10,739-75 1 (1990)



BARBARA F. C R A N D A L L ? ~ ~

Lkpartments of Psjchiairyt. Obsiefricsf , and Pediatrics$, University of California,Los Angeles and II Naiional Center for Hyperactive Children, Encino, California, U.S.A’.

SUMMARY In the majority of pregnancies involving a Down’s syndrome (DS) fetus, the level of alphafetoprotein (AFP) measured in maternal serum and amniotic fluid is reduced to about 70 per ceni. of the level attained in normal pregnancies. Causes of this decrease may include the pro’duction of an altered AFP molecule with modified turnover or transport properties, or a reduction in the level of AFP synthesis. We examined hepatic AFP mRNA transcripts and conipared AFP polypeptide isofoms in liver tissue samples obtained from a group of DS and normal abortuses. No difference was detected in the structure of the AFP mRNA transcript or in the charge or mass of AFP polypeptides in the two sample groups. Howlever, the hepatic AF:P level, expressed as pg AFP/mg protein, was significantly lower in a group of 28 DS cases relative to a group of 47 normal controls ( p =0.04). This difference in hepai.ic AFP concentration did not appear to be the result of a general reduction in the level of total protein or total RNA production. The greatest difference between the AFP levels of the DS and normal groups was observed in the earliest samples examined (i.e., at 17-19 weeks of age) where the median AFP levels differed by about 20 per cent. KEY WORDS


Down’s syndrome

Second trimester

INTRODUCTION An association has been established between reduced levels of AFP in maternal serum (MS) or amniotic fluid (AF) and fetal Down’s syndrome (DS, trisomy 21) (Merkatz et al., 1984; Cuckle et al., 1984, 1985; Davis et al., 1985; IHershey et al., 1986; Crandall et al., 1988; Kaffe et al., 1988). In pregnancies with an affected fetus, the median AFP level in both MS and AF is reduced to about 70 per cent of the normal level. Low MS AFP levels are now being used to indicate womlen at increased risk. for having a fetus with DS (Hershey et al., 1986; Cuckle et al., 1987; Di Maio e f a]., 1987), but the biochemical basis for the association between low AFP and DS has not been directly examined. The normal expression of the AFP gene, located on chromosome 4 (Minghetti et ~ ‘ 1 . 1983), . could be disturbed by the presence of an additional co-pyof chromosonie 21 in two ways. An increased level of chromosome 2 1 gene products may affect


*Portions of this work were presented at the Annual Meeting of the American Society of Human Genetics in October 1987and October 1989. YPresent address: Department of Obstetrics and Gynecology, Kaplan Hospital, Rehovot, 76100, Israel. Addressee for correspondence: Dr Kathryn E. Kronquist, Mental Retardation Research Center, Neuropsychiatric Institute, Room 48-241, U.C.L.A. School of Medicine, 760 Wesitwood Plaza, Los Angeles, CA 90024, U.S.A

0197--3851/90/110739-13SO6.50 0 1990 by John Wiley & Sons, Ltd.

Received I9 April I988 Final Revision 14 August 1990 Accepted 16 August 1990



enzymes involved in AFP messenger RNA or protein processing to produce a structurally modified AFP polypeptide with different turnover or transport properties. Alternatively, the production of an otherwise normal AFP polypeptide may be reduced or its time course of expression may be altered by a change in regulatory factors. The majority of AFP synthesized by the normal (non-DS) fetus after 10- 13 weeks of gestation is of hepatic origin (Gitlin et al., 1972; Gitlin, 1975). The present study was undertaken to examine fetal hepatic AFP in a group of second-trimester DS and normal abortuses. In this paper we describe the quantitation of AFP levels in the soluble fraction of liver homogenates and the comparison of AFP mRNA transcripts and polypeptides in the two sample groups. METHODS

