- Email: [email protected]
Carnitine Uptake Defect: Frameshift Mutations in the Human Plasmalemmal Carnitine Transporter Gene
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
252, 396 – 401 (1998)
RC989679
Carnitine Uptake Defect: Frameshift Mutations in the Human Plasmalemmal Carnitine Transporter Gene Anne-Marie Lamhonwah and Ingrid Tein1 Division of Neurology, Department of Pediatrics and Laboratory Medicine and Department of Pathobiology, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, M5G 1X8, Canada
Received October 14, 1998
The genetic defect associated with carnitine uptake is characterized by progressive infantile-onset carnitine responsive cardiomyopathy, weakness, recurrent hypoglycemic hypoketotic encephalopathy, and failure to thrive. The cDNA encoding the sodium iondependent, high-affinity human carnitine transporter (557 amino acids) has been recently cloned and mapped to human chromosome 5q31. We herein report the first molecular characterization of the mutations responsible for the carnitine uptake defect in two unrelated patients. RT-PCR analysis of patient lymphoblasts and fibroblasts followed by sequencing of PCR products and their subclones revealed frameshift mutations in the plasmalemmal carnitine transporter. In both patients, the abnormal transcripts showed a partial cDNA deletion of nucleotides 255–1649 resulting in a predicted truncated protein of 92 amino acids. Both patients are compound heterozygotes; in one patient the second mutant allele revealed a 19-bp insertion between nucleotides 874 and 875 resulting in a frameshift yielding a predicted truncated protein of 284 amino acids, while in the second patient the second mutant allele had a deletion of nucleotides 875–1046 resulting in a predicted truncated protein of 237 amino acids. © 1998 Academic Press
Carnitine (b-hydroxy-g-trimethylaminobutyric acid) is a small water-soluble quaternary amine that serves as an essential cofactor for the transport of long-chain fatty acids as acylcarnitines esters across the inner mitochondrial membrane and modulates the intramitochondrial acyl-CoA/CoA sulfhydryl ratio in mammalian cells, thereby providing the cell with a critical 1 To whom correspondence should be addressed at Division of Neurology, The Hospital for Sick Children, 555 University Ave., Toronto, Ontario, Canada M5G 1X8. Fax: (416) 813-6334. E-mail: [email protected]. Abbreviations used: CUD, carnitine uptake defect; DTT, dithiothreitol; FCS, fetal calf serum; MEM, minimal essential medium; 13 TBE, 90 mM Tris– borate, 2 mM EDTA, pH 8.0.
0006-291X/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
source of free CoA (1). Carnitine is of major importance in tissues such as muscle, heart, kidney and liver which rely heavily on efficient fatty acid oxidation as a major source of ATP production. In nonvegetarians, approximately 75% of body L-carnitine sources is dietary and 25% comes from de novo biosynthesis from lysine and methionine (2). Most tissues are capable of the synthesis of g-butyrobetaine, the immediate precursor of carnitine, which must then be exported to the liver for final hydroxylation to carnitine (1). The carnitine concentrations in tissues are normally 20- to 50fold higher than in serum. Human tissue concentrations (nmol/g) are heart (3500 – 6000) . muscle (2000 – 4600) . liver (1000 –1900) . brain (200 –500) (3). Uptake into tissues therefore occurs across a large concentration gradient which is maintained by an active transport system (1). The kidney is primarily responsible for regulating body stores and is capable of adjusting to wide variations in dietary carnitine because is has a threshold of 40 mmol/L, which is identical to the normal serum concentration (4). Kinetic studies have demonstrated similar Km values of 2– 6 mmol/L for carnitine transport in cultured skin fibroblasts (5– 8), muscle (9) and heart (10) suggesting that they share a common transporter, and much lower affinity Km values for human liver (500 mmol/L) and brain (.1000 mmol/L) (11). We have previously demonstrated the high-affinity carnitine transporter in cultured human skin fibroblasts to be highly dependent upon a large sodium potential across the plasma membrane and on the presence of free sulfhydryl groups (12). Previous work supports the hypothesis that primary systemic carnitine deficiency is due to a defect in the specific high-affinity carnitine transporter which is expressed in fibroblasts, muscle, heart, kidney and lymphoblasts (6 –9, 13, 14). This carnitine uptake defect (CUD) is a genetic disorder, presumed autosomal recessive in inheritance, which is characterized by carnitine-responsive cardiomyopathy with or without weakness, recurrent hypoglycemic hypoketotic encephalopathy, failure to thrive and extremely low plasma
396
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
