Rapid prenatal diagnosis by fluorescent in situ hybridization of chorionic villi: An adjunct to long-term culture and karyotype

Rapid prenatal diagnosis by fluorescent in situ hybridization of chorionic villi: An adjunct to long-term culture and karyotype Mark I. Evans, MD, Katherine W. Klinger, PhD, Nelson B. Isada, MD, Donna Shook, MD, Wolfgang Holzgreve, MD, Nancy McGuire, MD, and Mark P. Johnson, MD

Detroit, Michigan, Framingham, Massachusetts, and Munster, Germany OBJECTIVE: This series was designed to assess in a pilot study the feasibility of using fluorescence in situ hybridization on chorionic villi. STUDY DESIGN: We constructed probes derived from specific subregions of human chromosomes 21, 18, 13, X, and Y that give a single copylike signal when used in conjunction with suppression hybridization. RESULTS: In a blind series of 47 samples all, including one trisomy 21, were correctly identified. The samples were correctly classified as disomic for five chromosomes. CONCLUSIONS: The combination of chromosome-specific probe sets composed primarily of cosmid contigs and optimized hybridization and detection allowed accurate chromosome enumeration in uncultured human chorionic villi; these results are consistent with those obtained by traditional cytogenetic analysis and suggest a use for fluorescence in situ hybridization as an adjunct to karyotyping when rapid results are needed. (AM J OBSTET GYNECOL 1992;167:1522-5.)

Key words: Prenatal diagnosis, chorionic villus sampling, fluorescence in situ hybridization, molecular diagnosis, chromosome abnormalities

Prenatal detection of chromosomal abnormalities currently relies on analysis of banded metaphase chromosomes. Such analysis is accurate and reliable for the detection of aneuploidies and for more subtle abnormalities. However, standard cytogenetic analysis is technically demanding and can require 1 to 2 weeks to complete. Thus many people have perceived a need for simple methods for the rapid detection of chromosome abnormalities. I Fluorescence in situ hybridization has the potential to significantly decrease the time required to identify chromosomal abnormalities by allowing the analysis of interphase chromosomes. Each human chromosome occupies a discrete focal domain in the interphase nucleus. 2 • 3 The idea of using interphase chromosome analysis began with the analysis of the Barr and Y bodies to detect sex chromosome aneuploidies'" 5Ad_ vances in molecular biology have now made it possible to generate a wide variety of chromosome-specific deFrom the Division of Reproductive Genetics, Department of Obstetrics and Gynecology, Wayne State University/Hutzel Hospital, Integrated Genetics, and Westfalia Wilhelms-Universitat, Presented at the Thirty-ninth Annual Meeting of the Society for Gynecologic Investigation, San Antonio, Texas, March 18-21, 1992. Reprint requests: Mark I. Evans, MD, Director, Division of Reproductive Genetics, Department of Obstetrics and Gynecology, Hutzel Hospital/Wayne State University, 4707 St. Antoine Blvd., Detroit, MI48201. 6/6/41913

oxyribonucleic acid probes, opening up the field of interphase cytogenetics. In recent years a number of laboratories have worked to develop rapid cytogenetic assays on the basis of hybridization to cell or chromosome preparations of chromosome-specific DNA probes that can be visualized by fluorescence methods,6-13 A variety of probe sets were used in these studies, including (l) complex probes composed of the inserts from an entire chromosome library,IO-13 (2) ex-satellite repeat probes,14 and (3) composite probes composed of single-copy subclones" and single cosmids 11 and cosmid contigs." Clinical use of these assays has just begun. Considerable success has been achieved with fluorescence in situ hybridization for identification of marker chromosomes and translocations in metaphase analysis. In contrast, early attempts at aneuploidy detection in uncultured amniotic fluid cells suffered from limitations caused by probe design and assay conditions and did not result in clinically useful assays. 14, 15 More recently, we have shown that use of DNA probe sets that are based on cos mid contigs that are chromosome specific, have high signal-to-noise ratios, and have good spatial resolution of the fluorescent signals, coupled with careful attention to sample processing conditions, allows efficient prenatal detection of chromosomal aneuploidies in uncultured amniotic fluid cells. 11, 16 However, amniocentesis is only one of the methods used to obtain fetal samples for

