Improved prenatal detection of fra(X)(q27.3): methods for prevention of false negatives in chorionic villus and amniotic fluid cell cultures

The reliable detection of fra(X)(q27.3) in prenatal samples is important for providing genetic counseling. We have identified 5 new cases of prenatal fragile X [fra(X)] detection in 3 chorionic villus sample (CVS) and 2 amniotic fluid (AF) cell cultures. In 4 of the 5 cases, either excess thymidine (THY) or a combination of THY and 5-fluorodeoxyuridine (FUdR) was clearly superior to FUdR alone as fra(X) inducers. Amniocytes from one case were cultured only in RPMI-1640 and later exposed to FUdR or THY separately. They showed only 2% fra(X) while parallel cultures initiated in Chang medium and incubated in RPMI for at least 7 days (recovery) before fra(X) induction exhibited strikingly increased fra(X) frequencies. Chang medium alone will not allow fra(X) induction in AF (Jenkins EC, Brown WT [1986]: "Genetic Disorders and the Fetus: Diagnosis, Prevention and Treatment." New York: Plenum Press, pp 185-204). Now, using CVS cells, we report that only 1% and 0% fra(X) were detected using FUdR or THY in cells cultured in RPMI for 4 days after removal from Chang medium. Cells with 7 days "recovery" in RPMI exhibited increases from 2 to 6%. Therefore, we have found that Chang medium is very helpful when the appropriate recovery time in another medium is allowed before fra(X) induction. Some false negative reports can be attributed to: induction in Chang medium alone; lack of sufficient recovery time after initiating cells in Chang before induction; and unavailability of the excess THY fra(X) induction system.(ABSTRACT TRUNCATED AT 250 WORDS)     

Prenatal diagnosis of the fragile X syndrome: possible end of the experimental phase for amniotic fluid.

We have had experience with 260 prenatal diagnosis cases for the fragile X syndrome [fra(X)]; amniotic fluid was received in 230. There was a documented family history of fra(X) in 148 amniotic fluid cases. Our sample includes 91 males. Eleven were correctly identified as fra(X)-positive and 2 were false-negative. Eight of 57 females were fra(X) positive and one was a false-negative. CVS were received in 21 cases with a family history of fra(X), and there were 2 positive results in females and 3 false-negative results in males which were ultimately detected by means of molecular analysis or a subsequent amniotic fluid specimen. RFLPs were utilized in 29 cases (amniotic fluid and CVS); RFLPs identified 2 false-negative cytogenetic results in CVS. Two male fetuses were found to have a high probability of being affected by means of RFLPs, but on the basis of prenatal and postnatal negative fra(X) cytogenetic results and subsequent normal growth and development, they are either unaffected transmitting males or are double recombinants. Three female fetuses were also found to be cytogenetically negative in CVS but had a 90%, 93%, and 99% probability of being affected by RFLPs. On the basis of the data, it can be concluded: 1. Amniotic fluid experience is adequate to eliminate the "experimental" designation providing the limitations are understood and an experienced laboratory is involved. 2. Chorionic villus cells for cytogenetic analysis should still be considered experimental. 3. Negative results with CVS should be confirmed by molecular methods and/or by cytogenetic analysis of another tissue. 4. Multiple approaches can maximize reliability of fra(X) prenatal diagnosis.     

Prenatal cytogenetic diagnosis of the fragile X syndrome in amniotic fluid: calculation of accuracy.

We have completed over 350 prenatal diagnoses for the fragile X [fra(X)] syndrome using amniotic fluid, chorion villus specimen (CVS), fetal blood sampling and molecular methods. A total of 300 amniotic fluid specimens have been received for prenatal diagnosis of the fra(X) syndrome. There was a documented family history of fra(X) in 170/300 amniotic fluid cases, and 23/170 were correctly identified as cytogenetically fra(X) positive (16 male; 7 female). Three males were false-negative, and one female was fra(X) negative but identified as a probable carrier by RFLPs. No fra(X) positive or false-negative results were found in the absence of a fra(X) family history. Because the a priori risk for the fra(X) syndrome for each pregnancy was different and widely variable, the determination of the accuracy of the prenatal diagnosis results requires a consideration of these variables. On this basis, the calculated accuracy of prenatal cytogenetic diagnosis for the fra(X) syndrome is approximately 97%. This accuracy can be improved further with the simultaneous use of molecular methods, especially in view of recent developments.     

