PII: 0014-5793(91)80755-R Volume 284, number 1, 34-38 FEBS 09797 0 1991 Federation of European Biochemical Societies 00145793/91/S3.50 ADONIS 0014579391005043 3une 1991 Structure of the gene encoding the entire mature Ha subunit of human branched-chain a-keto acid dehydrogenase complex Nahid DariushI, Charles W. Fisher’, Rody P. Cox2 and David T. Chuangl Departments of ‘Biochemistry and 21nternal Medicine, The University of Texas Southwestern Medical Center. Dallas, TX 75235, USA Received 9 April 199 1 We report the isolation of a 22-kb human gcnomic clone (G7) that contains 8 exons encoding a partial mitochondrial presequence, the entire mature peptide and the complete 3’ untranslated region of the E Icr mRNA of human branched-chain cr-keto acid dehydrogenase complex. Based on this gene structure, exon 9 contains the Tytig3 -+ Asn mutation previously identified in the EIK subunit of Mennonite and other maple syrup urine disease (MSUD) patients. Moreover, the homozygous mutation appears to cause skipping of exon 6 in the mutant Elu transcript. The information on the gene structure for the entire mature Elol subunit will facilitate investigations into the molecular basis of MSUD involving this subunit. ELK gent structure (human); Maple syrup urine disease; Elr mutation; Exon skipping 1. INTRODUCTION The mammalian branched-chain a-keto acid de- hydrogenase (BCKAD) complex catalyzes the oxidaltive decarboxylation of the cr-keto acids derived from leucine, isoleucine and valine. The mitochondrial multi- enzyme complex consists of three catalytic components: a branched-chain ru-keto acid decarboxylase or E 1 com- prised two LY (Mr = 67000) and two ,8 (MI = 37 500) subunits, a dihydrolipoyl transacylase or E2 with a 24-mer structure (monomer MI = 46500) forming the core of the enzyme complex, and a dihydrolipoyl de- hydrogenase that exists as a homodimer (monomer Mr = 52000) and is a common component shared with pyruvate and a-ketoglutarate dehydrogenase complexes [ 11. In addition, the mammalian BCKAD complex con- tains two regulatory enzymes, a specific kinase and a specific phosphatase that control the activity of the en- zyme complex through reversible phosphorylation/de- phosphorylation mechanism [2]. Elc~ subunit of a compound heterozygous MSUD pa- tient [5]. The same mutation was later shown to be pres- ent in homozygous state in MSUD patients from a Men- nonite population [6,7]. To facilitate analysis of muta- tions in MSUD affecting the Elcv subunit, we have undertaken the cloning of the human Elcv gene. We report here the gene structure for the entire mature Elcr subunit of the human BCKAD complex, and the precise skipping of an exon in the mutant Elcr transcript. 2. MATERIALS AND METHODS 2. I, cDNA probes Partial human Ela cDNA clones used in the present study were described previously [4]. The regions of the human Ela cDNA se- quence present in these cDNA clones were as follows: hEla-I (0.9 kb), bases 840-1783; hEla-2 (1.5 kb), bases 191-1783 and hEla-4 (0.2 kb), bases l-190. cDNA probes were radiolabeled by the random priming method [8]. 