and tissue concentrations of carnitine (, 5% of controls), with lipid storage in muscle and liver, and a severe renal leak of carnitine (7). Recently, a new member of the organic cation transporter family, OCTN2 cDNA has been cloned from a human placental trophoblast cell line (15). The cDNA (;3.2 kb long) has a 59UTR region of 221 bp, an open reading frame (222– 1892) encoding a protein of 557 amino acids with twelve putative transmembrane domains followed by a long 39UTR. Homology search of the Genbank data base indicated that the human gene octn2 coding for OCTN2 has been sequenced in its entirety as a part of the Human Genome Project and mapped to human chromosome 5q31 (15). The function of the OCTN2 cDNA as an organic cation transporter has been assessed by the transport of tetraethylammonium in an heterologous expression system. Organic cation transporters function primarily in the elimination of cationic drugs and other xenobiotics in tissues such as the kidney, intestine and liver (16, 17) and presumably, these transporters would play similar role in the placenta by eliminating xenobiotics from the fetus (18). However, Tamai et al. (19) found that OCTN2 from a human kidney cDNA library shared a high homology (75.8%) with OCTN1 and identified OCTN2 as the physiologically important, high affinity sodium carnitine cotransporter in humans. This was based upon the observed tissue distribution of OCTN2 which was consistent with the reported distribution of carnitine transport activity and the functional characteristics of OCTN2 which coincided with those reported for plasma membrane carnitine transport. They found that in expression studies of OCTN2 in HEK293 cells, uptake of L-[3H]carnitine was strongly enhanced in a sodium-dependent manner with a Km value of 4.34 mmol/L. In addition, OCTN2-mediated L-[3H]carnitine transport was inhibited by the D-isomer, acetyl-D,Lcarnitine and g-butyrobetaine with high affinity and by glycinebetaine with lower affinity. This potent inhibition by g-butyrobetaine has been previously demonstrated in rat kidney (20) and fetal human heart myocytes (21). Furthermore, northern blot analysis showed that OCTN2 is strongly expressed in kidney, skeletal muscle, heart and placenta in adult humans. Since the CUD has been directly demonstrated in fibroblasts, lymphoblasts, and muscle (6 –9, 13, 14) and since no molecular characterization of the CUD has been reported, we studied the expression of OCTN2 in cultured fibroblasts and lymphoblasts from two unrelated patients in whom we had previously documented the CUD (cases #2 and #4) (7). MATERIALS AND METHODS Cell culture and RNA purification. Studies were performed with the approval of the Institutional Review Board of the Hospital for Sick Children, Toronto. The cell lines were cultured skin fibroblasts
and lymphoblasts from control individuals and patients with the CUD. Fibroblasts were grown in 100-mm2 dishes containing alphaMEM 1 10% fetal calf serum and lymphoblasts were grown in 25-cm2 Falcon tissue culture flasks in RPMI 1640 with glutamine and 15% FCS in the presence of 5% CO2 at 37°C. Total RNA was harvested using a Qiagen RNeasy total RNA isolation kit or using Trizol Reagent (Gibco BRL). RT-PCR analysis of mRNA. The carnitine transporter cDNA was synthesized with either specific antisense primer A4-1894 (59TTAGAAGGCTGTGCTTTTAAGGATTGT-39 or A5-1911 (59-CTTACTGGAAGCGATGTTAGAAGGCT-39 and sense primer S1-222 (59ATGCGGGACTACGACGAGGTGACC-39 using the SuperScript One-step RT-PCR system (Life Technologies). The following cycling conditions were employed using a DNA Thermal Cycler 2400 (Perkin–Elmer) (i) one cycle of 45°C for 50 min and 94°C for 2 min for cDNA synthesis and pre-denaturation, (ii) 40 cycles of denaturation at 94°C for 90 s, annealing at 60°C for 90 s and extension at 72°C for 2 min, and (iii) one final extension cycle of 72°C for 10 min. Total cellular RNA (5 mg) was also reverse-transcribed in a reaction volume of 20 ml using 500 ng oligo (T)15 and 200 units of SuperScript II reverse transcriptase (Gibco BRL), 0.5 mM each of the four dNTPs in 50 mM Tris (pH 8.3), 75 mM KCl, 3 mM MgCl2 and 10 mM DTT for 1 h at 42°C. The first strand cDNA (0.5 mg RNA equivalent) was then amplified in a 50-ml reaction volume containing 200 mM each of the four dNTPs, 2.5 units of Taq polymerase (BRL), 100 ng of each specific primer S1-222 and A5-1911 in 10 mM Tris (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, and 0.01% gelatin (22) to obtain the full coding cDNA (1690 bp). The sense primer S4-678 (59-GTGCTGTTGGGCTCCTTCATTTCA-39) and antisense primer A4-1894 were also used to amplify the DNA segment spanning nucleotides 678 –1911 with an annealing temperature 64°C. To analyze the cDNA amplicons, 3ml of loading dye were added to 15ml PCR products and the samples loaded on a 1% agarose (13 TBE). Subcloning and nucleotide sequencing of RT-PCR products. The cDNA amplicons were excised from the agarose gel and the DNA purified using the Geneclean II kit (Bio 101) prior to subcloning into the pCR2.1 vector using the TA cloning kit (Invitrogen). Plasmid DNA was isolated from transformed INVaF9 E. coli using the Qiaprep Spin Miniprep kit (Qiagen). Nucleotide sequencing of PCR inserts was done by the dideoxy chain termination method using the T7 sequencing kit (Pharmacia) and M13 universal and reverse primers. The sequencing reactions were analyzed on 6 or 8% sequagel (National diagnostics) and the dried gel exposed to BioMax X-ray film (Kodak). Analysis of nucleotide sequences was done using NCBI blast server, DNA Strider, and DNASis software.
