Molecular Cytogenetics: : PCR-based diagnosis of human trisomies using computer-assisted laser densitometry

MINI-REVIEW

Molecular Cytogenetics I: PCR-based diagnosis of human trisomies using computer-assisted laser densitometry

Sérgio D.J. Pena

Núcleo de Genética Médica de Minas Gerais (GENE) and Departamento de Bioquímica e Imunologia, Universidade Federal de Minas Gerais, Av. Afonso Pena, 3111/9, 30130-909 Belo Horizonte, MG. Fax: [phone]. E-mail: [e-mail]

INTRODUCTION

It has been estimated that as many as half of all human conceptuses may have chromosomal defects (Boué and Boué, 1973). Thus, the most common genetic diseases of man occur at the chromosomal level and cytogenetic procedures are the most frequent tests performed in clinical genetic laboratories. While constituting a time-honored procedure that provides reliable diagnoses, conventional cytogenetics has some inconveniences, namely:

• It depends on the availability of live dividing human cells that are generally only obtained after cell cultures.

• It is very time consuming, even in the best-equipped laboratories

• It depends on technical expertise that can only be developed after prolonged training and extensive experience.

The first item, need for dividing cells, is very limiting. Every cytogenetic laboratory has experienced failures in obtaining satisfactory cultures from routine samples such as peripheral blood lymphocytes or amniocytes. Failures are especially common in the case of tissues from abortuses and stillborns that are often contaminated with bacteria or may have been frozen or fixed in formalin or alcohol. Even when the tissue already has enough dividing cells in vivo, as in bone marrow aspirates, or when only very short-term cultures suffice, as in chorionic villus samples, failure to obtain adequate metaphase spreads may occur. Thus, there is a great need for procedures that will allow cytogenetic diagnoses in non-dividing human tissues, including specimens that have been processed for pathology. Moreover, it would be highly desirable that these procedures could be performed quickly and did not depend on inordinate expertise. I wish to show in this short review that with the utilization of computer-assisted laser densitometry it is possible to use PCR-based tests to achieve the rapid, simple and inexpensive molecular diagnosis of human chromosomal disorders in non-dividing and even in non-living human tissues.

DETECTION OF GENE DOSAGE BY PCR

The polymerase chain reaction (PCR) is a highly sensitive and specific methodology for detection of trace amounts of nucleic acid sequence but is not primarily a quantitative technique. This happens because PCR amplification is a complex exponential phenomenon and small differences in initial conditions may have a profound impact on the amounts of product obtained. However, in the past few years it has become clear that by co-amplification of the target sequence together with known amounts of an internal standard, PCR may become a useful comparative quantitative tool (reviewed by Zimmermann and Mannhalter, 1996). For that to happen, there is an absolute necessity that the primer recognition sites be shared between the specific template and the internal standard, thus generating a competitive kinetics. Thus, for instance, by using PCR co-amplification of the beta-amyloid precursor protein gene (APP; chromosome 21q21) and of a especially designed competitive internal standard, Celi et al. (1994) obtained by radioactive densitometry an adjusted mean dosage of 2.00 ± 0.29 for controls and of 3.05 ± 0.27 for 28 cases of trisomy 21. However, the necessity of introducing the artificial internal standard in every PCR reaction complicated significantly this procedure. It would be much preferable if one could take advantage of internal genomic standards for competitive quantitation.

In 1991, Mutter and Pomponio reported on a simple molecular technique for sex identification that made use of differences in the sequences of the ZFY gene on the Y chromosome and its homolog ZFX on the X chromosome to design a single primer pair that led to the amplification of both genes and yet produced differently sized X and Y amplicons. They showed that under these internal genomic competitive conditions it was possible to diagnose reliably individuals with a 47,XXY or 47,XYY chromosome complement, since the relative amounts of ZFX and ZFY products were in 2:1 and 1:2 ratios, respectively, in contrast with a 1:1 for normal controls. Obviously this very attractive approach, based on the peculiar homology of parts of the X and Y chromosomes, could not be applied to autosomes. However, Mansfield (1993) reasoned that in heterozygous individuals, human polymorphic microsatellites could provide a simple and direct means of establishing gene dosage, since one allele would serve as a competitive internal standard for the other. They used an automatic fluorescent DNA sequencer for computer-assisted laser densitometry (a term coined by Allingham-Hawkins et al., 1998) and showed that trisomic patients displayed, in informative microsatellite loci, three fragment peaks of equal intensity or two fragments at an average 2:1 dosage.