Fluorescent in situ hybridization of chorionic villi

Volume 167 I'\umber 6

1523

Table I. Results of 47 chorionic villus samples hybridized to five chromosome probes

Male Female

x

y

21

18

1 (97.0%) 2 (95.3%)

1 (95.5%) 0(98.0%)

2 (96.4%) Combined

Combined

genetic analysis. Chorionic villus sampling is widely used to obtain cells for prenatal diagnosis of genetic abnormalities in the first trimester of pregnancy. We herein report the results of a preliminary study in which, with the same probe set described above, we were able to detect the major clinically relevant chromosomes present in uncultured chorionic villi and thereby demonstrate the potential clinical use of fluorescence in situ hybridization for prenatal diagnosis in the first trimester. Material and methods

Probe set development labeling and hybridization have been described in great detail elsewhere l6 and will not be repeated at length here. For this study a single probe set was used for each chromosome that mapped to the following regions: 13q13 (DI3S6), ISq22-qter (MBP), and 21q22.3 (D21S71). Cosmid contigs were expanded as necessary to achieve a total sequence complexity of ~ 60 kbp; the chromosome 21 probe set is a three-cos mid contig containing SO kb of nonoverlapping DNA, and the chromosome IS probe set is a three-cosmid contig containing 109 kb of nonoverlapping DNA, whereas the chromosome 13 probe set is a three-cosmid contig containing - 97 kb of nonoverlapping DNA. The specificity of the probe sets was verified by fluorescence in situ hybridization to metaphase spreads. Chromosome X-specific cosmids were identified from a cosmid library constructed from DNA of a human-rodent somatic cell hybrid that contained chromosomes 13, 19, and X as the sole human components. A cos mid containing a pericentromeric repeat sequence was used as the X probe in this study. The Y probe used in this study was pDP97, a repetitive clone developed by D. Page (Whitehead Institute of Biomedical Research, Cambridge, Mass.). The specificity and sensitivity of individual cosmid clones and cosmid contigs was rigorously monitored and evaluated by fluorescence in situ hybridization against both metaphase spreads and interphase nuclei with short-term blood cultures and uncultured amniotic samples, respectively, during the development of the autosomal probe sets. Villi were dissociated with trypsin and collagenase to yield a uniform suspension and processed for hybridization as previously described. 16 Results

The chorionic villi analyzed in this study primarily represented excess tissue remaining after the sample

2 (94.9%)

13

2 (97.5%)

Combined

requirements for cytogenetic analysis were met. Fortynine samples were analyzed with the probes for 21, IS, 13, X, and Y. The number of nuclei in each sample that displayed zero, one, two, three, or four hybridization signals in each sample were recorded. Signal distribution was calculated as a percent for each sample, and the mean domain distribution was calculated for each probe set. The hybridization efficiency approached 100% in all hybridized samples. As shown in Table I, 96.4% of the hybridized nuclei in any disomic sample displayed two signals when analyzed with one of the autosomal probe sets. In contrast, one and three hybridization signals were detected in 3.1% and 0.6% of the nuclei, respectively. A similar distribution was seen for the sex chromosomes in which 95.3% of female nuclei showed two hybridization signals with the X chromosome probe, and one signal was seen each for the X and Y probe in 97.0% and 95.5%, respectively, of male nuclei. Trisomy 21 was detected in one sample and confirmed by cytogenetic analysis. However, it should be noted that one sample failed to hybridize to any probe, whereas four samples failed to hybridize to a single probe. The reason for these hybridization failures is currently unknown but is under investigation. Comment

