5727 JMed Genet 1995;32:572-575 Chromosomal localisation of a Y specific growth gene(s) J Med Genet: first published as 10.1136/jmg.32.7.572 on 1 July 1995. Downloaded from

Tsutomu Ogata, Keiko Tomita, Akiko Hida, Nobutake Matsuo, Yutaka Nakahori, Yasuo Nakagome

Abstract apparently large Yq terminal deletions are in- Although a Y specific growth gene(s) has variably sterile and occasionally have short been postulated in the Yqll region, the stature.24 However, since correlations between precise location has not been determined. genotype and stature have not been properly To localise the growth gene(s), we cor- examined, the precise location ofthe Y specific related genotype with stature in 13 Jap- growth gene(s) has not been determined. anese and four European non-mosaic In this paper, we attempt to localise the Y adult male patients with a partial Yq de- specific growth gene(s) on the basis of geno- letion. Fourteen patients preserving the type-phenotype correlations in patients with region between DYSll and DYS246 did Yq - chromosomes. not have short stature (11 Japanese, 165-180 cm; three Europeans, 165-173 cm) whereas the remaining three patients with Methods SELECTION OF PATIENTS the region deleted had short stature (two The patients analysed in the present study were Japanese, both 159 cm; one European, collected from a large series ofinfertile Japanese 157 cm). The results suggest that the region males ascertained from 1988 to 1993. The defined by DYS1I at interval 5C and by selection criteria used were: (1) measurement DYS246 at interval SD may be the critical of height between 20 and 50 years of age; region for the Y specific growth gene(s). (2) lack of associated disorders or therapeutic interventions that may affect stature; (3) ab- (J'Med Genet 1995;32:572-575) sence of demonstrable mosaicism; (4) lack of autosomal or X chromosomal abnormalities; The Y chromosome carries a Y specific growth (5) confirmation of a partial Yq deletion by gene(s) that significantly augments adult height molecular studies; and (6) exclusion of i(Yp) independently of the effect of gonadal sex ster- or idic(Y)(qll) chromosomes which are as- oids. This hypothesis is primarily based on sociated with two copies of a pseudoautosomal the white adult height comparisons between growth gene(s),5 but are not necessarily dis- patients with pure XY gonadal dysgenesis and tinguishable from non-fluorescent Yq- chro- http://jmg.bmj.com/ those with pure XX gonadal dysgenesis (172.0 mosomes by standard cytogenetic studies.6 In (SD 7.0) cm, n=24 v 164.3 (SD 7.7) cm, n= addition, male patients satisfying the above 22, p<0.01), between patients with complete selection criteria were taken from published testicular feminisation syndrome and normal reports. females (172.2 (SD 6.5)cm, n=23, v 162.2 (SD 6.0) cm, British standard, p<0.01), and HEIGHT ASSESSMENT between normal males and XX males (174.7 The adult heights of Japanese patients were on October 2, 2021 by guest. Protected copyright. (SD 6.7) cm, British standard v 166.4 (SD assessed by the Japanese height standards based n = In the three com- 7.4) cm, 33, p<0.01)."2 on the 1980 national survey (169.7 (SD 5.6) cm the sex chromosome is parisons, complement for males and 157.0 (SD 5.0) cm for females), Department of different but the effect of bioactive gonadal sex and those of white patients taken from pub- Paediatrics, Keio steroids is comparable. Furthermore, on the University School of basis of the adult height differences indicated lished reports were assessed by the British Medicine, 35 height standards reported by Tanner et al7 Shinanomachi, by the three comparisons, it has been assumed that the Y increases the (174.7 (SD 6.7) cm for males and 162.2 (SD Shinjuku-ku, Tokyo specific growth gene(s) 6.0) cm for females). When possible, adult 160, Japan adult height by 8-10 cm and thus accounts for T Ogata height was compared between the patient and N Matsuo about two-thirds of the sex difference in mean his normal sib(s) with a similar genetic and adult height (about 13 cm).