The human pyruvate dehydrogenase complex: a polymorphic region of the lipoate acetyl transferase (E2) subunit gene

The human pyruvate dehydrogenase complex: a polymorphic region of the lipoate acetyl transferase (E2) subunit gene

128 Biochimica et Biophysica Acta, 11)97 (1991) 128 132 ¢~ 1991 Elsevier Science Publishers B.V. All rights reserved 0925-4439/91/$03.51) ADONIS 0925...

407KB Sizes 0 Downloads 55 Views


Biochimica et Biophysica Acta, 11)97 (1991) 128 132 ¢~ 1991 Elsevier Science Publishers B.V. All rights reserved 0925-4439/91/$03.51) ADONIS 09254439911101119B

BBADIS 61061)

The human pyruvate dehydrogenase complex: a polymorphic region of the lipoate acetyl transferase (E2) subunit gene Lucky H. Moehario, L. Wang*, R.J. Devenish, I.R. Mackay and S. Marzuki Department 01"Biochernistry and Centrej~r Molecular Biology and Medicine, Monash Unil'ersity, Clayton (Australia)

(Received 4 March 1991)

Key words: Pyruvatedehydrogenasecomplex;Biliarycirrhosis; Polymorphism;eDNA A major issue in the study of the pathogenesis of primary biliary cirrhosis is whether the E2 subunit of the pyruvate dehydrogenase complex (PDH-E2), the major autoantigen in the disease, exists as a tissue-specific isoform, eDNA clones spanning a segment of the 3'-catalytic region of PDH-E2 (nt 1158-1361) have been isolated from human kidney, placenta and bile epithelium cells. Nucleotide sequence analysis of the clones showed differences consistent with the presence of normal variants of PDH-E2 in the human population. However, the existence of tissue-specific isoforms of PDH-E2 cannot yet be discounted.

Introduction The mitochondrial 2-oxo acid dehydrogenase complexes, in particular the pyruvate dehydrogenase complex (PDH), have come under immunological attention because subunit components of these enzymes have been identified as autoantigens in primary biliary cirrhosis (PBC), an autoimmune disease affecting the intra-hepatic biliary ductular cells of the liver [1]. The major autoantigen is the E2 subunit of PDH [2,3]; other autoantigens are the E2 subunit of the 2-oxo glutarate dehydrogenase and branched-chain 2-oxo dehydrogenase complexes, as well as component X of PDH [2,4,5] and PDH Elc~ and El/3 [6]. PDH is an enzyme complex located on the matrix surface of the inner mitochondrial membrane [7] and is present in all tissues, yet the disease focus is tissue (biliary ductular) specific. A major question relevant to the pathogenesis of PBC is whether the E2 subunit of the 2-oxo acid dehydrogenase complexes, particularly PDH, exist as tissue-specific isoforms, since an isoform specific to biliary tissue could be the primary antigenic target. We have previously reported the sequence of a eDNA from a human heart that encodes the C-termi-

* Present address: Australian Animal Health Laboratories, CSIRO, Geelong, Australia. Correspondence: S. Marzuki. Department of Biochemistry,,Monash University, Clayton, Victoria, Australia 3168.

hal portion of PDH-E2 [8]. A comparison of this heart cDNA sequence and the published placental PDH-E2 cDNA sequence [9], suggested that tissue-specific isoforms of PDH-E2 may exist. In this communication we present results of studies aimed at establishing whether the sequence differences previously observed represented tissue-specific isoforms, or simply normal variation in individuals within the human population. Materials and Methods Source o f c D N A s

Sources of human tissues for analysis were as follows: (a) kidney, human kidney eDNA library in Agtl 1 (Clontech Laboratories, Palo Alto, CA, U.S.A.); (b) placenta, full-term placenta from three normal individuals obtained 2-4 h after delivery (St. Vincent's Private Hospital, Melbourne, Australia); (c) bile duct, a dissection of bile duct adjacent to the gall bladder was obtained from a PBC patient with a gall stone who was undergoing cholecystectomy. The patient was a female aged 70 with PBC and serum antibodies to PDH-E2. The bile epithelium cells were scrapped gently from the epithelial surface of the duct tissue and collected after three to four washes in phosphate buffer saline (pH 7.2). Synthetic oligonucleotides

