`
`Minisatellite "lsoallele" Discrimination in Pseudohomozygotes
`by Single Molecule PCR and Variant Repeat Mapping
`
`DARREN G. MONCKTON AND ALEC J. JEFFREYS1
`
`Department of Genetics, University of Leicester, University Road, Leicester L£1 lRH, United Kingdom
`
`Received April 1, 1991
`
`The DlS8 hypervariable minisatellite MS32 has a het(cid:173)
`erozygosity of 97.5% based on detectable differences in al(cid:173)
`lele length using standard Southern blot analysis. It has
`previously been shown that the basic repeat unit is in itself
`variable and that this may be used to map the internal
`structure of an allele. This method has already been used to
`establish that alleles of the same length may have differing
`internal structures between nonrelated individuals. We
`now extend this approach to demonstrate that two appar(cid:173)
`ently homozygous individuals are in fact heterozygotes.
`For each individual the two comigratory alleles were sepa(cid:173)
`rated, without cloning, using single molecule dilution
`(SMD) of genomic DNA and recovery with PCR. Mapping
`of the variant repeat units revealed highly diverged inter(cid:173)
`nal structures and, for one individual, a size difference of
`one repeat unit (29 bp). SMD and PCR recovery provide an
`efficient system for separating co migratory alleles without
`prerequiremen.t for knowledge of sequence differences.
`© 1991 Academic Pre .. , Inc.
`
`Minisatellite or variable number tandem repeat
`(VNTR) loci have proven to be of considerable im(cid:173)
`portance in a wide range of genetic applications, espe(cid:173)
`cially in the areas of individual identification (Jef(cid:173)
`freys et al., 1985) and gene mapping (Nakamura et al.,
`1987). The primary reason for their widespread use as
`genetic markers is their often very high level of allele
`polymorphism and resulting heterozygosity (Wong et
`al., 1987; Nakamura et al., 1987). Allelic state is
`usually determined by the measurement of DNA frag (cid:173)
`ment length estimated from Southern blot hybridiza(cid:173)
`tion of genomic DNA. This system is however limited
`by the resolving power of agarose gel electrophoresis,
`not only for distinguishing small differences in allele
`sizes between individuals, which can lead to spurious
`
`1 To whom correspondenr.e should be 11ddressed. •
`
`departures from Hardy-Weinberg equilibrium (Dev(cid:173)
`lin et al., 1990), but also for discriminating between
`true and false homozygotes. However, allele length is
`not the only criterion by which minisatellite loci may
`be distinguished. DNA sequencing of human mini(cid:173)
`satellites has revealed repeat unit sequence variabil(cid:173)
`ity at almost all loci so far investigated (see Jeffreys et
`al., 1990). This additional level of polymorphism
`greatly extends the potential resolving power of mini(cid:173)
`satellite loci allowing alleles of similar, or even identi(cid:173)
`cal, size to be distinguished on the basis of internal
`structure.
`The DIS8 minisatellite (MS32) comprises a 29-bp
`repeat unit and has a reported heterozygosity of
`97.5% based on allele length (Wong et al., 1987). Se(cid:173)
`quencing of DIS8 has revealed an A toG transition in
`approximately 70% of the repeat units, resulting in
`the presence or absence of a Haeiii restriction site
`(Wong et al., 1987). In contrast, all repeat units are
`cut by the restriction enzyme Hinfi. The location of
`variant repeat units can be mapped in alleles ampli(cid:173)
`fied by PCR. Partial digestion of end-labeled alleles
`with Hinfi followed by gel electrophoresis and autora(cid:173)
`diography produces a continuous ladder of labeled
`DNA fragments from which the number of repeat
`units can be determined. Comparison of Haeiii and
`Hinfi partial digests enables each repeat unit to be
`scored as to whether or not it is cleaved by Haeili.
`Internal maps, or minisatellite variant repeat (MVR)
`haplotypes, thus generated can be simply encoded as a
`binary string of repeats cleaved or not cleaved by
`Haeiii (Jeffreys et al., 1990). We have previously
`shown that alleles of the same length shared by unre(cid:173)
`lated individuals may have widely differing internal
`structures, suggesting relatively distant genealogical
`origins (Jeffreys et al., 1990).
`Screening of D 1S8 across large panels of unrelated
`people revealed two individuals (AS89 and MACH, of
`Pakistani and Chinese descent, respectively) who by
`
`465
`
`GENOM!CS 11, 465-467 (1991)
`0888-7543/91 $3.00
`Copyright© 1991 by Academic Press, Inc.
`All rights of reproduction in any form reserved.
`
`The Johns Hopkins University Exhibit JHU2008 - Page 1 of 3
`
`
`
`466
`
`SHORT COMMUNICATION
`
`B
`
`A. 0 M A
`
`. .