DETECTION OF TRISOMIES BY PCR-BASED MICROSATELLITE ANALYSIS

Microsatellites are blocks of tandem repeat units of one to six base pairs that are ubiquitous, abundant and highly polymorphic in eukaryotic genomes due to variation in the number of repeating units between alleles (reviewed by Jeffreys and Pena, 1993). With the polymerase chain reaction it is possible to type microsatellites using primers designed from the unique sequence DNA flanking the tandem repeat arrays followed by polyacrylamide gel electrophoresis.

In selecting microsatellites for diagnosis of trisomies, several factors have to be taken into account. First and most obviously, the microsatellite must be located in the chromosome of interest and in the correct region. For instance, in the case of chromosome 21, markers located in 21q21 will permit not only the diagnosis of primary trisomies but also detection of secondary trisomies (unbalanced translocations). Nowadays thousands of tri- and tetra-nucleotide microsatellites have been sequenced and mapped in all human chromosomes (see, for instance, http://www.resgen.com/) and there is a great abundance to choose from. Second, the microsatellite has to be polymorphic and to have a level of heterozygosity as high as possible, since it will only be directly informative in heterozygous individuals. To maximize the chance of having informative results, multiplex sets of microsatellites from the same chromosome may be used (see below).

Microsatellite loci will be directly informative whenever they show more than one allele peak. Trisomic patients will display, in informative microsatellite loci, three fragment peaks of equal intensity or two fragments at an average 2:1 dosage (Figure 1). In the former case the diagnosis will be straightforward and more reliable while in the latter it will be purely quantitative and depend on statistic techniques such as discriminant analysis. The relative proportion of the two cases will depend primarily on the type of non-disjunction, although the heterozygosity level also is important. For a three-allele pattern to emerge it is necessary for the non-disjunction to occur in the first meiotic division, for the mother to be heterozygous at the relevant locus and for the allele carried by the sperm to be different from the two maternal alleles. A two-allele pattern will be observed in all cases in which non-disjunction occurs in the second meiotic division. For a given trisomy, if we know the proportion of cases due to non-disjunction at the first or the second meiotic division, it is possible, by using a simple extension of the Hardy-Weinberg law to trisomic states (Cavalli-Sforza and Bodmer, 1973; Pena, S.D.J., unpublished results), to calculate the relative rates of the 1-allele (non-informative), 2-allele and 3-allele patterns. For instance, let us take trisomy 21, where 74% of all cases are due to non-disjunction in the first meiotic division (Koehler et al., 1996) and consider the tetranucleotide microsatellite locus D21S11 (Sharma and Litt, 1992). This locus should be informative in 93% of individuals with trisomy 21 and should yield a 3-allele pattern in 42% of trisomics. For comparison, let us take trisomy 18, where only 31% of all cases are due to non-disjunction in the first meiotic division (Koehler et al., 1996) and a hypothetical microsatellite locus with the exact allele size distribution of D21S11. This hypothetical locus should be informative in 87% of patients with trisomy 18 and should yield a 3-allele pattern in only 18% of trisomic individuals.

Figure 1
- Detection of trisomy 21 by amniocentesis on the 16th week of pregnancy. One milliliter of amniotic fluid was spun and DNA extracted from the pellet. Approximately 20 ng of DNA was subjected to multiplex PCR of amelogenin and three microsatellite loci from chromosome 21: D21S1270, D21S11 and IFNAR. The upper primer of each pair was labeled with Cy5 and purified by HPLC. After PCR an aliquot of 1 ml was denatured and run on urea-PAGE on an ALF-Express automatic fluorescent DNA sequencer (Pharmacia Biotech). The data were analyzed using the Allelelinks software (Pharmacia Biotech). The upper and lower tracings show the results on a normal female fetus (note single peak at AMEL) and a male fetus (X and Y peaks at AMEL) with trisomy 21, respectively. In the former, all microsatellites on chromosome 21 are heterozygous and have normal dosage ratios that are shown within parentheses (the average dosage ratio varies with the microsatellite but is generally larger than one because the smaller allele has a slight competitive advantage in PCR amplification). In the latter, one can see a three-allele pattern at D21S11 (arrows), a two-allele pattern at D21S1270 with an abnormal gene dosage ratio of 2.6 and an uninformative single allele pattern at IFNAR. In this case the molecular diagnosis of trisomy 21 was made 24 h after the amniocentesis and was confirmed by conventional cytogenetics 9 days later.