Specific oligonucleotide primers used in the present study (shown in Table I) were synthesized in our laboratory using an Applied Biosystems Model 380S DNA

129 synthesizer. The oligonucleotides were purified by preparative polyacrylamide electrophoresis on a 15% polyacrylamide gel. Separated oligonucleotides were visualized directly under ultra violet light with silica backing placed under the gel (UV shadowing). The region containing full-length oligonucleotides was excised and the DNA was eluted. The concentration was estimated from the absorbance at 260 nm [12]. Isolation of phage DNA

a phage DNA from the human kidney cDNA library was prepared using a procedure described by Maniatis et al. (1982) [11]. The library was amplified to obtain a titer of 10m-1012 p f u / m l , followed by a large scale bacteriophage preparation and D N A isolation. The DNA obtained was purified using an Elutip-d column (Schleicher and Schuell, Dassel, F.R.G.). Isolation of poly(A) + m R N A

Isolation of poly(A) + m R N A was performed directly from tissues using a Fast track m R N A isolation kit (Invitrogen Corporation, San Diego, U.S.A.). 1 gram of tissue was lysed in the lysis buffer provided by the manufacturer and homogenized with a high speed Polytron homogenizer for 15-30 s. The lysate was then incubated at 45°C for 1-2 h; at the end of the incubation the salt concentration of the lysate was adjusted to 0.5 M by addition of 5 M NaCl. Poly(A) + mRNA was selected from the bulk of cellular RNA by using the affinity column oligo (dT) cellulose provided by the manufacturer. Finally the m R N A was eluted from the column, precipitated with 0.15 vols. of 2 M Na-acetate (pH 4.8) and 2.5 vols. absolute ethanol, and resuspended in 20 >1 sterile water. This procedure generally yielded 5 - 1 0 / x g mRNA per 1 g of tissue. Synthesis of cDNA

First strand cDNA synthesis was performed using a cDNA cycle - Red Module kit (Invitrogen Corporation,

San Diego, U.S.A.). An aliquot of 0.2-1/xg of poly(A) + m R N A was placed in an RNase free tube, methyl mercury hydroxide (CH3HgOH) added to a final concentration of 10 mM, and then incubated at room temperature (RT) for 5 min. After incubation, /3-mercaptoethanol was added to the mixture to 0.15 M to dissociate C H 3 H g O H from the mRNA. A specific oligonucleotide primer for PDH-E2, LF#1 (Table I), at a final concentration of 0.5-1 ~ M was used to prime the mRNA template for cDNA synthesis. The polymerization reaction was then set up as follows: the primed m R N A template, 1 /xl placental RNase inhibitor, 4 /~1 5 × buffer (as provided by the manufacturer), 1 ~zl 25 mM dNTPs and 5 units reverse transcriptase were combined and the total vol. adjusted to 20 ~zl before incubation at 42°C for 60 min. After incubation, RNA-cDNA hybrids were denatured by heating at 95°C for 3 min. A second round of synthesis was carried out by the addition of 5 units reverse transcriptase to the mixture, incubation at 42°C, and then heating at 95°C. The cDNA product was extracted with phenol/chloroform and precipitated with absolute ethanol. The cDNA pellet was then resuspended in 20 ~1 sterile water (100 ng//xl). Polymerase chain reaction amplification

Polymerase chain reaction (PCR) amplification was carried out using a thermostable Taq DNA polymerase [13]. The synthetic oligonucleotide primers employed are shown in Table I. Reactions were set up in a total vol. of 100 ~1 containing 200/xM dNTP, 2 mM MgC12, 10 mM Tris-HCl (pH 8.3), 50 mM KCI, 0.5-1 /xM primers and 2.5 units Taq DNA polymerase (Perkin Elmer Cetus, Norwalk, CT, U.S.A.) and conducted on a Gene Machine (Innovonics, Victoria, Australia) for 30 cycles with the following parameters: denaturation at 95°C for 60 s (except for the first cycle denaturation was for 5 min), annealing at 55°C for 90 s, elongation at 72°C for 180 s.