We have demonstrated that fluorescence in situ hybridization provides efficient and accurate prenatal detection of the five relevant chromosomal probes in uncultured cells from chorionic villi. In this preliminary study the one aneuploidy tested was easily identified. In addition, we determined the frequencies with which each hybridization pattern would occur in normal and trisomic samples, which allowed us to begin to establish baseline performance criteria for the assay. Our study shows that high efficiencies can be achieved with uncultured chorionic villi used as target cells. Hybridization efficiency and the extent to which the hybridization pattern reflects the correct genotype are a product of probe design, hybridization efficiency, and signal detection. The impact of parameters such as sample fixation, cell permeability, probe size, and complexity may vary with cell type, as has been previously shown for lymphocytes, uncultured amniocytes, and tumor tissue. I. 6. 7. II. 15. 17. 19. 20 In the current limited study hybridization efficiencies obtained with uncultured chorionic villi were higher than those obtained with uncultured amniocytes. This result is similar to

1524 Evans et al.

that seen in studies comparing Barr body detection in buccal epithelium and hair root cells and most likely represents the difference in analyzing a population of healthy, viable cells compared with a population of terminally differentiated epithelial cells, many of which are shed decidua. It is obvious that high hybridization efficiencies must be achieved for in situ hybridization assays to have clinical use in prenatal diagnosis because the hybridizationdetection efficiency of the assay greatly impacts the ability to detect the third signal in trisomic cells and therefore to accurately diagnose trisomies. If the aggregate probability of detecting a target chromosome is 0.9, then the probability of detecting two chromosomes in a nucleus is 0.9 x 0.9, or 0.81, while the probability of detecting three chromosomes is 0.9 x 0.9 x 0.9, or 0.729. This study shows that high efficiencies can be achieved with uncultured chorionic villi used as target cells. The most common chromosomal abnormalities in newborns are trisomy 21, with an incidence of 1 in 800; trisomy 18, with an incidence of 1 in 8000; trisomy 13, with an incidence of 1 in 20,000; monosomy X (Turner syndrome), with an incidence of 1 in 10,000; and other sex chromosome aneuploidies, with a combined incidence of 1 in 1000!' Together aneuploidies of the five chromosomes studied in this report can account for up to 95% of chromosome abnormalities in live births accompanied by birth defects in the child!2 and 67% of all chromosomal abnormalities, if balanced translocations are included. We therefore designed and tested probe sets targeted to these five chromosomes. The same hybridization conditions yield equivalent performance characteristics for all five probe sets, allowing them to be used for multicolor analysis when combined with multicolor fluorescence."' Probe sets for other target chromosomes could be designed, broadening the number of abnormalities detected by the assay. Methods that allow rapid and accurate detection of fetal aneuploidies provide additional time for thoughtful consideration of the test results and can aid in clinical decision-making when fetal abnormalities are seen on ultrasonography, when preterm labor occurs, etc. Analysis of interphase chromosomes with fluorescence in situ hybridization is extremely rapid. Given that there are two technologies for extremely rapid results (fluorescence in situ hybridization and direct chorionic villus sampling), which is likely to be better? Direct chorionic villus sampling has the advantage of a complete karyotype, albeit of poor quality, but most experienced centers would be extremely cautious about acting clinically on a nontypical abnormality on the basis of a direct chorionic villus sampling alone. Fluorescence in situ hybridization will certainly miss those unusual karyotypes, and its main question will center on the development of thorough statistics as to its reliability. Much more data are needed. However, the results of the current preliminary study support the use of fluo-

