Miller-Dieker syndrome

Nongenic regions of eukaryotic genomes include head-to-tail tandemly repeating DNA sequences. These sequences generally consist of an integral number of repeats, individual members of which share some common consensus sequence. The number of repeats within these variable number of tandem repeat (VNTR) regions varies between unrelated individuals; inheritance of an allele is generally Mendelian, and the location of each VNTR region in the genome is stable. Single-locus VNTR regions thus have particular use in genome mapping, in establishing biological relatedness between individuals, and in broad phylogenetic studies. Polymerase chain reaction (PCR) is especially suited to analysis of VNTR regions, but only if all alleles at the test locus are sufficiently small to be efficiently amplified.

The repeat unit at human VNTR locus D17S5 is 70 base pairs (bp) long (Wolff et al. 1988). The number of alleles at this locus and their size increments, as seen by enzyme restriction and Southern analysis (Odelberg et al. 1989), loosely correspond to data obtained by PCR-based procedures [e.g., Gecz (1991) and Batanian et al. (1990)]. However, rare long-length alleles detectable by the Southern blot hybridization of restriction-enzyme-digested DNA (Odelberg et al. 1989) are not as easily detected by PCR methods when amplification products are simply stained with ethidium or silver (Gecz 1991; Walsh et al. 1992). Most often the reason is preferential amplification of shorter over longer alleles in some heterozygote individuals, leading to so-called allele dropout. This is of particular concern if D17S5 is used in diagnosis of disease (Batanian et al. 1990), determination of parentage, or forensic studies (Lienert and Fowler 1992). In this survey of Australian Caucasoid and traditional aboriginal populations, PCR products were carefully examined by Southern analysis, and rare new alleles, considerably larger than those previously reported in PCR-based studies, were located.

MATERIALS AND METHODS

SOURCE AND ISOLATION OF DNA. The DNA samples were from unrelated Caucasoid and aboriginal individuals (gift of State Forensic Science, Adelaide, Australia). The Caucasoid samples were drawn from the South Australian population, and the aboriginal sample came from the central desert and north-central regions of South Australia and the Northern Territory, Australia.

The DNA was isolated from peripheral blood using either conventional proteinase K digestion and phenol/chloroform extraction or by using the automated GenePure 341 method (Applied Biosystems). The DNA from CEPH (Centre d'Etude du Polymorphisme Humaine) Utah families 982 and 983 was isolated by similar methods.

AMPLIFICATION OF D17S5. Amplification of the D17S5 locus using primers described by Horn et al. (1989) was performed according to the method of Lienert and Fowler (1992) with the following modifications. Each DNA sample (==50 ng) was incubated at 98degC for 10 min with 6-mu-g of bovine serum albumin (BSA) in 10 mu-l distilled sterile H sub 20. Forty microliters of master-mix was added after lowering the temperature to 72degC. The final 50 mu-l reaction mixture contained 67.0 mM Tris-HCl (pH 8.8), 16.6 mM (NH sub 4) sub 2SO sub 4, 10 mM P-mercaptoethanol, 0.2 mg/ml gelatin (Ajax 1080), 1.2 mM MgCl sub 2, 0.2 mM total dNTP, 0.22 CLM (100 ng) of each primer, and 1.5 U AmpliTaq (Perkin Elmer). The samples were cycled 32 times between 94degC (1 min), 55degC (1 s), and 72degC (1 min) before a final 7-min extension at 72degC.