TABLE 1 Synthetic oligonucleotides employed in the PCR amplification of PDH-E2

Primer LF#1




Position i 1343-1361

Length 25 met

Hybridize to sense-strand

866- 886

25 mer

non-sense strand

1 158-1 182

31 mer

non-sense strand



Eco R I




agtll #1218


24 mer

i Relative to placental sequence [9]: nucleotide A of the AUG translational initiation codon is designated 1.

130 Nucleotide sequencing Nucleotide sequence analysis was carried out using a T7 sequencing kit (Pharmacia LKB Biotechnology, Upssala, Sweden). The nucleotide sequence of cloned DNA in M13mpl8 and pBluescript KS was determined by the dideoxynucleotide chain termination method [14]. In the case of kidney cDNA, the 200 bp fragment was amplified asymmetrically by PCR and subjected to direct nucleotide sequencing using a specific oligonucleotide as the sequencing primer as described in Results and Discussion.






t~ A




C LF-1

Results and Discussion

The sequence differences described in our previous report [8] included the insertion of three nucleotides T, A and G in the heart cDNA sequence compared to the placental sequence, which in consequence resulted in: (a) the generation of a Clal restriction site in the heart cDNA sequence (hereafter this region is referred to as the Clal region) and (b) changes of amino acid sequence in this region, from 414[YLLSCKYGEV]423 for the placental PDH-E2 to 414[YLSIDVNMGEV]424 for the heart antigen. Another human cDNA coding for PDH-E2 isolated independently from a human liver cDNA library [10] showed identical nucleotide sequence in this region to that of the heart cDNA sequence. In the present study, the nucleotide sequence of the C-terminal ClaI segment of the PDH-E2 gene was determined for a wider range of tissues and individuals. First, a cDNA clone of 0.9 kb in length which includes the Clal region of PDH-E2 was isolated from a human kidney cDNA library. The strategy employed is shown in Fig. 1. First round PCR amplification was carried out on total purified A phage template DNA, prepared from an aliquot of the kidney cDNA library. The pair of oligonucleotide primers used were: LF#1, specific to the PDH-E2 sequence nucleotide (nt) 13431361, and Agtl 1 #1218, specific to the right arm of the Agtll, around 40 nucleotides 5' of the EcoRl cloning site (Table I). When the products were examined by electrophoresis on an agarose gel, a faint band corresponding to a 0.9 kb fragment was observed. This product of the PCR amplification reaction was then subjected to second and third rounds of amplification by PCR using the oligonucleotide primer pairs L F # 1 L F # 2 and L F # 1 - L F # 4 (Table 1) respectively in order to amplify fragments spanning the Clal region and to determine whether the original PCR product (0.9 kb fragment) was specific to the PDH-E2 gene. In this manner a 200 bp fragment that represented part of catalytic/E2 domain of PDH-E2 (nt 1158-1361) spanning the Clal region was successfully isolated. This DNA fragment was then subjected to asymmetric PCR amplification with the oligonucleotide primer pair


0.90 kb


0.49 kb


0.20 kb

Fig. 1. Isolation of the 'Clal region' DNA of PDH-E2. The approach employed in the isolation of the Clal region DNA of PDH-E2 from the human kidney library is diagrammed. An aliquot of 20 ng of purified phage DNA was subjected to PCR amplification using oligonucleotide primer pairs as follows: (A) first round PCR amplification using LF#I and Agtll #1218 detected a faint band corresponding to a 0.9 kb fragment; (B) second round PCR amplification using LF#1 and LF#2 showed a band of 0.49 kb as expected; and (C) third round PCR amplification using LF#I and EF#4 yielded a band of 0.2 kb in size as expected. The representation of the PDH-E2 gene shows the sequence blocks coding for the structural domain of human placental PDH-E2: leader sequence (L), outer and inner lipoyl binding domain (O/I-LBD), E l / E 3 binding region ( E l / E 3 ) and catalytic/E2 binding domain (Catalytic/E2) [15].

