311
Biochimica et Biophysica Acta, 1139 (1992) 311 - 314 © 1992 Elsevier Science Publishers B.V. All rights reserved 0925-4439/92/$05.00
BBADIS 61173
Multiple short direct repeats associated with single mtDNA deletions Nils-G6ran Larsson and Elisabeth Holme Department of Clinical Chemistry, Gothenburg University, Sahlgren's Hospital, Gothenburg (Sweden) (Received 26 February 1992)
Key words: mtDNA deletion; Kearns-Sayre syndrome; Pearson's syndrome; Mitochondrial encephalomyopathy
We have sequenced the breakpoints of deleted mtDNA in muscle from four children with mitochondrial myopathy and multisystem mitochondrial disorders. The deletions were 4884, 6067, 7663 and 7150 base pairs (bp) in size and affected several protein and transfer RNA genes. The sequences needed for transcription and replication of mtDNA were not affected in any case. The deletions were flanked by direct short repeats in all cases. Multiple repeats were found in case 1 and 4. Imperfect repeats were found in case 3 and 4 and this made it possible to distinguish the repeats 5' and 3' to the deletion. In both cases the 3' repeat was retained. The deletion of 7663 bp in case 3 has been reported in two other cases and may represent a second hotspot for mtDNA deletions in addition to the common deletion of 4977 bp found in one third of cases. A comparison of the breakpoint sequence of case 3 with the two other reported cases revealed that when a deletion is formed between the same repeats in different patients either the 5' or 3' repeat can be retained. This study shows that both single and multiple repeats can be associated with single mtDNA deletions and that both 5' and 3' repeated sequences can be retained. These findings are consistent with the slip-replication model for the generation of mtDNA deletions.
Introduction Single large deletions of m t D N A are associated with Pearson's syndrome [1,2] or other severe multisystem disorders of infants [3,4], and chronic progressive external ophthalmoplegia or Kearns-Sayre syndrome in children and adults [5-7]. Infants who survive the severe anemia of Pearson's syndrome may later in life develop neuromuscular and heart symptoms typical of KearnsSayre syndrome [8,9]. The patients always have heteroplasmy with a mixture of normal and deleted m t D N A in various proportions. The fraction of deleted m t D N A may vary widely between different tissues of the same individual but it is generally high in organs manifesting disease, e.g., muscle, brain or bone marrow [7,8,10]. The disorders with single m t D N A deletions are usually spontaneously occurring [11,12] and only one type of deletion is found in different tissues from the affected individual [7,8,10]. This strongly suggests that the m t D N A deletions occur de novo in the oocyte or at a very early stage of embryogenesis. The molecular mechanism involved in the formation of m t D N A deletions is unclear. Sequencing of the breakpoint region of
Correspondence to: Nils-G6ran Larsson, Department of Clinical Chemistry, Sahlgren's hospital, S-413 45 Gothenburg, Sweden.
deleted m t D N A from different patients has revealed short direct repeats precisely flanking the deletion in the majority of cases [2,13-19]. It has been suggested that these repeats mediate the formation of deletions by slipped mispairing during m t D N A replication [15,17]. The breakpoint sequences of deleted m t D N A from four children are presented and some unusual features of these sequences are discussed. Materials and Methods
Patients. Cases 1-3 were children with Kearns-Sayre syndrome. Case 3 recovered from severe sideroblastic anemia in infancy, typical of Pearson's syndrome, and later developed Kearns-Sayre syndrome. These patients have previously been described in detail [8]. Case 4 was a child who had a multisystem mitochondrial disorder with short stature, anemia, hypoparathyroidism, ptosis, dementia, sensorineural hearing loss, muscle weakness and mitochondrial myopathy [3]. Analysis of mtDNA. D N A was prepared from muscle and Southern blot analysis of m t D N A was performed as described [8]. The deletions were m a p p e d by reprobing the same blots with different cloned m t D N A fragments [8]. The polymerase chain reaction (PCR) was used to amplify fragments corresponding to the breakpoint region of the deleted mtDNA. The PCR reac-
