Volume 16 Number 23 1988
`
`Nucleic Acids Research
`
`Deletion screening of the Duchenne muscular dystrophy locus via multiplex DNA amplification
`
`Jeffrey S.Chamberlainl*, Richard A.Gibbs1, Joel E.Ranierl, Phi Nga Nguyen1'2 and C.Thomas
`Caskeyl2
`
`lInstitute for Molecular Genetics and 2Howard Hughes Medical Institute, Baylor College of Medicine,
`Houston, TX 77030, USA
`
`Received August 24, 1988; Revised and Accepted October 28, 1988
`
`Accession nos X13045-X13048 (incl.)
`
`ABSTRACT
`The application of recombinant DNA technology to prenatal diagnosis of
`many recessively inherited X-linked diseases is complicated by a high frequen-
`cy of heterogenous, new mutations (1).
`Partial gene deletions account for
`more than 50% of Duchenne muscular dystrophy (DMD) lesions, and approximately
`one-third of all cases result from a new mutation (2-5).
`We report the
`isolation and DNA sequence of several deletion prone exons from the human DMD
`We also describe a rapid method capable of detecting the majority of
`gene.
`deletions in the DMD gene.
`This procedure utilizes simultaneous genomic DNA
`amplification of multiple widely separated sequences and should permit dele-
`tion scanning at any hemizygous locus.
`We demonstrate the application of this
`multiplex reaction for prenatal and postnatal diagnosis of DMD.
`
`INTRODUCTION
`DMD is among the most common human genetic diseases, occurring approxi-
`mately once in every 3,500 male births (6).
`Although there is no cure or
`effective therapy, highly accurate prenatal diagnosis and carrier detection is
`now possible via Southern analysis using DMD cDNA and genomic clones (2,5,7-9;
`P.Ward et al., manuscript in preparation).
`Unfortunately, there are several
`major limitations that prevent widespread and routine use of Southern analysis
`Exons of this gene are located on at least 65 genomic Hind
`for DMD diagnosis.
`III restriction fragments (2,10), which necessitates hybridization of Southern
`blots with 7-9 separate cDNA subclones to resolve each exon-containing re-
`striction fragment for diagnosis of genomic alterations.
`Linkage analysis for
`DMD patients that do not display genomic alterations requires additional
`Southern blots using genomic clones as relatively few restriction fragment
`length polymorphisms (RFLPs) have been described that are detectable with the
`Furthermore, Southern analysis is an expensive, tedious, and time
`cDNAs.
`consuming technique that requires the use of radioisotopes making it unsuit-
`able for routine use in clinical laboratories.
`An alternative to Southern analysis for mutation diagnosis involves the
`polymerase chain reaction [PCR; (11)].
`In this method specific regions of a
`
`© I R L Press Limited, Oxford, England.
`
`11141
`
`Ariosa Exhibit 1026, pg. 1
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`gene can be amplified up to a million-fold from nanogram quantities of genomic
`DNA and then analyzed for the presence of mutant alleles either by direct DNA
`sequencing or by hybridization with allele-specific oligonucleotide (ASO)
`These techniques have proven useful in the diagnosis of several
`probes.
`diseases including B-thalassemia (12), hemophilia A (13), sickle-cell anemia
`(14), and phenylketonuria (15).
`The application of PCRs to DMD diagnosis has been limited since point
`However, partial gene
`mutations leading to DMD have not been identified.
`deletions occur in 50-60% of DMD patients and account for virtually all of the
`mutations that have been directly observed.
`These deletions typically span
`several hundred kilobases of the DMD locus and generally overlap 2 specific
`regions located approximately 0.5 and 1.2 megabases from the promoter (2,6,8).
`We have exploited these observations to develop a rapid and simple technique
`that is able to detect approximately 70% of all DMD gene deletions.
`In this
`method sequences flanking six deletion prone exons of the gene are simultan-
`Any of these regions absent from patient DNA fail
`eously amplified via PCR.
`to amplify and are readily identified via agarose gel electrophoresis of the
`Multiplex amplification reactions should be broadly
`reaction products.
`We demonstrate here the use
`applicable to the study of a number of problems.
`of this technique for prenatal and postnatal diagnosis of DMD and discuss the
`advantages and logistics of integrating these methods into diagnostic labora-
`tories.