RESULTS Identification of the mutations in the carnitine transporter. Figure 1 shows partial nucleotide sequence data around the site of mutation in the OCTN2 cDNA that encodes for the plasmalemmal carnitine transporter. Representative subclones are shown of the RT-PCR products from the cultured skin fibroblasts and lymphoblasts of a normal control and from patients #1 and #2 in whom the CUD has been previously demonstrated. Figure 1A shows the nucleotide sequence in a normal individual (sense strand) and the position G874 –G875 where the mutation occurs in both index cases. The 19-bp insertion “tatggccatcaggttggag” between nucleotides G874 and G 875 has been found in patient #1 and could result from a missplicing event since this sequence contains part of the donor intronic sequence. This 19-bp
397
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Carnitine transporter cDNA sequences in normal controls and in two patients with the carnitine uptake defect. Total fibroblast RNA (1 mg) was reverse-transcribed and amplified by the polymerase chain reaction using primers S4 (59-GTGCTGTTGGGCTCCTTCATTTCA-39, sense) and A4 (59-TTAGAAGGCTGTGCTTTTAAGGATTGT-39, antisense). RT-PCR products were subcloned into vector pCR2.1 (Invitrogen) and sequenced using T7 sequencing kit (Pharmacia). (A) Autoradiograph of nucleotide sequence around the site of mutation in control fibroblasts (normal) and in patient 1 (19-bp insertion between nucleotide 874 – 875) and patient 2 (deletion 875–1046). (B) Nucleotide sequences showing the 19-bp insertion in one mutant allele of patient 1 and the deletion in one mutant allele of patient 2 versus “normal” at the same site of the carnitine transporter cDNA, suggesting a “hot-spot” site for mutation.
insertion creates a frameshift resulting in a premature stop codon, thus yielding a truncated protein of 284 amino acids (see Fig. 2) while the functional carnitine transporter protein is 557 amino acids long. In patient #2, the nucleotide sequence data demonstrate the mutation in the carnitine transporter cDNA as a deletion of nucleotides 875–1046 (Fig. 1) which also results in a frameshift and yields a truncated protein of 237 amino acids (see Fig. 2). Interestingly, the deduced amino acid sequences of the mutant allele with the “19-bp insertion” and the mutant allele with the “deletion 875–1046” would encode a truncated protein that has only the first four putative transmembrane domains (Fig. 2). Figure 3 shows the nucleotide sequence data of the second mutant allele common to both patients #1 and #2. The entire coding cDNA for the carnitine trans-
porter was amplified as a 1690-bp fragment (data not shown), with A222 being the first nucleotide of the methionine initiator codon, in control RT-PCR studies. In both patients #1 and #2, a PCR product of about 250 bp was observed (data not shown), excised and subcloned in the pCR2.1 vector for further analysis. Nucleotide sequence analysis of the short cDNA product revealed a large deletion that spans nucleotides 255– 1649; its deduced amino acid sequence translates into a truncated 92-amino-acid protein suggesting that the second mutant allele of the carnitine transporter common to both patients would encode a null protein. Figure 2 clearly shows that the deduced amino acid sequence of the mutant allele harboring the large deletion 255–1649 does not contain any of the twelve putative transmembrane domains of the carnitine transporter.
FIG. 2. (A) Comparison of the deduced carnitine transporter amino acid sequences from normal and mutant proteins resulting from the specific frameshift mutations. The different transmembrane domains are boxed and the (:) shows homology of amino acids in the corresponding protein. The presumed ATP/GTP binding motif “GTEILGKS” is double-underlined and is missing in the mutant proteins due to the frameshift. (B) Diagram of the normal carnitine transporter protein with the twelve putative transmembrane domains and the predicted mutant proteins in patient 1 and patient 2 in whom only four transmembrane domains would be present. The sizes of the corresponding proteins are shown. The arrow points to the location of the putative ATP/GTP binding motif.
398
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Carnitine transporter cDNA sequences in the second allele in normal controls and in the two patients with the carnitine uptake defect. Total RNA (1 mg) from normal and carnitine deficient fibroblasts was reverse-transcribed and amplified with oligos spanning the entire coding cDNA to the carnitine transporter (1690 bp). A222 corresponds to the first nucleotide of the methionine initiator codon. In both patients 1 and 2, the PCR products are approximately 250 bp in size. (A) Autoradiograph of the nucleotide sequence revealing a large deletion (255–1649) in the cDNA responsible for a frameshift mutation with a premature stop codon resulting in a predicted 92 aa protein. (B) Nucleotide sequence of the normal allele and the mutant allele common to both patients 1 and 2 within the NH2-terminal of the carnitine transporter.
DISCUSSION Tamai et al. (19) have documented the functional biochemical and tissue distribution characteristics of OCTN2 to support its candidacy as the physiologically important, high affinity sodium-carnitine cotransporter in humans, although the plasmalemmal carnitine transporter had not been previously molecularly identified. Our molecular analysis of the octn2 gene in two patients with the CUD has provided further strong evidence to support the identity of OCTN2 as the sodium iondependent, high affinity carnitine transporter cDNA. We report three frameshift mutations in the OCTN2 cDNA resulting in a premature stop and truncated protein that would make the carnitine transporter non-functional in these two patients. Furthermore, our findings seem to demonstrate a “hot-spot” for mutations in the octn2 gene. Both deletion and insertion abrogate a presumed ATP/ GTP binding motif (GTEILGKS) in the truncated mutant proteins in these two unrelated patients of different ethnic origins and suggests heterogeneity of mutations. Patient #1 is a male of Italian descent while patient #2 is a female of Mexican descent with a family history of an affected brother who had died previously of cardiomyopathy (7). Both children had early onset myopathy, cardiomyopathy and failure to thrive with #5% of control car-