Experience with multiple approaches to the prenatal diagnosis of the fragile X syndrome: amniotic fluid, chorionic villi, fetal blood and molecular methods

We have had experience with 160 prenatal diagnosis cases for the fragile X syndrome [fra(X)] or Martin-Bell Syndrome. In 140, amniotic fluid was utilized; 98 had a documented family history of fra(X). The 94 completed cases included 4 no growth; 56 males of which 7 were fra(X)-positive and 2 false-negative; 38 females of which 5 were fra(X) positive. There was no fra(X) positive result when a family history of mental retardation was not documented as fra(X). Molecular methods (RFLPs) were utilized in 10 amniotic fluid and 5 chorionic villus specimens (CVS). Percutaneous umbilical blood sampling was used in 2 negative cases and 1 fra(X) positive case because of timing, tissue culture failure or confirmation of another method. CVS were received in 13 cases, and RFLPs were utilized in 5 of the CVS cases. There was no positive fra(X) CVS chromosome result in males, 1 positive result in a female, but 2 false negatives were detected by RFLPs. On the basis of the results, it can be concluded that cytogenetic and molecular methods are complementary and best used together and that multiple approaches can enhance the efficiency and reliability of fra(X) prenatal diagnosis.     

Second trimester prenatal diagnosis of the fragile X

The fra(X) chromosome was detected in 5 samples of amniotic fluid cells in a series of 23 pregnancies at risk. The prenatal results were confirmed in 2 male abortuses, one with a relatively high and one with a very low frequency of expression in both amniocytes and fetal tissue. In a third male fetus with low expression in amniocytes, the fra(X) was not detected in the fetal tissues tested. In another male with low expression in amniocytes the fra(X) was not detected after birth. In one female with a low expression in amniocytes, a very high frequency (28%) was detected in cord blood after birth. Low expression of the fra(X) was found in a 4-year-old normally developed girl, where the prenatal results had been negative. In 4 males and 4 females the negative prenatal diagnoses were confirmed after birth. This study indicates that prenatal diagnosis of the fragile X after amniocentesis may be complicated, either due to technical problems related to the use of amniotic fluid cells, or due to genetic heterogeneity, or both. Part of this heterogeneity could be due to the existence of normal male transmitters. Also, it seems that the frequency of expression in amniocytes from female carriers can not be used for the prediction of the frequency in blood after birth.     

The prenatal detection of the fragile X chromosome: review of recent experience

The fragile X chromosome has been identified in specimens from 17 male and 10 female fetuses in 11 laboratories throughout the world, obtained from at least 79 fetuses at increased risk for the fra(X) syndrome. Of these, 19 were confirmed, 6 were pending, 1 was negative and 1 could not be confirmed. Twenty-five of the 79 cases were studied in our laboratory (Institute for Basic Research [IBR]) and resulted in fra(X) demonstration in specimens from 3 male and 5 female fetuses. All 3 males and 2 of the 5 females have been confirmed. When amniocytes from the two confirmed female fetuses were exposed to FUdR after culturing in Chang medium, fra(X) frequencies were virtually negative indicating that Chang medium should not be used in fragile X studies at least when FUdR is used to induce fragility. Finally, amniocytes from a fra(X) male fetus studied in 3 different laboratories exhibited strikingly different frequencies. To date, we have experienced no false-positives or negatives, but the latter case was controversial. It is recommended that laboratories undertaking fra(X) prenatal detection use a combination of at least two different proven induction systems as well as complementary DNA marker studies to prevent false negative diagnosis.     