Origin of tissue samples All cases of trisomy 21 were identified by karyotypic analysis of cells from amniotic fluid or chorionic villi. Fetal liver tissue was obtained within 10-20 min of pregnancy termination. Tissue was frozen immediately in liquid N, and stored at - 70°C. The gestational age of abortuses was determined by measurement of fetal foot length (Barry, 1981). R N A isolation and analysis of AFP transcripts Total RNA was isolated from 1.5-g samples of frozen liver by the method of Chirgwin et al. (1979). A fraction enriched in polyadenylated [poly(A)+]RNA was obtained by affinity chromatography (Aviv and Leder, 1972). For Northern blot analysis, samples containing 2pg of poly(A)+ RNA were fractionated on 1 per cent agarose/3 per cent formaldehyde gels and transferred to nylon membranes. Hybridization was performed using a modification of the method of Church and Gilbert (1984), which employed 0 5 M sodium phosphate (pH 7.2), 1 mM EDTA, 5 per cent SDS, 0.1 per cent BSA, and 50 per cent formamide. Membranes were incubated at 42°C for 18 h with a m AFP-specific cDNA labelled with 32P by the random hexamer primer method of Feinberg and Vogelstein (1983). The probe, pHAF,, contained approximately 40 per cent of the AFP coding sequence and the 3’-untranslated region (Morinaga c7t al., 1983). Blots were washed four times at 55°C with 0.1 x SSC and 0.1 per cent SDS for a total of 2 h. Reverse transcriptase-PCR (RT-PCR) analysis was used to compare the length of specific regions of AFP mRNA from DS and normal individuals. For first strand synthesis, an oligonucleotide that bound in the extreme 3‘ region of AFP mRNA was annealed to 0.5pg of total cellular RNA and incubated with 600 units of Moloney murine leukaemia virus reverse transcriptase (Bethesda Research Laboratories), according to the supplier’s specifications, for 100 min at 37°C. Onetwentieth of the reaction product was added to a tube containing 50 mM Tris-HCl (pH 8.8), 75 mM KCl, 3 mM MgCiL,, 10 mM DTT, 0.2 mM dNTPs, 0.2 ,UM forward and reverse primers, described in Figure 2, and 1.5 units of Taq DNA polymerase in a total volume of 50 pl. Automated PCR was carried out using a variety of conditions to produce the six different amplification products. The reaction products


74 1

were analysed on 1.6 per cent agarose gels and visualized with ethidium bromide staining. Restriction enzyme analysis was performed on each product to confirm that the appropriate AFP cDNA region had been amplified.

Polyucrylamide gel electrophoresis and Western blot analysis

1 soelectric focusing (IEF) was performed using the method of Oi'Farrell (1975) with the following modifications: liver homogenate samples were boiled for 5 min in dissociation buffer containing 1 per cent SDS, 8 per cent (w/v) Nonildet P-40, and 5 per cent (v/v) 2-mercaptoethanol. When the samples had cooled, an equal volume of dissociation buffer containing 4 per cent (w/v) pH 3-10 ampholines (Serva Biochemicals), and 9~ urea was added. The IEF gels were prepareld as described previously (Kronquist et al., 1984). The second dimension gel was a linear 5-1 5 per cent SDS-polyacrylamide gradient slab gel made according to Laemmli (1 970). Polypeptides resolved on slab gels were transferred to nitrocellulose membranes (Towbin et al., 1979). AFP was detected with a 1 :800 dilution of rabbit anti-human AFP antiserum (Dako Laboratories) followed by incubation with [1'2sI]-proteinA (specific activity 10 pCi/pg). iLfcmurement oj"AFP and total protein in liver homogenates Frozen liver fragments were homogenized at a ratio of 0.1 g tissue/2 ml of 50 mM sodium phosphate buffer (pH 7.6), containing 0.1 M NaCl, 0.1 per cent (w/v) Triton X-100, and three protease inhibitors: 10 p~ benzamidine, 10 PM phenylmethylsulphonyl fluoride, and 100 p~ pepstatin. Tissue disruption was performed at 4°C using two 20 s homogenization treatments (Model PT10/35 homogenizer, Brinkman Instruments) and homogenates were centrifuged at 100 OOOg for I h at 4°C. We compared several homogenization protocols and found i.hat the above method produced maximal soluble AFP. The 100 000 g supernatants were diluted 1:40 or 1:80 with AFP-free serum and AFP was quantitated using an ELISA technique (EIA Diagnostic Kit, Abbott Laboratories). Intra- and inter-assay variation was less than 5 and 10 per cent, respectively. Triton X-100 and other components of the homogenization buffer had no effect on the AFP immunoassay so long as the 1OOOOOg supernatants were diluted at least 2.5-fold prior to assay. Total protein was measured by a modification of the Lowry procedure after diluting the 100 OOOg supernatant 1:40 with 0.1 M NaOH and 1 per cent SDS (Markwell et al., 1978). The Lowry protein determination was not afrected by the presence of the anionic detergent Triton X-100 at a final concentration of 0.00025 per cent. The AFP immunoassay and the total protein me;isurement for each sample were done in triplicate on homogeriates prepared fro:m three separate tissue fragments. Assay variability was 8 per cent for the AFP dekrmination and 5 per cent for the Lowry protein determination. RESULTS