`
`MATERIALS AND METHODS
`Isolation of genomic clones
`Genomic clones were isolated from a human genomic library prepared in
`EMBL 3 (16; kindly provided by John Weis) and from a 49XXXXY library prepared
`The murine Dmd cDNA clone
`in A DASH (Stratagene; T. Webster, unpublished).
`XD-1 has been described previously (17).
`Human cDNA probes were kindly
`Clones were isolated and characterized
`provided by Dr. Louis M. Kunkel (2).
`Exons were sequenced by the dideoxy
`essentially as described previously (18).
`chain termination method (19) following subcloning into Ml3mpl8 and 19 (20) or
`pTZ18 and 19R (Pharmacia) using vector and insert specific synthetic oligonu-
`cleotide primers.
`Multiplex genomic DNA amplification
`DNA was prepared from lymphoblasts, cultured amniotic fluid cells, or
`chorionic villus specimens using an Applied Biosystems model 340A DNA ex-
`Synthetic oligonucleotide primers were prepared by phosphoramidite
`tractor.
`
`11142
`
`Ariosa Exhibit 1026, pg. 2
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`chemistry (21) on an Applied Biosystems model 380B DNA synthesizer.
`Primers
`were purified by electrophoresis on 20% polyacylamide gels, followed by elec-
`troelution and desalting on a NENsorb column (Dupont).
`We have not attempted
`multiplex amplification with unpurified primers although we have recently ob-
`served that purification is not always necessary when amplifying a single se-
`Reaction conditions were modified from Kogan et al (13) to permit
`quence.
`Primers (1pM each final concentration)
`multiplex amplification as follows.
`were annealed at 550 for 45 sec, 210 sec extension times and 10 units Taq
`polymerase (Cetus) were used, 5OOng DNA was added as template, and amplifi-
`cation was performed for 25 cycles (except as indicated) on a Perkin-Elmer/
`Cetus automatic thermocycler using the 'step-cycle' function.
`On the final
`cycle extensions were performed for ten min.
`Reaction products can be stored
`at 40 for up to two months prior to analysis on agarose gels.
`Optimal resolu-
`tion was obtained by electrophoresing 15pl of the 1004p reactions through 1.4%
`agarose gels at 3.7V/cm for 100 min.
`Reaction mixes containing everything ex-
`cept template and enzyme appear quite stable at -70°, although we have not at-
`tempted storage for longer than one month.
`Manual amplifications using gly-
`cerin filled heat blocks required slightly different conditions.
`Those reac-
`tions were annealed at 47° for 30 sec., and extended for 180 sec.
`
`RESULTS
`Isolation of DMD genomic clones
`Development of DNA amplification based diagnostic methods required the
`isolation of genomic clones from deletion prone regions of the DMD gene.
`Since this gene is extremely large [> 2Mb (2, 22-24)] and contains at least 60
`exons, we sought to direct our library screening by using cDNA subclones known
`to hybridize with exons frequently deleted in DMD patients.
`Three cDNA probes
`were used to screen two human genomic libraries, and four of the isolated
`clones were used in the present study.
`Exon 17 of the DMD gene was isolated
`using the murine DMD cDNA XD-1 (17), which hybridizes with the DXS 164 (pERT
`87) locus (2).
`Two additional exons were isolated with the human DMD cDNA
`probe 7 (2).
`As the complete exon/intron structure of the DMD gene is not yet
`known, it is not clear how many additional exons lie
`5' of these two.
`These
`exons are within the 4.1 kb and 0.5 kb HindIII fragments detectable with cDNA
`probe 7 (2).
`Comparison of these exon sequences with the human DMD cDNA
`sequence (25) reveals that the 4.1 kb fragment lies 5' of the 0.5 kb HindIII
`fragment, and that these are adjacent exons on the reported DMD cDNA.
