XY chromosome behaviour in the germ-line of the human male: A FISH analysis of spatial orientation, chromatin condensation and pairing

Chromosome Research 1994, 2, 445-452

XY chromosome behaviour in the germ-line of the human male: a FISH analysis of spatial orientation, chromatin condensation and pairing

Susan J. Armstrong, Amanda J. Kirkham & Maj A. Hult6n

Received 8 June 1994; received in revised form 8 July 1994 Accepted for publication by H.C. Macgregor 8 July 1994

We have used multicolour fluorescence in situ hybridization to study the behaviour of the X and Y chromosomes in relation to a representative autosome, chromosome 1, on air-dried testicular preparations from normal fertile human males, In a proportion of Sertoli cells at interphase as well as spermatogonial metaphases there is an apparent selective undercondensation of the heterochromaUc block of the long arm of the Y, which may be of functional significance with respect to Y-specific gene activity, initiating and main- taining spermatogenesis; we suggest that this may involve a mechanism similar to heterochromatin position-effect variegation in Drosophila. In the supporting Sertoli as well as pre-meiotic and leptotene cells the X and Y occupy relatively restricted domains at opposite poles of the nuclear membrane, while the chromosome 1 centromere regions are located interstitially and appear prealigned. The XY pairing and 'sex vesicle' formation comprises a complex series of spatial move- ment and differential condensation patterns. On the basis of these observations we propose that: the XIST/Xist gene, known to be involved in somatic X inactivation, imposes a chromatin reorganization leading to bending at the X-inactivation centre both at first meiotic prophase in males and in the soma in females; and the differential X and Y segments are protected from potentially deleterious meiotic exchanges by their separate spatial orientation. In addition, there is an indication that the timing of pairing and first meiotic segregation of the sex chromosomes is different, and precocious in comparison to the pairing and segregation of the autosomes, which may explain the high incidence of sex chromosome aneuploidy in sperm.

Key words: meiosis, spatial organization, XY chromosomes

I n t r o d u c t i o n

Recent developments in fluorescence in situ hybridization (FISH) have produced a powerful technology,

which allows specific chromosomes to be identified at stages other than metaphase. The technique has been shown to have potential in the analysis of spatial organization of human somatic interphase cells (Lichter et al. 1988) and has given valuable insights into organization of specific chromosome or even whole genome domains during interphase in plant species and hybrids (reviewed by Heslop-Harrison & Bennett 1990). We have previously demonstrated that FISH analysis can be used not only to characterize meiotic stages in the human male, using chromosome paints for selected autosomes (Goldman & Hult6n 1992), but also for analysis of chromosome configurations and segregation in male reciprocal translocation carriers (Goldman & Hult6n 1993a,b, Armstrong & Hult6n 1994). Analysis of the sex chromosomal aneuploidy in sperm, using dual colour FISH (Goldman et al. 1993) suggested a precocious sex chromosomal first meiotic segregation in relation to autosomes. We have now used FISH to study the behaviour of both sex chromosomes in comparison to chromosome 1 in relevant germ-line stages, i.e. the Sertoli cells of the supporting lineage, spermatogonial metaphase, premeiotic interphase, first meiotic prophase and metaphase, second meiotic metaphase and postmeiotic products.

M a t e r i a l s and m e t h o d s

Sample preparation

Testicular biopsy material was obtained from .two chromosomally normal fertile men during reverse vasectomy procedures. Meiotic preparations were prepared as reported previously (Hult6n et al. 1992) and stored in air at -70°C for approximately 2 years before analysis. A preliminary analysis showed the only difference between the samples to be slightly better chromosome morphology in preparations from one

The authors are at the LFS Research Unit, West Midlands Regional Genetic Services, Birmingham Heartlands Hospital, Birmingham B5PX, UK. Fax: (+44) 21 766 5681. Address correspondence to M.A. Hultdn

© 1994 Rapid Communications of Oxford Ltd Chromosome Research Vol 2 1994 445

S. J. Armstrong et al.

case, which was therefore used for the more detailed analysis.