Distribution of autosomal fragile sites in specimens cultured for prenatal fragile X diagnosis

We reviewed the distribution of autosomal fragile sites (FS) and spontaneous chromosome breaks or gaps (CB) at chromosome locations other than those recognized as FS from 100 amniotic fluid samples (AF), 19 chorionic villus samples (CVS), and 5 percutaneous umbilical blood samples (PUBS) referred for fragile X [fra(X)] analysis. We present data on the degree of expression of autosomal fragility in AF, CVS, and PUBS samples, and the relationship between degree of expression and induction system. The most common observed FS were: 3p14, 9p32, and 6q26 in AF; 9q32, 3q27, and 8q22 in CVS; and 3p14, Xq22, and 16q23 in PUBS cases. Distribution of FS and CB, when compared by induction system, was not found to be identical. Our data also indicate that the presence of any particular FS cannot be used as an indicator for the effectiveness of the fra(X) induction system in prenatal samples.     

Prenatal cytogenetic diagnosis of the fragile X chromosome: feasibility and speed of in situ clonal method in amniotic fluid cell tissue culture.

We have had experience with over 300 amniotic fluid specimens for prenatal diagnosis for the fragile X chromosome [fra(X)], and the flask method of tissue culture has been routinely utilized requiring extended tissue culture periods of 3-4 weeks. The use of the in situ clonal method of tissue culture for routine prenatal cytogenetic diagnosis of amniotic fluid cells has shortened tissue culture time and resulted in more rapid reporting; however, it has not been widely employed for fra(X) prenatal diagnosis. The simultaneous use of both methods of tissue culture has resulted in the detection of 2 cytogenetically fra(X) positive cases in amniotic fluid, with more rapid reporting and satisfactory expression of the fra(X) with the in situ clonal method. Thus, the use of the in situ clonal method of tissue culture for fra(X) prenatal diagnosis in amniotic fluid cells is feasible, faster and can serve as a more rapid cytogenetic adjunct to the newer DNA testing methods.     

Prenatal detection of fra(X)(q27.3) in female identical twins: reliability of low level cytogenetic prenatal expression in females.

Recently, we detected fra(X)(q27.3) in amniocyte cultures from female identical twins. The pregnant woman did not exhibit fra(X)(q27.3) in whole blood cultures but was the sister of 2 affected brothers. DNA marker analyses showed that she was a carrier of FRAXA. Amniotic fluid cultures (AFCs) from twins A and B exhibited the fragile X [fra(X)] chromosome, but the level of cytogenetic expression was very low in twin A's AFCs. DNA marker studies indicated both twins were carriers of FRAXA. Peripheral umbilical blood sample (PUBS) cultures exhibited fra(X)(q27.3) at a frequency of about 10% for both twins. DNA fingerprinting indicated that the twins were identical, confirming the clinical impression, with a very thin separating amniotic membrane. To our knowledge, this is the only report of prenatal fra(X)(q27.3) detection in female identical twins, and the second report of identical twin detection [Rocchi et al., 1985]. We have diagnosed prenatally fra(X)(q27.3) in 5 female fetuses using AFCs. The average fra(X) frequency was 4% for these positive female fetuses with a range of 0.5% to 8.5%. Follow-up whole blood studies confirmed our original results at an average fra(X) frequency of 25%. In conclusion: 1. Low frequencies, perhaps 1 or 2%, or a few positive cells in AFCs, are likely to increase in magnitude when confirmed in whole blood cultures either pre- or postnatally. 2. It appears likely that the risk is low for false positive results in AFCs when low frequencies of fra(X)(q27.3) are encountered.     

The fragile X(q27) form of X-linked mental retardation: FUdR as an inducing agent for fra(X)(q27) expression in lymphocytes,fibroblasts, and amniocytes