`- The remaining environmental background. To allow for height Department of Human between sexes has been adult height difference comparisons between different races and be- Genetics, University of the notion that testicular an- Tokyo, 7-3-1 Hongo, explained by tween different standard deviation to increase adult sexes, height Bunkyo-Ku, Tokyo drogens have the potential score (SDS) was obtained using the equation: 113, Japan 3-5 cm while ovarian oestrogens have height by height SDS = (X - M)/SD, where X is an in- K Tomita no major effect on adult height.'-3 A Hida dividual person's height and M and SD are the has been as- Y Nakahori The Y specific growth gene(s) mean and the standard deviation of the normal Y Nakagome signed to Yqll, proximal to the gene(s) for population, respectively. on Correspondence to: spermatogenesis, the basis of karyotype- Dr Ogata. phenotype correlations.4 Non-mosaic adult Received 5 December 1994 male patients with apparently small Yq terminal MOLECULAR STUDIES Revised version accepted for not DNA was extracted from publication 15 February deletions are frequently sterile but do Genomic peripheral 1995 usually have short stature, whereas those with leucocytes of each Japanese patient and was Chromosomal localisation of a Y specific growth gene(s) 573 1 2 3 4 5 L-( < o~ a) - HIC | II 1 (O -4.n C _s n X E a >- ) mm:^> t en ) 0 LO OOO O-O <00 0CY) 000000000',- m >. > > >->-r- r- LLU- X N eeOM) U.rW> en en N _n.- -4-X < x LLx x> > > > > > 2 > > > > >--> > > > > > > >- J Med Genet: first published as 10.1136/jmg.32.7.572 on 1 July 1995. Downloaded from (1--N00000000<:000c0 000 0000001C l I I Figure 1 An example of Case 1 46, XY ._-M 1 180 +1.84 discrimination between 2 46,XY 1 175 +0.95 non-fluorescent Yq - 3 46,XY ~~mEI~-- 1 173 +0.59 chromosomes and i(Yp) or 4 46,XY ii-ii i ~~~- -m 171 +0.23 5 46,XY 171 +0.23 idic(Y)(qll) chromosomes. - mu - mum * - ~~~~~---m Shown are TaqI digests 6 46,XY 1 165 -0.84 7 46, X,Yq-(nf) 69 -0.13 hybridised with 47z 8 46, X,Yq-(nf) ~mI------175 +0.95 Japanese defining an Xq-Yp 9 46,XY 173 +0.59 homologous locus 10 46, X,Yq-(nf) ~~IE ~ ~ M 172 +0.41 (DXYS5) (lane 1: control 11 46,X.Yq-(nf) 176 +1.13 male; lanes 2, 3, and 5: 12 46, X,Yq-(nf) 159 -1.91 infertile male patients with 13 46,X.Yq-(nf) 159 -1.91 46,X, + mar; lane 4: XX male). The intensity ratio 14 46, X,Yq-(nf) 173 -0.26 between the X specific and 15 46. X,Yq-(nf) - 2 ** * *- - -- 172 -0.41 i Y specific bands is 16 46, X,Yq-(nf) * ~ ~~~~~~~* * *- 165 -1.46 Whites comparable between lanes 17 46, X,Yq- 157 -2.661i 1, 2, and 3, indicating that DXYSSY is present in a single copy in patients 2 I. and 3. Thus, the two ,, patients were interpreted as having a 46,X, Yq- karyotype and were u a 0 - included in the present fMI*l I Iw11 IH study (cases 11 and 12 in fig 2). By contrast, the comparison of band intensity between lanes 1 and 5 implies that DXYSSY is present in two Case 11 - - - - copies in patient 5. Thus, the patient was regarded as 137 - - -X having a 46,X,i(Yp) or a 46,X,idic(Y) (qll) karyotype and was excludedfrom the present study. The comparison of Critical region for band intensity between a Y specific growth gene(s) lanes 1 and 4 is consistent with the XX male patient Figure 2 Genotype-phenotype correlations of a Y specific growth gene(s) in non-mosaic patients with Yq - having two copies of chromosomes. The case numbers correspond to those in the table. Karyotype (nf= non-fluorescent), height, and height DXYSSX. SDS (standard deviation score) are shown for each case. The black segments represent the positive loci confirmed by http://jmg.bmj.com/ molecular studies, the striped segments represent the presumed positive loci inferredfrom interpolation or cytogenetic findings, the minus symbols represent the negative loci confirmed by molecular studies, the open segments represent the presumed negative loci inferredfrom interpolation or cytogenetic findings, and the asterisks represent the dosage unknown loci. The critical region for a Y specific growth gene(s) is shown by the arrow.