MOLECULAR CYTOGENETICS IN PRACTICE

Diagnosis of gender and of sex chromosome aneuploidies in males

In 1991, Nakahori et al. sequenced both the amelogenin gene on the X chromosome (AMELX) and an amelogenin-like sequence on the Y chromosome (AMELY) and used this information to design primers that led to the amplification of X and Y products differing in 177 bp. Soon afterwards Sullivan et al. (1993) used these AMEL sequences to design different primers that generate PCR products of 106-base pairs and 112-base pairs for the X and Y homologues, respectively (Figure 1). In GENE - Núcleo de Genética Médica we have been using the primers designed by Sullivan et al. (1993) with a 100% success rate in diagnosing the gender in DNA samples extracted from clinical or forensic samples, including post-mortem specimens. By incorporating the fluorescent dye Cy5 during PCR, followed by electrophoresis in a computer-assisted laser densitometer (ALF-Express; Pharmacia Biotech, Uppsala, Sweden) we can easily quantify the X and Y amplification products and reliably achieve the molecular diagnosis of Klinefelter syndrome (47,XXY) or 47,XYY.

Diagnosis of trisomy 21 and application in prenatal diagnosis

Several research groups have used PCR amplification of microsatellites from chromosome 21 coupled with computer-assisted laser densitometry to diagnose Down syndrome (Mansfield, 1993; Pertl et al., 1994, 1996; Toth et al., 1998; Verma et al., 1998) especially for purposes of prenatal diagnosis. Most of these groups use the microsatellite D21S11 that is hypervariable and localized in the ideal position 21q21 (Sharma and Litt, 1992). At GENE - Núcleo de Genética Médica we have developed a multiplex protocol that allows the simultaneous amplification of three different microsatellites, D21S11, D21S1270 and IFNAR together with AMEL (Figure 1). Since the past year we have been using this routinely as a rapid screening test for all the amniotic fluid samples received for prenatal diagnosis with conventional cytogenetics. In short, in all amniotic fluid specimens 1 ml is centrifuged and DNA is extracted from the cell pellet. Using primers labeled with the fluorescent dye Cy5 during PCR followed by electrophoresis in a computer-assisted laser densitometer (ALF-Express; Pharmacia Biotech) we achieve the result of the screening test within 24 h. In the vast majority of cases this preliminary result is normal and it is then communicated to the family (with appropriate explanation that this is a screening test for trisomy 21 only). This has led to a significant decrease in the anxiety level of the parents while waiting for the definitive results of the conventional chromosome analysis that takes 8-14 days. Thus far, we have restrained ourselves from communicating abnormal screening results, preferring to wait for the final karyotype. Very recently, Verma et al. (1998) studied 2139 samples of amniotic fluid, 2083 of which were not macroscopically blood-stained and thus considered adequate for molecular cytogenetics. Using three microsatellites they correctly identified all cases of trisomy 21 in their study population (no false-positives or false negative results). It is interesting to note that they have independently reported the use of the exact same three microsatellites that we have been using at GENE - Núcleo de Genética Médica since 1997, namely, D21S11, D21S1270 and IFNAR. Based on their results, they have proposed that molecular analysis might replace conventional cytogenetics for prenatal diagnosis of trisomy 21 by amniocentesis, at least when done as a mass screening procedure (Verma et al., 1998).

Of course, trisomy 21 is not the only chromosomal abnormality seen in prenatal diagnosis. In particular, sex-chromosome aneuploidies (especially 47,XXY, 47,XYY and 47,XXX), together with other autosomal abnormalities (trisomy 18 and trisomy 13), may together be as common as trisomy 21 (Verma et al., 1998). As described above, multiplexing AMEL together with the chromosome 21 microsatellites will allow the correct diagnosis of all sex chromosome aneuploidies in males. For the diagnosis of 47,XXX an X-linked microsatellite would be necessary. Likewise, if we were to use only molecular analysis for prenatal diagnosis, we would strongly recommend the study of microsatellites from chromosomes 18 and probably from chromosome 13 also (Pertl et al., 1996; Toth et al., 1998; see Figure 2). It should be noted here that 69% of all cases of trisomy 18 and 48% of trisomy 13 are due to non-disjunction in the second meiotic division (Koehler et al., 1996) and consequently the efficiency of their detection by molecular cytogenetics will be reduced in comparison with trisomy 21.