The effect of FUdR on the expression of fra(X)(q27) was examined in lymphocytes and/or fibroblasts from 16 affected males and 5 carriers from 10 families; six different culture media were used: F10, 5% serum, pH 7.3(37°C); medium 199, 5% serum, pH 7.6(37°C); folate-free 199, 5% serum, pH 7.6(37°C), and these three media with FUdR (0.05 μm). In lymphocytes there was no significant difference in the percentage of fra(X) expressing cells between any of the FUdR-containing media. The highest percentage of expressing cells seen in lymphocytes with FUdR was 56%. The average enhancement in males with FUdR in the 199 and folate-free 199 media was 30%. This relative enhancement with FUdR was very much higher in a few blood specimens delayed in transit and FUdR may prevent some of the false-negative results obtained from mailed specimens. FUdR did not induce the marker in four obligate carriers with previously negative results. The fibroblasts from affected males were grown in the six specific media for the last 48 hr. Two of the six media yielded reproducibly positive results. These were 199-FUdR and folate-free 199-FUdR with mean percentages of expressing cells of 12.8 ± 7.1% and 11.3 ± 6.1%, respectively. F10-FUdR, which contains thymidine, did not permit expression of the marker in fibroblasts and there was no difference in the percentage of fra(X) expression in 199-FUdR media with or without folate. It was concluded that FUdR shows promise as an agent to permit prenatal diagnosis of the condition and to enhance the detection of the marker in lymphocyte cultures.     

Prenatal diagnosis and carrier detection in fragile X.

Prenatal diagnosis was performed in 81 cases at risk for the fragile X syndrome. There were 12 fra(X)-positive cases, two of which showed low expression in cultured amniotic fluid cells. FUdR and high thymidine were used for induction of fra(X) (q27.3) expression in all cases. In 21 cases linkage studies were performed, 7 with probes for the loci DXS52, DXS98 and DXS105, 13 with probes for DXS369 and DXS296, DXS304 or DXS374 and one with the probe Do33 for DXS465. In 11 of these cases linkage analysis gave risk figures higher than 95% or lower than 5%, all in concordance with the cytogenetic findings. Discordance was found in three cases studied earlier, the two cases with low expression mentioned above and one cytogenetically normal case, which were now restudied with the new probes. RFLP-studies and linkage analysis was also performed for 24 cytogenetically fra(X)-negative females having a 50%, 25% or 12.5% risk of being carriers according to pedigree data. In 15 cases the risk dropped to 1% or less. Six of these women were pregnant and had asked for prenatal diagnosis but after genetic counseling prenatal diagnosis was avoided.     

Dialyzed fetal bovine serum increases cytogenetic fragile X expression.

Short-term whole blood cultures from 9 unrelated male individuals with the fragile X [fra(X)] syndrome were exposed to 5-fluorodeoxyuridine (FUdR). The fra(X) frequency was higher in 8 of 9 cases where the complete medium contained dialyzed fetal bovine serum (DFBS). In 3 of the cases, the fra(X) frequency nearly tripled (e.g., 12/100 to 33/100) while in 2 others, it nearly doubled (e.g., 15/100 to 29/100). When DFBS cultures from 2 other fra(X) individuals were exposed to increasing folic acid concentrations ranging from 2 to 4,000 x 10(-6) M, there was virtually no change in fra(X) expression. In 6 of 9 DFBS cultures, the mitotic index decreased, and it increased in 3. Therefore, although the fra(X) frequency increased, in most DFBS cultures the mitotic index decreased. Whether the reduction in mitotic index indicates an inverse correlation between reduced mitotic index and increased fra(X) expression, at least in cultures from some individuals, will be determined by additional studies. In conclusion: (1) medium supplementation with dialyzed fetal bovine serum should be considered when using FUdR for fra(X) identification in order to avoid potentially false negative results; (2) there appears to be no direct correlation between increased mitotic index and increased fra(X) expression in whole blood cultures; (3) increased folic acid concentrations do not affect fra(X) expression when FUdR fra(X) induction is employed; therefore requesting people to refrain from taking vitamins, including folic acid, before fra(X) testing (a practice that still persists in some places) appears unnecessary.     

Frequency of tri- or multiradial configurations in fragile X identification

Using the FUdR system for fragile X induction, we have observed no triradial or bisatellited configurations at fra (X) (q27.3) in over 5,000 fra(X) chromosomes examined from over 150 fra(X) individuals. Based on our observations, and those of Turner and Jacobs (1983) and Daniel et al (1984), we hypothesize that triradial configurations may not occur at Xq27 with FUdR induction. To test this hypothesis we cultured whole blood simultaneously in parallel folate-deficient and FUdR fra(X) induction systems, and systematically examined fra(X) chromosomes for triradials. Neither autosomes nor X chromosomes exhibited any apparent triradial figures in the FUdR system, while 1.4% of the fra(X) chromosomes in TC 199 exhibited a triradial. Also we observed one autosomal triradial at 4q35. We conclude that triradial configurations occur in low frequencies in the folate deficient system and seldom if ever in the FUdR system.     