Probes

To allow a comparison between the behaviour of the sex chromosomes in relation to a representative autosome, five probes were applied for this study:

1 Chromosome X biotinylated paint (Cambio Ltd, UK), detected with Texas Red;

2 Chromosome X centromere directly labelled with FITC dUTP (Cytocell Ltd, UK);

3 Chromosome Y heterochromatic block of long arm (Yqh) directly labelled with FITC dUTP (Cytocell);

4 Chromosome I heterochromatic block of long arm (lqh), directly labelled with rhodamine dUTP or FITC dUTP (Cytocell);

5 Chromosome 1 biotinylated paint (Cambio), detected with Texas Red and used in conjunction with the lqh probe, directly labelled with FITC dUTP (Cytocell).

Figure la and b shows chromosome identification of metaphases from a lymphocyte preparation of a chromosomally normal male. The Yqh displays a bright

green fluorescence by FITC. The small red band, located at the end of the short arm of the Y (Ypter), represents the region homologous with the end of the short arm of the X (Xpter). The X chromosome fluoresces brightly red by Texas Red, while the X centromere is yellow by FITC in combination with Texas Red. Chromosome 1 is clearly differentiated with FITC or rhodamine at the heterochromatin block lqh (Figure la); alternatively, the whole chromosome 1 can be visualized with Texas Red in conjunction with lqh labelled with FITC (Figure lb).

Slide preparation

Slides were warmed to room temperature before being placed at 67°C for 3 h, washed twice for 5 min in 2xSSC at room temperature, and denatured for 2 min at 70°C in 70% formamide/2xSSC solution. Finally they were dehydrated for 5 min in each stage of an ethanol series (70, 85 and 100%) at 4°C.

Probe preparation and hybridization

Probes were prepared and denatured according to the manufacturers' instructions and applied immediat61y

Figure 1. Multicolour FISH of 46,XY male blood lymphocytes a & b, c & d spermatogonial metaphases. Chromosomes X (long arrows), Y (short arrows) and 1 (arrowheads) are indi- cated, a & b The X is dual labelled with biotinylated whole chromosome paint detected by Texas Red and the centromere probe directly labelled with FITC (yellow). The Y is identified by the heterochromatic block Yqh in green by FITC. Note also that the tip of the short arm of the Y, which represents the primary pairing segment, homologous with the X, is labelled red accord- ingly. Chromosome 1 is identified in a by the probe to the heterochromatic block lqh, directly labelled with FITC, and in b by dual labelling with the biotinylated chromosome paint by Texas Red and the centromere, as in a yellow by FITC. c Chromosome identification as in b; note com- pacted Yqh of the Y. d Decon- densed Yqh in green by FITC; the X is labelled as in a-c, but the lqh blocks stained with by rhodamine.

446 Chromosome Research Vol 2 1994

FISH analysis of human germline X and Y chromosomes

onto the warm, denatured slides. A 22x22 mm cover- chromosomal organization similar to that observed slip was sealed over each preparation, using a vulca- in Sertoli cells, with the X and Y occupying domains nizing solution. The slides were incubated overnight in in the periphery of the cell, spatially separated at a humidified box in darkness at 37°C. opposite poles (Figure 2c).

Post-hybridization

After hybridization, the coverslips were carefully re- moved and the slides subjected to 5-min washes in 50% formamide/2 x SSC, at 45°C. Further 5-min washes were performed in 2xSSC, lxSSC and 4xSSC/0.05% Tween at 37°C.

Detection

Immediately after washing, detection was carried out using a Texas Red and FITC dual-colour painting kit (Cambio), following the manufacturer's instructions. Slides were counterstained with DAPI and treated with an ethanol series (70, 85 and 100%) at room temperature, before being mounted in an antifade buffer (Vectashield; Vectorstain Ltd, UK).

Microscopy and photography

The slides were observed using a x 100 objective on a Nikon fluorescent microscope with a triple filter (DAPI/TexasRed/FITC). Photographs were taken using a Fujichrome 400 ASA, rated at 1600 ASA for exposure.

Results

Chromosome identification

FISH allowed the X and Y chromosomes, in combination with either lqh alone or whole chromosome 1 and lqh together, to be identified unequivocally at relevant stages of spermatogenesis (Figures 1-3).

Sertoli cells and premeiotic germ cells

Diploid interphase nuclei, representing Sertoli cells of the supporting lineage, characterized by prominent nucleoli (detected by concomitant silver staining, not illustrated), showed the sex chromosomes to be located opposite each other, apparently bound to the nuclear membrane (Figure 2a & b). This positioning contrasts with that of chromosome 1, which as far as could be ascertained, is located within the nucleus. The X and Y in general occupy relatively discrete domains, but in 25% of cells (five of 20) there was some expansion of Yqh (Figure 2a).