`The
`fourth region cloned for these studies was isolated with human DMD cDNA probe
`
`11143
`
`Ariosa Exhibit 1026, pg. 3
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`5'tgactttcgatgttgagattactttcccttgctatttcagtgaaccaaacttaagtca
`gataaaacaattttatttggcttcaatatggtgctattttgatctgaaggtcaatctacc
`aacaagcaagaacagtttctcattattttcctttgccactccaagcagtctttactgaag
`A tctttcgagcaatgtctgacctctgtttcaatacttctcacagATTTCACAGGCTGTCAC
`CACCACTCAGCCATCACTAACACAGACAACTGTAATGGAAACAGTAACTACGGTGACCAC
`AAGGGAACAGATTACTGTGGATTCTGCTCAAGAGGAACTTCCACCACCACCTCCCCAAAA
`GAAGAGGCAGATTACTGTGGATTCTGAAATTA_GAAAAGgtgagagcatctcaagctt3'
`
`5'tgtccaaaatagttgactttctttctttaatcaataaatatattactttaaagggaaa
`aattgcaaccttccatttaaaatcagctttatattgagtatttttttaaaatgttgtgtg
`tacatcgtaggtgtgtatattaatttttatttgttacttgaaactaaactctgcaaatgc
`B aggaaactatcagagtgatatctttgtcagtataaccaaaaaatatacgctatltctcta
`taatctgttttacataatccatctatttttcttgatccatatgcttttacctg ragGCGA
`TTTGACAGATCTGTTGAGAAATGGCGGCGTTTTCATTATGATATAAAGATATTTAATCAG
`TGGCTAACAGAAGCTGAACAGTTTCTCAGAAAGACACAAATTCCTGAGAATTGGGAACAT
`GCTAAATACAAATGGTATCTTAAGgtaagtctttgatttgttttttcgaaattgtattta
`tcttcagcacatctggactctttaacttcttaaa-atcagqttctqaaqgqtqatggaaa
`ttacttttgactgttgttgtcatcattatattacdtagaaagaaaa3'
`
`5'tacaacatttcatagactattaaacatggaacatccttgtggggacaagaaatcgaat
`ttgctcttgaaaaggtttccaactaattgatttgtaggacattataacatcctctagctg
`acaagcttacaaaaataaaaactggagctaaccgagagggtgcttttttccctgacacat
`aaaaggtgtctttctgtcttgtatcctttggatatgggcatgtcagtttcatagggaaat
`C tttcacatggagcttttgtatttctttctttgccagtacaactgcatgtggtagcacact
`gtttaatcttttctcaaataaaaagacatggggcttcatttttgttttgcctttttggta
`tcttacagGAACTCCAGGATGGCATTGGGCAGCGGCAAACTGTTGTCAGAACATTGAATG
`CAACTGGGGAAGAAATAATTCAGCAATCCTCAAAAACAGATGCCAGTATTCTACAGGAAA
`AATTGGGAAGCCTGAATCTGCGGTGGCAGGAGGTCTGCAAACAGCTGTCAGACAGAAAAA
`AGAG iagggcgacagatctaataggaatgaaaacattttagcagactttttaagctt3'
`
`5'ttttgtagacggttaatgaataattttgaatacattggttaaatcccaacargtaata
`tatgtaaataatcaatattatgctgctaaaataacacaaatcagtaagattctgtaatat
`ttcatgataaataacttttgaaaatatatttttaaacattttgcttatgccttgagaatt
`atttacctttttaaaatgtattttcctttcagGTTTCCAGAGCTTTACCTGAGAAACAAG
`GAGAAATTGAAGCTCAAATAAAAGACCTTGGGCAGCTTGAAAAAAAGCTTGAAGACCTTG
`AAGAGCAGTTAAATCATCTGCTGCTGTGGTTATCTCCTATTAGGAATCAGTTGGAAATTT
`ATAACCAACCAAACCAAGAAGGACCATTTGACGTTLAGgtaggggaactttttgctttaa
`tatttttgtcttttttaagaaaaatggcaatatcactgaattttctcatttggtatcatt
`taaqgaagactttattcaqcataacca
`attaaagacaaaatattacttgttaaaqat
`caataggcacagggaccactgcaatggaStattacaggaggttggatagagagagattgg
`gctcaactctaaatacagcacagtggaagtaggaatttatagc3'
`
`Nucleotide sequence of four DMD gene exons and flanking introns.
`Figure 1:
`Exon sequences are in upper case, and intron sequences are in lower case.
`- 3' orientation of primers used for
`Arrows indicate the location and 5'
`amplification reactions. Forward arrows are above the sequence to which they
`correspond, reverse arrows are below the sequence to which they are complemen-
`(A) exon 17; EMBL/Genbank accession # X13045.
`tary (Table 1).
`(B) 4.1 kb
`HindIII fragment exon detectable with cDNA probe 7 (2); EMBL/Genbank accession
`# 13046. (C) 0.5 kb HindIII fragment exon detectable with probe 7;
`EMBL/Genbank accession # X13048.
`(D) 1.2/3.8 kb HindIII fragments exon
`detectable with probe 8 (2); EMBL/Genbank accession f X13047.
