J Med Genet 1992; 29: 897-901

Quantitative Southern blot analysis in thedystrophin gene of Japanese patients withDuchenne or Becker muscular dystrophy:a high frequency of duplications

Yoshiyuki Hiraishi, Shingo Kato, Tadayuki Ishihara, Toshiya Takano

AbstractEighty-four unrelated patients withDuchenne or Becker muscular dystrophyin Japan were studied by quantitativeSouthern blot analysis with dystrophincDNA probes. We found partial deletionsand duplications in 47 (56%) and 12 (14%)cases respectively by HindIII digestion.The duplications were confirmed byBgAI digestion and densitometric scan-ning. The frequency of duplications inthis study is significantly higher thanthose previously reported. This may bebecause of the small sample number,the racial difference, or our quantitativemethods. Our results suggest that attemptsto detect duplications are important for aprecise diagnosis. Both deletions andduplications clustered at the two hotspots as reported previously. Six caseswere exceptions to the 'reading framehypothesis'. We detected three types ofHindIII RFLP. Based on the resultsof one duplication case, we propose arevised sequential order of exons in thecDNA1O region of the dystrophin gene.(J Med Genet 1992;29:897-901)

Duchenne muscular dystrophy (DMD) is afatal X linked recessive disorder of children

84 85 86 87 88 89 90 N

-lmm wup0 -Mm -m

4m dab -~4 ~ a

_ _ _ _ _ _

_low _-SNW _- 10-0-86(JF)

_ - 7-0

_ 3.93.7

_0m _ - 3 1

--- - -- -^- - - ^1 6- 1.2

Figure 1 Typical Southern blot analysis of the patients with DMD. The DNA ofthe patients (79 to 90) and a normal male (N) were digested with HindIII and thefilter was hybridised with the probe cDNA8. The sizes of the exon containingfragments in kb are shown on the right side of the autoradiograph. A junction fragmentis shown by (JF).

with an incidence of approximately 1 in 3500male births.' Becker muscular dystrophy(BMD) is a less frequent, milder form andallelic to DMD. The 14 kb full length cDNAof the gene responsible for DMD and BMDhas been cloned,2 which encodes a membraneassociated protein of 430 kd called dystro-phin.2- The dystrophin gene is about 2 3 mil-lion bp in size and consists of more than 75exons.7 Partial deletions and, less frequently,partial duplications have since been reportedin this gene. Although the frequencies of mostof the reported deletions were in a range from50% to 67%,712 those of duplications variedfrom 0% to 6-7%.7'13' This variation may bein part because of difficulties in detectingduplications accurately by Southern blotanalysis. We studied Japanese patients withDMD or BMD using quantitative Southernblot analysis and detected duplications in 12(14%) of 84 cases examined.

Materials and methodsSUBJECTSEighty-four unrelated Japanese male patientsin the National Higashi-Saitama Hospital wereexamined, 79 with DMD, three with BMD,and two outliers. The criteria for clinical dia-gnosis were that patients become wheelchairbound before the age of 15 for DMD and afterthis age for BMD. For the outliers the criteriawere becoming wheelchair bound before theage of 15 but surviving beyond the age of 30without any life support systems.

PROBESDystrophin cDNA probes (cDNA1-2a, 2b-3,4-5a, 5b-7, 8, 9-14)2 were obtained from theAmerican Type Culture Collection. The probecDNA9-14 was divided into five parts andsubcloned; cDNA9 (7875-9080 bp), cDNA10(9080-9713 bp), cDNAl 1 (9713-10 905 bp),cDNA12 (10 905-11 510 bp), and cDNA13-14(11 510-13973bp). The probe cDNA13-14was not used as it contained repetitive se-quences.

SOUTHERN BLOT ANALYSISDNA was extracted from peripheral bloodleucocytes of patients as previously de-scribed.'5 DNA concentrations were deter-mined with a spectrophotometer. DNA was

Department ofMicrobiology, KeioUniversity School ofMedicine, 35Shinanomachi,Shinjuku-ku, Tokyo160, Japan.Y HiraishiS KatoT Takano

NationalHigashi-SaitamaHospital, Kurohama,Hasuda, Saitama349-01, Japan.T Ishihara

Correspondence to Dr Kato.Received 11 April 1992.Accepted 27 May 1992.