Other diploid interphase nuclei, including the premeiotic germ cells, appeared to have a spatial sex

Spermatogonial metaphases

Diploid metaphases of the germinal epithelium are generally accepted to be spermatogonial. In 72% of these (18 of 25) chromosome morphology and condensation were similar to those of in vitro cultured lymphocytes, with a highly condensed Yqh heterochromatic block (Figure lc). In the remaining 28% of cells (7 of 25) this segment showed a marked undercondensation and was the same length as the whole X (Figure ld).

First meiotic prophase

The first indication of differentiation of spermatocytes is an apparent stretching of the X across the nucleus towards the Y with the end of the X long arm (Xqter) still anchored in the nuclear membrane. At the same time there is a similar stretching of the short arm of the Y towards the X but with its long arm remaining condensed and situated in the nuclear membrane (Figure 2d). At this initial 'leptotene' stage the lqh blocks appear to become pre-aligned, located interstitially within the nucleus.

The short arms of the X and Y eventually come in close proximity to allow pairing to be initiated between them (Figure 2e). The exact extent of the XpYp pairing is difficult to interpret by this technology, but we could ascertain that it did not involve the heterochromatic block of the long arm of the Y. Following pairing initiation, the short arm of the X seemed to be com- paratively over-stretched and undercondensed. Con- comitantly with the XY pairing, the heterochromatic blocks lqh become more closely aligned, while the chromosome 1 arms still appear unpaired. We have interpreted this as a transitional stage between late zygotene and early pachytene, as homologous XpYp pairing has taken place, but autosomal synapsis is still incomplete. There was an indication of telomeric autosomal pairing (Figure 2f), but technological limitations made this difficult to assess.

During later pachytene, when the two lqh blocks are closely aligned, the XY pair transforms structurally with 'sex vesicle' formation. The Xq telomere loses its anchoring to the nuclear membrane, while Yqh still remains fixed in its original position. Formation of the 'sex vesicle' is seemingly initiated by folding about the Xq near to the centromere in conjunction with condensation of the whole X (Figure 2g). Further chromatin condensation and bending of the X results in a loop with the telomeres of the long arms of the X and Y in close proximity at the nuclear membrane. The

Chromosome Research Vol 2 1994 447

S. J. Armstrong e t a l .

a

d

b

~ ? i i i i l ¸

e

©

T

Figure 2. Behaviour of chromosomes X (long arrows), Y (short arrows) and 1 (arrowheads) in a & b in Sertoli cell interphase nuclei, ¢ a diploid presumptive premeiotic interphase nucleus and d-i early first meiotic prophase. The X chromosome is identified by Texas Red and its centromere by FITC in yellow; the heterochromatic block of the Y (Yqh), is differentiated by FITC in green, a Sertoli cell with apparently decondensed Yqh. Note interstitial lqh blocks and positioning of X and Y at opposite poles, b Sertoli cell with compact Yqh. ¢ Diploid presumptive premeiotic interphase, showing the X and Y at opposite poles and the lqh blocks also well separated, d Early first meiotic spermatocyte nucleus, showing the X and Y situated at the nuclear membrane. Note the apparent decondensation of the short arm of the Y (lower short arrow)..The lqh blocks are located interstitially and appear prealigned, e First spermatocyte illustrating pairing initiation between the X and Y. Note undercondensation of the short arm of the X, and also the shorter distance between the lqh blocks in comparison with ¢. Both termini of X and Y are anchored in the nuclear membrane, f First spermatocyte with X and Y synapsed at the primary XpYp pairing segment as in e. Note the alignment of the lqh blocks shown in green by FITC, although large segments of the chromosome arms as detected by Texas Red are not paired, g First spermatocyte at a slightly later stage with closer alignment of the lqh blocks and pairing of the short arms of the X and Y. Note the long arm of the X losing anchoring to the nuclear membrane and bending, h First spermatocyte at pachytene with pairing of the lqh blocks and the 'sex vesicle' composed of a more condensed ring, closed by the primary XpYp and secondary XqYq pairing segments. Note the sharp bending of the X adjacent to the centromere, i First spermatocyte at approximately the same stage as in h, showing the homologues 1 fully paired and the 'sex vesicle' as a closed ring with a typical sharp bend adjacent to the centromere.