Malernal serum and amnioticjuid AFP levels in the DSgroup The MS and A F AFP levels, expressed as a multiple of the mediain (MOM), and the gestational age at termination of the 28 cases of fetal DS are summarized in



Table 1. AFP measurement in 28 pregnancies involving a fetus with Down’s syndrome Maternal serum

Amniotic fluid

18 67 1.69-0.23 0.72 0.87 18G22.0

12 83 1.2 1-0.34 0.72 0.73 174-2 1.4


Number of cases Percentage less than 1 MOM*( ‘ X ) AFP range (MOM) Medium AFP Mean AFP Fetal age range (weeks)

*The MOM was calculated by dividing the mother’s AF or MS AFP level by the median level obtained for a group of 100 age-matched control samples with known nornial outcomes (Crandall e t a / . , 1983).

Table 2. Measurement of hepatic RNA, soluble protein,* and AFP in Down’s syndrome and normal sample groups?

DS (n) Normal (n)


Total RNA (mg/g tissue)

Total protein (mg/ml homogenate)

Pg AFPImg protein

3.67k0.61 (25) 3.69f0.84 (35) 0.92

4.03 f0.53 (28) 4-03t- 0.45 (47) 0.95

2.34 f0.47 (28) 2.71 f 1.03 (47) 0.04

*Protein in the 100 OOOg fraction of liver homogenates. ?Values expressed as means standard deviation. $Determined using a two-sample /-test (one-tailed).

Table 1. In two pregnancies, AFP levels were measured in both MS and A F For all the other affected pregnancies, information was available on only one of the two parameters. Among 18 women whose MS AFP level was known, 12 (67 per cent) had a MOM below 1.0, with a median of 0.72 and a mean of 0.87. Of 12 women whose A F AFP level was available, 10 (83 per cent) had an AFP level below 1.0 MOM, with a median of 0.72 and a mean of 0.73. Thus, in the pregnancies from which we obtained the affected fetal1liver samples, we observed the typical reduction in the median AF and MS AFP level associated with fetal trisomy 21. Affected fetuses ranged in age from 17.4 to 22.7 weeks of completed gestation. Examinarion o f A F P messenger RNA Hepatic total cellular RNA was isolated from 25 DS and 35 control cases of comparable gestational age. The mean yield of RNA per gram of tissue 5 SD was 3.67 50.61 mg for the DS cases and 3.69 5 0.84 mg for the controls (Table 2). Poly(A)+ RNA isolated from 25 DS and 13controls was examined by Northern blot analysis. Figure 1 shows a representative blot of DS (lanes 3-7) and normal samples (lanes 1 and 2) probed with a hunian AFP cDNA. All samples contained a single hybridizing AFP mRNA band with a length of 2.4 kilobases.