79 80 81 82 83

897

on March 30, 2020 by guest. P

rotected by copyright.http://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.29.12.897 on 1 Decem

ber 1992. Dow

nloaded from

Hiraishi, Kato, Ishihara, Takano

digested to completion with restriction endo-nuclease HindIII and, for duplication cases,

BglII. Five micrograms of the digested DNAwere electrophoresed in 0O7% agarose gel. Theamount ofDNA in each lane was equalised bycomparing the intensity of the DNA smear

stained with ethidium bromide on test runs.

DNA in the gel was transferred to Hybond-N + membrane (Amersham) and hybridisedwith 32p labelled probes as previously de-scribed.'6 The probes were labelled by therandom hexanucleotide primed method'7 and50 ng of the probe was used for one hybridisa-tion. Post-hybridisation washes were carriedout at 650C in 1 x SSPE (0 15 mol/I sodiumchloride, 10 mmol/l sodium phosphate,pH 7 4, 1 mmol/l EDTA) containing 1% SDS.Membranes were exposed to x ray film at- 70°C with intensifying screens for one to twodays. The autoradiograph was scanned with a

transmission densitometer (model CS-9000,Shimadzu, Kyoto, Japan). The membraneswere rehybridised three to four times after the

1-2a 2b-3 4-5a

JF-

-

JFFU..

JF

IIII

_

a

__==_M_

_

I(J

JF)

JF

5b-7 8 9 | 10 11

Figure 2 Location and extent of deletions and duplications in the patients wit)or BMD on the map of the dystrophin cDNA. The sequential order of the exon

containing HindIII fragments is shown at the bottom; the order in the region ojcDNA10 was revised by this study (see text). Open, closed, and hatched boxesrepresent DMD, BMD, and outliers, respectively. Cases carrying a junction frare shown by JF on either the right or left side, depending on which end was

responsible for the generation of the junction fragment, or by (JF) when the sidresponsible was unable to be determined. The cases which are exceptions to the rframe hypothesis (see Discussion) are shown by arrows.

removal of probes by incubating in 0 4 mol/lNaOH at 42°C.

ResultsDELETION ANALYSISWe analysed HindIII digested DNA of thepatients with each of the cDNA probes exceptfor cDNA13-14. In 84 unrelated patients, wedetected 47 intragenic deletions (56%) includ-ing one in BMD and two in outliers. Fig 1

shows a typical example with the cDNA8probe. Certain hybridising bands were notdetected in cases 80, 86, 88, and 89 as com-

pared with the bands of a normal male. Case 79had lost all the exons involved in cDNA8. Incase 80, two bands of 10-0 and 1-2 kb were

deleted and a new band of 8-6 kb appeared.This fragment was thought to be a junctionfragment, generated by a deletion of one of therestriction enzyme sites flanking an exon con-

taining fragment. Eight junction fragments(17%) were found in the deletions. Fig 2summarises the location and the extent of thedeletions on the map of exon containing Hin-dIII fragments.7"' The deletions were clus-

NM' tered in two regions: minor and major hotspots in the regions of cDNAl-2a andcDNA8, respectively. These results were con-

sistent with those reported so far.27'8

DUPLICATION ANALYSIS

By quantitative Southern blot analysis, we

detected 12 intragenic duplications (14%), allof which were in patients with DMD. Fig 3shows typical results of cases carrying duplica-tions. When HindIII digested DNA was ana-

lysed, the bands of 10-5 and 7-5 kb withcDNAI-2a in case 90, those of 1 6 and 3-7 kbwith the probe cDNA8 in case 58, and those of6-6, 6-0, 3 5, and 2-8 kb with cDNA1O in case

70 were denser than the corresponding bandsof a normal male control. These denser bandswere contiguous on the cDNA map in eachcase (fig 2). To confirm these findings, we

digested the DNA of these cases with anotherrestriction endonuclease, BglII, and per-formed similar experiments. The partial cor-

respondence between the HindIII and BglIIexon containing fragments has been estab-lished.'8 The BglII bands corresponding to thedenser HindIII bands were also denser in all of12 cases in which duplications were suggestedby HindIII digestion. To determine the inten-sity of these bands, we scanned the autoradio-graphic films with a transmission densit-ometer. Typical densitometric scanning

noXot results and intensities of hybridisation bands

12are shown in fig 4 and the table, respectively.