4 4 8 Chromosome Research Vol 2 1994

a

FISH analysis of human germline X and Y chromosomes

e

C

~iii~i~i, ~ i !~''I

h

Figure 3. Spermatocytes at a-c late first meiotic prophase, d-f first metaphase, g & h second metaphase and ipostmeiotic 'dyads'. Chromosomes X (long arrows) and Y (short arrows) are identified as in Figures 1 & 2; chromsome t (arrowheads) is detected in a, b, d & i by the lqh in green by FITC or in c, e, f -h by the combination with the whole chromosome paint, a First spermatocyte at pachytene, showing the 'sex vesicle' as a condensed ring and the lqh blocks paired, b & c First spermatocyte at late pachytene with further chromatin condensation of the 'sex vesicle' and the lqh blocks paired. Note the positioning of the X centromere interstitially, while the Xq terminus is still anchored in the nuclear membrane. Chromosomes 1 identified by lqh blocks b alone and c in combination with chromosome 1 paint, d & e Spermatocyte at first metaphase showing XY bivalent and chromosome 1 bivalent, identified by lqh blocks, d alone or • in combination with the whole chromosome paint, f Spermatocyte at first metaphase, showing univalents X and Y together with bivalent 1. g Spermatocyte at second metaphase, showing single X and chromosome 1. h Spermatocyte at second metaphase, showing single Y together with chromosome 1. i Postmeiotic haploid nuclei with Ys together with a single homologue 1 identified by its qh or single Xs.


S. J. Arms t rong et al.

intranuclear centre of the loop is located near to the centromere on Xq. The folding process continues until the telomeres of the long arms are associated and the XY appears as a ring (Figure 2h & i). Chromosome 1 is now fully synapsed.

Further extensive condensation causes shortening, and the closing of the XY ring culminates with the tightly condensed 'sex vesicle' formation, characteristic of late pachytene. We did not observe any pachytene cells failing to form an XY body. In nearly all such 'sex vesicles' examined, the X centromere and the Yqh were situated at opposite poles. The Yqh remains at the periphery, while the X centromere is located interstitially (Figure 3a-c).

Meiotic metaphases and postmeiotic products

At first metaphase (MI) the XY bivalent is held together by a presumptive chiasmatic association at XpYp (Figure 3d & e) with an undercondensation of Xp apparent in some cells. In 14% of MI cells (seven of 50) the X and Y appear as univalents (Figure 3f). Chromosome 1 bivalent, in contrast, is held together by two or more chiasmata, with clear separation of the heterochromatic blocks (Figure 3d-f). At second metaphase, segregation of the XY and autosomal bivalents is apparent, as illustrated by cells containing a single chromosome 1 plus an X (Figure 3g) or a Y (Figure 3h). Postmeiotic haploid products often appeared as 'dyads', each pair containing the XY segregation products (Figure 3i).

Discussion

We have used multicolour FISH to demonstrate the behaviour of the X and Y chromosomes in the germ line of the human male in relation to the representative autosome chromosome 1. FISH technology allows tag- ging of specific chromosome segments throughout meiosis, and this has provided a unique opportunity to study the X and Y chromosomes at relevant stages of the meiotic process.

Y heterochromatin decondensation in Sertoli cells and spermatogonial metaphases

We have demonstrated that a proportion of interphases of Sertoli cells (25%) as well as spermatogonial metaphases (28%) show a selective undercondensation of the heterochromatic block of the long arm of the Y (Yqh). Speed et al. (1993) and Guttenbach et al. (1989, 1993) made a similar observation with respect to Sertoli cells in humans and mice, respectively, suggesting that this undercondensation may be of relevance for spermatogenesis gene activity. There is some discrepancy be-

tween our own findings and those of Speed et al. (1993), both in the extent of Yqh expansion and the frequency of Sertoli cell interphases involved: they found a much more extensive Yqh undercondensation, occurring in the majority of cells (86%). The causes of this discrepancy are not known, and will require further study.