1 2 3 4 5 6 7 8 -



I I - - - -

Figure 1. Northern blot analysis of hepatic RNA from DS and normal samples. Poly(A)+ RNA (2 pg/ lane) a a s fractionated on agarose/formaldehyde gels, transferred to nylon membranes, and hybridized to an AFP-specific DNA probe. Lanes 1 and 2: normal samples at 19 and 20 weeks of gestation. Lanes 3-7: DS samples at 18.4, 18.6, 18.8, 21.0 and 22.7 weeks of gestation, respectively. Lane 8: poly(A)+ RNA from adult rat liver (negative control). The length of the AFP transcript was calculated from the mig a t ion of RNA standards (New England Nuclear, Boston, MA). The autoradiographic exposure was for 1 h a t -70°C

The AFP gene contains 15 exons, some of which are under 60 bp long (Saiki et al., 1985). Putative splicing anomalies in DS individuals could result in m a l l changes in AFP mRNA length that might not be detected on Northern blots. We therefore conducted an examination of the transcripts from two normal and five DS cases by amplifying six overlapping regions of the AFP message, illustrated in Figure 2A, using RT-PCR. This type of analysis might not detect AFP mRNA splicing analmalies occurring within exons 1 or 15, where loss of a primer binding site would result in no product being formed. However, structural differences within the 13 internal exons would be detected, assuming that the alternative rnRNA species rep,resented at least 10-15 per cent of the AFP mRNA population, so that its amplification product could be detected by ethidium bromide staining, and as long as the novel product differed in length from the normal product by at least 30 bp so thal the forms could be resolved electrophoretically.




309 435-





a 540


Figure 2. Comparison of six regions of hepatic AFP mRNA in DS and normal cases. (A) Representation of the AFP message with exon boundaries delimited as in Sakai et a/. (1985). Open and closed boxes indicate the position of 5’ and 3’ primers, respectively. The length of each amplified DNA fragment, the corresponding nucleotides, numbered as in the cDNA sequence of Morinaga e / a/.(1983). and the primer sequences are as follows. 557 bp fragment (nucleotides 2-558). 5’ primer: 5‘ TATTGTGCTTCCACCACTGCCAATAAC 3’; 3’ primer: 5’ AGGGATGCCTTCTTGCTATCTCATAAA 3‘. 435 bp fragment (nucleotides 453-887), 5‘ primer: 5’ CCACTTTTCCAAGTTCCAGAACCTGTC3‘; 3’ primer: 5’ CCCATCCTGCAGACAATCCAGCACATC 3’. 614 bp fragment (nucleotides 532-1 145), 5 primer: TTTATGAGATAGCAAGAAGGCATCCCT 3’;3’ primer: 5’ GACAGCAAGCTGAGGATGTCTTCTTGA 3’. 540 bp fragment (nucleotides 96&1499), 5‘ primer: 5’ ACCACGCTGGAACGTGGTCAATGTA 3’; 3 primer: 5’ ACATAAGTGTCCGATAATAATGTCAGC 3’. 676 bp fragment (nucleotides 1209-1884), 5’ primer: 5’ AACCCTCTTGAATGCCAAGATAAAGGA 3’; 3’ primer: 5’ CTGAAGTAATTTAAACTCCCAAAGCAG 3’. 3!)9 bp fragment (nucleotides 1533-1 93 I), 5’ primer: 5‘ GTTGGCCAGTGCTGCACTT’CTTCATATAT3‘; 3‘ primer: 5’ GAGAAAAGTTCACACCGAATGAAAGAC 3’. (B-G)Agarose gel electrophoresis of PCR amplification products. In each panel, lane 1 contains DNA standards; their length in base pairs is indicated on the left. Lanes 2-8 contain amplified products from the same normal and DS cases described in the legend to Figure 1. Lanes 2 and 3: normal cases; lanes 4-8: DS cases. Lane 9 is a control with no added template. The length in base pairs of the amplified product is ]indicated on the right of each panel





67 43 -

946743 30 -

20 14I


P' Figure 3. Two-dimensional immunoblot analysis of AFP. Aliquots of liver homogenate containing 200 ng of AFP were subjected to isoelectric focusing and SDS gel electrophoresis. Polypeptides were

transferred to nitrocellulose membranes and incubated with anti-human AFP followed by 'Z51-labelled protzin A. Panels A and B show entire blots; the film exposure time was 24 h in the presence of one intensifying screen. Panels C-H show only the portion ofeach blot containing the AFP complex; the film exposure time was 8 h with one intensifying screen. (A, C , D, E, F) DS cases obtained al. 19,16,18.8, 19.1, and 19 9 weeks of gestational age, respectively; (B, G, H) normal samples obtained at 21.4,20, and 15.2 weeks of gestational age, respectively. The molecular mass of AFP was calculated based on the migration of a set of protein standards (Pharmacia, Piscataway, NJ). The apparent pI was determined by cutting matched isoelectric focusing gels into 0.25 cm pieces, extracting the ampholines from each piece in 1 ml of boiled water, and measuring the pH. The normally accepted value for the isoelectric point of AFP is 4.8 (Ruoslahti and Seppala, 1971). We observed a more basic PI due to the presence of SDS in the isoelectric focusing gel. The added SDS improves protein solubility but has the general effect of increasing the apparent PI of all sample polypeptides by approximately 1 pH unit