12 The bands in the suspected duplication were

found to be twice as dense as the normal bands.h DMD Thus, we concluded that these denser bands

represented partial gene duplications. The re-sults are summarised on the map of exon

agment containing HindIII fragments in fig 2. The

le duplications were clustered at the same hot

reading spots as the deletions, but less frequently in themajor hot spot.

rations

JF.

Duplic

JJFN

Deletions

I:1a IJF)

1-

ml-- -

.- -

MO

C4Ul)"C.LL?7 9(p LI) LI)C? (Pr,:.,? mu)-C'4CN(? -LO LC) ID IN (D(P r.,7"qc"qcpq.: A---- - M'

898

=M

on March 30, 2020 by guest. P

rotected by copyright.http://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.29.12.897 on 1 Decem

ber 1992. Dow

nloaded from

Quantitative Southern blot analysis in the dystrophin gene ofJapanese patients

cDNA1-2aHindlil Bg/llN 90 90 N

cDNA8Hindlll BgIlIN 58 58 N

cDNA10HindlIl BgIlIN 70 70 N

A

1 *5.5

7.5- 4 *

3 2532~3-1

- -13 5

_* _-49

_Em mo - 3-7_so- _ --29

o.o - 4_ -

7*0 - - -

L u - _ -*_ - 9 1

66 - _ * _* - - 566-0 -- _-*_M* _ - 4.1

379= _* mO-3-5321 - rn4M

-v-* - - 2-6

3 5 - _2.8 -_255-2 4

_*

3* -* - - 231

16- _ _ *-2 -

_ - 35W -n- 1 25

Figure 3 Typical Southern blot analysis of the patients carrying a duplication. The DNA of the patients (58, 70,and 90) and a normal male (N) were digested with HindIII or BglII. The cDNA probes used for hybridisationwere cDNA1-2a for case 90, cDNA8 for case 58, and cDNAOI for case 70. The sizes of the exon containingfragments are shown at the side of the autoradiographs. The duplicated bands are shown by asterisks on the right.

Case 70-

A1

t --

.

Normal' - : ./E,1 \

1i-2+ -.o\Xo.. ..

J.". --p

. .. . .. .

4i ......,, . x

1,, 1.

. - 6.. ..

.. ,. ., ...... ... ;.

.:4 1

.. 1

.A1A.---. -. -1 -- .,.. A1 . . . . . I . .1 1- 4

-§ I, X 7---1 \ 5 t- t.A--:---.--- . -:-----\ |e' 4 A I* w z 4

1.t=A TL.. : 4 : : : . : : : . : : : 4. :_...... 4

Figure 4 Densitometric scanning results for case 70 and a normal male. Theautoradiographic film in fig 3 was used. The bands are numbered in order of size.

In two of the 12 cases carrying duplications,we detected HindIII bands of altered sizes.These bands were thought to be junction frag-ments accompanying duplications, since theprobes used contained both the normal and theduplicated fragments and since the BglII

bands which corresponded to the duplicatedHindIII bands and one of the flanking singletHindIII bands were duplicated in both cases.The latter results gave us information onwhich end of the duplication regions was re-sponsible for generation of the junction frag-ments (fig 2).

SEQUENTIAL ORDER OF HindIII FRAGMENTSIN cDNA1OIn case 70, the duplication extended into theregion of cDNA10. The 6-6, 6-0, 3 5, and2-8 kb HindIII bands were obviously dupli-cated, while the other bands of 12-0, 2-55, and12-0 kb had normal intensities, as shown in fig3. Den Dunnen et al' reported that the 5' and3' order of the HindIII fragments in cDNA10were 6-0, 3 5, 2-4, 12-0, 6-6, 2-8, and 2-55 kb.According to the order of HindIII fragmentsthey proposed, case 70 must carry two inde-pendent duplications within a small region,but this seems unlikely. The 6 6 and 2-8kbfragments should precede those of 2-4 and120Okb. We propose a revised order of Hin-dIII fragments in this region, as shown in fig 2.

