Interactions between C-bands of chromosomes 1 and 9 in recurrent reproductive loss

Hum Genet (1983) 63:58-62 © Springer-Verlag1983 Interactions Between C-Bands of Chromosomes 1 and 9 in Recurrent Reproductive Loss Judith H. Ford, D. F. Callen, Cynthia G. Roberts, and Adrianne B. Jahnke Cytogenetics Dept., Queen Elizabeth Hospital, Woodville, South Australia, 5011 Summary. The size of the heterochromatic regions of chromosomes 1 and 16 was measured in a Test group of women with histories of recurrent spontaneous abortion and a Control group of fertile women. Measurements were made on Giemsa banded preparations and the euchromatic regions of lq and 16q were used to correct for between-cell contraction. For each chromosome pair, the larger and smaller chromosome was identified and populations of each were compared between the two subject groups. For chromosome 1, the smaller chromosome of the Test group was significantly smaller than that of the Control group (P< 0.001) and the size of the pair difference was larger in the Test than in the Control group (P< 0.01)i For chromosome 16, the smaller chromosome of the Test group was smaller than that of the Control group (5% level). The interaction of chromosome 1 and chromosome 9 heterochromatin in each individual has been analyzed. The combined score for the smaller chromosome 1 and the larger chromosome 9 shows a bimodal distribution and allows discrimination between the two subject groups. Various possible ways in which this interaction might affect reproductive outcome are discussed. be the result of some interaction between the lqh and 9qh regions. Furthermore, the two homologous chromosomes of a pair may act synergistically or independently. Recently Ford et al. (1982) studied the possible associations of alterations in chromosome 9 heterochromatin with reproductive loss, and found that the larger chromosome 9 of each pair was significantlylarger in subjects with histories of recurrent foetal loss than in a control group. However the authors were unable to determine whether this association was due to the size of the pair difference or to the size of the larger chromosome of each pair. This paper examines the distributions of the sizes of the regions lqh and 16qh from chromosome pairs of the same two groups of subjects as Ford et al. (1982). Where possible the data for chromosome 1 and the data for chromosome 9 are correlated in each subject, with the data published for chromosome 9, with the aim of defining either a threshold heterochromatic content or a specific inter-chromosomal interaction, which might account for the deleterious reproductive effect. Materials and Methods Introduction The pericentromeric "heterochromatic" regions on the human chromosomes 1, 9, and 16 are polymorphous and many investigations have sought to discover a role for the frequently observed variations in these chromosomal regions. Large Cbanding heterochromatic regions or pericentric inversions of this material have been shown to be associated with reproductive loss (e.g. Bou6 et al. 1975; Ford 1977), congenital abnormalities (e.g. Wang and Hamerton 1979) and cancer (Atkin and BritoBabapulle 1981). However these associations have usually shown only marginal statistical significance. The results of such studies have often been received with scepticism, since it is difficult to accept that a variant is responsible for a pathological condition, when the normal population exhibits a moderate frequency of this variant. Thus, if the significantly increased frequency of the variants in the affected sub-populations are not to be denied, more information must be found to support the data and discover its biological meaning. For example, if lqh+ and 9qh+ each show significant associations with a certain pathology, then the resultant pathology could be influenced independently by lqh and 9qh, or Offprint requests to: J. H. Ford Subjects Control Group. Blood cultures were established from women attending this hospital for first trimester termination of pregnancy. The first 100 with suitable chromosome preparations were