nitine concentrations in muscle and had a dramatic improvement in growth, power and cardiac function following institution of high dose oral carnitine supplementation. Patient #1 was shown to have a striking decrease in renal reabsorption of carnitine (52%; normal .95%) despite low serum carnitine concentrations. Also, his muscle carnitine concentration increased to only 13 % of control after carnitine supplementation, suggesting impaired carnitine transport into the muscle. However, this was sufficient to result in a resolution of the lipid storage and a restoration of motor power. On carnitine uptake studies, both children had minimal or no uptake of carnitine throughout the entire range of physiologic concentrations of carnitine precluding the calculation of Km or Vmax values; at a carnitine concentration of 5 mmol/L, their mean rate of uptake was 2% of controls. The parents of these two children, had normal Km values (5.0 –5.7 mmol/L) but reduced Vmax values for carnitine uptake (13–44% of controls), suggesting that obligate carriers have a reduced number of normally functioning transporters (7). Of interest the father of patient #1 and the mother of patient #2 both had very low Vmax values of 13 and 17% of controls respectively, whereas the mother of patient #1 had a higher Vmax of 44%. This may suggest that the carriers with the lowest Vmax values are het-
400
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
erozygous for the large deletion spanning nucleotides 255–1649 which results in a null protein, but this remains to be confirmed in future studies. The early diagnosis and treatment of patients with the CUD with high dose oral carnitine supplementation is critical, given the otherwise progressive and lethal nature of this disorder and the striking family histories of affected siblings including sudden infant death syndrome (7, 8). Carnitine supplementation is life-saving and reverses the myopathy and cardiomyopathy, presumably by utilizing a low-affinity, high-concentration, nonspecific-diffusion uptake of carnitine into the cells, thereby bypassing the specific carrier-mediated transporter (7, 8). Although, at present, carnitine uptake studies in cultured skin fibroblasts or cultured lymphoblasts are considered to be both highly sensitive and specific tests for the diagnosis of affected patients and the screening of siblings and obligate carriers, they are expensive, time-consuming and somewhat invasive. Though there may be a number of different mutations underlying the CUD, the identification of common mutations (e.g., hot spot for mutations) may facilitate the development of a panel of specific molecular probes which may significantly reduce the time and expense to diagnosis and treatment. Furthermore, molecular studies of the octn2 gene in patients with the CUD will allow us to pinpoint the molecular defects and may provide insight into the critical structure-function relationships of the normal human carnitine transporter. ACKNOWLEDGMENTS This work was supported by an operating grant from the Heart and Stroke Foundation of Ontario (T-3746). I.T. is a recipient of a Medical Research Council of Canada Scholarship.
REFERENCES 1. Bremer, J. (1983) Physiol. Rev. 63, 1420 –1480. 2. De Vivo, D. C., and Tein, I. (1990) Internat. Pediatr. 5, 134 –140.
3. Stanley, C. A. (1987) Adv. Pediatr. 34, 59 – 88. 4. Engel, A. G., Rebouche, C. J., Wilson, D. M., Glasgow, A. M., Romshe, C. A., and Cruse, R. P (1981) Neurology 31, 819 – 825. 5. Rebouche, C. J., and Engel, A. G. (1982) In Vitro 18, 495–500. 6. Treem, W. R., Stanley, C. A., Finegold, D. N., Hale, D. E., and Coates, P. M. (1988) New Engl. J. Med. 319, 1331–1336. 7. Tein, I., De Vivo, D. C., Bierman, F., Pulver, P., De Meirleir, L. J., Cvitanovic-Sojat, L., Paigon, R. A., Bertini, E., Dionisi-Vici, C., Servidei, S., and Dimauro, S. (1990) Pediat. Res. 28, 247–255. 8. Stanley, C. A., DeLeeuw, S., Coates, P. M., Vianey-Liaud, D., Divry, P., Bonnefont, J.-P., Saudubray, J.-M., Haymond, M., Trefz, F. K., Breningstall, G. N., Wappner, R. S., Byrd, D. J., Sansaricq, C., Tein, I., Grover, W., Valle, D., Rutledge, S. L., and Treem, W. R. (1991) Ann. Neurol. 30, 709 –716. 9. Pons, R., Carozza, R., Tein, I., Walker, W. F., Addonizio, L. J., Rhead, W., Miranda, A. F., DiMauro, S., and De Vivo, D. C. (1997) Pediatr. Res. 42, 583–587. 10. Bahl, J. J., and Bressler, R. (1987) Annu. Rev. Pharmacol. Toxicol. 27, 257–277. 11. Bieber, L. L. (1988) Annu. Rev. Biochem. 57, 261–283. 12. Tein, I., Bukovac, S. W., and Xie, Z.-W. (1996) Arch. Biochem. Biophys. 329, 145–155. 13. Eriksson, B. O., Gustafson, S., Lindstedt, S., and Nordin, I. (1989) J. Inher. Metab. Dis. 12, 108 –111. 14. Tein, I., and Xie, Z.-W. (1996) Clin. Chim. Acta 252, 201–204. 15. Wu, X., Prasad, P. D., Leibach, F. H., and Ganapathy, V. (1998) Biochem. Biophys. Res. Commun. 246, 589 –595. 16. Pritchard, J. B., and Miller, D. S. (1993) Physiol.Rev. 73, 765– 796. 17. Ullrich, K. J. (1994) Biochim. Biophys. Acta 1197, 45– 62. 18. Zevin, S., Schaner, M. E., Illsley, N. P., and Giacomini, K. M. (1997) Pharmacol. Res. 14, 401– 405. 19. Tamai, I., Ohashi, R., Nezu, J., Yabuuchi, H., Oku, A., Shimane, M., Sai, Y., and Tsuji, A. (1998) J. Biol. Chem. 273, 20378 – 20382. 20. Huth, P. J., and Shug, A. (1980) Biochim. Biophys. Acta 602, 621– 634. 21. Molstad, P., Bohmer, T., and Eiklid, K. (1977) Biochim. Biophys. Acta 471, 296 –304. 22. Saiki, R. K., Scharf, S., Faloona, F., Mullism, K. B., Horn, G. T., Erlich, H. A., and Arheim, N. (1985) Science 230, 1350 – 1354.