It is also of interest that we observed a substantial selective undercondensation of Yq heterochromatin in a proportion of spermatogonial metaphases (28%), as previously noted by Hult6n & Lindsten (1973) with quinacrine staining. Since it is commonly thought that decondensation of chromatin may be linked to gene expression (Weintraub & Groudine 1976), these observations may be important with respect to Yq genes involved in spermatogenesis. One possibility is a mechanism similar to the so-called heterochromatin position-effect variegation, first discovered in Drosophila and reviewed by Shaffer and co-workers (Shaffer et al. 1993). It is suggested that all genes in Drosophila may be under this influence, so that their degree of transcription can be moderated by the distance to heterochromatic blocks. We propose that a modification of this mechanism may be operating in a tissue-specific manner, as exemplified by the selective Yq heterochromatin decondensation in Sertoli cells and in particular in spermatogonial metaphases, leading to an activation of adjacent genes, located in the Yq euchromatin.

XY spatial orientation, chromatin condensation and pairing at first meiosis

There can be no doubt that the X and Y chromosomes exhibit a different spatial orientation to autosomes such as chromosome 1. At the initiation of meiosis the sex chromosomes occupy relatively restricted domains at opposite poles of the nuclear membrane, while the lqh blocks are situated interstitially and appear to become pre-aligned. The synapsis of the relatively short homologous XY primary pairing segment, comprising 2.6 Mb (Petit et al. 1988) and required for chiasma formation, is an exacting process (Hult6n & Pearson 1971). Our FISH analysis shows that the Yq heterochromatin remains condensed and anchored in the nuclear membrane-- and we were able to confirm this by subsequent FISH analysis of histopathological testis sections (results not shown). However, there is, on the other hand, a selective undercondensation and apparent stretching of the short arms of both sex chromosomes towards each other. Whether or not this may have any functional significance with respect to gene expression (Weintraub & Groudine 1976) remains to be shown.

Following the synapsis of the XpYp homologous segments, the long arm of the X, which hitherto has remained relatively contracted, is precociously condensed, when folding down, and finally reaching the tip of the long arm of the Y. The net result of this complex pattern of differential condensation and intra-


F I S H analys i s of h u m a n g e r m l i n e X and Y chromosomes

nuclear movements leads to the 'sex vesicle' becoming located adjacent to the Y but at the opposite pole of the nucleus in relation to the original position of the X.

This behaviour of the sex chromosomes may not be trivial and may have a functional significance: we suggest that the spatial separation of the main X and Y domains could provide a protective mechanism, limiting the opportunity for synapsis and chiasma formation in these segments.

XqYq secondary pairing, synapsis and exchange

A secondary association between the long arms of the X and Y was seen in all spermatocytes analysed by FISH, and the same has been previously noted in 50% of cells on electron microscope analysis of synaptonemal com- plexes (Chandley et al. 1984). We have proposed that this constitutes a secondary pairing segment and thus provides the potential for homologous synapsis and chiasma formation (Johannisson et al. 1988). Indeed, a stretch of X and Y homology of 0.4 Mb near their telomeres has recently been identified by Freije et al. (1992). It has been suggested that this XqYq homology may have arisen as a result of ectopic recombinations, mediated by interspersed repeated (LINE) sequences, originally present in the differential non-homologous stretches of X and Y DNA, with XqYq exchanges limited to gene conversion events (Kvaloy et al. 1994). It is not yet known whether both chiasma formation and gene conversion may take place within the secondary XqYq pairing segment. Either way, the XqYq association is not maintained at first meiotic metaphase when, at least in air-dried preparations, the XY bivalent is seen as a straight rod with a presumptive chiasma at the XpYp segment or, alternatively, as univalents X and Y. In this case sex chromosomal univalence at MI comprised 14%, and it may be as high as 28% (Hult6n et al. 1966). We have previously suggested that this high rate of univalence may be the result of a precocious sex chromosomal disjunction in relation to autosomal first meiotic anaphase, and may explain the high rates of aneuploidy observed in a FISH study of X and Y in sperm (Goldman et al. 1993).

'Sex vesicle' formation and gene inactivation

The bending of the X that we have observed to take place at pachytene of first male meiosis in the process of 'sex vesicle' formation produces a configuration similar to that seen by Walker et al. (1991) in the somatic Barr body. They used a combination of X centromeric and telomeric probes to demonstrate that the Barr body is a looped X chromosome structure with its telomeres associated at the nuclear membrane and the centromere located near the middle of the loop. It is presumably a retention of this interphase X loop formation, which can be seen as a selective bending of the inactive X in somatic metaphases (Flejter et al. 1984).