`
`2.45-
`
`A
`
`kb
`4.361-
`
`2.322-
`2.027-
`
`lOx 6pg DNA
`lOx 3pg DNA
`lOx zero DNA
`kb ~~~
`--+++---+--+-++---+-----------
`
`AS89
`
`1··:.' • •
`
`MACH
`
`!Ox 6pg DNA
`lOx 3pg DNA
`lOx zero DNA
`kb ~~~
`-+-+-+-+-+------+-------------
`
`2.45-
`
`I • • • • •
`
`FIG. l. Separation of minisatellite alleles in pseudohomozygotes by single molecule dilution and PCR. (A) DlSB alleles amplified from
`total genomic DNA. For each individual 20 ng blood DNA was PCR amplified in a 10-)ll reaction using the nested primers Cl and D
`corresponding to the fianking region of the D1S8 minisatellite (Ref. (3)). Amplified allele~> were electrophoresed through a 1% agarose gel
`;>.. Hindlli marker DNA; 0, zero DNA negative control; M, amplified alleles from
`and visualized by ethidium bromide staining. Lanes: >-.,
`individual MACH; A, amplified alleles from individual AS89. (B) Allele separation by single molecule dilution. For each individual, genomic
`DNA was diluted in 5 mMTris-HCl (pH 7.5) in the presence ofO.l !lM PCRprimers. Ten 10-111 PCRreactions containing either 6, 3, or 0 pg
`DNA were amplified for 28 cycles using the D1S8 minisatellite flanking primers A plus B. Products were detected by Southern blot
`hybridization. Primer sequences are given in Ref. (3). Derivatives Cl and Dl incorporate a 5' extension containing nn EcoRI restriction site
`that was used to generate end labeled PCR products.
`
`Sau3AI restriction digestion and Southern blot analy(cid:173)
`sis appeared to be homozygous for an approximately
`2.5-kb allele at D1S8 (data not shown), PCR amplifi(cid:173)
`cation of DlS8 from total genomic DNA revealed a
`single band on an ethidium bromide-stained gel for
`each individual (Fig. lA). To determine if these indi(cid:173)
`viduals were indeed true humozygotes we attempted
`to map the internal structures of these alleles. Inter(cid:173)
`nal mapping fro!ll total genomic DNA produced an
`ambiguous autoradiograph with widely differing band
`intensities, presumably reflecting a composite inter(cid:173)
`nal map derived from two comig:ratory but discrete
`alleles (Fig. 2, lanes marked T). Since the alleles were
`inseparable by standard agarose gel electrophoresis,
`the individual alleles were separated by single mole(cid:173)
`cule dilution (SMD) of genomic DNA (Jeffreys et al.,
`1990; Ruano et al., 1990) and recovery with PCR
`(Saiki et al., 1988) \ISing PCR primer pair A plus B,
`which correspond to the flanking sequence of the
`D1S8 minisatellite and allow amplification of the en(cid:173)
`tire minisatellite allele (Jeffreys et al., 1990). After 28
`cycles of amplification a 5-~1 aliquot was removed,
`electrophoresed, and Southern blot hybridized (Fig.
`1B). Internal nested primers C plus D were used to
`reamplify the alleles from each of the presumptive
`single molecule-positive reactions up to a level visible
`on an ethidium-stained gel. All four alleles were com(cid:173)
`pletely mapped from both ends using DNA derived
`from at least three separate single molecule reactions
`for each allele (Fig. 2).
`
`Individual AS89 was found to have two alleles of
`identical size, containing 71 repeat units, but of
`widely differing internal structures. One allele shares
`a 5' MVR haplotype common to many individuals
`(unpublished data). The other allele belongs to an(cid:173)
`other previously characterized set of homogeneous al(cid:173)
`leles composed almost entirely of repeats cleaved by
`Haelll. This individual is therefore a true heterozy(cid:173)
`gote at this locus. Analysis of the second individual
`MACH, of Chinese origin, revealed this person to be a
`compound heterozygote at this locus, each allele hav(cid:173)
`ing widely diverged internal structures and a length
`difference of 1 repeat unit (allele lengths 72 and 73
`repeat units). Both alleles describe new 5' haplotypes
`previously unseen in our studies on Caucasian individ(cid:173)
`uals (Jeffreys et al., 1990).
`Our method for internal mapping does not involve
`cloning of single amplified molecules (minisatellites
`are frequently unstable on cloning in Escherichia
`coli), but samples the average properties of the entire
`amplified pool of PCR products. For each allele at
`least three separate single molecule amplifications
`were performed and in no case was any MVR map
`discrepancy found between them. As noted in pre(cid:173)
`vious experiments (Jeffreys et al., 1990), it appears
`that Taq polymerase-induced misincorporation
`errors are not a problem with single molecule mini(cid:173)
`satellite mapping. Finally, in each case the superim(cid:173)
`position of the two separated alleles gave rise to the
`same composite map as derived from total genomic
`
`The Johns Hopkins University Exhibit JHU2008 - Page 2 of 3
`
`
`
`SHORT COMMUNICATION
`
`467
`
`AS89
`
`MACH
`
`T
`A
`B
`,.........,,.........,,.........,
`F lf F H FH
`
`B
`A
`T
`·~~~
`FH
`I'll FH
`
`A
`
`B
`
`A589 allele B
`
`tYuuu.auuua.aoa• ........ ua-.aa.aaao.a&&euu~u~·.uaataaatt
`
`MA.Cti allele A
`
`tttaatu.~~tautu.t.taattt:aaauaaatantt41•t.taaatt.4a attaaatt.atttattatataatat
`
`DNA, confirming that the single molecule alleles
`recovered were the true alleles present in each indi(cid:173)
`vidual.