HindlIl RFLPsIn Southern blot analysis of HindIII digestedDNA of the patients, we found three types ofRFLP in the dystrophin gene: 2-5, 2-7, and5.3 kb fragments in three, two, and onepatients with the probes cDNAI-2a, 11, and11, respectively. The major alleles correspond-ing to each one were 3-1, 2-55, and 6-8kb insize. One of the patients carried two minoralleles of 2-5 and 2-7 kb. The sizes of BglIIfragments corresponding to these RFLPs werenot changed. The cases containing these minoralleles had a deletion or a duplication of dif-ferent regions, suggesting that these fragmentsare not junction fragments. This is the secondreport of HindIII RFLPs in the dystrophingene. A HindIII RFLP with cDNA1-2a wasreported with alleles 8 3/7 5 kb.'9 However, weare not able to rule out the possibility that a

-s --P---P-

899

'4. 'IFAW - - 0q. * -N. 7-4

on March 30, 2020 by guest. P

rotected by copyright.http://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.29.12.897 on 1 Decem

ber 1992. Dow

nloaded from

Hiraishi, Kato, Ishihara, Takano

Intensity of hybridising bands in case 70 and a normal male in fig 4. The area undereach peak which was automatically calculated in arbitrary units with the densitometeris shown.

Band Size (kb) Normal Case 70 Case 70/normal

1 12-0 9570 7262 0-762 6.6 15 010 27 332 1-823 6-0 7159 16287 2-274 3-5 32 493 62 700 1-935 2-8 15 918 32 347 2-036 2-55 40244 36453 0-917 2 4 16 802 15 070 0 90

deletion in the intron generated the 5-3 kbHindIII fragment in cDNAl 1.

PHENOTYPE AND READING FRAMEWe studied the correlation between diseasephenotype and the changes in reading framecaused by deletions or duplications in our

cases. The cases with a junction fragment wereexcluded from this analysis since it was notpossible to identify the deleted or duplicatedexon.10 In DMD, the reading frame was dis-rupted in 43 cases, including 37 with a deletionand six with a duplication, but maintained inthe other three cases carrying a deletion (fig 2).In all of the three BMD cases and outliers, thereading frame was disrupted (fig 2). It was

noteworthy that one BMD case and one outliercarried a deletion of the same HindIII frag-ments as deleted in DMD cases (fig 2).

DiscussionWe performed quantitative Southern blotanalysis to detect deletions and duplications inthe dystrophin gene of 84 Japanese patientswith DMD or BMD. The frequency of dupli-cations in this study, 14%, was remarkablyhigher than those reported by others: 6-7%and 5°5% by Den Dunnen et al7 and Hu et al,.4respectively. This difference may be becauseof the small sample number in our study, thesensitive and quantitative methods we used forthe analysis, or the racial characteristics of theJapanese. Sugino et aP0 and Asano et aP'

reported similar studies on Japanese patients,but they did not set out to find duplications.Our results as well as others showed thatduplications comprise a significant fraction ofthe genetic lesions inDMD and BMD. There-fore, the detection of these in the dystrophingene should not be ignored in the moleculardiagnosis ofDMD and BMD. Recently, mul-tiplex PCR has been developed, which is ableto amplify several deletion prone exons simul-taneously.324 This technique will be very use-

ful for the first screening of DMD/BMDpatients. However, it is very difficult to dia-gnose patients with duplications and carriers,since the present technique of PCR cannotdetermine the quantity of the target DNAaccurately.We detected neither deletions nor duplica-

tions in 30% of the patients examined, whichis similar to other studies in which both dele-tions and duplications were analysed.7914

About 70% is probably the upper limit for thedetection of these genetic lesions by Southernblot analysis with cDNA probes. What are thegenetic causes in the remaining patients?Recently, point mutations causing prematuretranscriptional termination were reported inthe dystrophin gene of patients with DMD.2526Matsuo et aP7 detected a 52 bp deletion in exon19. Other possible genetic disorders, for ex-ample, mutations in the promoter region,introns, or unidentified exon sequences, aswell as gene inversion, may cause DMD andBMD. Alternatively, lesions in certain genesother than the dystrophin gene may affect thefunction or the stability of dystrophin result-ing in the onset ofDMD and BMD.Monaco et aF8 proposed the reading frame