401
252, 396 – 401 (1998)
RC989679
Carnitine Uptake Defect: Frameshift Mutations in the Human Plasmalemmal Carnitine Transporter Gene Anne-Marie Lamhonwah and Ingrid Tein1 Division of Neurology, Department of Pediatrics and Laboratory Medicine and Department of Pathobiology, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, M5G 1X8, Canada
Received October 14, 1998
The genetic defect associated with carnitine uptake is characterized by progressive infantile-onset carnitine responsive cardiomyopathy, weakness, recurrent hypoglycemic hypoketotic encephalopathy, and failure to thrive. The cDNA encoding the sodium iondependent, high-affinity human carnitine transporter (557 amino acids) has been recently cloned and mapped to human chromosome 5q31. We herein report the first molecular characterization of the mutations responsible for the carnitine uptake defect in two unrelated patients. RT-PCR analysis of patient lymphoblasts and fibroblasts followed by sequencing of PCR products and their subclones revealed frameshift mutations in the plasmalemmal carnitine transporter. In both patients, the abnormal transcripts showed a partial cDNA deletion of nucleotides 255–1649 resulting in a predicted truncated protein of 92 amino acids. Both patients are compound heterozygotes; in one patient the second mutant allele revealed a 19-bp insertion between nucleotides 874 and 875 resulting in a frameshift yielding a predicted truncated protein of 284 amino acids, while in the second patient the second mutant allele had a deletion of nucleotides 875–1046 resulting in a predicted truncated protein of 237 amino acids. © 1998 Academic Press
Carnitine (b-hydroxy-g-trimethylaminobutyric acid) is a small water-soluble quaternary amine that serves as an essential cofactor for the transport of long-chain fatty acids as acylcarnitines esters across the inner mitochondrial membrane and modulates the intramitochondrial acyl-CoA/CoA sulfhydryl ratio in mammalian cells, thereby providing the cell with a critical 1 To whom correspondence should be addressed at Division of Neurology, The Hospital for Sick Children, 555 University Ave., Toronto, Ontario, Canada M5G 1X8. Fax: (416) 813-6334. E-mail: [email protected]. Abbreviations used: CUD, carnitine uptake defect; DTT, dithiothreitol; FCS, fetal calf serum; MEM, minimal essential medium; 13 TBE, 90 mM Tris– borate, 2 mM EDTA, pH 8.0.
0006-291X/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
source of free CoA (1). Carnitine is of major importance in tissues such as muscle, heart, kidney and liver which rely heavily on efficient fatty acid oxidation as a major source of ATP production. In nonvegetarians, approximately 75% of body L-carnitine sources is dietary and 25% comes from de novo biosynthesis from lysine and methionine (2). Most tissues are capable of the synthesis of g-butyrobetaine, the immediate precursor of carnitine, which must then be exported to the liver for final hydroxylation to carnitine (1). The carnitine concentrations in tissues are normally 20- to 50fold higher than in serum. Human tissue concentrations (nmol/g) are heart (3500 – 6000) . muscle (2000 – 4600) . liver (1000 –1900) . brain (200 –500) (3). Uptake into tissues therefore occurs across a large concentration gradient which is maintained by an active transport system (1). The kidney is primarily responsible for regulating body stores and is capable of adjusting to wide variations in dietary carnitine because is has a threshold of 40 mmol/L, which is identical to the normal serum concentration (4). Kinetic studies have demonstrated similar Km values of 2– 6 mmol/L for carnitine transport in cultured skin fibroblasts (5– 8), muscle (9) and heart (10) suggesting that they share a common transporter, and much lower affinity Km values for human liver (500 mmol/L) and brain (.1000 mmol/L) (11). We have previously demonstrated the high-affinity carnitine transporter in cultured human skin fibroblasts to be highly dependent upon a large sodium potential across the plasma membrane and on the presence of free sulfhydryl groups (12). Previous work supports the hypothesis that primary systemic carnitine deficiency is due to a defect in the specific high-affinity carnitine transporter which is expressed in fibroblasts, muscle, heart, kidney and lymphoblasts (6 –9, 13, 14). This carnitine uptake defect (CUD) is a genetic disorder, presumed autosomal recessive in inheritance, which is characterized by carnitine-responsive cardiomyopathy with or without weakness, recurrent hypoglycemic hypoketotic encephalopathy, failure to thrive and extremely low plasma
396
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
and tissue concentrations of carnitine (, 5% of controls), with lipid storage in muscle and liver, and a severe renal leak of carnitine (7). Recently, a new member of the organic cation transporter family, OCTN2 cDNA has been cloned from a human placental trophoblast cell line (15). The cDNA (;3.2 kb long) has a 59UTR region of 221 bp, an open reading frame (222– 1892) encoding a protein of 557 amino acids with twelve putative transmembrane domains followed by a long 39UTR. Homology search of the Genbank data base indicated that the human gene octn2 coding for OCTN2 has been sequenced in its entirety as a part of the Human Genome Project and mapped to human chromosome 5q31 (15). The function of the OCTN2 cDNA as an organic cation transporter has been assessed by the transport of tetraethylammonium in an heterologous expression system. Organic cation transporters function primarily in the elimination of cationic drugs and other xenobiotics in tissues such as the kidney, intestine and liver (16, 17) and presumably, these transporters would play similar role in the placenta by eliminating xenobiotics from the fetus (18). However, Tamai et al. (19) found that OCTN2 from a human kidney cDNA library shared a high homology (75.8%) with OCTN1 and identified OCTN2 as the physiologically important, high affinity sodium carnitine