I 15









GESTATIONAL AGE (weeks) Figure 4. Hepatic AFP levels in DS and normal samples. The pg AFP/mg protein in the 100 OOOg supernatant fraction ofliver homogenater. from 28 DS and 66 normal cases is plotted as a function of the gestational age. (0)Normal samples; (a)DS cases. The figure shows the division of the DS cases into four quartiles. Dotted lines indicate the median AFP level of normal cases, and solid lines indicate the median AFP level of DS cases

Our results (Figure 2, B-G) indicate that, for each of the six regions amplified, only products of the appropriate predicted size were obtained from both the DS and the normal individuals. No widence was found for any unusual AFP mRNA processing in Down's syndrome. Comparison of A FP polypeptides

To test for possible differences in the post-transcriptional processing of AFP that might affect its mass or charge, liver homogenate proteins from affected and normal individuals were separated by isoelectric focusing followed by SDS gel electrophoresis, and AFP was detected with a specific antibody. In the electrophoretic system employed, AFP resolves into a group of charge isomorphs having an average mass of 69 kD. Figure 3 shows a comparison among eight individuals: five DS cases (panels A, C , D, E, and F) and three controls (panels B, G, and H). The AFP complex appeared very similar in all the individuals examined; minor variations were not consistently associated with DS individuals, leading to the conclusion that the average charge and mass of AFP polypeptides were not altered in DS.



Table 3 . Hepatic alpha-fetoprotein levels as a function of the gestational age Age range (weeks)

Quartile 1 Quartile 2 Quartile 3 Quartile 4

15.2-17.3 17.4-18'8 18.9-20.1 20.2-21.0 21.1-22.7 22.8-24.0

pg AFP/mg protein Normal (n) DS (n)

4.37 (10) 3.35 (1 1) 2.58 (10) 2.40 (8) 2.40(18) 2.08 (9)

ND* 2.64 (8)t 2.40 (7)f 2.28 ( 8 ) f 2.08 ( 5 ) f


*ND = N o data. TSignificant a t p =0,03 by Wilcoxon rank sum test. tNot significant atpE0.05 by Wilcoxon rank sum test

Comparison of hepatic protein and AFP levels Total protein and AFP were measured in the 100 000 g supernatant fraction of 28 DS and 47 normal liver homogenates (Table 2). The control group consisted of cases with no known genetic or medical complication that were of comparable gestational age to the DS group. The two groups appeared to be identical with respect to their total protein concentration (p=0.95). However, the mean value far pg AFP/mg protein in the DS group was 2.34k0.47 while that in the control group was 2.711 & 1.03. A two-sample t-test indicated a significant difference in th'e AFP content o f t he two groups ( p = 0.04). To examine the effect of development on hepatic AFP levels, individual values were plotted as a function of the gestational age (Figure 4). Nineteen samples obtained outside the age range of the DS group were included in the graph to give a clearer indication of the normal developmental trend. The DS samples were divided as equally as possible into quartiles. The median units of AFP/mg protein for DS and normal samples were compared within each quartile and the results are presented in Table 3. Median values, rather than means, are reported because the median is considered to be a more reliable indicator of central tendency for small groups. Across the developmental period depicted in Figure 4, the AFP level of both sample groups displayed a tendency to decline with increasing gestational age. In all fou:r quartiles, the median distribution of the DS samples was lower than that of the normals. The greatest difference between the two groups was seen in quartile 1, where the DS and normal medians differed by about 20 per cent. We used a distribution-free rank sum test (Wilcoxon) (Hollander and Wolfe, 1973) to compare AFP levels in the DS and normal groups. This non-parametric statistical method was used because of the small number of samples in each quartile. In quartile 1, a value of p = 0.03 indicated a significant difference between DS and nor-mals during this gestational period. However, in quartiles 2-4, the difference between DS and normals was not significant at a level ofp = 0.05.