`The nucleotide
`differing from the reported cDNA sequence (25) is underlined.
`All splice
`sites have been confirmed by sequencing the exons both 5' and 3' of the ones
`shown (JSC unpublished).
`
`11144
`
`Ariosa Exhibit 1026, pg. 4
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`Table 1.
`Exon and Sizel
`
`Summary of DMD gene multiplex amplificatioti pnmer sets.
`Amplified3
`
`Primer Sec uence2
`
`Deleted4
`
`A. Exon 8
`182bp; (probe lb)
`
`B. Exonl7
`178bp; (probe 3)
`
`C. Exon 19
`88bp; (probe 3)
`D. 4.1Kb HindlIl
`148bp; (probe 7)
`
`E 0.5Kb Hndlll
`176bp; (probe 7)
`F. 1213.8Kb Hindlil
`186bp; (probe 8)
`
`F- GTCCMACACACMACCTGTTGAG
`R- GGCCTCATTCTCATGTTCTAATTAG
`F- GACMCGATGTTGAGATTACMCCC
`R- AAGCTTGAGATGCTCTCACCTTTTCC
`F- TTCTACCACATCCCA1TMTCTTCCA
`R- GATGGCAAAAGTGTTGAGAAAAAGTC
`F- CTTGATCCATATGC1TMACCTGCA
`R- TCCATCACCCTTCAGAACCTGATCT
`F- AMCATGGAACATCCTTGTGGGGAC
`R- CATTCCTATTAGATCTGTCGCCCTAC
`F- TTGAATACATTGGTTAAATCCCAACATG
`R- CCTGMTAAAGTCTTCCTTACCACAC
`
`360 bp
`
`416 bp
`
`459 bp
`
`268 bp
`
`547 bp
`
`506 bp
`
`11.3%
`
`9.4%
`
`10.3%
`
`4.0/o
`
`8.4%
`
`18.2%
`
`Total5:
`
`37%/o
`
`1 Each exon is designated A,B,C,D,E, or F. When known the exon number is listed, when not known the
`size of the genomic Hind IlIl fragment that the exon is located on is listed. Also shown is the human DMD
`cDNA probe that hybridizes with each exon (2), as well as the size of the exon in base pairs (bp).
`2 Shown is the sequence in 5-3' orientation for the PCR primers used to amplify each region. F: forward
`primer, hybridizes 5' of the exon; R: reverse primer, hybridizes 3' of the exon.
`3 The size of the amplified fragment obtained with each primer set.
`4 The percentage of DMD patents analyzed that are deleted for each indicated exon. This data is derved
`from references 2 and 6.
`5 The percentage of DMD patients analyzed that are deleted for any of the exons A-F (2,6). This number is
`less than the sum of the individual exon deletion frequencies as many deletions encompass multiple exons.
`
`8 (2).
`Sequence analysis revealed the exon on these clones to overlap both
`the 1.2 kb and 3.8 kb Hind III restriction fragments detectable with probe 8.
`This exon sequence contained a single base mismatch from the reported human
`cDNA sequence (Fig ID, underlined) which results in an amino acid change from
`glutamine to lysine.
`It is not yet known whether this represents a cloning
`artifact or a polymorphism.
`Figure 1 shows the DNA sequences of these four
`exons and their flanking introns.
`Two previously published exon/intron sequences have also been incorpo-
`The first is exon 8 (26) and the second is exon 19
`rated into this study.
`(27).
`These exons are located on 7.5 kb (exon 8) and 3.0 kb (exon 19) HindIII
`restriction fragments detectable with cDNA probes lb, and 3, respectively (2).
`Analysis of data obtained from DNA deletions that were observed in a group of
`over 200 DMD patients indicates that one or more of the six exons referred to
`above is missing from 70X of the deletions studied (2,6).
`
`11145
`
`Ariosa Exhibit 1026, pg. 5
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`p 4
`
`XJi1.1
`
`p87
`
`DMD Locus
`JBir
`
`pio
`
`J66
`
`754
`
`5 l*Ss-ia<
`EXONS:
`
`af
`
`>U.N@;sASsg; ;-g--p
`df et
`bTtc
`
`Cen-
`
`g
`
`3-
`
`C7
`
`Tel
`
`approximate scale (Mb): 0
`
`0.5
`
`1.0
`
`1.5
`
`2.0
`
`The approximate size of
`Figure 2:
`Schematic representation of the DMD gene.