2.2. Genontic screening and subcloning Maple syrup urine disease (MSUD) is an autosomal recessive disorder, in which the activity of the BCKAD complex is deficient leading to ketoacidosis, mental retardation and a high mortality among patients. The disease is genetically heterogeneous. Deficiencies in dif- ferent subunits of the BCKAD complex have been shown using cultured cells from MSUD patients j3,4]. A T-A transition resulting in a Tyr-Asn missense mutation was described by Zhang ct al. to occur in the A human leukocyte genomic library in AEMBL-3 (Clontech) was screened by plaque hybridization. Plaques producing positive signals were purified, and phage DNA prenxred from minilysates as describ- ed previously [9]. The purified phage DNA was digested with restric- tion enzymes, followed by fractionation in agarose gels. Exon- containing fragments were identified by Southern blotting analysis, and subcloned into the Bluescript KS vector (Stratagene). 2.3. Nucleotide sequencing Cbrres/Jondrnce address: D. Chuang, Dcpartmcnt of Biocl:cmistry, UlliWrsitY of Texas Southwestern Medical Center, 5323 Harry I-lines Blvd., Dallas, ‘I-X 75235-9038, USA. Fax: (I) (214) 688.8856. Plasmids containing genomic DNA or cDNA inserts were obtained from the minipreparations of transformed E. culi XL-1 Blue cells [9]. Nuclrotide sequencing was carried out by the dideoxy chain termina- !inn method [lo] using T7 DNA polymerase (Pharmacia) on double- stranded tcmplatcs [ 1 l] I 2.4. cDNA synrlresis arrd primm Poly(A)’ RNA was isolated from cultured fibroblasts [12) and 34 PuOlislrad 6.s Blsekr Sclorcc Publishers .!I. 1’. Volume 284, number I FEB.5 LETTERS .Fuae 1991 E 3.3 Lb E I .J E 5.5 hb E I J S 4.5 kb 6 I- i B 11.2 kb S I kb Fig. 1. Restriction map and exon organization of the human Elr~ gene. The exons (2-9) present in the G7 clone (22 kb) are represented by closed boxes, and numbered after the putative exon 1. Restriction fragments (horizontat bars) were generated with SolI (S). EcoRI (E) and BarnHI (B). cDNA synthesized and amplified by the polymerase chain reaction (PCR) [13]. The primers used in the amplification of different regions of the Elm cDNA (5’ --+3 ‘) [4] and restriction sites contained were as folIows: 1: CGGGGCGATCGATGCAGCGA (CM) (bases l-20), 1’ : CTCATAGAGGATCCGGTCCA (Brrinli I) (bases 339-358), 2: TGCTGCAGCTCTACAAGAGCA (Psri) (bases 300-320), 2’ : GTAACAGATGTCGACCCCTGTT (Sue (bases 680-700), 3 : ACTCGCCACGCGGATCCCT (BumHI) (bases 618-637), 3’ : GCTCAGCAGATAGTGCTGCTGCAG (PsrI) (bases 1085-l 105), 4:GGGTACCACAGCACCAGTGAC (Kpnl) (bases 99%lols), 4' : TCTCGGGGTACCTGAGGATGG (bases (Kpnl) 1356-1376). 3. RESULTS AND DISCUSSION A tota1 of 2.4 x lo6 pfu from the EMBL-3 genomic library were screened using hElru-1 cDNA (bases 840-1783) as a probe. One of the positive clones (G7) purified contained a 22-kb insert for the Ela! gene. A restriction map of the entire genomic insert is shown in Fig. 1. Digestion of the G7 clone witfi EcoRI produced gene fragments of 3.3 and 5.5 kb. Hybridization with hEla-2 cDNA (bases 191- 1783) [4], and subsequent nucleotide sequencing showed that these two J!ZCORI fragments in