Somatic X-inactivation is currently thought to be related to the activity of a gene, exclusively expressed on the inactive X, and its RNA has been identified by FISH with the Barr body of somatic cells (Brown et al. 1992). Expression of XIST (human) or Xist (mouse) has, however, also been reported to occur in male germ cells (McCarrey & Dilworth 1992, Salido et al. 1992, Richler et al. 1992). Its been concluded that the same mechanism(s) of X-inactivation may be involved at male first meiosis and in the female soma. The similarities between the spatial orientation of the X in the 'sex vesicle' demonstrated here and the somatic Barr body (Walker et al. 1991) support this interpretation. We propose that the XIST/Xist gene acts in initiation of X-inactivation in both situations by imposing a chromatin reorganization, leading to bending at the X-inactivation centre (XIC). We suggest that the XIST/Xist-induced X-bending provides an opportunity for the non-homologous chromosome association in the process of heterochromatinization, presumably involving heterochromatin specific proteins. A detailed examination of the loca- tion of the bending point in the 'sex vesicle' showed this to be in the long arm of the X, at a site expected to correspond to the human XIST locus, the XIC.

One difference between the meiotic 'sex vesicle' and the somatic Barr body is, of course, the involvement of the Y chromosome in the 'sex vesicle'. It seems likely that, following the initiation of X-inactivation via the activity of the XIST/Xist gene and its proposed bending action, the inactivation process may spread to the Y. This heterochromatinization will also imply a protec- tion from XY exchanges between their differential segments, which could have potentially deleterious effects (McKee & Handel 1993) with respect to, for example the testis determining factor (TDF), located in the differential segment of Xp. The exact mechanism or mechan- isms of the heterochromatinization process are unknown. It is, however, of interest to note that the major X segment known to escape inactivation in the female soma, i.e. the tip of the short arm, corresponds to the male meiotic primary pairing segment at XpYp. The question then arises as to whether the mechanism of somatic X-inactivation is similar to the autosomal genomic imprinting (Moore & Haig 1991), where we have previously suggested that memory of parental meiotic pairing is an essential feature (Hult4n & Hall 1990). In this respect it would be of further interest to find out if the same correlation exists with respect to the shorter secondary meiotic pairing segment at XqYq, but, as far as we are aware, this area has not yet been studied in detail.

References

Armstrong SJ, Hult6n MA (1994) Positional control of meiotic recombination and interference: increased chiasma frequency in the interstitial segments of two human male translocation carriers involving chromosome 9,


S. J. Armstrong et al.

t(9;10)(p22;q24) and t(9;21)(q21;q11.2) Third Inter- national Workshop on Chromosome 9. Am J Hum Genet 58: 201, 1994.

Brown CJ, Hendrich BD, Rupert JL et aL (1992) The human XIST gene: analysis of a 17 Kb inactive X-specific RNA that contains conserved repeats and is highly localised within the nucleus. Cell 71: 527-542.

Chandley AC, Goetz P, Hargreave TB, Joseph AM, Speed RM (1984) On the nature and extent of XY pairing at meiotic prophase in man. Cytogenet Cell Genet 38: 241-247.

Flejter WL, Van Dyke DL, Weiss L (1984) Bends in human mitotic metaphase chromosomes, including a bend marking the X-inactivation center. Am J Hum Genet 36: 218-226.

Freije D, Helms C, Watson MS, Donis-Keller H (1992) Iden- tification of a second peudoautosomal region near the Xq and Yq telomeres. Science 258: 1784-1787.

Goldman ASH, Hult6n MA (1992) Chromosome in situ suppression hybridization in human male meiosis. J Med Genet 29: 98-102.

Goldman ASH, Hult6n MA (1993a) Meiotic analysis by FISH of a human male 46,XY, t(15;20)(q11.2;q11.2) translocation heterozygote: quadrivalent configuration, orientation and first meiotic segregation. Chromosoma 102: 102-111.

Goldman ASH, Hult4n MA (1993b) Analysis of chiasma frequency and first meiotic segregation in a human male reciprocal translocation heterozygote t(1;11)(p36.3;q13.1) using fluorescence in situ hybridisation. Cytogenet Cell Genet 63: 16-23.