analysed by Southern blotting and the poly- 52d (DYF27/A-C), 87-4 (AMGL), 12f3 on October 2, 2021 by guest. Protected copyright. merase chain reaction (PCR). For Southern (DYSl1), 87-28 (DYS140), 87-26 (DYS139), blot analysis, genomic DNA was digested with 87-31 (DYS134), 87-19 (DYS132), 87-6 EcoRI, TaqI, or StuI, and was hybridised with (DYS135), 49f (DYSI), and pHY10 the following probes: pDP1007 (ZFY), 47z (DYZ1).5-" For PCR analysis, genomic DNA (DXYS5Y), 87-27 (DXYS73Y), 50f2 (DYS7/ was amplified with following primers: PABA/ A-E), 87-17 (DYS131/A,B), 87-7 (DYS13O), PABB/PABC (PABX/PABY), SRY-lF/SRY-2R (SRY), sY78 (DYZ3), sY81 (DYS271), sY82 (DYS272), sY83 (DYSli), sY84 (DYS273), Adult male patients used in the present study sY87 (DYS275), sY86 (DYS 148), sY85 Case Reference (DY274), sY165 (DYS246), sY90 (DYS278), sY91 (DYS 136), sY94 (DYS279), and sY95 Japanese 1 Unpublished 2 Nagafuchi et all5 (patient No 1411) (DYS280). The primer sequences were as de- 3 Nagafuchi et al" (patient No 1542) scribed previously'2'4 and the annealing tem- 4 Nagafuchi et al" (patient No 1469) 5 Nagafuchi et al'5 (patient No 1508) perature was 54°C for PABA/PABB/PABC, 6 Nagafuchi et al" (patient No 1561) 65°C for SRYlF/SRY-2R, and 60°C for the 7 Unpublished 8 Unpublished remaining primers. 9 Nagafuchi et al'5 (patient No 1560) The order and interval assignment of loci 10 Unpublished 11 Nakahori et al'0 (patient No 1068) examined by Southern blot analysis were based 12 Nakahori et al'o (patient No 1069) on the reports ofNakahori et al'0 and Nagafuchi 13 Unpublished White 14 Bardoni et al'8 (patient No 22) et all5 and those of loci examined by PCR 15 Bardoni et al'8 (patient No 13) analysis were based on the report of Affara et 16 Bardoni et al'8 (patient No 14) 16 17 Skare et al19 and Vollrath et al'4 (patient WHT al. To integrate the order of Yq loci analysed 1165) in the present study, the yeast artificial chro- 574 Ogata, Tomita, Hida, Matsuo, Nakahori, Nakagome mosome (YAC) contig reported by Foote et Discussion al'7 was screened with 87-28, 87-26, 87-31, The present study suggests that the region and 87-19. between DYSI1 at interval 5C and DYS246 at interval 5D may be the critical region for the Y specific growth gene(s) (fig 2). All of the

14 patients with Yq - chromosomes preserving J Med Genet: first published as 10.1136/jmg.32.7.572 on 1 July 1995. Downloaded from * **< Results the critical region were free from short stature SELECTION OF PATIENTS (cases 1 to 11 and cases 14 to 16), and all the Thirteen Japanese patients, including eight pre- three patients with Yq - chromosomes missing viously reported cases,'015 met the selection the critical region had short stature (cases 12, criteria; seven had an apparently normal fluor- 13, and 17). Although the short stature in the escent Y chromosome and six had a non- three cases could also be partly accounted for fluorescent Yq - chromosome. Discrimination by growth disadvantage resulting from chro- between non-fluorescent Yq- chromosomes mosome imbalance,' it appears unlikely that and i(Yp) or idic(Y)(ql1) chromosomes was the short stature is totally attributable to this: carried out by dosage blottings (fig 1). In ad- case 11 was normal in height in the presence dition, four white patients were identified in of chromosome imbalance similar to that of --- published reports~~~~~~~141819; one hadha a fluorescent the three cases. Rather, the coexistence of IFigure 3 Representative Yq - chromosome'4 19 and three had a chromosome imbalance may explain why the rresults of Southern blot non-fluorescent Yq - chromosome. 