cotransporter in humans. This was based upon the observed tissue distribution of OCTN2 which was consistent with the reported distribution of carnitine transport activity and the functional characteristics of OCTN2 which coincided with those reported for plasma membrane carnitine transport. They found that in expression studies of OCTN2 in HEK293 cells, uptake of L-[3H]carnitine was strongly enhanced in a sodium-dependent manner with a Km value of 4.34 mmol/L. In addition, OCTN2-mediated L-[3H]carnitine transport was inhibited by the D-isomer, acetyl-D,Lcarnitine and g-butyrobetaine with high affinity and by glycinebetaine with lower affinity. This potent inhibition by g-butyrobetaine has been previously demonstrated in rat kidney (20) and fetal human heart myocytes (21). Furthermore, northern blot analysis showed that OCTN2 is strongly expressed in kidney, skeletal muscle, heart and placenta in adult humans. Since the CUD has been directly demonstrated in fibroblasts, lymphoblasts, and muscle (6 –9, 13, 14) and since no molecular characterization of the CUD has been reported, we studied the expression of OCTN2 in cultured fibroblasts and lymphoblasts from two unrelated patients in whom we had previously documented the CUD (cases #2 and #4) (7). MATERIALS AND METHODS Cell culture and RNA purification. Studies were performed with the approval of the Institutional Review Board of the Hospital for Sick Children, Toronto. The cell lines were cultured skin fibroblasts
and lymphoblasts from control individuals and patients with the CUD. Fibroblasts were grown in 100-mm2 dishes containing alphaMEM 1 10% fetal calf serum and lymphoblasts were grown in 25-cm2 Falcon tissue culture flasks in RPMI 1640 with glutamine and 15% FCS in the presence of 5% CO2 at 37°C. Total RNA was harvested using a Qiagen RNeasy total RNA isolation kit or using Trizol Reagent (Gibco BRL). RT-PCR analysis of mRNA. The carnitine transporter cDNA was synthesized with either specific antisense primer A4-1894 (59TTAGAAGGCTGTGCTTTTAAGGATTGT-39 or A5-1911 (59-CTTACTGGAAGCGATGTTAGAAGGCT-39 and sense primer S1-222 (59ATGCGGGACTACGACGAGGTGACC-39 using the SuperScript One-step RT-PCR system (Life Technologies). The following cycling conditions were employed using a DNA Thermal Cycler 2400 (Perkin–Elmer) (i) one cycle of 45°C for 50 min and 94°C for 2 min for cDNA synthesis and pre-denaturation, (ii) 40 cycles of denaturation at 94°C for 90 s, annealing at 60°C for 90 s and extension at 72°C for 2 min, and (iii) one final extension cycle of 72°C for 10 min. Total cellular RNA (5 mg) was also reverse-transcribed in a reaction volume of 20 ml using 500 ng oligo (T)15 and 200 units of SuperScript II reverse transcriptase (Gibco BRL), 0.5 mM each of the four dNTPs in 50 mM Tris (pH 8.3), 75 mM KCl, 3 mM MgCl2 and 10 mM DTT for 1 h at 42°C. The first strand cDNA (0.5 mg RNA equivalent) was then amplified in a 50-ml reaction volume containing 200 mM each of the four dNTPs, 2.5 units of Taq polymerase (BRL), 100 ng of each specific primer S1-222 and A5-1911 in 10 mM Tris (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, and 0.01% gelatin (22) to obtain the full coding cDNA (1690 bp). The sense primer S4-678 (59-GTGCTGTTGGGCTCCTTCATTTCA-39) and antisense primer A4-1894 were also used to amplify the DNA segment spanning nucleotides 678 –1911 with an annealing temperature 64°C. To analyze the cDNA amplicons, 3ml of loading dye were added to 15ml PCR products and the samples loaded on a 1% agarose (13 TBE). Subcloning and nucleotide sequencing of RT-PCR products. The cDNA amplicons were excised from the agarose gel and the DNA purified using the Geneclean II kit (Bio 101) prior to subcloning into the pCR2.1 vector using the TA cloning kit (Invitrogen). Plasmid DNA was isolated from transformed INVaF9 E. coli using the Qiaprep Spin Miniprep kit (Qiagen). Nucleotide sequencing of PCR inserts was done by the dideoxy chain termination method using the T7 sequencing kit (Pharmacia) and M13 universal and reverse primers. The sequencing reactions were analyzed on 6 or 8% sequagel (National diagnostics) and the dried gel exposed to BioMax X-ray film (Kodak). Analysis of nucleotide sequences was done using NCBI blast server, DNA Strider, and DNASis software.
RESULTS Identification of the mutations in the carnitine transporter. Figure 1 shows partial nucleotide sequence data around the site of mutation in the OCTN2 cDNA that encodes for the plasmalemmal carnitine transporter. Representative subclones are shown of the RT-PCR products from the cultured skin fibroblasts and lymphoblasts of a normal control and from patients #1 and #2 in whom the CUD has been previously demonstrated. Figure 1A shows the nucleotide sequence in a normal individual (sense strand) and the position G874 –G875 where the mutation occurs in both index cases. The 19-bp insertion “tatggccatcaggttggag” between nucleotides G874 and G 875 has been found in patient #1 and could result from a missplicing event since this sequence contains part of the donor intronic sequence. This 19-bp
397
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Carnitine transporter cDNA sequences in normal controls and in two patients with the carnitine uptake defect. Total fibroblast RNA (1 mg) was reverse-transcribed and amplified by the polymerase chain reaction using primers S4 (59-GTGCTGTTGGGCTCCTTCATTTCA-39, sense) and A4 (59-TTAGAAGGCTGTGCTTTTAAGGATTGT-39, antisense). RT-PCR products were subcloned into vector pCR2.1 (Invitrogen) and sequenced using T7 sequencing kit (Pharmacia). (A) Autoradiograph of nucleotide sequence around the site of mutation in control fibroblasts (normal) and in patient 1 (19-bp insertion between nucleotide 874 – 875) and patient 2 (deletion 875–1046). (B) Nucleotide sequences showing the 19-bp insertion in one mutant allele of patient 1 and the deletion in one mutant allele of patient 2 versus “normal” at the same site of the carnitine transporter cDNA, suggesting a “hot-spot” site for mutation.