combination contained the 3’ end of exon 2 (bases 190 to 283 in the Ela, cDNA) and exons 3,4 and 5. The EcoRI site in exon 2 was identical to the one pres- ent in the human E~CX cDNA [4]. Double digestion of the G7 clone with San and BarnHI resulted in gene fragments of I .7, 11.2 and 4.5 kb (Fig. 1). Wybridiza- tion with hEI@-4 cDNA (bases I-190) indicated that the 1.7 kb ~~~I-~~~~1 fragment contained exons in the 3so 360 370 ValMetG~uALaPheGluGlnAlaCluArgLysProLysProAs~roAsnLeuLeuPh~SerAspValTyrGlnGluMet acagCTGATGGA~CTTTGAGCAGGCCGAGC~~GCCC~CCCAACCCCAACCTGCTCTTCTCAGACGTGTATCAGGAGATG 1243 380 390 ProAlaGlnLeuArgLysGlnGlnCluSerLeuAlaArgH~sLeuGlnThrTyrGlyGluI~i~~yrProLeuAspHisPheAsp CCCGCCCAGCTCCGC.~GCAGCAGGAGTCGCCCCCCACCTGCAGACCTACGGGGAGCAC~ACCCACTGGATCACTTCGAT 1327 400 Lys*** AAGTGAGACCTGCTCAGCCCACCCCCACCCRTCCTCCTCAGC~ACCCC~AGAGGTA~CCCCACTCT~GGGGAGCAGGGGGACCTGA 1411 CAGCACACCACTGTCTTCCCCAGTCAGCTCCCTCT~TACTCA~GGCCA~GGCGGCTGCCA~TCTTCACCCCTGCTCCTCC 1495 CGGCTGTTACATTGTCAGGGGACAGCATCTGCAGCAGTTGCTGA~CTCCGTCAGCCCCCTCTTCACCTGTTGTTACAGTGCCT 1$79 TCTCCCAGCCGCTGCGTGAC;CGCACATTCAGCACTAG~GCCCCTCTGGGCE,TGGGGTGGACATGGCAGGTCAGCCTGTGG~C 1663 TTGCGCACGTGCGAGTCGCCAGCAGAGGTCACGAATAAACTGCATCTCTGCGCCTGGCTCTCT 1726 Fig. 2. Nuclsulidc and &~LIWJ tllliillu acid sequences UT the ehc)m of llUlllilll Elcv yer~o. EUJII~~ seyucllccs ate in caWls, atld Ilat ciitl ill(lullir: SC~IICIICCS at cxon/intron boundaricsorc in lower case. Theapproximatc sizes of introns (except intron I) are indicated. The highly COnscrVCd region (undcrlh~cd) flanking the two phosphoscrinc (asterisk) r&dues is encoded by cxon 8. The complete 3’ ->---” end of the gene, which Includes the polyadenylatioa signal (AATAAA) (underlined) tcrmhlatcs at base 1726. Nuclcotide numbers arc accordine 10 IhE Ilulllan Lila CDNA scqlleilcc 14). 35 Volume 283, number 1 FEBS LETTERS June 1991 aa- Q_/) 'y=jy-J (1) - "3' Fig. 3. Subclones of the amplified regions of the Elcv cDNA from a Mennonite MSUD patient (P.K.). Open boxes represent the coding region and solid lines the non-coding region of the Elm cDNA. Stippled boxes depict primer pairs used for PCR amplification as described in section 2.4. The number in parentheses indicates the number of subclones isolated and sequenced. The deleted exon 6 is shown by a peak line. The T-A conversion at base 1307 is present in the amplified 2-4’ region. 5’ portion of the gene. Probing with hElcr-1 cDNA showed that the 4.5 kb WI-BarnHI fragment carried exons in the 3’ region. Further restriction analysis, subcloning and sequencing established that exons 2 and 3 were located in the 1.7 kb WI-BarnHI fragment and exons 6,7,8, and 9 in the 4.5 kb SalI-BumHI fragment. The organization of exons 2 to 9 of the Elcv gene is depicted in Fig. 1. Fig. 2 shows the complete coding sequences for exons 2-9 and exon-intron boundaries of the Elcv gene. The 8 exons encode a partial mitochondrial presequence from His-(-g) to Phe-(-I), the entire mature Elcv subunit (Ser-1 to Lys-400) and the complete 3’ un- translated region including the single polyadenylation signal (AATAAA). The sizes of exon are as follows: ex- on 2 (180 