312 tions were performed as described previously [20]. PCR with primers corresponding to nucleotides (hi) 62016220 (NG5) of the heavy (H) strand of m t D N A and nt 14961-14980 (NG6) of the light (L) strand were used to amplify a 1117 bp fragment in case 3. This fragment was digested with the restriction endonucleases Barn H1 and HindlII, which cut this fragment at nt 14258 and 6203, respectively. The resulting 391 bp fragment corresponding to the breakpoint region was isolated, ligated to BamHI and HindllI digested p T Z 1 8 R plasmid (Pharmacia, Uppsala, Sweden) and cloned in HB 101 bacteria using standard methods. The plasmid insert was sequenced using the M13 reverse sequencing primer (Pharmacia) and the Sequenase version 2.0 sequencing kit (United States Biochemical, Cleveland, OH, USA). Three different plasmid clones were sequenced. The asymmetric primer method [21] was used in cases 1, 2 and 4 to generate single-stranded m t D N A for direct sequencing as described [20]. A primer corresponding to nt 10711-10730 of the H-strand (NG9, 4 pmol) and a primer corresponding to nt 16061-16080 of the L-strand (NG8, 50 pmol) were used in case 1. A primer corresponding to nt 8701-8720 of the H-strand of m t D N A (NG7, 4 pmol) and a primer corresponding to nt 14981-15000 of the L-strand ( N G I 0 , 50 pmol) were used for asymmetric PCR in case 2. A primer corresponding to nt 8161-8180 of the H-strand (NG25, 4 pmol) and a primer corresponding to nt 16015-16034 of the L-strand (NG17, 50 pmol) were used in case 4. Primers corresponding to nt 10735-10754 (NG16), 8731-8750 (NG18) and nt 8475-8494 (NG52) of the H-strand were used for sequencing the single-stranded D N A from cases 1, 2 and 4, respectively. Sequence analysis of mtDNA. Numbering of m t D N A nucleotides was done according to the Cambridge human m t D N A sequence [22]. The direct repeat with the lowest nucleotide numbers is referred to as the 5' repeat and the repeat with the highest nucleotide numbers is referred to as the 3' repeat. The MacVector sequence analysis program (IB'I, New Haven, CT, USA) was used to analyze the breakpoint sequences. Results Sequencing of the breakpoint regions of deleted m t D N A in case 1 - 4 showed that the size of the deletions were 4884, 6067, 7663 and 7150 bp, respectively. The different m t D N A deletions affected several protein and transfer R N A genes, Fig. 1. The promoter for transcription of the heavy (HSP) and light strand (LSP) of m t D N A and the origin of replication of the heavy (OH) and light strand (OL) were not affected. The breakpoint regions were in all cases situated within protein coding genes. The deletion interrupted the translational reading frame shortly after the breakpoint
I
I
I
0
1
2
OH
1 ]
I 4
T--T--~F-T ~
6
,'
~-T-l-r-y-~---~ 8
'~
!0
'I
:
'
'4
I~
I
HSP
L /\\\
LSP
OL
002
ATP8 ATP6 ND3
ND41
bj ND6
Case 1 Case 2 Case 3 Case 4
Fig. I. Map of mitochondrial D N A showing lhe location of the deletions in case 1-4, filled boxes. The following genes tire indicated by boxes: N A D H dehydrogenase subunits ( N D I - b and ND4L), cytochrome b (cyt b), cytochrome c oxidasc subunits 1-.3 (CO1 3). ATP synthase subunits 6 and 8 (ATP6 and 8,) and large (16 S) and small (12 S) ribosomal RNA. The transfer R N A genes arc not indicated. The arrows indicate the origin of heavy (OlD and light (OL) strand replication and the promoters for mmscription of the heavy' (HSP) and light (LSP) strand of mtDNA.
in case 1 and 3 and truncated the genes for N A D H - d e hydrogenase subunit 4 (ND4) and cytochrome c oxidase subunit 1 (CO1), respectively. The deletion in case 2 caused a translational in frame fusion of the 5' part of the A T P synthase subunit 6 (ATP6) gene and the 3' part of the cytochromc b (cyt h) gene. The deletion in case 4 caused an in frame fusion of the 5' part of the A T P synthase subunit 6 (ATP 6) gene and the 3' part of the cyt b gene. The sequences flanking either side of the 5' and 3' breakpoints of the different deletions were aligned, as shown in Fig. 2. Short directly repeated sequences precisely flanking the breakpoints were found in all cases. Three different repeats of 5, 9 and 5 bp were found in case 1. The breakpoint sequence of case 1 showed that the 5' breakpoint of the deletion was at the 9 bp repeat at nt 10952-10961 and the 3' breakpoint of the deletion was at the 9 bp repeat at nt 15837-15845. It could not be determined if the 5' or 3' 9 bp repeat was retained. In case 2 a single repeat of 10 bp at nt 8829-8838 and nt 14896-149(15 was found precisely flanking the deletion and it could not bc determined if the 5' or the 3' repeat was retained. A single large imperfect repeat of 16/17 bp was found in case 3. The 5' repeat at nt 6325-6341 contains a guanine (G) that is lacking in the 3' repeat at nt 13989-14004. Sequencing showed that the breakpoint sequence was identical to the 3' repeat. In case 4 an imperfect repeat of 11/14 bp and a perfect repeat of 8 bp were found flanking the deletion. The breakpoint sequence of case 4 showed that the 5' breakpoint of the deletion was at the 14 bp repeat at nt 8574-8587 and the 3' breakpoint was at the 11 bp repeat at nt
313 I
10940
I
10950
D .