`the locus, the position of the amplified fragments, and the location of the
`genomic probe markers are inferred from references 2,22-24,26,27,40,41.
`(a)
`exon 8; (b) exon 17; (c) exon 19; (d) exon from Fig. 1B; (e) exon from Fig.
`(f) exon from Fig. 1D.
`1C;
`
`Development of multiplex genomic DNA amplification reactions
`25-28 base oligonucleotide primers were synthesized to correspond to se-
`quences flanking these six DMD exons.
`Primers were chosen to overlap or lie
`outside of exon splice site regions to ensure that a deletion that removed a
`splice site but left the exon intact would be detectable.
`This also provided
`greater flexibility in choosing the size of the region to be amplified, fa-
`cilitating resolution of each fragment by agarose gel electrophoresis follow-
`ing multiplex reactions.
`The location of the sequences chosen are shown in
`Figure 1 for the four exons cloned in this study.
`Table 1 shows the sequence
`of all 6 primer sets, the size of each exon and the amplified fragment ob-
`tained, a description of each exon, and the frequency that each has been
`observed to be deleted in patients studied.
`For the remainder of this manus-
`cript these exons will be referred to as A,B,C,D,E, and F (Table 1).
`Figure 2
`shows a schematic representation of the DMD locus illustrating the relative
`location of each of these six exons which span approximately 1 megabase of the
`gene.
`
`Exon 17, the initial isolate, was used to demonstrate that deletions
`could be detected via PCR and that the reactions could successfully detect
`deletions in amniotic fluid cells (28).
`However, PCR amplification of single
`exons would require a cumbersomely large number of reactions for analysis.
`We
`therefore combined multiple primer sets in a single reaction.
`Addition of
`each extra primer set frequently required modification of primer annealing
`temperatures, time of annealing, polymerase extension times, and the amount of
`The conditions that permitted multiplex amplification with
`enzyme required.
`all six primer sets are listed in materials and methods.
`
`11146
`
`Ariosa Exhibit 1026, pg. 6
`IPR2013-00276
`
`

`

`H:tL Lr LL
`
`Nucleic Acids Research
`
`-1),t.
`-cl-a(
`
`A. I
`
`jI..
`
`Pf TIE'N Tl
`
`(L) I L) #
`
`I
`L'
`
`Ct t
`LOL3
`"I
`
`L'.
`
`L')
`
`A
`
`B
`
`> - -
`N-
`
`N.
`
`:
`
`Figure 3.
`Multiplex DNA amplification of lymphoblast DNA from 20 unrelated
`male DMD patients.
`A and B show two sets of ten samples each.
`DRL # refers
`to the R.J. Kleberg Center for Human Genetics Diagnostic Research Laboratory
`MW: HaeIII-digested 0 X174 DNA. -
`family number.
`: no template DNA added to
`the reaction.
`To the right of Panel A is indicated which amplified fragment
`corresponds to each region indicated in Fig. 2.
`Each sample was amplified as
`described in materials and methods then stored at 4° until analysis.
`Shown is
`an ethidium bromide stained 1.4% agarose gel through which 15pl of each 10Opl
`reaction was electrophoresed.
`
`Deletion detection using DNA from male DMD patients
`DNA from a number of male patients containing deletions previously de-
`lineated using human cDNA probes were initially used as templates for multi-
`plex amplifications.
`The results demonstrated that deletions involving any or
`all of the six regions assayed were detectable via multiplex amplifications
`and that the relative location or size of the deletions did not affect the
`results (data not shown).
`Analysis of amplification products utilizing
`
`11147
`
`Ariosa Exhibit 1026, pg. 7
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`control and deletion containing DNAs indicated that under the conditions used
`only regions specifically targeted were amplified and that primer sets cor-
`responding to deleted regions never amplified other regions of the genome
`(Figures 3 and 4, and data not shown).
`To test the general utility of this method DNA from 20 unrelated male DMD
`patients that had not been characterized by Southern analysis were analyzed by
`Figure 3 shows the amplification products obtained
`multiplex amplification.
`Deletions were detected in 12/20 of the samples, thus
`with each DNA sample.
`Figure 3 also illustrates
`identifying the mutations in 60% of the patients.
`that a variety of deletions were detected, ranging from a single region to all
`Each of the DNAs used was subsequently examined by the R.J. Kleberg Center
`6.
`for Human Genetics-DNA Diagnostic Laboratory at Baylor College of Medicine
`Those results, which took
`using Southern analysis with human DMD cDNA probes.