DISCUSSION Numerous studies have described an association between DS and reduced levels of AFP in maternal serum and amniotic fluid. It is unlikely that the reason for this association is because affected fetuses are simply smaller than their normal counterparts. Librach et al. (1988) found no significant difference in the weight distribution between a group of 50 second-trimester DS abortuses and a group of 65 normal controls, adjusted for age. One goal of the present study was to look directly in the liver from DS and normal fetuses for possible differences in the transcription or post-translational modification of AFP that might account for its observed reduction in fetal or maternal body fluids. Examination of poly(A)+ RNA by Northern blot hybridization and comparison of amplified regions of the AFP mRNA transcript in the two sample groups failed to reveal any detectable differences in the structure of the AFP message. Normal events in the post-transcriptional processing of AFP include the cleavage of a peptide leader sequence and the covalent attachment of an asparagine-linked carbohydrate side-chain. We employed two-dimensional SDS-gel electrophoresis, an analytical method that resolves proteins by isoelectric point and molecular mass, in order to detect possible alterations in the post-transcriptional modification of AFP, but were unable to detect any consistent differences between the charge or mass of AFP isoforms in DS and !normalindividuals. This negative finding reduces, but does not dismiss, the possibility that some aspect of post-transcriptional AFP processing is altered in DS. Glycosylation is an important modification that could potentially affect the intracellular transport or circulatory half-life of AFP, and small alterations in the neutral sugar composition of the AFP oligosaccharide would not produce a change in the mobility of the AFP polypeptide on the twodimensional gels. The structure of'the oligosaccharide side-chain on hepatic AFP in the normal and DS sample groups is, therefore, an area that may require further study, possibly by utilizing lectiris to discriminate among different carbohydrate configurations. A positive finding in the present study was the demonstration of a significant reduction in the hepatic AFP level of affected fetuses, in the apparent absence of any general effects on the level of RNA or total protein (Table 2). This result tends to support the hypothesis of Cuckle and Wald that lower maternal serum AFP levels in DS pregnancies may be due to rcduced fetal AFP production (Cuckle and Wald, 1986). Studies in several laboratories have shown that AFP gene expression is controlled principally at the level of transcription (for reviews see Tilghman (1985) and Nahon (1987)), so it would be reasonable to expect that the observed reduction in hepatic AFP level would be associated wi1:h a decreased level of AFP mRNA in the affected sample group. We have measureld relative AFP mRNA levels in DS and normal cases by slot blot analysis, normalizing the level of AFP-specific hybridization to that of a constitutively expressed gene, but have so far been unable to detect a statistically significant difference between the two groups. We believe that this is most likely due to technical reasons. There is a relatively high variability of 10-12 per cent associated with the hybridization analysis, and we currently have available



only a small number of samples (eight DS and ten normal cases) jiom the most informative age group between 17 and 19 weeks of gestation. A meaningful study of AFP mRNA levels will probably have to await the collection of many more normal and affected cases. Our data indicate that there may be a difference in the time course of expression of AFP in the DS group. In the period between 17.4 and 18.8 weeks of gestation, the disiribution of pg AFP/mg protein in the DS group was significantly lower than that of the normal group, but later in gestation, between 18.8 and 22.7 weeks, a significant difference between the two groups was not observed. A related effect of gesLationalage on the reduction of amniotic fluid AFP levels in a group of 11 1 DS pregnancies was noted by Crandall et al. (1988). They reported that the greatest reduction in AFP level was found at the earliest gestational ages examined; 94 per cent of the cases at 17 weeks or earlier fell below the median compared with 66 per cent of the cases at 18 weeks and later. It is known that during gestation the total quantity of AFP synthesized by the normal (non-DS) fetus increases most rapidly in the period between 14 and 20 weeks of gestation. After the 20th week, a plateau is reached and the total quantity of AFP synthesized remains at a constant level until around 32 weeks of gestation, at which time i t declines sharply prior to birth (Gitlin, 1975). We observed a significant difference between the hepatic AFP level of DS and normal fetuses at II 7-19 weeks of gesl.ation, but not later at 19-23 weeks. At term, when the level of total AFP synthesis is again changing rapidly, a statistically significant decrease in cord serum AFP levels in a group of 22 DS newborns was found by Cuckle andl Wald (1986). However, at mid-gestation, Scioscia et al. (1988) did not observe a significant difference in the distribution of cord serum AFP levels in a group of 1 1 DS abortuses, nine of which were obtained after 19 weeks of gestation. It may be that the greatest diff’zrencebetween the AFP levels of the normal and DS groups is manifested during the two periods when a rapid change in total AFP synthesis is occurring. During these critical times at 14-20 weeks of gestation and, later, between 32 weeks and term, the regulation of AFP levels in DS individuals may be particularly subject to perturbation by factors associated either directly or indirectly with the additional copy of chromosome 21. ACKNOWLEDGEMENTS