ELECTROPHORESIS AND SOUTHERN ANALYSIS. The PCR products were electrophoresed in 2% agarose gel using TAE buffer [10 mM Tris-HCl (pH 8.0), 5 mM NaAcetate, 0.5 mM EDTA] for 2.5 hr at 6 V/cm. The gels were stained in ethidium bromide and photographed, and the results were evaluated. Samples of the PCR product (about 0.11 volumes) were suitably diluted with buffer and electrophoresed in similar agarose gels and blotted for 3 hr by dilute alkaline (0.04 M NaOH) transfer to Hybond N sup + membrane (Amersham). The membranes were neutralized by washing twice for 1 min each in 2X SSC (0.3 M NaCl and 0.03 M sodium tricitrate) and prehybridized at 55degC for 1 hr in a solution of 0.1% N-lauroylsarcosine, 0.02% SDS, and 1% blocking reagent (Boehringer) in 5X SSC. The membrane was then hybridized for 1 hr at 55degC with a biotinylated 22-bp probe (10 ng/ml) in the hybridization mixture. The synthetic probe sequence 5 'biotin-CTGGGGCAGGGCTGTGAGACC3' is directed to part of the VNTR unit (Wolff et al. 1988). Unbound biotinylated probe was removed by first washing four times for 5 min each in a solution of 2X SSC and 0.1% SDS, followed by one washing for 20 min in a solution of 0.5X SSC and 0.1% SDS at 55degC. The site of probe attachment was detected by chemiluminescence using a strep-avidin alkaline phosphatase conjugate and dioxetane substrate, as described by the supplier (Bresatec Ltd., South Australia).

ALLELE SIZING, DATA COLLATION, AND STATISTICAL ANALYSIS. The alleles were initially sized against a 123-bp ladder standard (BRL) and either biotin-labeled SPP1/EcoRI fragments (Bresatec Ltd.) or biotin-labeled phi-X 174/HinfI fragments (BRL). A D17S5 allelic length standard was prepared by mixing products from individually amplified DNA samples. This was routinely run in gels intended for Southern blotting and chemiluminescent detection of PCR products. Alleles were sized using automated image capture and fragment-sizing software (Tracktel Version 2.0, Vision Systems, South Australia). Alleles were named on the basis of repeat number, allele 1 being 170 bp (70-bp repeat plus a 100-bp flanking sequence), allele 2 being 240 bp (2 70-bp repeats plus a 100bp flank), etc.

Calculations of heterozygosity, discriminating power, and evaluation of Hardy-Weinberg equilibrium were made using the Genotype Analysis program (version 1.10; G.R. Morgan, Brisbane, Australia). Two-way contingency tests (X sup 2 test and G statistics) were used to determine the statistical significance of population sample homogeneity (program provided by G. Carmody, Carleton University, Ottawa, Canada).

RESULTS

The DNA from 201 Caucasoid and 166 traditional aboriginal individuals was amplified for locus D17S5. Eighteen different-length alleles were observed, ranging from 170 bp (1 repeat unit) to 1430 bp (19 repeat units). An allele of 18 repeat units was not found in this survey. A small number of rare and unusually long alleles were seen following oligoprobing. Examples of these are shown in Figure 1. Alleles 15, 16, 17, and 19 (Figure 1; Table 1) are all longer than examples previously described (Gecz 1991). (Figure 1; Table 1 omitted.) Each of these measured to within 10 bp of their theoretical length [(70 x number of repeats) + 100) bp]. Few of these long alleles were initially visible following ethidium staining, particularly if the size difference between alleles was large (e.g., samples in lanes 4, 5, 7, and 8 of Figure 1). These samples are the most likely to be incorrectly typed as homozygotes, and it was only following Southern transfer and oligoprobing that the larger alleles were clearly visualized. Probing of the sample in lane 6 (Figure 1) also shows the in vitro production of low copy numbers of shorter allele products, typical of out-of-phase priming of repeat arrays.

The distribution of D17S5 alleles in each surveyed population is recorded in Table 1. In the Caucasian population the alleles at locus D17S5 display high heterozygosity (h = 86.4%) and discriminating power (0.963). In the aboriginal population lower heterozygosity (h = 80.8%) and discriminating power (0.942) were encountered because of the relative dominance of allele 1. For each population the number of observed genotypes was in agreement with the expected genotypes and thus conformed to Hardy-Weinburg equilibrium expectations (see Table 1). Statistical comparison between these distinct populations showed them to be different: p = 0.00 in the chi-square test and G statistic. Notably, alleles 1, 2, and 10 are dominant in the aboriginal population, compared with alleles 2, 3, and 4 being dominant in Caucasoids.