used as controls; no other selection criteria were used. Test Group. Women from couples referred to this department for chromosome analysis, because of histories of recurrent abortion, were included in the Test group if they met the following criteria: (i) chromosome preparations were of suitable standard, and (ii) the reproductive history included three or more abortions, irrespective of whether there were any live-born children. Fifty four women had preparations in which both chromosomes 1 and 16 were suitable for analysis. Subjects for Combined Analysis of Chromosomes 1 and 9 Only 35 of the subjects in the test group (above) had been studied by Ford et al. (1982) for chromosome 9. Thus only these 35 subjects are considered in the combined analysis. The first 35 women in the control group, who had been reported in both studies, were used as controls. No other selection procedures were used. 59 HC HC EC 16 EC Fig. 1. Chromosomal regions measured. The heterochromatic region is depicted HC and the enchromatic region is depicted EC. The arrows indicate the limits of the regions measured Culture Peripheral b l o o d was cultured for 48 h in H a m s F10 with 20% calf serum, a n d the cells harvested with 0 . 0 7 5 M KC1 a n d fixed with 3 : 1 m e t h a n o l : acetic acid. Slides were air-dried a n d treated with a saline : trypsin G i e m s a b a n d i n g technique. C - b a n d i n g was induced with Ba(OH)2. Cells were p h o t o g r a p h e d u n d e r 100X oil i m m e r s i o n o n K o d a k 3 5 m m film a n d printed at c o n s t a n t magnification (6.2 ×). Measurements M e a s u r e m e n t s were m a d e in the m a n n e r described previously for c h r o m o s o m e 9 ( F o r d et al. 1982). F o r c h r o m o s o m e s 1 a n d 16, where the limits of the h e t e r o c h r o m a t i c region can be recognized in G - b a n d e d p r e p a r a t i o n s , m e a s u r e m e n t s were expressed as ( H C + E C ) / E C where H C is the h e t e r o c h r o m a t i c region a n d E C is the e u c h r o m a t i c region of l q a n d 16q, as depicted in Fig. 1. E C is included as the c o n s t a n t or n o n - v a r i a b l e region, a n d is used to correct for between-cell c h r o m o s o m e contraction. The error of m e a s u r e m e n t was d e t e r m i n e d for b o t h c h r o m o s o m e 1 a n d 16 in the same way as described previously ( F o r d et al. 1982), a n d in each case the error did n o t exceed 2%. In the analysis therefore, each m e a s u r e m e n t was a s s u m e d to have a m a x i m u m error of 2%. Analysis F o r each h o m o l o g o u s pair of c h r o m o s o m e s , the larger a n d smaller c h r o m o s o m e of each pair was identified a n d the various tests p e r f o r m e d as given in Table 1. Table 1. Scores calculated and compared between the test and control populations. The Mann-Whitney 2-tailed test was used to analyse the data. This test is appropriate to data of this type since it makes no assumptions about normality, and is sensitive to differences in variances and medians Test number Score Abbreviation 1 2 3 4 5 6 7 8 9 10 11 Larger chromosome 1 Smaller chromosome 1 Pair difference ch. 1 Larger chromosome 16 Smaller chromosome 16 Pair difference ch. 16 Larger 1 minus Larger 9 Larger 1 minus Smaller 9 Smaller 1 minus Larger 9 Smaller 1 minus Smaller 9 Pair difference ch. 1 plus Pair difference ch. 9 1L 1S ( 1 L - IS) 16L 16S ( 1 6 L - 16S) (1L-gL) ( 1 L - 9S) (1S-9L) (1S-9S) ( 1 L - I S ) + (9L-9S) Results F o r c h r o m o s o m e s 1 a n d 16 the medians a n d ranges for the larger a n d smaller c h r o m o s o m e s in the two subject groups, a n d the results of the M a n n - W h i t n e y statistical analysis are presented in Table 2. A scatter diagram plotting the size of the pair difference for c h r o m o s o m e 1 against the size of the pair difference for 60 Table 2. Medians, ranges and statistical analyses for c h r o m o s o m e s 1 and 16 Score Control Test Relationship 1L x = 1.28 x = 1.24 No difference (1.15-1.89) (1.15-1.98) x = 1.23 x = 1.17 (1.08-1.39) (1.10-1.29) x = 1.45 x = 1.43 (1.23-2.24) (1.24-1.75) x = 1.37 x = 1.33 (1.15-1.67) (1.20-1.56) IS 16L 16S chromosome 9, in each individual is shown in Fig. 2; and a scatter diagram plotting the size of the smaller chromosome 1 against the size of the larger chromosome 9 in each individual, is shown in Fig. 3. Both of these diagrams show some degree of clustering within each subject group, but this clustering is more marked in the small 1/large 9 chromosomes than in analysis of pair differences. Clustering effects were analysed by comparing the composite scores for the various equations relating chromosomes I and 9, in the two populations. The equations were formulated as (size of chromosome 1 - size of chromosome 9) to allow determination of any interaction between small and large chromosomes. The results of these analyses are shown in Table 3. Clearly the relationship 1 S - 9L is the most statistically significant, and the comparison of this result with the three other equations relating chromosomes 1 and 9 emphasizes that the large chromosome 9 is having a more marked effect on statistical significance than the small chromosome 1. Control > Test*** No difference Control > Test* * P < 0.05 *** P < 0.001 See Table 1 for full description of scores .16 x x oContro[ xTest .14 E .12 o X .10 X la n X X X .o8 • • X X • X • X Q .oE X .04 X X X X X X X • • .02 X X X • )O J~/~2~ 14 0 .0 d2 .06 .04 .08 .1 .12 .14 36 Difference -chromosome 1 .18 .20 .22 Fig. 2. Scatter diagram of pair differences for c h r o m o s o m e s 1 and 9. Each individual is represented by a single point on the diagram; those of the Test g r o u p by a cross and those of the Control group by a solid circle ,24 • Control. x Test 1.40 - - 1.35 1.30 o u 1.25 X X X 1.20 X eoee 1.15 1.10 x x •80 ~3 .85 Xo X Ooo *" ®0 X X • X X xox 3 X x X X X XxX .90 .95 Larger chromosome 9 *I 1.00 t.05 Fig. 3. Scatter diagram of smaller c h r o m o s o m e , 1 and larger c h r o m o s o m e 9. Each individual is represented by a single point on the diagram; those of the Test group are represented by a cross and those of the Control group by a solid circle 61 Table 3. Results of Mann-Whitney test for scores calculated and compared between the two populations Test no. Score 1 2 3 4 5 6 7 8 9 10 11 Significance 1L N.S. 1S 1 L - 1S 16L 16S 16L- 16S 1L - 9L 1L-9S 1S - 9L lS-9S ( 1 L - 1S) + ( 9 L - 9S) P = < 0.001 0.01 >P>0.001 N.S. 0.05 > P > 0.01 N.S. 0.01 > P > 0.001 N.S. P= < 0.001 N.S. 0.01 >P>0.001 may have masked differences between the groups. Only a marginally significant difference is seen for chromosome 16 and this is found in the smaller chromosome populations. Interactions Between Chromosomes 1 and 9 See Table 1 for key to score Test Contro[ IS:smoLLer chromosome 1 9L: I.Qrger chromosome 9 Of the interactions between chromosomes 1 and 9 analysed, three are statistically significant (nos. 7, 9, 11, Table 3). However of these, the interaction of the smaller chromosome 1 and the larger chromosome 9 gives the greatest discrimination between the two groups of subjects. Moreover, when this interaction is examined in each individual either by a scatter diagram or by a combined score equation, the distribution is found to be bimodal with only limited overlap between the two groups. Some degree of overlap is expected due to the underlying heterogeneity of the populations since some "recurrent aborters" may be undetected in the Control group, and some of the Test group will have poor reproductive histories for other than chromosomal causes. The score for the combined pair differences for the two chromosomes (no. 11) is significantly larger in the Test than in the Control group. However the two groups show considerable overlap in distribution and cannot be separated by this parameter. Biological Effect of Heterochromatic In teraction 6 2 0t .10 .15 .20 .25 .30 IS-9L .35 .40 .45 .50 Fig. 4. Frequency distribution of 1S-9L. Abbreviations are described in Table 1. Test group is represented by dots and Control group by crosshatching The distributions of 1 S - 9 L for the two populations are shown in Fig. 4. The two