I
.
I
I 15830
18820 .... 2 w.t.
C T C
TAGC TAG
18840 ATGC ATGC
-
6340 l
l
14000
18580
l 15730
TAAC T T A A C C T G TAACC TGAC l
13990
C
A A
I14910
A ACC A A CC TAGACC
13980
GG G C
T T
6330 l
T G T G C TACT
i
w.t. case 4 w.t.
T
I14900
6320 l
C TG
15850
18830
114890
w,t. case 3 w.t.
I 15840
TCT GGA
A A
10960
18590
1 lq
G C C G C A G T A G C A G A C G CAGAC
G C CT CCT
l 15740
18600 T ATT ATT
AT AA AA l 15750
Fig. 2. Sequence alignment of the wild type (w.t.) mitochondrial DNA (mtDNA) with deleted mtDNA from case 1-4. The first and third lines show the wild type sequence around the 5' and 3' break points and the second line show the breakpoint sequence of deleted mtDNA from case 1-4.
15727-15737. Sequencing showed that the breakpoint sequence was identical to the 3' repeat.
Discussion We have found multiple repeats flanking single m t D N A deletions in two children. In most previously published cases single perfect repeats precisely flanking the deletion have been reported [2,13-19]. In a minority of cases either no repeats or imprecisely located repeat elements have been found [17,19]. Most patients have unique deletions but in one third of cases an identical 4977 bp deletion flanked by a perfect 13 bp repeat is found [13]. The molecular mechanism involved in the formation of these spontaneously occurring single m t D N A deletions is unknown. The finding of perfect repeats precisely flanking the breakpoint in the majority of cases indicates that recombination or slipped mispairing may be of importance. Several different possible mechanisms have been extensively reviewed by Mita et al. [17]. Computer analysis of different breakpoint sequences has not detected any common sequence motifs [17] and this makes it less likely that a specific recombinase is involved. Shoffner et al. [15] suggested a slip-replication m t D N A deletion mechanism, where the 5' and 3' repeats base pair during mtDNA replication and the intervening sequence and one of the repeats is deleted. Usually it can not be determined if the 5' or 3' repeat is retained, since it is impossible to distinguish identical repeats, as demonstrated by case 2. The unusual find-
ing of imperfect repeats in case 3 and 4 made it possible to distinguish the repeats and showed that the 3' repeat was retained in both cases. The deletion in case 3 flanked by an imperfect 16/17 bp repeat has been described in two other patients [14,17] and may represent a second hotspot for mtDNA deletions besides the common 4977 bp deletion. The case described by Johns et al. [14] and case 3 of this study both had retained the 3' repeat, whereas the case described by Mita et al. [17] had retained the 5' repeat. These three cases demonstrate that when a m t D N A deletion is formed between repeats either the 5' or the 3' repeat can be retained. In case 1 the deletion was flanked by three different repeats of 5, 9 and 5 bp and sequencing showed that the first 5' repeat and the third 3' repeat was retained (Fig. 2). This case demonstrates that when multiple repeats are present repeats both 5' and 3' to the deletion can be retained. A comparison of the breakpoint sequences of case 1-4 with other published sequences showed that the 3' imperfect repeat in case 3 overlaps with a 9 bp repeat involved in the formation of a smaller deletion [18]. In case 1 the 5' repeats overlap with a 10 bp repeat involved in the formation of a smaller deletion [17] and the 3' repeats overlap with a 10 bp repeat involved in the formation of a larger deletion [19]. In summary, this study shows that both single and multiple repeats can be involved in the formation of mtDNA deletions and that both 5' and 3' repeated sequences can be retained. These findings are consistent with the slip replication model for the generation of single mtDNA deletions.
Acknowledgements This study was supported by grants from the Swedish Medical Research Council, project 585, and the Sven Jerring Foundation.