Figure 2 -
Molecular detection of chromosomal abnormalities in 5 pregnancy losses. From top to bottom are respectively a male fetus without visible abnormalities, a male first trimester abortus with trisomy 16 (arrows indicate a three-allele pattern), a male stillborn with trisomy 18 (arrows indicate a three-allele pattern), a female second trimester abortus with trisomy 21 (arrows indicate a three-allele pattern) and a first trimester XXY abortus with triploidy. In the latter, observe the abnormal gene dosage ratios at microsatellites on chromosomes 16, 13, 21 and AMEL (within parentheses) as well as a three-allele pattern at chromosome 18.

Multiplex diagnosis of several autosomal trisomies and application in the study of fetal losses

Chromosomal defects cause 50-60% of all fetal losses in the first trimester of pregnancy (Boué and Boué, 1973; Hassold, 1986). Monosomy X (45,X) corresponds to 20-25% of all cases, triploidy to 15-20% and the several trisomies to approximately 50%. Of the latter, trisomy 16 is the most commonly seen in first trimester abortuses. As pregnancy advances, the proportion of fetal losses due to chromosomal defects drops progressively. Among stillbirths, 5-12% present chromosomal defects, with trisomies 13, 18 and 21 being the most frequent aneuploidies (reviewed by Pena et al., 1996). Based on this knowledge, at GENE - Núcleo de Genética Médica we have developed a multiplex PCR reaction for the study of DNA extracted from tissues of fetal wastage products that examines AMEL and microsatellites in chromosomes 13, 16, 18 and 21 (Figure 2). In addition, we have several back up loci for the eventuality that any of the microsatellite loci in the primary multiplex test is not informative and for confirmation of abnormal results. Moreover, when no abnormalities are seen in the screening test and the fetal sex is feminine, we then study several microsatellites from the human X chromosome, searching for a loss of heterozygosity that would indicate the presence of monosomy X (45,X; Turner's syndrome). This test will be discussed in detail in a forthcoming mini-review on the PCR-based diagnosis of chromosomal deletions using computer-assisted laser densitometry (Pena, 1998). At any rate, using this multiplex set we have been able to diagnose chromosomal abnormalities in dozens of abortuses and stillbirths that could not have been properly assessed by conventional cytogenetics because they had been frozen, fixed in formalin or alcohol, or failed to grow in culture due to maceration or bacterial contamination. Obviously the same multiplex set can be used for the very rapid diagnosis of chromosomal defects in newborns with congenital malformations, including ambiguous genitalia.

CONCLUSIONS

The molecular detection of human trisomies using PCR and computer-assisted laser densitometry has emerged as a powerful new tool for rapid cytogenetic diagnosis. It should be used as an ancillary procedure to conventional cytogenetics in prenatal diagnosis centers because it is rapid, simple and relatively inexpensive. Moreover, it can be used as the sole technique for molecular diagnosis of human chromosomal disorders in non-dividing and even in non-living human tissues, when conventional cytogenetics is not applicable. In this fashion it is especially valuable for the cytogenetic investigation of abortuses and stillbirths, but can also be used to study pathological specimens, including tumors. Many of the applications of the PCR-based detection of trisomies overlap with those of multicolor interphase fluorescence in situ hybridization (FISH; reviewed by Adinolfi and Crolla, 1994). However, besides being as reliable as FISH, it is much quicker, simpler and cheaper and has the tremendous added advantage of being easily automated (Verma et al., 1998). The PCR-based method requires very small amounts of DNA and can potentially be used for single cell diagnosis in preimplantation embryos (Sherlock et al., 1998). Moreover, it may become the method of choice for non-invasive prenatal diagnosis on fetal cells present in maternal blood, once the appropriate isolation techniques are perfected (Verma et al., 1998).

ACKNOWLEDGMENTS

The experiments with original results described in this mini-review were performed at GENE - Núcleo de Genética Médica, by Elen E.R.F. Carvalho, Márcia C.B.N. Campos, Riva P. Oliveira and Anna Izabel R. de Melo. The research reported here was fully supported by GENE - Núcleo de Genética Médica.