December 1992 Am J Obstet Gynecol

to enhance standard cytogenetics, allowing accurate identification of trisomic chromosome constitution in uncultured chorionic villi in significantly less time. Analysis of the complete karyotype ensures that less common chromosomal abnormalities will also be detected. REFERENCES 1. Lichter P, Jauch A, Cremer T, Ward DC. Detection of Down syndrome by in situ hybridization with chromosome 21 specific DNA probes. In: Patterson D, ed. Molecular genetics of chromosome 21 and Down syndrome. WileyLiss: New York, 1990:69-78. 2. Cremer T, Cremer C, Baumann H, et al. Rabl's model of the interphase chromosome arrangement tested in Chinese hamster cells by premature chromosome condensation and laser-UV-microbeam experiments. Hum Genet 1982;60:46-56. 3. Hens L, Baumann H, Cremer T, Sutter A, Cornelis J], Cremer C. Immunocytochemical localization of chromatin regions UV-microirradiated in S-phase or anaphase: evidence for a territorial organization of chromosomes during the cell cycle of Chinese hamster cells. Exp Cell Res 1983; 149:257-69. 4. Barr ML, Bertram EG. A morphological distinction between neurones of the male and female, and the behaviour of the nucleolar satellite during accelerated nucleoprotein synthesis. Nature 1949;163:676-7. 5. Pearson PL, Bobrow M, Vosa CG. Detection of chromosome aberrations in the human interphase nucleus by visualization of specific target DNAs with radioactive and non-radioactive in situ hybridization techniques: diagnosis of trisomy 18 with probe Li.84. Hum Genet 1986;74:34652. 6. Julien C, Bazin A, Guyot B, Forestier F, Feffos F. Rapid prenatal diagnosis of Down's syndrome with in situ hybridization of fluorescent DNA probes. Lancet 1986;2:863-64. 7. Lichter P, Cremer T, Tang C-JC, Watkins PC, Manuelidis L, Ward DC. Rapid detection of human chromosome 21 aberrations by in situ hybridization. Proc N ad Acad Sci U SA 1988;85:9664-8. 8. Pinkel D, LandegentJ, Collins C, et al. Fluorescence in situ hybridization with human chromosome-specific libraries: detection of trisomy 21 and translocations of chromosome 4. Proc Nad Acad Sci USA 1988;85:9138-42. 9. Cremer C, Lichter P, Borden J, Ward DC, Maneulidis L. Detection of chromosome aberrations in metaphase and interphase tumor cells by in situ hybridization using chromosome-specific library probes. Hum Genet 1988;80: 235-6. 10. WiegantJ, Ried T, NederlofPM, Van der Ploeg M, Tanke HJ, Raap AK. In situ hybridization with fluoresceinated DNA. Nucleic Acids Res USA 1991;19:3237-41. 11. Klinger KW, Harvey R, Dackowski W, et al. Interphase cytogenetics: improved prenatal detection of aneuploidy in uncultured fetal cells. New results from a large blinded study. Am J Hum Genet 1991;49:23. 12. Klinger KW, Harvey R, Dackowski W, et al. Prenatal analysis of chromosomal abnormalities using fluorescent in situ hybridization. In: Proceedings of the first bioltechnology winter symposium, 1991:95. 13. Klinger KW, Dackowski W, Leverone B, et al. Prenatal detection of aneuploidy of 21, 18, 13, X or Y by interphase in situ hybridization. Am] Hum Genet 1990;47:A224. 14. Kuo W-L, Tenjin H, Segraves R, Pinkel D, Golbus MS, Gray J. Detection of aneuploidy involving chromosomes 13, 18, or 21 by fluorescence in situ hybridization (FISH) to interphase and metaphase amniocytes. Am J Hum Genet 1991;49:112-9. 15. Tkachuk DC, Pinkel D, Kuo W-L, Weier H-U, Gray]W. Clinical applications of fluorescence in situ hybridization. Genet Anal Tech Appl 1991;8:67-74. 16. Klinger KW, Lande~ G,.Shook D, t;t al. ~apid.detectio~ of

Volume 167 Number 6

fluorescence in situ hybridization (FISH). Am J Hum Genet 1992;51:55-65. 17. Yu L-C, Bryndorf T, Christensen B, et al. Enumeration of specific chromosomes in uncultured fetal cells using fluorescence in situ hybridization (FISH). Am J Hum Genet 1990;47:A1l32. 18. Kuo W-L, Tenijin H, Segraves R, Pinkel D, Golbus MS, Gray J. Detection of aneuploidy involving chromosomes 13, 18, or 21 by fluorescence in situ hybridization (FISH) to interphase and metaphase amniocytes. Am J Hum Genet 1991;49:112-9. 19. Jordan CA. In situ hybridization in cells and tissue sections: a study of myelin gene expression during CNS myelination and remyelination. In: Chesselet M-F, ed. In situ hybridization histochemistry. Boca Raton, Florida: CRC, 1990:39-70.