8 Dis- degree of short stature in the three cases is 6analysis. Upper panel: crimination between non-fluorescent Yq- apparently greater than the estimated deletional 1TaqI digests probed with criminatio b n on-locent Yq - 452d (DYF27). DYF271A chromosomes and i(Yp) or idic(Y)(ql 1) chro- effect of the Y specific growth gene(s) (8 to con Yq is deletedfrom cases mosomes was based on dosage blottings. 10 cm). Interestingly, the critical region consists jUYand 12, whereas Thus, a total of 17 patients was used in the ofY specific single copy DNA sequences rather I and than of X homologous or Y specific repetitive urr,C,DYF27*Bip "It PJ{oavi-v".IYF27rCLu-CvUr present study (table). panel: TaqI digests probed DNA sequences.'7 This agrees with the growth with 87-28 defining the X gene(s) postulated on the critical region acting specific locus (DXYS74X) and the Y specific locus in a Y specific manner, and implies the growth (DYS140). DYS140 is GENOTYPE-PHENOTYPE CORRELATIONS gene(s) being single rather than several in num- present in case 11 but is The results are summarised in fig 2. Southern ber. Furthermore, according to the detailed absent from case 12. blot analysis for the Japanese patients (cases genetic map reported by Affara et al'6 the critical 1-13) showed that 11 patients with interstitial region is estimated to be about 1 Mb in physical or terminal Yq deletions accompanied by the length. It is expected, therefore, that the Y breakpoints distal to DYS140 were normal in specific growth gene(s) is located in a fairly height (cases 1 to 11) whereas the remaining small region. two patients with terminal Yq deletions ac- However, there are several problems in the companied by the breakpoints proximal to present study. First, it was impossible to allow DYS 140 had short stature (cases 12 and 13) for several influences on stature, such as par- (fig 3). In cases 11 and 12, the sib's adult ental height (genetic height potential), socio-

height was available; despite the similar size of economic status, and secular trend.2 22 Second, http://jmg.bmj.com/ deletion between cases 11 and 12, case 11 had latent mosaicism can not be excluded. Third, a height SDS comparable to that of the normal there may be undetected complex deletions. sib (sister 164 cm, SDS + 1.39), and case 12 Lastly, the patients analysed were few in num- had an obviously lower height SDS than the ber. normal sibs (brother 175 cm, SDS + 0.95; sis- Furthermore, the possibility that another Y ter 159 cm, SDS + 0.20). Similarly, three white specific growth gene(s) is present on Yp has

patients with terminal Yq deletions with break- not been excluded formally. Since the mean on October 2, 2021 by guest. Protected copyright. points distal to STSP were normal in height adult height of white Yp(+)XX males con- (cases 14 to 16) whereas the remaining one firmed by molecular analysis (166.2 (SD patient with an interstitial Yq deletion with 5.8) cm, n = 14)2 is consistent with loss ofthe Y breakpoints proximal to DYS 11 and distal to specific growth gene(s), and the adult height of DYS 1 had short stature (case 17). In case a white Yp(-) XY female shown by molecular 16, the sib's adult