Fra(X)(q27.2), the common fragile site, observed in only one of 760 cases studied for the fragile X syndrome.

Cell cultures from 760 whole blood, amniotic fluid, chorionic villus sample, and peripheral umbilical blood sample specimens were exposed to multiple fra(X)(q27.3) induction systems (none had aphidicolin). Fifty-three exhibited the rare fragile site, fra(X)(q27.3) or FRAXA, none of which demonstrated the common fragile site or FRAXD at band Xq27.2. Only one cell in one of the negative whole blood FUdR-treated cultures from a mentally retarded male showed FRAXD. Therefore, it appears that FRAXD occurs very rarely in cultures treated to induce FRAXA since only one positive cell was observed in over 88,000 analyzed. It appears that very low frequencies of fra(X)(q27) can be accounted for only in part by the presence of the common fragile site since only one of 9 cases, each with one fra(X)(q27) positive cell, exhibited FRAXD and the others were FRAXA. After confirmation of FRAXA with direct DNA testing in a large number of low frequency cases, it should be possible to rely on the detection of very low frequencies of fra(X)(q27.3), e.g., 1% with at least 2 positive cells.     

Fragile X expression increased by low cell-culture density

The effect of cell density on expression of fra(X) was studied. A lymphoblast cell line from one fra(X) individual and whole blood from another individual was tested at various cell densities using RPMI-1640 with FUdR (0.1 microM) 24 hrs before harvest. Densities from 0.25 X 10(5) to 2 X 10(6) cells/ml were tested. Chromosomes were G-banded and analyzed for fra(X) frequency. Increased density caused fra(X) frequency to decline in lymphoblasts and whole blood. In the established line low density fra(X) frequency was 51.2% and decreased to 6.5% at the high density. In blood fra(X) frequency was 34.7% at low density and decreased to 18% fra(X) in high density. We suspect that decay of the FUdR effect may explain the results. Our results suggest that to maximize fra(X) frequency, cultures should be inititated at low density. This may be important in analysis of low-percentage fra(X) patients, nonexpressing female carriers, and obligate nonpenetrant males.     

Progress toward an internal control system for fragile-X induction by 5-fluorodeoxyuridine in whole-blood cultures

We have been attempting to develop a consistently reliable internal control to assure the effectiveness of the 5-fluorodeoxyuridine (FUdR) fragile-X [fra(X)] induction system. We carried out a systematic study of whole-blood specimens cultured from 56 individuals from two different laboratories. An analysis of nearly 9,000 cells demonstrated: (1) the importance of establishing baseline levels of fragile sites in each laboratory, and (2) that a combination of common fragile sites (different for each laboratory) could serve as a consistently reliable indicator of the effectiveness of the FUdR fra(X) induction system. It was suggested that a non-FUdR culture(s) should be incorporated into a laboratory's fra(X)-screening protocol, so that if there are any doubts about the effectiveness of the FUdR system a comparison to background or spontaneously occurring fragile sites can be made within the laboratory. Repeat cultures are recommended where no increase in common fragile-site frequency is observed in the FUdR induction system, and where fra(X) was strongly suspected but not found. In addition, the necessity of using more than one fra(X) induction system in whole-blood cultures was demonstrated, including the effectiveness of an FUdR/excess thymidine double-induction system. Finally, 2 cases of apparent mosaicism for Klinefelter syndrome in fra(X) individuals were observed.     