We thank D r James T. McMahon, Medical Director of Eve Surgicid Centers; Dr Morton W. Barke of Inglewood Women’s Hospital; Dr Karen Blanchard of the Women’s Medical Group, Santa Monica, California; and Dr Oliver W. Jones, Division of Medical Genetics, School of Medicine, University of California, San Diego, for valuable tissue samples. The human AFPcDNA probe, pHAF2, was the gift of D r Taiki Tamaoki, Department of Medical Biochemistry, University of Alberta, Edmonton, Alberta, Canada. We also thank Mr Myles Matsiimoto and Ms Caroline French for performing the AFP immunoassays. REFERENCES Aviv, H., Leder, P. (1972). Purification of biologically active globin mRNA by chromatography on oligothymidilicacid-cellulose,Proc. Natl. Acad. Sci. U.S.A.,69, 1408-1412.



Barry, C.L. (198 1). Pediatric Pathology, Berlin: Springer-Verlag, 3. Chirgwin, J., Przybyla, A., MacDonald, R., Rutter, W. (1979). Isolation of biologically active ribonucleic acid from sources enriched in ribonuclase, Biochemistry, 18,5294-5299. Church, G., Gilbert, W. (1984). Genomic sequencing, Proc. Nail. Acad. Sci. C‘.S.A.,81, 1991-1995. Crandall, B.F., Robertson, R., Lebherz, T., King, W., Schroth, P. (1983). Maternal serum alpha-fetoprotein screening for the detection of neural tube defects: report of a pilot program, West. J . Med., 138,524--530. Crandall, B.F., Matsumoto, M., Perdue, S. (1988). Amniotic fluid AFP in Down syndrome and other chromosomal abnormalities, Prenut. Diugn., 8,255-262. Cuckle, H., Wald, N.J. (1986). Cord serum alpha-fetoprotein and Down’s syndrome, Br. J . Obstet. Gynaecol., 93,408-4 10. Cuckle, H., Wald, N., Lindenbaum, R. (1984). Maternal serum AFP measurement: a screening test for Down syndrome, Lancer, i, 926-929. Cuckle, H., Wald, N., Lindenbaum, R., Jonasson, J. (1985). Amniotic fluid AFP levels and Down syndrome, Lancet, i, 290-201. Cuckle, H., Wald, N., Thompson, S. (1987). Estimating a woman’s risk of having a pregnancy associated with Down’s syndrome using her age and serum alpha-fetoprotein, Br. J . Obstet. Gynaecol., 94,387-402. Davis, R., Cosper, P., Huddleston, J.. Bradley, E., Finley, S., Finley, W., Milunsky, A. (1985). Decreased levels of amniotic fluid AFP associated with Down syndrome, Am. J. Obsrer. Gynecol., 153,541-544. Di Maio, M., Baumgarten, A., Greenstein, R., Saal, H., Mahoney, M. (1987). Screening for fetal Down syndrome in pregnancy by measuring maternal serum alpha-fetoprotein levels, N. Engl. J . Med., 317,342-346. Feinberg, P., Vogelstein, B. (1983). A technique for radiolabeling restriction fragments to high specific activity, Anal. Biochern., 132,6-13. Gitlin, D. (1975). Normal biology of alpha-fetoprotein, Ann. N . Y . Acad. Sci., 259,7-15. Gitlin, D., Pericelli, A,, Gitlin, G. (1’972).Synthesis of AFP by liver, yolk sac and intestinal tract of the human conceptus, Cancer Res., 32,979-982. Hershey, D., Crandall, B.F., Perdue, S. (1986). Combining maternal age and serum AFP to predict risk of Down syndrome, 