`a further week to obtain, confirmed in all 20 cases the presence or absence of
`The cDNA probes detected
`each of the six regions analyzed (data not shown).
`small deletions in 2 of the 8 samples that did not reveal deletions via the
`amplification method, both of which were in regions outside of the six origin-
`Thus the multiplex method successfully detected 12/14 dele-
`ally analyzed.
`tions (86%) and less than one day was required to obtain the results.
`Prenatal diagnosis via multiplex amplification
`Successful diagnosis of deletions in the DMD gene of male patients led us
`to ask if the same techniques could be utilized for rapid prenatal diagnosis.
`During the course of this study, six families referred to the R.J. Kleberg Jr.
`In each case a female who
`Center for Human Genetics were chosen for analysis.
`had previously given birth to an affected male requested prenatal diagnosis of
`DNA was prepared from cultured amniotic fluid cells and used as
`a fetus.
`template for multiplex amplification in parallel with lymphoblast DNA isolated
`Identical deletions were
`from the affected male in each family (Figure 4).
`detected with the fetal and affected male DNAs in 2 of the six families, diag-
`In two cases a deletion was detected in the
`nosing the fetuses as affected.
`affected male DNA but not with the fetal DNA, diagnosing the fetuses as un-
`No deletions were detected in either the fetal or affected male
`affected.
`DNAs in the remaining 2 cases.
`Southern analysis confirmed the presence or absence of the six regions in
`every case (data not shown) demonstrating that multiplex amplification works
`well with amniotic fluid cells and that the deletions were not masked due to
`amplification from any contaminating maternal cells (see below).
`In one of
`the two families where no deletion was detected via amplification (DRL # 469),
`
`11148
`
`Ariosa Exhibit 1026, pg. 8
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`FAMILY (DRL) #
`43C 483
`485 469
`531
`
`521
`
`~~~......
`
`Multiplex DNA amplification for prenatal diagnosis of DMD.
`Figure 4.
`Shown
`are the results of amplification using DNA from an affected male (AM; lympho-
`blast DNA) and a male fetus (MF; cultured amniotic fluid cell DNA) from six
`Analysis was as described in Figure 3.
`Both the affected
`different families.
`male and the fetal DNAs of DRL #s 521 and 531 display a deletion of region f
`In DRL # 43C the affected male
`(Fig. 2) diagnosing these fetuses as affected.
`is deleted for all regions except f, while the fetus is unaffected.
`The af-
`fected male in DRL # 483 is deleted for region a, while the male fetus is un-
`Neither of the samples from DRL #s 485 or 469 displayed a deletion
`affected.
`with this technique. Analysis was as described in Fig. 3.
`
`Southern analysis did detect a small deletion located outside of the six re-
`gions amplified in both the affected male and fetal DNAs.
`Therefore multiplex
`amplification detected 2 of 3 deletions in the fetal samples providing a diag-
`nosis of affected, two were diagnosed as normal, and two were indeterminate
`(only one of which was further clarified by Southern analysis).
`Frequently chorionic villus specimens (CVS) rather than amniotic fluid
`cells are used for DNA diagnosis.
`Since CVS are potentially contaminated with
`greater amounts of maternal tissue than are amniotic fluid cells we tested
`whether multiplex amplification from CVS derived DNA might fail to detect
`deletions due to amplification from the normal allele of any contaminating
`decidual DNA.
`Chorionic villus specimens received by the R.J. Kleberg Center
`for Human Genetics for diagnostic studies are microscopically dissected to
`remove decidual tissue prior to study, and cytogenetic analysis rarely detects
`any maternal XX karyotypes in dissected male specimens.
`Multiplex amplification was performed with fetal DNA prepared from two
`chorionic villi and with lymphoblast DNA from an affected brother from one of
`No living affected male was available for the second case.
`the families.
`
`11149
`
`Ariosa Exhibit 1026, pg. 9
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`FAMILY (DRL) #
`92
`120
`MF AM MF-
`
`Multiplex DNTA amplification from CVS DNA.
`Figure 5.
`Both the affected male
`(AM; lymphoblast DNA) and the male fetus (MF; CVS DNA) from DRL # 92 display a
`deletion of regions e and f (Fig. 2), diagnosing the fetus as affected.
`CVS
`DNA from DRL 1/ 120 did not display a deletion with this technique, and no
`living affected relative was available for comparison.