distributions show remarkable separation with only a relatively small region of overlap. Discussion Heterochromatic Region o f Chromosome 1 In contrast to the results for chromosome 9 (Ford et al. 1982), the larger of the number 1 chromosomes of the two subject groups shows no significant difference in size. However in the populations of the smaller chromosomes, the test group chromosomes are significantly smaller ( P < 0.001) than those in the control group, and the size of the pair difference in the Test group is significantly larger ( P < 0.01) than the size of the pair differences in the Control group. Heterochromatic Region of Chromosome 16 The results for chromosome 16 are very variable, particularly in the larger chromosomes of the control group and this variability Since little is known of the function of heterochromatin, predictions about the possible role of heterochromatic interaction are speculative. However the observed high correlation between the variant complement and maternal history suggests that an effect is likely to occur at female meiosis. Such an effect during meiosis could be by a mechanism involving chromosome pairing or segregation although it is difficult to conceive just how an interaction of 1 and 9 might operate in either case. Individuals carrying heterochromatic variants show an increased frequency of mitotic non-disjunction (Ford and Lester 1978), but direct evidence of susceptibility to meiotic misdivision is not available. Approximately 60% of first trimester abortus specimens show chromosomal abnormalities (Boue and Boue 1973a), and when two consecutive abortuses were analysed from 41 women who were non-translocation carriers (Boue and Boue 1973b), 68% of the specimens were chromosomally abnormal. In 51% of the women, both specimens showed either aneuploidy or triploidy and in a further 34% of the women, only one of the specimens was chromosomally abnormal. Since a high proportion of recurrent aborters carry significant variants or variant combinations, and 85% of women with two consecutive abortuses have at least one chromosomally abnormal foetus, it is inferred that individuals carrying variants have a high susceptibility to meiotic chromosome non-disjunction. Some women with recurrent pregnancy loss have chromosomally normal offspring yet chromosomally normal foetuses may be morphologically abnormal (Warburton et al. 1980). In such cases there may be an interference with differentiation pathways and the interaction between the heterochromatic elements of chromosomes 1 and 9 may influence this by disturbing some critical synthetic activity. The findings presented here suggest interactive effects of heterochromatin between chromosomes 1 and 9. This effect needs further documentation but offers the possibility of predicting the reproductive status of a w o m a n from an analysis of her heterochromatic variants. 62 References Atkin NB, Brito-Babapulle V (1981) Heterochromatin polymorphism and human cancer. Cancer Genet Cytogenet 3:261-272 Bou6 J, Bou6 A (1973a) Anomalies chromosomiques dans les avortments spontanes. In: Boue and Thibault (eds) Les accidents chromosomiques de la reproduction. I.N.S.E.R.M.: 29-55 Bou6 J, Bou6 A (1973b) Chromosomal analysis of two consecutive abortuses in each of 43 women. Humangenetik 19:275-280 Bou6 J, Taillemite JL, Hazael-Massieux P, Leonard C, Bou6A (1975) Association of pericentric inversion of chromosome 9 and reproductive failure in ten unrelated families. Humangenetik 30:217-224 Ford JH (1977) Cytogenetics of infertility and habitual abortion. Records of Adelaide Children's Hospital 1 : 287-293 Ford JH, Lester P (1978) Chromosomal variants and non-disjunction. Cytogenet Cell Genet 21:300-303 Ford JH, Callen DF, Jahnke AB, Roberts CG (1982) Within pair differences of human chromosome 9 C-bands associated with reproductive loss. Hum Genet 336:1-4 Whang HS, Hamerton JL (1979) C-band polymorphism of chromosomes 1, 9 and 16 in four subgroups of mentally retarded patients and a normal control population. Hum Genet 51:269-275 Received November 2 / Revised December 16, 1982