height was recorded; height analysis (168cm)23 is comparable to that of SDS was comparable between case 16 and the patients with pure XY gonadal dysgenesis, it normal sib (brother 163 cm, SDS - 1.76). appears that a Y specific growth gene(s) is PCR analysis for cases 11 to 13 showed absent from the distal to the middle part of Yp that the breakpoint was between DYS246 and (in such patients, the pseudoautosomal growth DYS274 in case 11, between DYS273 and gene(s) is expected to be present in two copies DYS 11 in case 12, and between DYS271 and after abnormal Xp;Yp interchanges2). How- DYZ3 in case 13 (fig 4). In addition, the ever, it might be possible that a Y specific proximal breakpoint in case 17 was found to growth gene(s) is actually transferred onto Xp be between DYS271 and DYZ3. in several Yp(+) XX males but is subject to The screening of the Y chromosomal YAC random or non-random X inactivation, re- contig with four probes defining Yq loci as- sulting in insufficient statural effect. Similarly, signed DYS140 to interval 5C (between only a single adult height is available for Yp(-) DYS 11 and DYS273), DYS 139 to interval XY females (primarily because of loss of a 5G (between DYS 136 and DYS279), DYS 134 gene(s) for Turner stigmata on Yp24). In ad- to interval 5K (distal to STSP), and DYS 132 dition, the possibility that a Y specific growth to interval 50 (distal to STSP) (fig 2) (data gene(s) exists in the proximal part ofYp remains not shown). tenable. Chromosomal localisation of a Y specific growth gene(s) 575 Despite the above caveats, our results offer termining region of the human Y chromosome encodes a Va) finger protein. Cell 1987;51:1091-104. E the first clue to the cloning of the Y specific 9 Vergnaud G, Page DC, Simmler MC, et al. A deletion map E ofthe human Y chromosome based on DNAhybridization. -CN CD) growth gene(s). In summary, although geno- 0- AmJHum Genet 1986;3:109-24. a) Q a type-phenotype analysis of the Y specific 10 Nakahori Y, Tamura T, Nagafuchi S, et al. Molecular cloning C U) en U) 0 O co X growth gene(s) is still not conclusive, we pro- and mapping of 10 new probes on the human Y chro- ucJ U Q u( mosome. Genomics 1991;9:765-9. pose that the Y specific growth gene(s) may be 11 Nakahori Y, Mitani K, Yamada M, Nakagome Y. A human J Med Genet: first published as 10.1136/jmg.32.7.572 on 1 July 1995. Downloaded from Y-chromosome specific repeated DNA family (DYZ1) present in the critical region defined by DYS 1 1 consists of a tandem array of pentanucleotides. Nucleic and DYS246. Further accumulation of in- Acids Res 1986;14:7569-80. DYZ 3 12 Ellis N, Taylor A, Bengtsson BO, Kidd J, Rogers J, Good- formative patients will permit a more precise fellow P. Population structure of the human pseudo- 4-170 bp localisation. autosomal boundary. Nature 1990;344:663-5. 13 Nakagome Y, Seki S, Fukutani K, Nagafuchi S, Nakahori Y, Tamura T. PCR detection of distal Yp sequences in an 4, XX true hermaphrodite. AmJMed Genet 199 1;41: 112-14. We would like to thank Drs Hideki Fuse, Mikio Namiki, No- 14 Vollrath D, Foote S, Hilton A, et al. The human Y chro- DYS 271 buyuki Kondo, Teruaki Iwamoto, Ichiro Taniuchi, and Hidenori mosome: a 43-interval map based on naturally occurring 4-209 bp Sumitani for providing us with blood samples and clinical deletions. Science 1992;258:52-9. data. This work was supported by Grant-In-Aid for Scientific 15 Nagafuchi S, Namiki M, Nakahori Y, Kondoh N, Okuyama Research from the Ministry ofEducation, Science, and Culture, A, Nakagome Y. A minute deletion of the Y chromosome by Grant for Paediatric Research from the Ministry of Health in men with azoospermia. Jf Urol 1993;150:1155-7. DYS 11 and Welfare, and by Grant from Vehicle Racing Commemorative 16 Affara NA, Lau YFC, Briggs H, et aL Report of the first Foundation. international workshop on Y chromosome mapping 1994. 