hypothesis for the aetiology of DMD andBMD; the translational reading frame of thedystrophin gene is disrupted, resulting in theproduction of truncated and inactive dystro-phin in DMD, while the reading frame is notdisrupted and the function of dystrophin par-tially remains in BMD. We identified six caseswhich were exceptions to this hypothesis in the46 patients carrying a deletion or a duplicationbut no junction fragment, three cases ofDMDwhose reading frames were not disrupted andthree cases of BMD and outliers in whichframeshifts occurred. Similar exceptions havealso been reported by others.7'029 In addition,we found two cases of BMD and outlier thatcarried a deletion of the same exons as deletedin DMD. Forrest et aP and Baumbach et al8reported similar observations. These findingsshow that the disease phenotype is not depend-ent only on the deletion of specific exons.

This study was supported in part by a Re-search Grant (2A-3) for Nervous and MentalDisorders from the Ministry of Health andWelfare, Japan.

1 Emery AEH. Duchenne muscular dystrophy. In: MotulskyAG, Harper PS, Bobrow M, Scriver C, eds. Oxfordmonographs on medical genetics. No 15. Oxford: OxfordUniversity Press, 1987.

2 Koenig M, Hoffman EP, Bertelson CJ, Monaco AP, FeenerC, Kunkel LM. Complete cloning of the Duchennemuscular dystrophy (DMD) cDNA and preliminarygenomic organization of the DMD gene in normal andaffected individuals. Cell 1987;50:509-17.

3 Hoffman EP, Brown RH Jr, Kunkel LM. Dystrophin: theprotein product of the Duchenne muscular dystrophylocus. Cell 1987;51:919-28.

4 Sugita H, Arahata K, Ishiguro T, et al. Negative immuno-staining of Duchenne muscular dystrophy (DMD) andmdx muscle surface membrane with antibody againstsynthetic peptide fragment predicted from DMD cDNA.Proc Japan Acad 1988;64,Ser B:37-9.

5 Watkins SC, Hoffman EP, Slayter HS, Kunkel LM. Immu-noelectron microscopic localization of dystrophin in myo-fibres. Nature 1988;333:863-6.

6 Zubrzycka-Gaarn EE, Bulman DE, Karpati G, et al. TheDuchenne muscular dystrophy gene product is localizedin sarcolemma of human skeletal muscle. Nature1988;333:466-9.

7 Den Dunnen JT, Grootscholten PM, Bakker E, et al.Topography of the Duchenne muscular dystrophy(DMD) gene: FIGE and cDNA analysis of 194 casesreveals 115 deletions and 13 duplications. Am J HumGenet 1989;45:835-47.

8 Baumbach LL, Chamberlain JS, Ward PA, Farwell NJ,Caskey CT. Molecular and clinical correlations of dele-tions leading to Duchenne and Becker muscular dystro-phies. Neurology 1989;39:465-74.

9 Gillard EF, Chamberlain JS, Murphy EG, et al. Molecularand phenotypic analysis of patients with deletions withinthe deletion-rich region of the Duchenne muscular dys-trophy (DMD) gene. Am J Hum Genet 1989;45:507-20.

10 Koenig M, Beggs AH, Moyer M, et al. The molecular basisfor Duchenne versus Becker muscular dystrophy: correla-tion of severity with type of deletion. Am Y Hum Genet1989;45:498-506.

900

on March 30, 2020 by guest. P

rotected by copyright.http://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.29.12.897 on 1 Decem

ber 1992. Dow

nloaded from

Quantitative Southern blot analysis in the dystrophin gene ofJapanese patients

11 Cooke A, Lanyon WG, Wilcox DE, et al. Analysis ofScottish Duchenne and Becker muscular dystrophy fami-lies with dystrophin cDNA probes. J Med Genet 1990;27:292-7.

12 Upadhyaya M, Smith RA, Thomas NST, Norman AM,Harper PS. Intragenic deletions in 164 boys withDuchenne muscular dystrophy (DMD) studied with dys-trophin cDNA. Clin Genet 1990;37:456-62.

13 Hu X, Burghes AHM, Ray PN, Thompson MW, MurphyEG, Worton RG. Partial gene duplication in Duchenneand Becker muscular dystrophies. J Med Genet 1988;25:369-76.