insertion creates a frameshift resulting in a premature stop codon, thus yielding a truncated protein of 284 amino acids (see Fig. 2) while the functional carnitine transporter protein is 557 amino acids long. In patient #2, the nucleotide sequence data demonstrate the mutation in the carnitine transporter cDNA as a deletion of nucleotides 875–1046 (Fig. 1) which also results in a frameshift and yields a truncated protein of 237 amino acids (see Fig. 2). Interestingly, the deduced amino acid sequences of the mutant allele with the “19-bp insertion” and the mutant allele with the “deletion 875–1046” would encode a truncated protein that has only the first four putative transmembrane domains (Fig. 2). Figure 3 shows the nucleotide sequence data of the second mutant allele common to both patients #1 and #2. The entire coding cDNA for the carnitine trans-
porter was amplified as a 1690-bp fragment (data not shown), with A222 being the first nucleotide of the methionine initiator codon, in control RT-PCR studies. In both patients #1 and #2, a PCR product of about 250 bp was observed (data not shown), excised and subcloned in the pCR2.1 vector for further analysis. Nucleotide sequence analysis of the short cDNA product revealed a large deletion that spans nucleotides 255– 1649; its deduced amino acid sequence translates into a truncated 92-amino-acid protein suggesting that the second mutant allele of the carnitine transporter common to both patients would encode a null protein. Figure 2 clearly shows that the deduced amino acid sequence of the mutant allele harboring the large deletion 255–1649 does not contain any of the twelve putative transmembrane domains of the carnitine transporter.
FIG. 2. (A) Comparison of the deduced carnitine transporter amino acid sequences from normal and mutant proteins resulting from the specific frameshift mutations. The different transmembrane domains are boxed and the (:) shows homology of amino acids in the corresponding protein. The presumed ATP/GTP binding motif “GTEILGKS” is double-underlined and is missing in the mutant proteins due to the frameshift. (B) Diagram of the normal carnitine transporter protein with the twelve putative transmembrane domains and the predicted mutant proteins in patient 1 and patient 2 in whom only four transmembrane domains would be present. The sizes of the corresponding proteins are shown. The arrow points to the location of the putative ATP/GTP binding motif.
398
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Carnitine transporter cDNA sequences in the second allele in normal controls and in the two patients with the carnitine uptake defect. Total RNA (1 mg) from normal and carnitine deficient fibroblasts was reverse-transcribed and amplified with oligos spanning the entire coding cDNA to the carnitine transporter (1690 bp). A222 corresponds to the first nucleotide of the methionine initiator codon. In both patients 1 and 2, the PCR products are approximately 250 bp in size. (A) Autoradiograph of the nucleotide sequence revealing a large deletion (255–1649) in the cDNA responsible for a frameshift mutation with a premature stop codon resulting in a predicted 92 aa protein. (B) Nucleotide sequence of the normal allele and the mutant allele common to both patients 1 and 2 within the NH2-terminal of the carnitine transporter.
DISCUSSION Tamai et al. (19) have documented the functional biochemical and tissue distribution characteristics of OCTN2 to support its candidacy as the physiologically important, high affinity sodium-carnitine cotransporter in humans, although the plasmalemmal carnitine transporter had not been previously molecularly identified. Our molecular analysis of the octn2 gene in two patients with the CUD has provided further strong evidence to support the identity of OCTN2 as the sodium iondependent, high affinity carnitine transporter cDNA. We report three frameshift mutations in the OCTN2 cDNA resulting in a premature stop and truncated protein that would make the carnitine transporter non-functional in these two patients. Furthermore, our findings seem to demonstrate a “hot-spot” for mutations in the octn2 gene. Both deletion and insertion abrogate a presumed ATP/ GTP binding motif (GTEILGKS) in the truncated mutant proteins in these two unrelated patients of different ethnic origins and suggests heterogeneity of mutations. Patient #1 is a male of Italian descent while patient #2 is a female of Mexican descent with a family history of an affected brother who had died previously of cardiomyopathy (7). Both children had early onset myopathy, cardiomyopathy and failure to thrive with #5% of control car-