`Samples were analyzed
`as described in Fig. 3.
`
`Figure 5A demonstrates that a deletion was detected in DRL # 92 fetal DNA and
`in DNA from his affected brother, but not in the DRL # 120 fetal DNA.
`Southern analysis with cDNA probes confirmed the veracity of these results,
`but detected a deletion of two exons located outside of the amplified regions
`of the DRL # 120 fetal DNA.
`Thus although the DRL # 120 deletion was not
`detectable with this method the results with the DRL # 92 sample indicate that
`maternal DNA did not lead to a false positive amplification under the condi-
`tions used.
`To further explore potential problems associated with contaminating
`maternal DNA reconstruction experiments with deleted and normal DNA were
`performed.
`We have observed that low level contamination of DNA samples with
`non-deleted DNA is not apparent at early rounds of amplification, but that as
`the level of amplification of the test DNA approaches saturation the contami-
`nating DNA continues to be amplified and eventually can produce similar
`quantities of reaction products as does the test DNA.
`In addition to the
`number of cycles performed, the relative amount of contamination from ex-
`ogenous or maternal DNA will also affect the results.
`To estimate the amount
`of contamination which can be tolerated before ambiguous results are obtained
`we mixed a partially deleted DNA sample with various amounts of non-deleted
`DNA and then performed increasing cycles of multiplex amplification.
`Figure 6
`demonstrates that levels up to 3-5% contamination can be tolerated when
`amplification is limited to 25 cycles.
`Fewer cycles do not produce sufficient
`
`11150
`
`Ariosa Exhibit 1026, pg. 10
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`3lacyles
`28cycles
`22cycles
`25cycles
`o 2 -I o 9to 9 to 0
`C) -
`-C_O -4 -4 C
`-4 '_ C
`---
`
`of cyces
`% contamination
`
`4- exon d
`
`5OOng of DNA
`Effect of contamination on multiplex amplification.
`Figure 6.
`from DRL # 519 (Fig. 3B) was mixed with 50ng (10%), 5ng (1%), or 0.5ng (0.1%)
`Analysis was as
`of normal control female (46XX) DNA prior to amplification.
`described in Fig. 3 except that 15pl aliquots were removed from each reaction
`after the indicated number of amplification cycles and stored on ice until
`DRL # 519 is deleted only for region d.
`By 25 cycles the sample
`analysis.
`with 10% contamination has begun to display the exon d fragment.
`Note that
`the contaminant contains two X chromosomes; exogenous male DNA would be ex-
`pected to produce half the observed signals.
`
`quantities of DNA for visual analysis, whereas additional cycles allow detec-
`tion of decreasingly smaller quantities of contaminating DNA.
`
`DISCUSSION
`Multiplex amplification methods have recently been integrated into the
`DMD diagnostic protocols of the R.J. Kleberg Center for Human Genetics at
`Baylor College of Medicine. Currently all DNA samples received for analysis
`are initially analyzed by multiplex amplification.
`This procedure requires
`five hours from start-up to photography of results, and as currently practiced
`eliminates the need for further Southern analysis of approximately 37% of the
`samples (Table 1).
`Cases in which deletions are not detected are subsequently
`examined via Southern analysis using full-length DMD cDNAs subdivided into 7-9
`separate probes.
`Those samples that do not display genomic alterations with
`cDNAs must then be examined by Southern analysis with multiple genomic probes
`using family member DNAs digested with numerous restriction enzymes for lin-
`kage analysis (29).
`Complete Southern analysis of samples not diagnosed via
`multiplex amplification requires one to two weeks for hybridization of multi-
`ple blots with up to 16 separately labelled probes for each family referred
`The ability to eliminate 37% of the cases from the necessity
`for diagnosis.
`for Southern analysis clearly illustrates the tremendous savings of time and
`effort which can be achieved via a one day multiplex amplification protocol.
`While it is difficult to estimate the dollar savings which would result, the
`reduction in labor from one or two weeks to less than one day, and the lower
`
`11151
`
`Ariosa Exhibit 1026, pg. 11
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`reagent costs (including no radioisotopes) should make multiplex amplification
`significantly less expensive than Southern analysis.
`Additional features of amplification techniques contribute further to
`Whereas Southern analysis requires extensively purified high
`their utility.
`molecular weight DNA, PCR procedures can use crude and partially degraded
`samples.
`We have observed that DNA samples too degraded to produce detectable
`signals upon Southern analysis are suitable templates for multiplex amplifica-
`PCR procedures require only that the average molecular
`tion (data not shown).