14 Hu X, Ray PN, Murphy EG, Thompson MW, Worton RG.Duplicational mutation at the Duchenne muscular dys-trophy locus: its frequency, distribution, origin, and phe-notype genotype correlation. Am J Hum Genet 1990;46:682-95.

15 Herrmann BG, Frischauf AM. Isolation of genomic DNA.Methods Enzymol 1987;152:180-3.

16 Kato S, Tachibana K, Takayama N, Kataoka H, YoshidaMC, Takano T. Genetic recombination in a chromosomaltranslocation t(2;8)(pl l;q24) of a Burkitt's lymphoma cellline, KOBK1I1. Gene 1991;97:239-44.

17 Feinberg AP, Vogelstein B. A technique for radiolabelingDNA restriction endonuclease fragments to high specificactivity. Anal Biochem 1983;132:6-13.

18 Darras BT, Blattner P, Harper JF, Spiro AJ, Alter S,Francke U. Intragenic deletions in 21 Duchenne muscu-lar dystrophy (DMD)/Becker muscular dystrophy(BMD) families studied with the dystrophin cDNA:location of breakpoints on HindIII and BglII exon-containing fragment maps, meiotic and mitotic origin ofthe mutations. Am J Hum Genet 1988;43:620-9.

19 Davies KE, Mandel JL, Monaco AP, Nussbaum RL, Wil-lard HF. Report of the committee on the genetic constitu-tion of the X chromosome. Cytogenet Cell Genet 1991;58:853-966.

20 Sugino S, Fujishita S, Kamimura N, et al. Molecular-genetic study of Duchenne and Becker muscular dystro-phies: deletion analyses of 45 Japanese patients and segre-gation analyses in their families with RFLPs based on the

data from normal Japanese females. Am J Med Genet1989;34:555-61.

21 Asano J, Tomatsu S, Sukegawa K, et al. Gene deletions inJapanese patients with Duchenne and Becker musculardystrophies: deletion study and carrier detection. ClinGenet 1991;39:419-24.

22 Chamberlain JS, Gibbs RA, Ranier JE, Nguyen PN, Cas-key CT. Deletion screening of the Duchenne musculardystrophy locus via multiplex DNA amplification. Nuc-leic Acids Res 1988;16:11141-56.

23 Beggs AH, Koenig M, Boyce FM, Kunkel LM. Detectionof 98% of DMD/BMD gene deletions by polymerasechain reaction. Hum Genet 1990;86:45-8.

24 Chamberlain JS, Gibbs RA, Ranier JE, Caskey CT. Mul-tiplex PCR for the diagnosis of Duchenne musculardystrophy. In: Innis MA, Gelfand DH, Sninsky JJ,White TJ, eds. PCR protocols: a guide to methods andapplications. San Diego: Academic Press, 1990:272-81.

25 Bulman DE, Gangopadhyay SB, Bebchuck KG, WortonRG, Ray PN. Point mutation in the human dystrophingene: identification through Western blot analysis. Geno-mics 1991;10:457-60.

26 Roberts RG, Bobrow M, Bentley DR. Point mutations inthe dystrophin gene. Proc Natl Acad Sci USA 1992;89:2331-5.

27 Matsuo M, Masumura T, Nakajima T, et al. A very smallframe-shifting deletion within exon 19 of the Duchennemuscular dystrophy gene. Biochem Biophys Res Commun1990;170:963-7.

28 Monaco AP, Bertelson CJ, Liechti-Gallati S, Moser H,Kunkel LM. An explanation for the phenotypic dif-ferences between patients bearing partial deletions of theDMD locus. Genomics 1988;2:90-5.

29 Malhotra SB, Hart KA, Klamut HJ, et al. Frame-shiftdeletions in patients with Duchenne and Becker musculardystrophy. Science 1988;242:755-9.

30 Forrest SM, Cross GS, Speer A, Gardner-Medwin D,Burn J, Davies KE. Preferential deletion of exons inDuchenne and Becker muscular dystrophies. Nature1987;329:638-40.

901

on March 30, 2020 by guest. P

rotected by copyright.http://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.29.12.897 on 1 Decem

ber 1992. Dow

nloaded from