References 1 R6tig, A., Colonna M., Bonnefont, J.P., Blanche, S., Fischer, A., Saudubray, J.M. and Munnich A. (1989) Lancet i, 902-903. 2 R6tig, ,A., Cormier, V., Blanche, S., Bonnefont, J.P., Ledeist, F., Romero, N., Schmitz, J., Rustin, P., Fischer, A., Saudubray, J.M. and Munnich, A. (1990) J. Clin. Invest. 86, 1601-1608. 3 Holme, E., Fahleson, P., Kristiansson, B., Larsson, N.-G., Oldfors, A. and Tulinius., M. (1990) Proceedings of the 28th Annual Symposium of the Society for the Study of Inborn Errors of Metabolism (SSIEM), P-82 (Abstr.). 4 Majander, A., Suomalainen, A., Vettenranta, K., Sariola, H., Perkki6, M., Holmberg, C. and Pihko, H. (1991) Pediatr. Res. 30, 327-330. 5 Lestienne, P. and Ponsot, G. (1988) Lancet i, 885. 6 Zeviani, M., Moraes, C.T., DiMauro S., Nakase, H., Bonilla, E., Schon, E.A. and Rowland L.P. (1988) Neurology 38, 1339-1346. 7 Moraes, C.T., DiMauro, S., Zeviani, M., Lombes, A., Shanske, S., Miranda, A.F., Nakase, H., Bonilla, E., Werneck, L.C., Servidei,
314
8 9
10
11
12 13 14 15
S., Nonaka, I., Koga, Y., Spiro, A.J., Brownell, A.K.W., Schmidt, B., Schotland, D.L., Zupanc, M., DeVivo, D.C., Schon, E.A.. and Rowland, L.P. (19891 N. Engl. J. Med. 320, 1293-1299. Larsson, N.-G., Holme, E., Kristiansson, B., Oldfors, A., and Tulinius, M. (1990) Pediatr. Res. 28, 131-136. McShane, M.A., Hammans, S.R., Sweeney, M., Holt, l.J., Beattie, T.J., Brett, E.M. and Harding, A.E. (1991) Am. J. Hum. Genet. 48, 39-42. Shanske, S., Moraes, C.T., Lombes, A., Miranda, A.F., Bonilla, E, Lewis, P., Whelan, M.A., Ellsworth, C.A. and DiMauro. S. (1990) Neurology 40, 24-28. Rowland, L.P., Hausmanowa-Petrusewicz, I., Bardurska, B., Warburton, D., Nibroj-Dobosz, I., DiMauro, S., Pallai, M. and Johnson, W.G. (1988) Neurology 38, 1399-1402. Larsson, N.-G., Eiken, H.G., Boman, H., Holme, E., Oldfors, A. and Tulinius, M.H. (1992) Am. J. Hum. Genet. 50, 360-363. Schon, E.A., Rizzuto, R., Moraes, C.T., Nakase, H., Zeviani, M. and DiMauro, S. (1989) Science 244, 346-349. Johns, D.R., Rutledge, S.L., Stine, O.C. and Hurko, O. (1989) Proc. Natl. Acad. Sci. USA 86, 8059-8062. Shoffner, J.M., Lott, M.T., Voljavec, A.S., Soueidan, S.A., Costi-
16 17
18
19
20 21 22
gan D.A. and Wallace, I).U. (1t~89) Proc. Natl. Acad. Sci. USA 86, 7952-7956. Holt, I.J., Harding, A.E. and Morgan-Hughes, J.A. (1989) Nucl. Acids Res. 17, 4465-4469. Mita, S., Rizzuto, R., Moraes, C.T., Shanskc, S., Arnaudo, E,, Fabrizi G.M.. Koga, Y., DiMauro, S. and Schon E.A. (1990) Nucleic Acids Res. 18, 561-567. R6tig, A., Cormier, V., Koll, F., Mize, C.E., Saudubray, J.M., Veerman, A., Pearson. tt.A. and Munnich, A. (19911 Genomics 10, 502-504. Degoul, F., Nelson, l., Amselem, S. Romero, N., ObermaicrKusser, B. Ponsot, G., Marsac, C. and Lestienne, P. (1991) Nucl. Acids Res. 19, 493-496. Larsson, N.-G., Andersen, O., Holme, E., Oldfors, A. and Wahlstr6m, J. (1991) Ann. Neurol. 30, 701 708. Gyllensten, U.B. and Ehrlich, H.A. (19881 Proc. Natl. Acad. Sci. USA 85, 7652-7656. Anderson, S., Bankier, A.T., Barell, B.G., DeBruijn, M.H.L., (oulson, A.R., Drouin, J., Eperon, I.C., Nierlich, D.P.. Roe, B.A., Sanger, F., Schreier, P.H,, Smith, A.J.tt., Staden, R. and Young, I.G. (1981) Nature 290, 457-465.