0,bstet. Gynecol., 68, 177-180. Hollander, M., Wolfe, D.A. (1973). In: Nonpararnetric Sfatistical Methods, New York: John Wiley, 67-69. Kaffe, S., Perlis, T., Hsu, L. (1988). Amniotic fluid alpha-fetoprotein levels and prenatal diagnosis of autosomal trisomies, Prenat. Diugn., 8, 183-187. Kronquist, K.E., Crandall, B.F., Cosico, L.G. (1984). Detection of novel fetal polypeptides in human amniotic fluid using two-dimensional gel electrophoresis, Tumor Biol., 5, 15-32. Laemmli, U.K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature, 227,680-685. Librach, C., Hogdall, C., Doran, T. (1988). Weights of fetuses with autosomal trisomies at termination of pregnancy: an investigation of the etiologic factors of low serum alphafetoprotein levels, Am. J . Obstet. Gynecol., 158,290-293. Markwell, M., Hass, S., Bieber, L., Tolbert, N. (1978). A modification of the Lowry procedure to simplify protein determination on membrane and lipoprotein samples, Anal. Biochem., 6,539-546. MerkatzJ., Nitowski, H., Macri, J., Johnson, W. (1984). An association between low maternal serum alpha-fetoprotein and fetal chromosomal abnormalities, A m . .I. Obsiet. Gynecol., 148,886-894. Minghetti, P., Harper, M., Alpert, EL, Dugaiczyk, A. (1983). Chromosomal structure and localization of the human alpha-fetoprotein gene, Ann. N . Y. Acad. Sci., 414, 1-1 2. Morinaga, T., Sakai, M., Wegmann, T., Tamaoki, T. (1983). Primary structures of human alpha-fetoprotein and its mRNA, Proc. Natl. Acad. Sci. U.S.A., 80,4604- 4608. Nahon, J.-L. (1987). The regulation of albumin and alpha-fetoprotein gene expression in mammals, Biochimie, 69,445-459.


75 1

O’Farrell, P. (1975). High resolution two-dimensional electrophoresis of proteins, J . Biol. Ciiern., 256,4007- 4021. Ruoslahti, E., Seppala, M. (1971). Studies of carcinofetal proteins. Physical and chemical properties of human alpha-fetoprotein, Znt. J . Cancer, 8,374-383. Sakai, M., Morinaga, T., Urano, Y., Watanabe, K., Wegman, T., Tamaoki, T. (1985). The human alpha-fetoprotein gene: sequence organization and the 5’ flanking region, J. Biol. Ciiern., 260,50 5S--SO60. Scio:;cia, A,, Blakemore, K . , Inati, M., Robinson, J., Grannum, P., Hobbin:;, J., Mahoney, M., Baumgarten, A. (1988). Mid-trimester fetal serum AFP levels in nonmal and Down syndrome fetuses, Am. J . Hum. Genet., 43, (suppl.), A250. Tilghman, S. (1985). The structure and regulation of the alpha-fetoproteiin and albumin genes. In: Maclean, N. (Ed.). Oxford Survey on Eukaryotic Genes, Vol. 11, Oxford: Oxford University Press, 161-206. Towbin, H., Staehlin, T., Gordon, J . (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets. Procedure and some applications, Proc. Natl. Ac-ad. Sci. U.S.A., 76,4350-4354.

Reduced fetal hepatic alpha-fetoprotein levels in Down's syndrome.

In the majority of pregnancies involving a Down's syndrome (DS) fetus, the level of alpha-fetoprotein (AFP) measured in maternal serum and amniotic fl...
786KB Sizes 0 Downloads 0 Views