–759. homeostasis in man. Anal. Biochem. 231: 315–322. 4. Cederblad, G. (1976) Plasma carnitine and body composition. Clin. 23. Vaz, F. M., Ofman, R., Westinga, K., Back, J. W. & Wanders, R.J.A. Chim. Acta 67: 207–212. (2001) Molecular and biochemical characterization of rat ⑀-N-trimethyllysine 5. Rebouche, C. J. (1982) Sites and regulation of carnitine biosynthesis hydroxylase, the first enzyme of carnitine biosynthesis. J. Biol. Chem. 276: in mammals. Fed. Proc. 41: 2848–2852. 33512–33517. 6. Rebouche, C. J. (1983) Effect of dietary carnitine isomers and ␥-bu- 24. Galland, S., Le Bourgne, F., Bouchard, F., Georges, B., Clouet, P., Grand- tyrobetaine on L-carnitine biosynthesis and metabolism in the rat. J. Nutr. 113: Jean, F. & Demarquoy, J. (1999) Molecular cloning and characterization of the 1906–1913. cDNA encoding the rat liver gamma-butyrobetaine hydroxylase. Biochim. Bio- 7. McGarry, J. D., Robles-Valdes, C. & Foster, D. W. (1975) Role of phys. Acta 1441: 85–92. carnitine in hepatic ketogenesis. Proc. Natl. Acad. Sci. U.S.A. 72: 4385–4388. 25. Raff, T., van der Giet, M., Endemann, D., Wiederholt, T., & Paul, M. 8. Brass, E. P. & Hoppel, C. L. (1978) Carnitine metabolism in the fasting (1997) Design and testing of beta-actin primers for RT-PCR that do not co- rat. J. Biol. Chem. 253: 2688–2693. amplify processed pseudogenes. Biotechniques 23: 456–460. 9. Paul, H. S., Gleditsch, C. E. & Adibi S. A. (1986) Mechanism of 26. Vaz, F. M., van Gool, S., Ofman, R., Ijlst, L. & Wanders, R.J.A. (1998) increased hepatic concentration of carnitine by clofibrate. Am. J. Physiol. 251: Carnitine biosynthesis. Purification of ␥-butyrobetaine hydroxylase from rat liver. E311–E315. Adv. Exp. Med. Biol. 466: 117–124. 10. Robles-Valdes, C., McGarry, J. D. & Foster, D. W. (1976) Maternal- 27. Galland, S., Le Bourgne, F., Gutonnet, D., Clouet, P. & Demarquoy, J. fetal carnitine relationship and neonatal ketosis in the rat. J. Biol. Chem. 251: (1998) Purification and characterization of the rat liver gamma-butyrobetaine 6007–6012. hydroxylase. Mol. Cell. Biochem. 178: 163–168. 11. Pande, S. V. & Parvin, R. (1980) Clofibrate enhancement of mitochon- 28. Vaz, F. M., Fouchier, S. W., Ofman, R., Sommer, M. & Wanders, R.J.A. drial carnitine transport system of rat liver and augmentation of liver carnitine and (2000) Molecular and biochemical characterization of rat ␥-trimethylaminobu- ␥-butyrobetaine hydroxylase activity by thyroxine. Biochim. Biophys. Acta 617: tyraldehyde dehydrogenase and evidence for the involvement of human aldehyde 363–370. dehyrogenase 9 in carnitine biosynthesis. J. Biol. Chem. 275: 7390–7394.