Variability of thymidylate synthase activity in whole blood cultures treated with FUdR

Induction of some fragile sites including fragile X [fra(X)] depends on the depletion of thymidine monophosphate (TMP) from the culture medium. This can be accomplished by use of inhibitors such as 5-fluorodeoxyuridine (FUdR) and by culturing cells in medium deficient in folate and TMP. FUdR inhibits the activity of thymidylate synthase (TS), thereby depleting cells of TMP. To determine the degree of FUdR inhibition of TS under routine cytogenetic culture conditions, we modified the tritiated dUMP TS method for use in short-term whole blood cultures stimulated with phytohemagglutinin. TS inhibition was highly variable across whole blood cultures from 30 individuals exposed to FUdR during the last 24 hours of a 4 day culture. If an additional dose of FUdR was added 12 hours before harvest, TS inhibition usually increased. These findings have a potential impact on the use of FUdR for the diagnosis of the fra(X) syndrome.     

Prenatal diagnosis of the fra(X) syndrome

The fra(X) syndrome is one of the most common causes of mental retardation, and validation of the reliability and feasibility of making the prenatal diagnosis of this disorder is important for genetic counseling and prevention. We have received a total of 74 amniotic fluid specimens for prenatal diagnosis of fra(X) from worldwide sources. Results were obtained on 68 specimens of which 43 had a documented family history of the fra(X) syndrome. Of the 43 specimens, 23 were male and 4 were prenatally diagnosed as being affected. On the basis of the results, several conclusions follow: 1.) At least 3 different tissue culture methods should be utilized. 2.) At least 150 cells should be scored, preferably 50 from each of 3 different tissue culture methods or 100 from each method if less than 3 methods are used. 3.) While the test appears to be reliable, it should still be considered to be experimental until larger numbers are obtained with completed follow-up of cases.     

Analysis of parent of origin specific DNA methylation at SNRPN and PW71 in tissues: implication for prenatal diagnosis.

Prader-Willi syndrome (PWS) and Angelman syndrome (AS) are distinct developmental disorders caused by absence of paternal or maternal contributions of the chromosome region 15q11-q13, resulting from deletions, uniparental disomy (UPD), or rare imprinting mutations. Molecular cytogenetic diagnosis is currently performed using a combination of fluorescence in situ hybridisation (FISH), DNA polymorphism analysis, and DNA methylation analysis. Only methylation analysis will detect all three categories of PWS abnormalities, but its reliability in tissues other than peripheral blood has not been examined extensively. Therefore, we examined the methylation status at the CpG island of the small nuclear ribonucleoprotein associated polypeptide N (SNRPN) gene and at the PW71 locus using normal and abnormal lymphoblast (LB) cell lines (n = 48), amniotic fluid (AF) cell cultures (n = 25), cultured chorionic villus samples (CVS, n = 17), and fetal tissues (n = 18) by Southern blot analysis with methylation sensitive enzymes. Of these samples, 20 LB cell lines, three AF cultures, one CVS, and 15 fetal tissues had been previously diagnosed as having deletions or UPD by other molecular methods. Methylation status at SNRPN showed consistent results when compared with FISH or DNA polymorphism analysis using all cell types tested. However, the methylation pattern for PW71 was inconsistent when compared with other tests and should therefore not be used on tissues other than peripheral blood. We conclude that SNRPN, but not PW71, methylation analysis may be useful for diagnosis of PWS/AS on LB cell lines, cultured amniotic fluid, or chorionic villus samples and will allow, for the first time, prenatal diagnosis for families known to carry imprinting centre defects.     

Array-CGH characterization of a prenatally detected de novo 46,X,der(Y)t(X;Y)(p22.3;q11.2) in a male fetus

We report on an apparently normal 5-month-old boy with a X;Y complex rearrangement identified first on prenatal diagnosis and found on array-CGH to have a 7.6?Mb duplication of Xp22.3 chromosome and a deletion of Yq chromosome, distal to the AZFa locus. Karyotype analysis on amniotic fluid cell cultures revealed a de novo homogenous chromosome marker that we interpreted as an isochromosome Yp. FISH analysis using SRY probe revealed only one signal on the derivative Y chromosome. The final karyotype was interpreted as 46,X,der(Y)t(X;Y)(p22.31;q11.22). Translocation Xp22;Yq11 in male are very rare event and only 4 cases have been published, all showing mental retardation and malformations. Herein we discussed some possible explanation for this apparent phenotypic variability.