L F # 1 and LF#4; the L F # 1 primer was present in limiting quantity in order to provide template for direct sequencing by the dideoxy method [14]. The nucleotide sequencing was performed with the use of L F # 1 oligonucleotide as the sequencing primer. Thus, this segment of the kidney cDNA clone was found to have identical sequence with that of the heart [8] and liver [10] cDNAs (Fig. 2). To investigate more specifically whether the placental sequence [9] represented tissue-specific variation or polymorphic variation within the human population, the sequence of the Clal region from three independent sources of placenta was analyzed. Poly(A) + mRNA was freshly isolated from the placental tissues [11] and first strand cDNAs synthesized using reverse transcriptase and the L F # 1 oligonucleotide as primer. An aliquot of 100 ng of cDNA was subjected to PCR amplification to isolate fragments spanning the Clal region from nt 1158 to 1361. The approach used was

131 L





i! i:


a Heart







* ......

* ...........

i i i i GTrGG * ..............


Kidney Racenta






Bile e Epithelium


A ..........

d ..................

T ................

A .......................


.............................................. ..............................................

Fig. 2. Nucleotide sequence of 'Clal region'. The diagram shows the nucleotide sequences in the Clal region of the catalytic/E2 domain of PDH-E2, between nt 1158-1361. The inserted nucleotides T, A and G previously found in the heart sequence (a: Ref. 8) are present in all clones. Additionally, in placenta I and placenta II cDNAs two base substitutions are present compared to the published placental sequence (b: Ref. 9), heart, as well as liver (c: Ref. 10). (d) Three independent clones showed identical sequence; (e) Two independent clones showed identical sequence; (*) Nucleotides which are deleted in the placenta sequence.

the same as that employed for the second and third rounds of PCR amplification of the kidney eDNA library (Fig. 1), and involved the sequential use of the specific oligonucleotide primer pairs L F # 1 - L F # 2 and L F # 1 - L F # 4 . The 200 bp fragments ultimately obtained were then ligated into one of the vectors pBluescript KS or M13mp18. Nucleotide sequence analysis of these clones (Fig. 2) showed that the three single nucleotide insertions (T, A and G) found previously in the heart sequence [8] are present in each of the three placental derived DNA fragments. In addition, in the fragments isolated from placenta I and placenta II, there are two additional base substitutions in each case, compared to the published placental sequence and also to those of the heart, liver and kidney. These nucleotides differences are G12s1 to A and G1261 t o T (placenta I) and G1251 t o A and T]273 to C (placenta II), resulting in Val to Ash, Gly to Ile and Val to Asn amino acid changes compared to that of the heart, respectively (Table II). Three independent clones were sequenced for placenta TABLE II

Amino acid sequence of the 'Clal region' in 3'-Catalytic portion of PHD-E2 Heart [8] Placenta [9] Liver [10] Kidney Placenta I Placenta II Placenta III Bile epithelium





. . . . .











. . . . .

. . . . . . .



. . . . . . .


. . . . . . . .



. . . . .