`weight of template DNAs be slightly greater than the largest fragment ampli-
`fied.
`This feature of the technique should prove advantageous for the analy-
`sis of low quality DNA resulting from suboptimal purification, storage, or
`Procedures have also been described for PCR amplification from
`shipping.
`paraffin embedded biopsy samples stored for years at room temperature (30).
`Such samples should allow retrospective examination of DMD cases diagnosed
`before the isolation of DMD clones, which can facilitate future prenatal
`In addition future pregnancies
`diagnosis involving other family members.
`within non-carrier DMD families where a lesion was thought to have arisen via
`new mutation can be rapidly screened by PCR amplification of a region deleted
`from the gene of the affected male to guard against rare cases of mosaic germ
`line mutations (31,32).
`As more sequence data for the DMD gene becomes available additional
`primer sets can be added to the assay to increase the frequency of deletion
`With 6 primer sets we detected deletions in 54% (15/28) of the
`detection.
`samples analyzed in this study, or 79% (15/19) of the deletions observed with
`Examination of the frequency that individual exons were observed
`cDNA probes.
`to be deleted from the DNA of over 200 randomly surveyed male DMD patients
`(2,6) indicates that the multiplex amplification procedure would have detected
`deletions in 37% of those patients, or 70% of the frequency observed with cDNA
`The higher detection rate displayed in Figures 3-5 presumably re-
`clones.
`flects a normal variation which should be expected from analyzing the smaller
`The deletion data
`number of cases referred to our laboratory for this study.
`previously obtained from DMD patients indicates that by cloning and subse-
`quently including 2 more exons in this assay [exon 4 and the 3.7 kb HindIII
`fragment detectable with human cDNA probe 8 (2)] a deletion detection rate of
`almost 90% would be achieved (2,6).
`Inclusion of other additional exons would
`result in very small increases in detection frequency, as most deletions
`encompass numerous exons and the few that do not are usually located in
`deletion prone regions of the gene (2,6).
`
`11152
`
`Ariosa Exhibit 1026, pg. 12
`IPR2013-00276
`
`

`

`Nucleic Acids Research
`
`The multiplex amplification technique can also be adapted to polymorphism
`detection for linkage analysis.
`We have observed that intron sequences con-
`tain a high frequency of polymorphisms, most of which do not occur at restric-
`tion enzyme recognition sites (unpublished observation).
`These polymorphisms
`can be detected via amplification of intron containing sequences (as is done
`with this method) followed by hybridization with ASO probes, or in the case of
`Techniques are currently
`RFLPs, via restriction enzyme digestion (1,13,14).
`under development which should permit polymorphism detection directly during
`Integration of these
`PCR amplification (RAG, manuscript in preparation).
`future modifications could permit analysis of almost all DMD diagnostic cases
`via amplification based techniques.
`The rapidity and simplicity of these methods should permit routine DMD
`diagnosis at clinical laboratories without the need for highly trained re-
`Currently we are using this assay in a kit form.
`search personnel.
`Primers
`and buffer are pre-mixed and aliquoted into tubes for storage at -70°.
`For
`analysis a tube is thawed, DNA and enzyme are added, the reaction is performed
`on an automatic thermocycler (Perkin Elmer/Cetus), and the results are deter-
`mined by agarose gel electrophoresis.
`The use of fluorescent primers will
`permit automatic analysis of reaction products on DNA sequencing apparatus
`such as is available from Applied Biosystems (33).
`We envision that clinical
`laboratories could routinely screen DNA samples for deletions, then forward
`only those cases in which deletions were not detected to a research oriented
`DNA diagnostic laboratory for Southern analysis.
`In extremely rare cases
`where every amplified region is deleted (e.g. DRL # 957 and 484, Fig. 3) the
`deletion should be confirmed via Southern analysis.
`Successful use of these multiplex amplification methods requires that DNA
`from maternal or exogenous origin not interfere with deletion detection.
`We
`have demonstrated that cultured amniotic fluid cells and chorionic villus
`specimens dissected of decidual tissue produce clearly manifested deletions as
`long as amplification reactions do not approach saturation (Figs. 4 and 5).
`Furthermore, after 25 cycles of amplification levels of non-deleted DNA at up
`to 3-5Z of the total do not interfere with interpretation of the results (Fig.
`6).
`As with any sensitive technique careful pipetting of reagents