Biochimica et Biophysica Acta, 1139 (1992) 311 - 314 © 1992 Elsevier Science Publishers B.V. All rights reserved 0925-4439/92/$05.00
BBADIS 61173
Multiple short direct repeats associated with single mtDNA deletions Nils-G6ran Larsson and Elisabeth Holme Department of Clinical Chemistry, Gothenburg University, Sahlgren's Hospital, Gothenburg (Sweden) (Received 26 February 1992)
Key words: mtDNA deletion; Kearns-Sayre syndrome; Pearson's syndrome; Mitochondrial encephalomyopathy
We have sequenced the breakpoints of deleted mtDNA in muscle from four children with mitochondrial myopathy and multisystem mitochondrial disorders. The deletions were 4884, 6067, 7663 and 7150 base pairs (bp) in size and affected several protein and transfer RNA genes. The sequences needed for transcription and replication of mtDNA were not affected in any case. The deletions were flanked by direct short repeats in all cases. Multiple repeats were found in case 1 and 4. Imperfect repeats were found in case 3 and 4 and this made it possible to distinguish the repeats 5' and 3' to the deletion. In both cases the 3' repeat was retained. The deletion of 7663 bp in case 3 has been reported in two other cases and may represent a second hotspot for mtDNA deletions in addition to the common deletion of 4977 bp found in one third of cases. A comparison of the breakpoint sequence of case 3 with the two other reported cases revealed that when a deletion is formed between the same repeats in different patients either the 5' or 3' repeat can be retained. This study shows that both single and multiple repeats can be associated with single mtDNA deletions and that both 5' and 3' repeated sequences can be retained. These findings are consistent with the slip-replication model for the generation of mtDNA deletions.
Introduction Single large deletions of m t D N A are associated with Pearson's syndrome [1,2] or other severe multisystem disorders of infants [3,4], and chronic progressive external ophthalmoplegia or Kearns-Sayre syndrome in children and adults [5-7]. Infants who survive the severe anemia of Pearson's syndrome may later in life develop neuromuscular and heart symptoms typical of KearnsSayre syndrome [8,9]. The patients always have heteroplasmy with a mixture of normal and deleted m t D N A in various proportions. The fraction of deleted m t D N A may vary widely between different tissues of the same individual but it is generally high in organs manifesting disease, e.g., muscle, brain or bone marrow [7,8,10]. The disorders with single m t D N A deletions are usually spontaneously occurring [11,12] and only one type of deletion is found in different tissues from the affected individual [7,8,10]. This strongly suggests that the m t D N A deletions occur de novo in the oocyte or at a very early stage of embryogenesis. The molecular mechanism involved in the formation of m t D N A deletions is unclear. Sequencing of the breakpoint region of
Correspondence to: Nils-G6ran Larsson, Department of Clinical Chemistry, Sahlgren's hospital, S-413 45 Gothenburg, Sweden.
deleted m t D N A from different patients has revealed short direct repeats precisely flanking the deletion in the majority of cases [2,13-19]. It has been suggested that these repeats mediate the formation of deletions by slipped mispairing during m t D N A replication [15,17]. The breakpoint sequences of deleted m t D N A from four children are presented and some unusual features of these sequences are discussed. Materials and Methods
Patients. Cases 1-3 were children with Kearns-Sayre syndrome. Case 3 recovered from severe sideroblastic anemia in infancy, typical of Pearson's syndrome, and later developed Kearns-Sayre syndrome. These patients have previously been described in detail [8]. Case 4 was a child who had a multisystem mitochondrial disorder with short stature, anemia, hypoparathyroidism, ptosis, dementia, sensorineural hearing loss, muscle weakness and mitochondrial myopathy [3]. Analysis of mtDNA. D N A was prepared from muscle and Southern blot analysis of m t D N A was performed as described [8]. The deletions were m a p p e d by reprobing the same blots with different cloned m t D N A fragments [8]. The polymerase chain reaction (PCR) was used to amplify fragments corresponding to the breakpoint region of the deleted mtDNA. The PCR reac-