bases), exon 3 (87 bases), exon 4 (109 bases), exon 5 (106 bases), exon 6 (206 bases), exon 7 (143 bases), exon 8 (172 bases) and exon 9 (564 bases). The introns range from 106 bases (between exons 2 and 3) to 5 kb (between exons 4 and 5) in length. All intronic 5’ and 3’ splice sites conform to the gt-ag rule [14]. The G7 clone does not contain the genomic sequence encoding the 5’ untranslated region of the Elcv mRNA that is not present in the cDNA. The upstream mitochondrial presequence from Gly-(-43) to Ala- (-10) in the Ela cDNA is also absent in the genomic clone (Fig. 2). These two regions are putatively encoded by exon 1 by analogy with the gene for thefl subunit of human mitochondrial ATP synthase [15]. In the past two years we have isolated and purified approximately 80 additional clones from four different genomic libraries, but none of these clones contained the putative exon 1 and the 5’ flanking regions of the Elcv gene. The reason for the inability to clone the extreme 5’ region is presently unknown. One possibility is that this region contains inverted repeats forming a cruciform structure, which prevents cloning in the A phage vector [16]. Based on the gene structure for the mature Elcv subunit, the homozygous T-A transition that causes MSUD in Mennonites [6,7] is present in exon 9. This mutation appears to affect the assembly of the Ela and El,0 subunits into an cr& structure (Fisher et al., in preparation). Moreover, during amplification of dif- ferent regions of the mutant Elcv cDNA from a Men- nonite MSUD patient (P-K.), a deletion that cor- responds precisely to exon 6 was found (Fig. 3). The deleted mutant transcript is a minor species, and ap- pears to be a secondary event related to the homozygous primary mutation. The mechanism for this exon dele- tion in the mutant Ela transcript remains to be elucidated. However, one can speculate that the primary mutation in exon 9 may have disruptive effects on the entire secondary structure of the mutant Elc~ pre-mRNA. Exon deletions may arise as a result of the failure of the splicing complex to recognize the correct splice site. Acknow/edger,le,frs: This work was supported by Grants DK-26758 and DK-37373 from the National Institutes of Health, and Grant l-l 149 from the March of Dimes Birth Defects Foundation. REFERENCES [l] Pettit, F.H., Yeaman, S.J. and Reed, L.J. (1978) Proc. Natl. Acad. Sci. USA 75, 4881-4885. [2] Randle. P.J., Fatania, H.R. and Lau, KS. (1984) in: Enzyme Regulation by Reversible Phosphorylation-Advances (Cohen, P. ed.) pp. l-26, Elsevier, Amsterdam, New York. [3J Danner, D.J. and Elsas, L.J. (1989) in: The Metabolic Basis of Inherited Disease (&river, C.R., Beaudet, A.L., Sly, W.S. and Valle. D. eds) 6th edn., pp. 671-692, McGraw-Hill, New York. -- 4 Fig. 3. Subcloncs of the amplified regions of the Ela cDNA from a Mennonite MSUD patient (P.K.). Open boxes represent the coding region and solid lines the non-coding region of the Elo cDNA. Stippled boxes depict primer pairs used for PCR amplification as dcscrlbcd in section 2,4. The number in parcnthsscs indicates the nutnbcr of subcloncs isolated and scqucnccd. The deleted cxon 6 is shown by n peak linr. Tbc T---+A conversion at base 1307 is prcssnr in the amplified 2-4’ region. 