Goldman ASH, Formina Z, Knights PA, Hill CJ, Walker AP, Hult6n MA (1993) Analysis of the primary sex ratio, sex chromosome aneuploidy and diploidy in human sperm using dual colour fluorescence in situ hybridization. Eur J Hum Genet 1: 325-334.

Guttenbach M, Schmid M, Jauch A, Vogt P (1989) The Y chromosome of the mouse is decondensed in Sertoli cells. Chromosoma 97: 429-433.

Guttenbach M, Winking H, Schmid M (1993) Organization of the Y chromosome in testis cells of fetal, subadult and adult mice as determined by in situ hybridization. Chromosoma 102: 618-622.

Heslop-Harrison JS, Bennett MD (1990) Nuclear architecture in plants. Trends Genet 6: 401-405.

Hult6n M, Pearson PL (1971) Fluorescent evidence for spermatocytes with two Y chromosomes in an XYY male. Ann Hum Genet 34: 273.

Hult6n M, Lindsten J (1973) Cytogenetic aspects of human male meiosis. Adv Hum Genet 4: 327-387.

Hult6n M, Hall J (1990) Proposed meiotic mechanism of genomic imprinting. Chromosomes Today 10: 157-162.

Hult6n M, Lindsten J, Ming Pen-Ming L, Fraccaro M (1966) The XY bivalent in human male meiosis. Ann Hum Genet 30: 119-123.

Hult6n M, Goldman ASH, Saadallah N, Wallace BNM, Creasy MR (1992) Meiotic studies in man. In: Rooney DE, Cze- pulddouski BH, eds. Human Cytogenetics. A Practical Ap- proach. Oxford: IRL Press, pp 193-221.

Johannisson R, Froster-Iskenius U, Hult6n M (1988) Sperma- togenesis in the fragile X mental retardation syndrome: II. Meiosis. J Hum Genet 79: 231-234.

Kvaloy K, Galvagni F, Brown WRA (1994) The sequence organization of the long arm pseudoautosomal region of the human sex chromosomes. Hum Mol Genet 3: 771-778.

Lichter P, Cremer T, Borden J, Manuelidis L, Ward DC (1988) Delineation of individual human chromosomes in metaphase and interphase cells by in situ suppression hybridization using recombinant DNA libraries. Hum Genet 80: 224-234.

McCarrey JR, Dilworth DD (1992) Expression of Xist in mouse germ cells correlates with X-chromosome inactivation. Nat- ure Genet 2: 200-203.

McKee BD, Handel MA (1993) Sex chromosomes, recombination and chromatin conformation. Chromosoma 102: 71-80.

Moore T, Haig D (1991) Genomic imprinting in mammalian development: a parental tug-of-war. Trends Genet 7: 45-49.

Petit C, Levilliers J, Weissenbach J (1988) Physical mapping of the human pseudoautosomal region: comparison with genetic linkage map. EMBO 7: 2369-2376.

Richler C, Soreq H, Wahrman J (1992) X inactivation in mammalian testis is correlated with inactive X-specific transcription. Nature Genet 2: 192-195.

Salido EC, Yen PH, Mohandas TK, Shapiro LJ (1992) Expres- sion of the X-inactivation associated gene Xist during spermatogenesis. Nature Genet 2: 196-199.

Shaffer CD, Wallrath LL, Elgin SCR (1993) Regulating genes by packaging domains: bits of heterochromatin in euchromatin? Trends Genet 9: 35-37.

Speed RM, Vogt P, K6hler MR, Hargreave TB, Chandley AC (1993) Chromatin condensation behaviour of the Y chromosome in the human testis. I. Evidence for decondensation of distal Yq in germ cells prior to puberty with a switch to Sertoli cells in adults. Chromosoma 102: 421-427.

Walker CL, Cargile CB, Floy KM, Delannoy M, Migeon BR (1991) The Barr body is a looped X chromosome formed by telomere association. Proc Natl Acad Sci USA 88: 6191-6195.

Weintraub H, Groudine M (1976) Chromosomal subunits in active genes have an altered conformation. Science 193: 848-856.


Documents

XY chromosome behaviour in the germ-line of the human male: A FISH analysis of spatial orientation, chromatin condensation and pairing