312 tions were performed as described previously [20]. PCR with primers corresponding to nucleotides (hi) 62016220 (NG5) of the heavy (H) strand of m t D N A and nt 14961-14980 (NG6) of the light (L) strand were used to amplify a 1117 bp fragment in case 3. This fragment was digested with the restriction endonucleases Barn H1 and HindlII, which cut this fragment at nt 14258 and 6203, respectively. The resulting 391 bp fragment corresponding to the breakpoint region was isolated, ligated to BamHI and HindllI digested p T Z 1 8 R plasmid (Pharmacia, Uppsala, Sweden) and cloned in HB 101 bacteria using standard methods. The plasmid insert was sequenced using the M13 reverse sequencing primer (Pharmacia) and the Sequenase version 2.0 sequencing kit (United States Biochemical, Cleveland, OH, USA). Three different plasmid clones were sequenced. The asymmetric primer method [21] was used in cases 1, 2 and 4 to generate single-stranded m t D N A for direct sequencing as described [20]. A primer corresponding to nt 10711-10730 of the H-strand (NG9, 4 pmol) and a primer corresponding to nt 16061-16080 of the L-strand (NG8, 50 pmol) were used in case 1. A primer corresponding to nt 8701-8720 of the H-strand of m t D N A (NG7, 4 pmol) and a primer corresponding to nt 14981-15000 of the L-strand ( N G I 0 , 50 pmol) were used for asymmetric PCR in case 2. A primer corresponding to nt 8161-8180 of the H-strand (NG25, 4 pmol) and a primer corresponding to nt 16015-16034 of the L-strand (NG17, 50 pmol) were used in case 4. Primers corresponding to nt 10735-10754 (NG16), 8731-8750 (NG18) and nt 8475-8494 (NG52) of the H-strand were used for sequencing the single-stranded D N A from cases 1, 2 and 4, respectively. Sequence analysis of mtDNA. Numbering of m t D N A nucleotides was done according to the Cambridge human m t D N A sequence [22]. The direct repeat with the lowest nucleotide numbers is referred to as the 5' repeat and the repeat with the highest nucleotide numbers is referred to as the 3' repeat. The MacVector sequence analysis program (IB'I, New Haven, CT, USA) was used to analyze the breakpoint sequences. Results Sequencing of the breakpoint regions of deleted m t D N A in case 1 - 4 showed that the size of the deletions were 4884, 6067, 7663 and 7150 bp, respectively. The different m t D N A deletions affected several protein and transfer R N A genes, Fig. 1. The promoter for transcription of the heavy (HSP) and light strand (LSP) of m t D N A and the origin of replication of the heavy (OH) and light strand (OL) were not affected. The breakpoint regions were in all cases situated within protein coding genes. The deletion interrupted the translational reading frame shortly after the breakpoint
I
I
I
0
1
2
OH
1 ]
I 4
T--T--~F-T ~
6
,'
~-T-l-r-y-~---~ 8
'~
!0
'I
:
'
'4
I~
I
HSP
L /\\\
LSP
OL
002
ATP8 ATP6 ND3
ND41
bj ND6
Case 1 Case 2 Case 3 Case 4
Fig. I. Map of mitochondrial D N A showing lhe location of the deletions in case 1-4, filled boxes. The following genes tire indicated by boxes: N A D H dehydrogenase subunits ( N D I - b and ND4L), cytochrome b (cyt b), cytochrome c oxidasc subunits 1-.3 (CO1 3). ATP synthase subunits 6 and 8 (ATP6 and 8,) and large (16 S) and small (12 S) ribosomal RNA. The transfer R N A genes arc not indicated. The arrows indicate the origin of heavy (OlD and light (OL) strand replication and the promoters for mmscription of the heavy' (HSP) and light (LSP) strand of mtDNA.
in case 1 and 3 and truncated the genes for N A D H - d e hydrogenase subunit 4 (ND4) and cytochrome c oxidase subunit 1 (CO1), respectively. The deletion in case 2 caused a translational in frame fusion of the 5' part of the A T P synthase subunit 6 (ATP6) gene and the 3' part of the cytochromc b (cyt h) gene. The deletion in case 4 caused an in frame fusion of the 5' part of the A T P synthase subunit 6 (ATP 6) gene and the 3' part of the cyt b gene. The sequences flanking either side of the 5' and 3' breakpoints of the different deletions were aligned, as shown in Fig. 2. Short directly repeated sequences precisely flanking the breakpoints were found in all cases. Three different repeats of 5, 9 and 5 bp were found in case 1. The breakpoint sequence of case 1 showed that the 5' breakpoint of the deletion was at the 9 bp repeat at nt 10952-10961 and the 3' breakpoint of the deletion was at the 9 bp repeat at nt 15837-15845. It could not be determined if the 5' or 3' 9 bp repeat was retained. In case 2 a single repeat of 10 bp at nt 8829-8838 and nt 14896-149(15 was found precisely flanking the deletion and it could not bc determined if the 5' or the 3' repeat was retained. A single large imperfect repeat of 16/17 bp was found in case 3. The 5' repeat at nt 6325-6341 contains a guanine (G) that is lacking in the 3' repeat at nt 13989-14004. Sequencing showed that the breakpoint sequence was identical to the 3' repeat. In case 4 an imperfect repeat of 11/14 bp and a perfect repeat of 8 bp were found flanking the deletion. The breakpoint sequence of case 4 showed that the 5' breakpoint of the deletion was at the 14 bp repeat at nt 8574-8587 and the 3' breakpoint was at the 11 bp repeat at nt
313 I
10940
I
10950
D .