Segregation of alleles in two three-generation pedigrees (CEPH Utah families 982 and 983) was Mendelian. One unusual result was noted, however. In family 982, individual 7057 consistently showed three bands (lane 15, Figure 2). (Figure 2 omitted.) The reason is uncertain, but sample contamination has been eliminated. This result may be a rare example of polymerase slippage. Alternatively, somatic mutation in the transformed cell line from which the DNA was extracted may have occurred. Such a result has been observed by us (in a different individual) when testing the Utah families at another VNTR locus and has also been seen infrequently during analysis of microsatellite VNTR regions in CEPH pedigrees (Webber et al. 1992).

DISCUSSION

Locus D17S5 has a 70-bp repeat length, a high GC base composition (64%), and a nearly tenfold variation in allele length in the tested populations. These characteristics limit the ability of PCR to create equal copy numbers of alleles that are co-inherited in heterozygote individuals, especially if the alleles differ markedly in length. Preferential amplification can be minimized by improved PCR conditions [e.g., Holm et al. (1991) and Walsh et al. (1992)]. In our experience preincubation of template DNA in 0.05% BSA in a hot-start procedure was found to enhance amplification of all alleles, without the subsequent need for long extension times (Gecz 1991) while also using short (1 s) annealing times.

Even by using efficient PCR methods, the results of this survey show the clear advantage of examining the amplification products with an oligomeric probe. Not only is the sequence of the PCR product authenticated, but also any preferential amplification of small alleles is counterbalanced by the preferential detection of large alleles. This is because the number of bound oligoprobes increases linearly with allele length, giving an increasing chemiluminescent signal (Figure 1). As a consequence, this survey better represents the full extent of large allele variation at this locus. However, although the risk of allele dropout is much reduced by these methods, it is not eliminated. All homozygote results should be treated with caution if locus D17S5 is used in linkage or diagnostic studies where, for example, the loss of an allele may be either a deletion of this region of chromosome 17 [e.g., Miller-Dieker syndrome (Batanian et al. 1990)] or a technical artifact.

The results from a human population perspective are significant in demonstrating how remarkably different the allele size ranges and allele frequencies are in these two distinct populations (Table 1). The differences in heterozygosity and discriminating power calculated for the two populations reflect significant differences in the relative dominance of their major alleles. The shortest allele is predominant in aboriginals, but the rare long alleles also tend to be found more in the aboriginal population than in the Caucasoid population. These results suggest the lengthy geographic separation of these populations and perhaps an ancient divergence. The mutation and evolution of locus D17S5 has thus followed different paths in these populations, such changes emerging from an apparently nonvariant monomorphic form present in nonhuman higher primates [our unpublished results testing gorilla and chimpanzee confirm the findings of Wolff et al. (1991)].

Excluding the larger alleles reported here, the allele frequencies of the Australian Caucasoid population are not significantly different from a surveyed American Caucasoid population (Robertson and Kronick 1991): p = 0.559 in a chi-square test; p = 0.564 in a G statistic. However, the Australian Caucasoid population appears to be significantly different from a survey of native Finns (Sajantila 1992): p = 0.028 in a chi-square test; p = 0.027 in a G statistic.

Deka et al. (1992) have also studied locus D17S5 using PCR-based methods, testing native individuals from northeast India (Kachari), Canada (Dogrib Indian), and Papua New Guinea (New Guinea Highlander). From limited data (less than 50 individuals in each case), chi-square and G statistical tests suggest that the allele frequency distribution for aboriginals at locus D17S5 is different from that of the other three anthropologically distinct groups. Unlike this study, Deka and colleagues found no alleles larger than about 1 kb using PCR methods, but they found one extra-long allele by Southern blot hybridization of restriction-enzyme-digested DNA.

The existence of especially long alleles at locus D17S5 is not altogether unexpected. Theoretical evaluation of diversity at locus D17S5 (Chakraborty and Daiger 1991) and the limited data from restriction-based Southern analysis (Odelberg et al. 1989) indicate that there may be as many as 20 alleles at this locus. Our survey of 377 individuals has located 18 alleles.