4-i275i*-275 bp Cytogenet Cell Genet 1994;67:359-401. 17 Foote S, Vollrath D, Hilton A, Page DC. The human Y chromosome: overlapping DNA clones spanning the 1 Ogata T, Matsuo N. Comparison of adult height between euchromtic region. Science 1992;258:60-6. DYS 273 patients with XX and XY gonadal dysgenesis: support for 18 Bardoni B, Zuffadi 0, Guioli S, et al. A deletion map of a Y specific growth gene(s). JMed Genet 1992;29:539-41. the human Yql 1 region: implications for the evolution of 4-326 bp 2 Ogata T, Matsuo N. Sex chromosome aberrations and the Y chromosome and tentative mapping of a locus stature: deduction of the principal factors involved in involved in spermatogenesis. Genomics 1991;11:443-51. the determination of adult height. Hum Genet 1993;91: 19 Skare J, Drwinga H, Wyandt H, van der Spek J, Troxler R, Milunsky A. Interstitial deletion involving most ofYq. Am YS 274 3 Varrela551-62.J, Alvesalo L, Vinkka H. Body size and shape in 46, J Med Genet 1990;36:394-7. 6-369 bp XY females with complete testicular feminization. Ann 20 Mueller WH. The genetics of size and shape in children Hum Biol 1984;11:291-301. and adults. In: Falkner F, Tanner JM, eds. Human growth. 4 Biuhler EM, A synopsis of the human Y chromosome. Hum 2nd ed. Vol 3. New York: Plenum Press, 1985:145-68. DYS 246 Genet 1980;55:145-75. 21 Eleveth PB. Population differences in growth: environmental 5 Ogata T, Petit C, Rappold G, Matsuo N, Matsumoto T, and genetic factors. In: Falkner F, Tanner JM, eds. Human 4-2(07 bp Goodfellow P. Chromosomal localisation of a pseudo- growth. 2nd ed. Vol 3. New York: Plenum Press, 1985: autosomal growth gene(s). J Med Genet 1992;29:624-8. 221-39. ttive 6 Ferguson-Smith MA. Genotype-phenotype correlations in 22 van Wieringen JC. Secular growth changes. In: Falkner F, Figure 4 Represen individuals with disorders of sex determination and de- Tanner JM, eds. Human growth. 2nd ed. Vol 3. New York: results of PCR anal{ysis.vesis. velopment including Turner's syndrome. Dev Biol 1991; Plenum Press, 1985: 307-31. Case 11 is positive ifor 2:265-76. 23 Muller U, Kirkels VGH, Scheres MJ. Absence of Turner DYZ3, DYS271, LDYSIJ, 7 Tanner JM, Whitehouse RH, Takaishi M. Standards from stigmata in a 46,XYp-female. Hum Genet 1992;90:239-42. DYS273, and DY'S274, birth to maturity for height, weight, height velocity, and 24 Ogata T, Matsuo N. Turner syndrome and female sex and negative for DIYS246. weight velocity: British children, 1965, Part II. Arch Dis chromosome aberrations: deduction of the principal fac- Case 12 Child 1966;41:613-35. tors involved in the development of clinical features. Hum is positive jfor 8 Page DC, Mosher R, Simpson EM, et al. The sex-de- DYZ3, DYS271, aIna7ind Genet (in press). DYS11, and negatiyve for DYS273 and DYS274. Case 13 is positive jfor DYZ3 and negativee for

DYS271, DYS11, and http://jmg.bmj.com/ DYS273. on October 2, 2021 by guest. Protected copyright.