Fluorescent in situ hybridization of chorionic villi

20. Lichter P, Boyle AL, Cremer T, Ward DC. Analysis of genes and chromosomes by nonisotopic in situ hybridization. GATA 1991;8:24-35. 21. Drugan A, Johnson MP, Evans Ml. Principles of inheritance. In: Evans MI, ed. Reproductive risks and prenatal diagnosis. Norwalk, Connecticut: Appleton & Lange, 1992:3-23. 22. Whiteman DAH, Klinger K. Efficiency of rapid in situ hybridization methods for prenatal diagnosis of chromosome abnormalities causing birth defects. Am J Hum Genet 1991 ;49:AI279. 23. Klinger KW, Landes G, Dackowski W, et al. Multicolor fluorescence in situ hybridization for the simultaneous detection of probe sets for chromosomes 13, 18,21, X and Y in uncultured amniotic fluid cells. Hum Mol Genet 1992;1:307-13.

The smaller than expected first-trimester fetus is at increased risk for chromosome anomalies Arie Drugan, MD, Mark P. Johnson, MD, Nelson B. Isada, MD, Wolfgang Holzgreve, MD, Ivan E. Zador, PhD, Mitchell P. Dombrowski, MD, Robert J. Sokol, MD, Mordechai Hallak, MD, and Mark I. Evans, MD Detroit, Michigan, and Haifa, Israel OBJECTIVE: Intrauterine growth retardation associated with fetal chromosome anomalies is usually documented on ultrasonography late in the second trimester. However, we believe and attempt to document here that the impact of aneuploidy on fetal growth is evident much earlier (Le., the aneuploid fetus may appear smaller than dates on ultrasonography even in the first trimester). STUDY DESIGN: For the population referred to our center for chorionic villus sampling from January 1988 to July 1991, we compared gestational age as calculated from the last menstrual period to that derived from fetal size as measured by crown-rump length. A cutoff of 7 days was chosen to select the study group. The remainder of our chorionic villus sampling population in which fetal size was expected was used as controls. We also divided those chorionic villus sampling patients by when a fetal death was observed by size. RESULTS: In the study period 3194 chorionic villus sampling procedures were performed and in 277 (8.7%) fetal length was smaller than expected by at least 7 days. Sixty (1.9%) chromosome anomalies were diagnosed by first trimester chorionic villus sampling in the study period. The frequency of chromosome anomalies was 4.3% in the study group and 1.7% in controls (p < 0.004). The more aberrant the karyotype on "postmortem chorionic villus sampling," the greater the growth retardation tended to be. CONCLUSIONS: In our chorionic villus sampling population a fetal crown-rump length smaller than dates is associated with a significant increase in risk of chromosome anomalies. Moreover, the larger the size-dates discrepancy, the higher the possibility that the aneuploidy affecting that pregnancy is of the severe or lethal type. (AM J OSSTET GVNECOL 1992;167:1525-8.)

Key words: Chromosome anomaly, first trimester, fetal size From the Center for Fetal Diagnosis and Therapy, the Departments of Obstetrics and Gynecology and Molecular Biology and Genetics, Hutzel Hospital/Wayne State University School of Medicine. Presented at the Thirty-ninth Annual Meeting of the Society for Gynecologic Investigation, San Antonio, Texas, March 18-21,1992. Reprint requests: Mark I. Evans, MD, Director, Division of Reproductive Genetics, Department of Obstetrics and Gynecology, Hutzel Hospital/Wayne State University, 4707 St. Antoine Blvd., Detroit, MI48201.

The diagnosis of symmetric intrauterine growth retardation (IUGR) in the second trimester indicates the need for further invasive diagnostic testing (e.g., amniocentesis, late chorionic villus sampling, cordocentesis) to exclude fetal chromosome anomalies. Up to 25% of fetuses with severe early onset IUGR are aneuploid, and such a diagnosis has a major impact on pregnancy