The data show that locus D17S5 has a greater heterozygosity and discriminating power than locus D1S80, another hypervariable VNTR locus amenable to PCR analysis (Sajantila et al. 1992). The limitation inherent to PCR-based analysis of locus D17S5 is the risk of preferential amplification of small alleles in heterozygotes, leading to mistyping of genotype. This risk has been largely overcome by using the nonisotopic oligoprobing method described, which has the additional advantage of having extreme sensitivity suitable for forensic analyses. Within-generation mutation of locus D17S5 has been reported (Wolff et al. 1988), but we have seen no such examples in the pedigrees and parentage cases we have tested thus far.

FOOTNOTES

1. School of Biological Sciences, Flinders University of South Australia, Sturt Road, Bedford Park, Adelaide, South Australia, Australia 5042.

2. State Forensic Science, 21 Divett Place, Adelaide, South Australia, Australia 5000.

LITERATURE CITED

Batanian, J.R., S.A. Ledbetter, R.K. Wolff et al. 1990. Rapid diagnosis of Miller-Dieker syndrome and isolated lissencephaly sequence by the polymerase chain reaction. Hum. Genet. 85:555-559.

Chakraborty, R., and S.P. Daiger. 1991. Polymorphisms at VNTR loci suggest homogeneity of the white population of Utah. Hum. Biol. 63:571-587.

Deka, R., S. De Croo, L.M. Yu et al. 1992. Variable number of tandem repeat (VNTR) polymorphism at locus D17S5 (YNZ22) in four ethnically defined human populations. Hum. Genet. 90:86-90.

Gecz, J. 1991. PCR amplification of large VNTR alleles of D17S5 (YNZ22) locus. Nucleic Acids Res. 19:5906.

Holm, T., C. Terry, and M. Georges. 1991. In vitro amplification of a set of VNTR loci for forensic science. Crime Lab. Dig. 18:187-189.

Horn, G.T., B. Richards, and K.W. Klinger. 1989. Amplification of a highly polymorphic VNTR segment by the polymerase chain reaction. Nucleic Acids Res. 17:2140.

Lienert, K., and J.C.S. Fowler. 1992. Analysis of mixed human/microbial DNA samples: A validation study of two AMP-FLP typing methods. Biotechniques 13:276-281.

Odelberg, S.J., R. Plaetke, J.R. Eldridge et al. 1989. Characterization of eight VNTR loci by agarose gel electrophoresis. Genomics 5:915-924.

Robertson, J.M., and M. Kronick. 1991. Automating DNA fingerprinting: A multifluorophore approach to reduce chance in match calling. Crime Lab. Diag. 18:179-182.

Sajantila, A. 1992. DNA Analysis in Forensic Medicine: Application of the Polymerase Chain Reaction (PCR) to the Identification of Individuals. Publications of the National Public Health Institute A3/1992. Helsinki, Finland; National Public Health Institute, 39.

Sajantila, A., B. Budowle, M. Strom et al. 1992. PCR amplication of alleles at the D1S80 locus: Comparison of a Finnish and a North American Caucasian population sample and forensic casework evaluation. Am. J. Hum. Genet. 50:816-825.

Walsh, P.S., H.A. Erlich, and R. Higuchi. 1992. Preferential PCR amplification of alleles: Mechanisms and solutions. PCR 1:241-250.

Webber, J., Z. Wang, and P. Wilkie. 1992. Human linkage mapping with short tandem repeat polymorphisms. In Conference Proceedings, The Australian Gene Mapping Workshop, T. Hetzel, L. Jazwinska, and L. McIntyre, eds. Brisbane, Australia: University of Queensland Press, 71.

Wolff, R.K., Y. Nakamura, and R. White. 1988. Molecular characterization of a spontaneously generated new allele at a VNTR locus: No exchange of flanking DNA sequence. Genomics 3:347-351.

Wolff, R.K., Y. Nakamura, S. Odelburg et al. 1991. Generation of variability at VNTR loci in human DNA. In DNA Fingerprinting: Approaches and Applications, T. Burke, G. Dolf, A.J. Jeffreys et al., eds. Basel, Switzerland: Birkhauser Verlag, 20-38.

Copyright Wayne State University Press Apr 1994
Provided by ProQuest Information and Learning Company. All rights Reserved