I
.
I
I 15830
18820 .... 2 w.t.
C T C
TAGC TAG
18840 ATGC ATGC
-
6340 l
l
14000
18580
l 15730
TAAC T T A A C C T G TAACC TGAC l
13990
C
A A
I14910
A ACC A A CC TAGACC
13980
GG G C
T T
6330 l
T G T G C TACT
i
w.t. case 4 w.t.
T
I14900
6320 l
C TG
15850
18830
114890
w,t. case 3 w.t.
I 15840
TCT GGA
A A
10960
18590
1 lq
G C C G C A G T A G C A G A C G CAGAC
G C CT CCT
l 15740
18600 T ATT ATT
AT AA AA l 15750
Fig. 2. Sequence alignment of the wild type (w.t.) mitochondrial DNA (mtDNA) with deleted mtDNA from case 1-4. The first and third lines show the wild type sequence around the 5' and 3' break points and the second line show the breakpoint sequence of deleted mtDNA from case 1-4.
15727-15737. Sequencing showed that the breakpoint sequence was identical to the 3' repeat.
Discussion We have found multiple repeats flanking single m t D N A deletions in two children. In most previously published cases single perfect repeats precisely flanking the deletion have been reported [2,13-19]. In a minority of cases either no repeats or imprecisely located repeat elements have been found [17,19]. Most patients have unique deletions but in one third of cases an identical 4977 bp deletion flanked by a perfect 13 bp repeat is found [13]. The molecular mechanism involved in the formation of these spontaneously occurring single m t D N A deletions is unknown. The finding of perfect repeats precisely flanking the breakpoint in the majority of cases indicates that recombination or slipped mispairing may be of importance. Several different possible mechanisms have been extensively reviewed by Mita et al. [17]. Computer analysis of different breakpoint sequences has not detected any common sequence motifs [17] and this makes it less likely that a specific recombinase is involved. Shoffner et al. [15] suggested a slip-replication m t D N A deletion mechanism, where the 5' and 3' repeats base pair during mtDNA replication and the intervening sequence and one of the repeats is deleted. Usually it can not be determined if the 5' or 3' repeat is retained, since it is impossible to distinguish identical repeats, as demonstrated by case 2. The unusual find-
ing of imperfect repeats in case 3 and 4 made it possible to distinguish the repeats and showed that the 3' repeat was retained in both cases. The deletion in case 3 flanked by an imperfect 16/17 bp repeat has been described in two other patients [14,17] and may represent a second hotspot for mtDNA deletions besides the common 4977 bp deletion. The case described by Johns et al. [14] and case 3 of this study both had retained the 3' repeat, whereas the case described by Mita et al. [17] had retained the 5' repeat. These three cases demonstrate that when a m t D N A deletion is formed between repeats either the 5' or the 3' repeat can be retained. In case 1 the deletion was flanked by three different repeats of 5, 9 and 5 bp and sequencing showed that the first 5' repeat and the third 3' repeat was retained (Fig. 2). This case demonstrates that when multiple repeats are present repeats both 5' and 3' to the deletion can be retained. A comparison of the breakpoint sequences of case 1-4 with other published sequences showed that the 3' imperfect repeat in case 3 overlaps with a 9 bp repeat involved in the formation of a smaller deletion [18]. In case 1 the 5' repeats overlap with a 10 bp repeat involved in the formation of a smaller deletion [17] and the 3' repeats overlap with a 10 bp repeat involved in the formation of a larger deletion [19]. In summary, this study shows that both single and multiple repeats can be involved in the formation of mtDNA deletions and that both 5' and 3' repeated sequences can be retained. These findings are consistent with the slip replication model for the generation of single mtDNA deletions.
Acknowledgements This study was supported by grants from the Swedish Medical Research Council, project 585, and the Sven Jerring Foundation.