III and the same nucleotide sequence was obtained for each of the clones. This suggests that the variations observed are not associated with sequence heterogeneity within the eDNA population, either as a result of PCR-induced error or of multiple templates as a manifestation of the presence of m R N A transcripts from a multigene family. Taken collectively, these data indicate that the observed differences represent normal variants of PDH-E2 in the human population. The original placental sequence may represent an uncommon polymorphic variant. Finally, and in keeping with our particular interest in the pathogenesis of PBC, we determined the nucleotide sequence of the ClaI region in a eDNA encoding PDH-E2 using DNA isolated from cells derived from the bile duct epithelium of a patient with PBC. The strategy employed for this was precisely the same as that described above for the placental samples. PCR-amplified eDNA obtained was cloned into M13mp18 and the nucleotide sequence was determined. The nucleotide sequence derived from two independent clones showed identical sequence to that of the heart PDH-E2 eDNA sequence [8] (Fig. 2) and therefore the corresponding inferred amino acid sequence for biliary ductular PDH-E2 also was identical to that of heart PDH-E2 (Table II). While the sequence variations observed in the present study would seem to be the manifestation of normal polymorphic variation for PDH-E2, the possible existence of tissue-specific isoforms of PDH-E2 cannot yet be excluded on the data presented here. In this context, we have investigated human genomic DNA obtained from peripheral blood mononuclear cells of a normal individual by Southern blot/hybridization analysis. An oligonucleotide specific to the heart PDH-E2 sequence in the ClaI region, E2-h (nt 1243-1264), was used as the hybridization probe. The probe detected two bands in DNA digested with either EcoRI, HindlIl or HincH (data not shown) suggesting the presence of at least two PDH-E2 related sequences in the human genome. As we have also already observed [8] there is an indication from the literature that the N-terminus of the precursor PDH-E2 protein of placenta may differ markedly from that of the liver precursor protein. The significance of the reported differences in the N-terminal leader sequence between the PDH-E2 precursors of these tissues remains unclear. The determination of the complete nucleotide sequence from the tissue sources examined here will be necessary to establish whether there is indeed a different leader sequence for PDH-E2 in different tissues.

Acknowledgements We thank Dr. B. Loveland for his assistance in obtaining the bile epithelium cells. This work was sup-

132 ported by grant 880249 from the National Health and Medical Research Council of Australia.


References 10 I Kaplan, M.M. (1987) N. Engl. Med. 36, 521-528. 2 Yeaman, S.J., Fussey, S.P.M., Danner, D.J., James, O. F.W., Mutimer, D.J. and Bassendine, M. (1988), Lancet i, 1067-1070. 3 Baum, H. and Palmer, C. (1985) Molec. Aspect Med. 8, 201 236. 4 Fussey, S.P.M., Guest, J.R., James, O.F.W., Bassendine, M.F. and Yeaman, S.J. (1988) Proc. Natl. Acad. Sci. USA 85, 8654 8658. 5 Surh, C.D,, Roche, E.T., Danner, D.J., Ansari, A., Coppel, R.L., Prindville, T., Dickson, E.R. and Gershwin, M.E. (1989) Hepatology 10, 127-133. 6 Fussey, S.P.M., Bassendine, M.F., Fines, D., Turner, I. B., James, O.F.W. and Yeaman, S.J. (1989) Clin. Science 77, 365-368. 7 Yeaman, S.J. (1986) Trend Biochem. Sci. 11,293-296. 8 Moehario, L.H., Smooker, P.M., Devenish, R.J., Mackay, I.R.,




14 15

Gershwin, M.E. and Marzuki, S. (1990) Biochem. lnternat. 20, 417-422. Coppel, R.L.. Jane McNeilage, L., Surh, C.D., Van de Water. J., Spithill, T.W., Whittingham, S. and Gershwin, M.E. (1988) Proc. Natl. Acad. Sci. USA 85, 7317 7321. Thekkumkara, T.J., Ho, L., Wexler, I.D., Pons, G., Liu, T.C'. and Patel, M.S. (1988) FEBS Lett. 240, 45-48. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular cloning: A laboratory mannual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular cloning: A laboratory mannual, second edition. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Saiki, R.K., Gelfand, D.It., Stoffel, S., Scharf, S.J., Higuchi, R., ttorn. G.T., Mullis, K.B. and Erlich, H.A. (1988) Science 239, 487-491. Sanger, F., Nicklen, S. and Coulson, A.R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463 5467. Surh, C.D., Coppel, R. and Gershwin, M.E. (1990) J. Immunol. 144, 3367-3374.