36 Volume 284, number 1 FEBSLETTERS June 1991 -43 -40 -30 -20 GlyAlaIfeAlaAlaAlaArgValTrpArqLeuAs~rgGlyLeuSerGl~l~laLeuLeuLeuLe~r~lnProGlyAls CGGGGCGATCGCTGCAGCGAGGGTCTGGCGGCTPLAACCCCT 85 -10 1 ArgGlyLeuAlaArgSer XisProProArgGlnGlnGlnGlnPheSerSerLeuAspAspLysPro CGGGGACTGGCTAGATCT...... ..ctcttccccagCACCCCCCCAGRGGCAGCAGCAGCAGTTTTCATCTCT~ATGAC~GCCC 151 10 20 30 GlnPheProGlyAlaSerAlaGluPheIleAspLysLeuGluPheI~eGl~roAsnValIleSer~ly~leProIleTyrArg CAGTTCCCAGGGGCCTCGGCGGAGTTTATAGTTGGTCCCCATCTACCGC 235 40 50 ValMetAspArgGlnGlyGlnIleIleAsnProSerGluAspProHis GTCATGGACCGCCAAGGCCAGATCATCAACCCCCAGCGAGGACCCCCACgtgagaggcggcctcccccact~cccgtgcccccca 283 Leu?roLysGluLysValLeu cgcccaggcc...../O.l kb/... CcaactgccccacgtctatctgtgcctccacccycagCTCGGTGCTG 304 60 70 80 LysLenTyrL:rsSerMetTrLeuLeuAsnThrMetAspArgIleLeuTyrGluSerGl~rgGln hRGCTCTACAAGAGCATGACACTGCTT~CACCATGGACCGCATCCTCTATGAGTCTCAGCGGCAGgtgcgtgg..../3.1 370 90 100 GlyArgIlePhePheTyrMetThrAsnTyrGlyGluG~uGlyThrHisValGlySerAla I&/...... .ccactccacccccagGGCCGGATCTCCTTCTACATGACC~CTATGGTGA~AGGGCACGCACGTGGGGAGTGCC 430 110 AlaAlaLeuAspAsnThrAspLeuValPheGlyGlnTyrArgGluAlaG GCCGCCCTGGACAACACG;ACCTGGTGTTTGGCCAGTACCGGGAGGCAGgtacgtct...../5.0 kb/......ctcctcccct 479 120 130 140 lyValLeuMetTyrArgAspTyrProLeuGluLeuPheMotAlaGlnCy~TyrGlyAsnIleSerAspLeuGlyL~sGly cctagGTGTGCTGATGTATCGGGACTACCCCCTACATCAGTGACTTGGGCAAGGGG 559 150 160 170 ArgGlnMetProValHisTyrGlyCysLysCluArgHi~PheVal~~~IleSerSerProLeuAlaThrGl~IlePr~G~~ CGCCAGATGCCTGTCCACTACGGCTGCAACCAACGCCACTTCGTCACTATCTCCTCTCCACTGGCCACGCAGATCCCTCAGGgtg 641 la0 laValGlyAlaAlaTyrAlaAlaLysArgAlaAsnAlaAsnArgVal aggat.... ./3.1 kb/..... tctcatcccctgcagCGGTGGGGGCGGCGTACGCAACAGGGTC 668 190 200 210 VaiIleCysTyrPheGlyGluGlyAlaALaSarGluGlyAsphlaHisAlaGlyPheAsnPh~Al~laThrLeuGluCysPro GTCATCTGTTACTTCCGCGAGGGGGCAGCCAGGGGGCTTCGCTGCCACACTTGAGTGCCCC 772 220 230 IleIlePhePheCysArgAsnAsnGlyTyrAlaIleSerThrProThrSerGluGlnTyrArgGlyAspG~yIleA ATCATCTTCTTCTGCCGGAACAATCGCTACGCCATCTCCACGCCCACCTCTGAGCAGTATCGCGGCGATGGCATTGgta~gggc.. 848 240 250 laAlaArgGlyProGl~yrGlyIleMetSerXleArgVa~A~p~lyA~~sp . ../0.3 kb/...... tctgtgtccccacagCAGCACGAGGCCCCGGGTATGGCATCATGTC~TCCGCG~GGATGG3~TGAT 901 260 270 280 ValPheAlaValTyrAsnAlaThrLysGluAlaArgArgArgAlaValAlaGluAsnGlnProPheLeuIleGluAlaMe~Thr GTCTTTGCCGTATACAACGCCACAAAGGAGGCCCGACGGCGGGCTGTGGCAGAG~CCAGCCCTTTCTCATCGAGGCCATG~CC 985 290 * 300 TyrAr ~1l~Gly~i~~i~SerThrSerAs~As~SerSerAlaTyr TACAGgtgcctgc...../C.2 kb/..... cttgcccctgtgcagGATCGGGCACCACAGCACCAGTGACGACAGTTC.4~CGTAC 1030 * 310 320 Ar~SerValAs~GluValAsnTyrTrpAsDLYsGlnAspHisProI~eSerArgLeuArgHis~yrLeuLeuSerGlnGlyTrp CGCTCGGTGGATGhGGTC~~T~ACTGGGAT~CAGGACChCCCC~TCTCCCGGCTGCGGCACTAT~rGCT~AG~~AA~~~~~~ 1114 330 340 TrpAspGluGluGlnGluLysAlaTrpArgLysGlnSerArgArgLys TGGGATGACCAGCAGGAGAAGGCCTGGAGTCCCGgtg~gg~~.,.. ./l.i ti/. . . ..cccatqtcccc 1162 37 Vo!umc 284 , number 1 FEES LETTERS June 1991 [4] Fisher, C.W., Chuang, J.L., Griffin, T.A., Lau, KS., Cox, R.P. and Chuang. D.T. (1989) J. Biol. Chem. 264.3448-3453. IS] Z