References 1 R6tig, A., Colonna M., Bonnefont, J.P., Blanche, S., Fischer, A., Saudubray, J.M. and Munnich A. (1989) Lancet i, 902-903. 2 R6tig, ,A., Cormier, V., Blanche, S., Bonnefont, J.P., Ledeist, F., Romero, N., Schmitz, J., Rustin, P., Fischer, A., Saudubray, J.M. and Munnich, A. (1990) J. Clin. Invest. 86, 1601-1608. 3 Holme, E., Fahleson, P., Kristiansson, B., Larsson, N.-G., Oldfors, A. and Tulinius., M. (1990) Proceedings of the 28th Annual Symposium of the Society for the Study of Inborn Errors of Metabolism (SSIEM), P-82 (Abstr.). 4 Majander, A., Suomalainen, A., Vettenranta, K., Sariola, H., Perkki6, M., Holmberg, C. and Pihko, H. (1991) Pediatr. Res. 30, 327-330. 5 Lestienne, P. and Ponsot, G. (1988) Lancet i, 885. 6 Zeviani, M., Moraes, C.T., DiMauro S., Nakase, H., Bonilla, E., Schon, E.A. and Rowland L.P. (1988) Neurology 38, 1339-1346. 7 Moraes, C.T., DiMauro, S., Zeviani, M., Lombes, A., Shanske, S., Miranda, A.F., Nakase, H., Bonilla, E., Werneck, L.C., Servidei,
314
8 9
10
11
12 13 14 15
S., Nonaka, I., Koga, Y., Spiro, A.J., Brownell, A.K.W., Schmidt, B., Schotland, D.L., Zupanc, M., DeVivo, D.C., Schon, E.A.. and Rowland, L.P. (19891 N. Engl. J. Med. 320, 1293-1299. Larsson, N.-G., Holme, E., Kristiansson, B., Oldfors, A., and Tulinius, M. (1990) Pediatr. Res. 28, 131-136. McShane, M.A., Hammans, S.R., Sweeney, M., Holt, l.J., Beattie, T.J., Brett, E.M. and Harding, A.E. (1991) Am. J. Hum. Genet. 48, 39-42. Shanske, S., Moraes, C.T., Lombes, A., Miranda, A.F., Bonilla, E, Lewis, P., Whelan, M.A., Ellsworth, C.A. and DiMauro. S. (1990) Neurology 40, 24-28. Rowland, L.P., Hausmanowa-Petrusewicz, I., Bardurska, B., Warburton, D., Nibroj-Dobosz, I., DiMauro, S., Pallai, M. and Johnson, W.G. (1988) Neurology 38, 1399-1402. Larsson, N.-G., Eiken, H.G., Boman, H., Holme, E., Oldfors, A. and Tulinius, M.H. (1992) Am. J. Hum. Genet. 50, 360-363. Schon, E.A., Rizzuto, R., Moraes, C.T., Nakase, H., Zeviani, M. and DiMauro, S. (1989) Science 244, 346-349. Johns, D.R., Rutledge, S.L., Stine, O.C. and Hurko, O. (1989) Proc. Natl. Acad. Sci. USA 86, 8059-8062. Shoffner, J.M., Lott, M.T., Voljavec, A.S., Soueidan, S.A., Costi-
16 17
18
19
20 21 22
gan D.A. and Wallace, I).U. (1t~89) Proc. Natl. Acad. Sci. USA 86, 7952-7956. Holt, I.J., Harding, A.E. and Morgan-Hughes, J.A. (1989) Nucl. Acids Res. 17, 4465-4469. Mita, S., Rizzuto, R., Moraes, C.T., Shanskc, S., Arnaudo, E,, Fabrizi G.M.. Koga, Y., DiMauro, S. and Schon E.A. (1990) Nucleic Acids Res. 18, 561-567. R6tig, A., Cormier, V., Koll, F., Mize, C.E., Saudubray, J.M., Veerman, A., Pearson. tt.A. and Munnich, A. (19911 Genomics 10, 502-504. Degoul, F., Nelson, l., Amselem, S. Romero, N., ObermaicrKusser, B. Ponsot, G., Marsac, C. and Lestienne, P. (1991) Nucl. Acids Res. 19, 493-496. Larsson, N.-G., Andersen, O., Holme, E., Oldfors, A. and Wahlstr6m, J. (1991) Ann. Neurol. 30, 701 708. Gyllensten, U.B. and Ehrlich, H.A. (19881 Proc. Natl. Acad. Sci. USA 85, 7652-7656. Anderson, S., Bankier, A.T., Barell, B.G., DeBruijn, M.H.L., (oulson, A.R., Drouin, J., Eperon, I.C., Nierlich, D.P.. Roe, B.A., Sanger, F., Schreier, P.H,, Smith, A.J.tt., Staden, R. and Young, I.G. (1981) Nature 290, 457-465.