`Single molecule dilution and PCR recovery is a
`quick and efficient method for separating comigra(cid:173)
`tory alleles of the same locus. Moreover it has several
`further advantages over cloning direct from a geno(cid:173)
`mic library, requiring a very minimal quantity of in(cid:173)
`put DNA, as well as a suitability for isolation of prod(cid:173)
`ucts unstable in a bacterial host. Likewise, it offers
`advantages over cloning from a PCR-derived library
`in that it has very low probability of incorporating
`polymerase errors and removes the potential problem
`of heteroduplex-derived chimeric «alleles." It also as(cid:173)
`sumes no knowledge of internal sequence difference
`and thus allows isolation of uncharacterized polymor(cid:173)
`phisms.
`Internal mapping of minisatellites greatly en(cid:173)
`hances the effective resolution of an already highly
`informative set of markers, as demonstrated here by
`its ability to distinguish pseudohomozygotes from
`true homozygotes. D1S8 is one of the few minisatel(cid:173)
`lites for which an efficient internal mapping strategy
`has been developed so far, but work is in progress to
`increase the number of loci to which internal mapping
`can be applied. Internal mapping not only improves
`the general informativeness of minisatellite loci, but
`also provides important clues as to the evolution and
`generation of new alleles at these, some of the most
`invaluable, markers in the human genome.
`
`Internal mapping of PCR-amplified D1S8 alleles. In(cid:173)
`l<'IG. 2.
`ternal mapping of all alleles was performed as previously described
`(3). (A) End-labeled products were partially digested with Hinfl
`(F) or Hae!II (H), electrophoresed on a 1.2% agarose gel, dried, and
`autoradiographed. Lanes marked A and B are single molecule-de(cid:173)
`rived alleles; lanes marked T are internal maps derived from total
`genomic DNA. (B) Encoded maps of all four single molecule-de(cid:173)
`rived alleles, where A = repeat unit cleaved by Haelll and T = re(cid:173)
`peat unit not cleaved by Haeiii.
`
`ACKNOWLEDGMENTS
`
`Thanks are given to John Keyte for oligonucleotide synthesis,
`Mao Chen for supplying blood samples, and John Armour for tak(cid:173)
`ing them. This work was supported by grants to A.J.J. from the
`Medical Research Council and Wolfson Foundation. A.J.J. is a
`Lister Institute Research Fellow. D.G.M . is a Medical Research
`Council Research Student.
`
`REFERENCES
`
`2.
`
`1. DEVLIN, B., RISCH, N ., AND ROEDER, K. (1990) . No excess of
`homozygosity at loci used for DNA fingerprinting. Science
`249: 1416-1420.
`JEFFREYS, A. J., WILSON, V., AND THEIN, S. L. (1985). Individ(cid:173)
`ual-specific "fingerprints" of human DNA. Nature 316: 76-
`79.
`JEFFREYS, A. J., NEUMANN, R., AND WILSON, V. (1990) . Re(cid:173)
`peat unit sequence variation in minisatellites: A novel source
`of DNA polymorphism for studying variation 11nd mutation
`by single molecule analysis. Cell60: 473-485.
`4. NAKAMURA, Y., LEPPERT, M., O'CONNELL, P., WOLFF, R.,
`HOLM, T., CULVER, M., MARTIN, c., FUJIMOTO, E., HOFF, M.,
`KUMLIN, E ., AND WHITE, R. {1987). Variable number of tan-
`
`3.
`
`dem repeat markers for human gene mapping. Science 235:
`1616-1622.
`5. RuANO, G., Kmo, K. K., AND STEPHENS, J. C. (1990). Haplo(cid:173)
`type of multiple polymorphisms resolved by enzymatic am(cid:173)
`plification of single DNA molecules. Proc. Natl. Acad. Sci.
`USA 87: 6296- 6300.
`6. SAIKI, R. K., GELFAND, D. H ., STOFFEL, s., ScHARF, s. J.,
`HIGUCHI, R., HORN, G. T ., MULLIS, K. B ., AND ERLICH, H. A.
`(1988). Primer-directed enzymatic amplification of DNA
`with a thermostable DNA polymerase. Science 239: 487-491.
`7. WONG, Z., WILSON, V., PATEL, 1., POVEY, S., AND JEFFREYS,
`A. J. (1987). Characterization of a panel of highly variable
`minisatellites cloned from human DNA. Ann. Hum. Genet.
`51: 269-288.
`
`The Johns Hopkins University Exhibit JHU2008 - Page 3 of 3
`
`