(Received August 4, 1998)

• Adinolfi, M. and Crolla, J. (1994). Non-isotopic in situ hybridization. Adv. Hum. Genet. 22: 187-256.
• Allingham-Hawkins, D.J., McGlynn-Steele, L.K., Brown, C.A., Sutherland, J. and Ray, P.N. (1998). Impact of carrier status determination for Duchenne/Becker muscular dystrophy by computer-assisted laser densitometry. Am. J. Med. Genet 75: 171-175.
• Boué, A. and Boué, J. (1973). Evaluation des erreurs chromo-somiques au moment de la conception. Biomedicine 18: 372-377.
• Cavalli-Sforza, L.L. and Bodmer, W.F. (1973). The Genetics of Human Populations W.H. Freeman, San Francisco.
• Celi, F.S., Cohen, M.M., Antonarakis, S.E., Wertheimer, E., Roth, J. and Shuldiner, A.R. (1994). Determination of gene dosage by a quantitative adaptation of the polymerase chain reaction (gd-PCR): rapid detection of deletions and duplications of gene sequences. Genomics 21: 304-310.
• Hassold, T. (1986). Chromosome abnormalities in human reproductive wastage. Trends Genet. 2: 105-110.
• Jeffreys, A.J. and Pena, S.D.J. (1993). Brief introduction to human DNA fingerprinting. In: DNA Fingerprinting: State of the Science (Pena, S.D.J., Chakraborty, R., Epplen, J.T. and Jeffreys, A.J., eds.). Birkhäuser, Basel, pp. 1-20.
• Koehler, K.E., Hawley, R.S., Sherman, S. and Hassold, T. (1996). Recombination and nondisjunction in humans and flies. Hum. Mol. Genet. 5: 1495-1504.
• Mansfield, E.S. (1993). Diagnosis of Down syndrome and other aneuploidies using quantitative polymerase chain reaction and small tandem repeat polymorphisms. Hum. Mol. Genet. 2: 43-50.
• Mutter, G.L. and Pomponio, R.J. (1991). Molecular diagnosis of sex chromosome aneuploidy using quantitative PCR. Nucleic Acids Res. 19: 4203-4207.
• Nakahori, Y., Hamano, K., Iwaya, M. and Nakagome, Y. (1991). Sex identification by polymerase chain reaction using X-Y homologous primer. Am. J. Med. Genet. 39: 472-473.
• Pena, S.D.J. (1998). Molecular Cytogenetics II: PCR-based diagnosis of human chromosomal deletions and microdeletions. Genet. Mol. Biol. (in press).
• Pena, S.D.J., Martins-da-Costa, H.B.B.L. and Tiritilli, S.F. (1996). Investigaçăo genética das perdas fetais. In: Medicina Fetal (Isfer, E.V., Sanchez, R.C. and Saito, M., eds.). Editora Revinter, Săo Paulo, pp. 14-21.
• Pertl, B., Yau, S.C., Sherlock, J., Davies, A.F., Mathew, C.G. and Adinolfi, M. (1994). Rapid molecular method for prenatal detection of Down's syndrome. Lancet 343: 1197-1198.
• Pertl, B., Weitgasser, U., Kopp, S., Kroiser, P.M., Sherlock, J. and Adinolfi, M. (1996). Rapid detection of trisomies 21 and 18 and sexing by quantitative fluorescent PCR. Hum. Genet. 98: 55-59.
• Sharma, V. and Litt, M. (1992). Tetranucleotide repeat polymorphism at the D21S11 locus. Hum. Mol. Genet. 1: 67.
• Sherlock, J., Cirigliano, V., Petrou, M., Tutschek, B. and Adinolfi, M. (1998). Assessment of diagnostic quantitative fluorescent multiplex polymerase chain reaction assays performed on single cells. Ann. Hum. Genet. 62: 9-23.
• Sullivan, K.M., Mannucci, A., Kimpton, C.P. and Gill, P. (1993). A rapid and quantitative DNA sex test: fluorescence-based PCR analysis of X-Y homologous gene amelogenin. Biotechniques 15: 636-638.
• Toth, T., Findlay, I., Papp, C., Toth-Pal, E., Marton, T., Nagy, B., Quirke, P. and Papp, Z. (1998). Prenatal detection of trisomy 21 and 18 from amniotic fluid by quantitative fluorescent polymerase chain reaction. J. Med. Genet. 35: 126-129.
• Verma, L., Macdonald, F., Leedham, P., McConachie, M., Dhanjal, S. and Hulten, M. (1998). Rapid and simple prenatal DNA diagnosis of Down's syndrome. Lancet 352: 9-12.
• Zimmermann, K. and Mannhalter, J.W. (1996). Technical aspects of quantitative competitive PCR. Biotechniques 21: 268-279.

Publication Dates

• Publication in this collection
  23 Feb 1999
• Date of issue
  Sept 1998

History

• Received
  04 Aug 1998