nitine concentrations in muscle and had a dramatic improvement in growth, power and cardiac function following institution of high dose oral carnitine supplementation. Patient #1 was shown to have a striking decrease in renal reabsorption of carnitine (52%; normal .95%) despite low serum carnitine concentrations. Also, his muscle carnitine concentration increased to only 13 % of control after carnitine supplementation, suggesting impaired carnitine transport into the muscle. However, this was sufficient to result in a resolution of the lipid storage and a restoration of motor power. On carnitine uptake studies, both children had minimal or no uptake of carnitine throughout the entire range of physiologic concentrations of carnitine precluding the calculation of Km or Vmax values; at a carnitine concentration of 5 mmol/L, their mean rate of uptake was 2% of controls. The parents of these two children, had normal Km values (5.0 –5.7 mmol/L) but reduced Vmax values for carnitine uptake (13–44% of controls), suggesting that obligate carriers have a reduced number of normally functioning transporters (7). Of interest the father of patient #1 and the mother of patient #2 both had very low Vmax values of 13 and 17% of controls respectively, whereas the mother of patient #1 had a higher Vmax of 44%. This may suggest that the carriers with the lowest Vmax values are het-
400
Vol. 252, No. 2, 1998
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
erozygous for the large deletion spanning nucleotides 255–1649 which results in a null protein, but this remains to be confirmed in future studies. The early diagnosis and treatment of patients with the CUD with high dose oral carnitine supplementation is critical, given the otherwise progressive and lethal nature of this disorder and the striking family histories of affected siblings including sudden infant death syndrome (7, 8). Carnitine supplementation is life-saving and reverses the myopathy and cardiomyopathy, presumably by utilizing a low-affinity, high-concentration, nonspecific-diffusion uptake of carnitine into the cells, thereby bypassing the specific carrier-mediated transporter (7, 8). Although, at present, carnitine uptake studies in cultured skin fibroblasts or cultured lymphoblasts are considered to be both highly sensitive and specific tests for the diagnosis of affected patients and the screening of siblings and obligate carriers, they are expensive, time-consuming and somewhat invasive. Though there may be a number of different mutations underlying the CUD, the identification of common mutations (e.g., hot spot for mutations) may facilitate the development of a panel of specific molecular probes which may significantly reduce the time and expense to diagnosis and treatment. Furthermore, molecular studies of the octn2 gene in patients with the CUD will allow us to pinpoint the molecular defects and may provide insight into the critical structure-function relationships of the normal human carnitine transporter. ACKNOWLEDGMENTS This work was supported by an operating grant from the Heart and Stroke Foundation of Ontario (T-3746). I.T. is a recipient of a Medical Research Council of Canada Scholarship.
REFERENCES 1. Bremer, J. (1983) Physiol. Rev. 63, 1420 –1480. 2. De Vivo, D. C., and Tein, I. (1990) Internat. Pediatr. 5, 134 –140.
3. Stanley, C. A. (1987) Adv. Pediatr. 34, 59 – 88. 4. Engel, A. G., Rebouche, C. J., Wilson, D. M., Glasgow, A. M., Romshe, C. A., and Cruse, R. P (1981) Neurology 31, 819 – 825. 5. Rebouche, C. J., and Engel, A. G. (1982) In Vitro 18, 495–500. 6. Treem, W. R., Stanley, C. A., Finegold, D. N., Hale, D. E., and Coates, P. M. (1988) New Engl. J. Med. 319, 1331–1336. 7. Tein, I., De Vivo, D. C., Bierman, F., Pulver, P., De Meirleir, L. J., Cvitanovic-Sojat, L., Paigon, R. A., Bertini, E., Dionisi-Vici, C., Servidei, S., and Dimauro, S. (1990) Pediat. Res. 28, 247–255. 8. Stanley, C. A., DeLeeuw, S., Coates, P. M., Vianey-Liaud, D., Divry, P., Bonnefont, J.-P., Saudubray, J.-M., Haymond, M., Trefz, F. K., Breningstall, G. N., Wappner, R. S., Byrd, D. J., Sansaricq, C., Tein, I., Grover, W., Valle, D., Rutledge, S. L., and Treem, W. R. (1991) Ann. Neurol. 30, 709 –716. 9. Pons, R., Carozza, R., Tein, I., Walker, W. F., Addonizio, L. J., Rhead, W., Miranda, A. F., DiMauro, S., and De Vivo, D. C. (1997) Pediatr. Res. 42, 583–587. 10. Bahl, J. J., and Bressler, R. (1987) Annu. Rev. Pharmacol. Toxicol. 27, 257–277. 11. Bieber, L. L. (1988) Annu. Rev. Biochem. 57, 261–283. 12. Tein, I., Bukovac, S. W., and Xie, Z.-W. (1996) Arch. Biochem. Biophys. 329, 145–155. 13. Eriksson, B. O., Gustafson, S., Lindstedt, S., and Nordin, I. (1989) J. Inher. Metab. Dis. 12, 108 –111. 14. Tein, I., and Xie, Z.-W. (1996) Clin. Chim. Acta 252, 201–204. 15. Wu, X., Prasad, P. D., Leibach, F. H., and Ganapathy, V. (1998) Biochem. Biophys. Res. Commun. 246, 589 –595. 16. Pritchard, J. B., and Miller, D. S. (1993) Physiol.Rev. 73, 765– 796. 17. Ullrich, K. J. (1994) Biochim. Biophys. Acta 1197, 45– 62. 18. Zevin, S., Schaner, M. E., Illsley, N. P., and Giacomini, K. M. (1997) Pharmacol. Res. 14, 401– 405. 19. Tamai, I., Ohashi, R., Nezu, J., Yabuuchi, H., Oku, A., Shimane, M., Sai, Y., and Tsuji, A. (1998) J. Biol. Chem. 273, 20378 – 20382. 20. Huth, P. J., and Shug, A. (1980) Biochim. Biophys. Acta 602, 621– 634. 21. Molstad, P., Bohmer, T., and Eiklid, K. (1977) Biochim. Biophys. Acta 471, 296 –304. 22. Saiki, R. K., Scharf, S., Faloona, F., Mullism, K. B., Horn, G. T., Erlich, H. A., and Arheim, N. (1985) Science 230, 1350 – 1354.
401