10TH ANNIVERSARY ARTICLE Physical Mapping of the Human Chromosome 11q23 Region Containing the AtaxiaTelangiectasia Locus Shan Wei, Mariano Rocchi, Nicoletta Archidiacono, Nicoletta Sacchi, Giovanni Romeo, and Richard A. Gatti

A B S T R A C T : Two breakpoints within chromosome 11q23 were characterized with 29 DNA probes to

establish a physical map of the region. This region is notable in that it contains at least 14 functional genes which are also syntenic in the mouse (chromosome 9). Chromosome 11q23 includes these markers: STMY, CLG, NCAM, DRD2, APOA1, APOC3, APOA4, CD3E, CD3D, CD3G, PBGD, THY1, ets-1, and cbl-2. The two breakpoints, herein called "X;11" and "4;11," defined a region of approximately 8 cM containing the APO and CD3 complexes as well as the polymorphic marker DllS29. DRD2 localized centromeric to the X;11 breakpoint despite evidence for close genetic linkage to DllS29, suggesting that DRD2 lies close to the X;11 breakpoint. THY1, PBGD, and cbl-2 localized telomeric to the 4;11 breakpoint and thus to the [1911S29--APO--CD3] grouping as well. The physical map helps to correlate the cytogenetic and linkage maps of this region. It also suggests that the human 1 lq23 syntenic grouping is inverted with respect to its murine counterpart. Based on this physical map and on our primary linkage map of the 1 lq23 region, we are able to confirm a preliminary localization of the gene for ataxiatelangiectasia group A (ATA) to a region centromeric to the interval defined by DllS144 (pYNB3.12) and THY1.

INTRODUCTION W e r e c e n t l y l o c a l i z e d a gene for ataxia-telangiectasia group A [ATA] to c h r o m o s o m e 11q22-23 by linkage analyses u s i n g t w o genetic markers, THY1 and pYNB3.12 [1]. D N A f r o m a large A m i s h p e d i g r e e was u s e d to s c r e e n 171 markers, or a p p r o x i m a t e l y 35% of the g e n o m e , before this linkage was found. The most likely o r d e r of the three loci p l a c e d A T A o u t s i d e of THY1 and pYNB3.12, closer to pYNB3.12 (pYNB3.12 is hereafter t e r m e d D l l S 1 4 4 . It is also t e r m e d MCT128.1, D l l S 1 4 8 and D l l S 2 8 5 in the literature.) A d d i t i o n a l p o l y m o r p h i c markers in this r e g i o n h a v e since b e e n tested. H o w e v e r , w e c a n n o t c a l c u l a t e l o c a t i o n scores for A T A u n t i l the p r o p e r o r d e r and d i s t a n c e s b e t w e e n t h o s e markers h a v e b e e n established. W e define h e r e i n w h e t h e r v a r i o u s markers on distal 11q l o c a l i z e c e n t r o m e r i c

From the Department of Pathology, University of California at Los Angeles (S. W., R. A. G.), Los Angeles. California, Department of Human Genetics, Gaslini Institute, Genoa (M. R., N. A., G. R.), Italy, and the National Cancer Institute (N. S.), Frederick, Maryland. Address reprint requests to: Dr. R. A. Gatti, M.D., UCLA School of Medicine, Department of Pathology, Los Angeles, CA 90024. Received July 13, 1989; accepted August 10, 1989.

1 © 1990 Elsevier Science Publishing Co., Inc. 655 Avenue of the Americas, New York, NY 10010

Cancer Genet Cytogenet 46:1-8 (1990}

0165-4608/90/$03.50

2

s. Wei et al. or telomeric to each of two breakpoints. These breakpoints define a constitutional abnormality t(X;11) [2] and a translocation t(4;11) of an acute leukemia [3].

MATERIALS AND METHODS Hybrids The somatic cell hybrid HY.87Z4, as described by Rocchi et al. [2], was obtained by fusing the HPRT-deficient Chinese hamster ovary (CHO) cell line YH.21 with the human fibroblast cell line GM3552, obtained from NIGMS Human Mutant Cell Repository (Camden, NJ). The fibroblast line GM3552 is from a patient with the following chromosome constitution: 46,X,t(X;11)(Xpter-*Xq26::11q23-~11qter;11pter-* 11q23::Xq26-*Xqter). The hypoxanthine aminopterin thymidine (HAT)-selected clone HY.87Z4 retained the derivative chromosome 11pter~11q23::Xq26~Xqter as well as autosomes 1, 2, 4, 5, 6, 12, 15, and 21. The cytogenetic characterization of this hybrid was confirmed using appropriate X-linked probes whose regional mapping was known from the literature. Cytogenetic studies were repeated just before the DNA used in these studies was extracted. HY.87Z4 derives from the same t(X;ll)-containing human fibroblast, GM3552, as the microcell hybrid cell line MC-1, described by Scott et al. [4] and characterized by Maslen et al. [5] and retains the same derivative chromosome. Results described herein using HY.87Z4 are identical to those we obtained using MC-1 (courtesy of Tom Glaser). A panel of somatic cell hybrids containing the derivatives of the chromosome translocation t(4;11) (q21;q23) in a Chinese hamster background (fibroblast strain A3 ) was also used. Reciprocal fragments of the 4;11 translocation of cell line RS4;11 [3] were present in hybrids 27B (containing 11q+) and 11B (containing 4 q - ) [see reference 6]. In this way, the hybridization results with each probe had to be positive on one hybrid and negative on the other or vice versa. This was indeed observed in all experiments.

DNA Probes and Southern Blotting The 29 DNA probes used in these studies were obtained from sources shown in Table 1. Southern blotting was performed as previously described [7].

RESULTS Figure 1 shows a typical Southern blot used in these experiments (eg, in Fig. 1, an NCAM probe was hybridized to DNA from HY.87Z4 in lane 1). The probe crosshybridized strongly with CHO DNA (lane 3) but gave no signal for those sized bands in lane 4 containing only human DNA. Conversely, the band in lane 4 was also visualized in lanes I and 2, indicating that the neural cell adhesion molecule (NCAM) gene is present on human chromosome 11 and localizes centromeric to the breakpoint at chromosome 11q23. Chromosome 11q23 probes were localized relative to the X;11 and 4;11 breakpoints in similar Southern blotting experiments. As shown in Fig. 2, the results allowed us to locate the 4;11 breakpoint telomeric to the X;11 breakpoint. The 4;11 breakpoint is bracketed by the APO gene cluster and by THY1. Our data showed that: 1) DllS144 ( - pYNB3.12 -- MCT128.1 = DllS148 = DllS285) localized centromeric to DllS29 (= L7), 2) DllS98 localized not near DllS83 but centromeric to the X;11 breakpoint, 3) NCAM localized centromeric to both breakpoints, in contrast to the mouse genome,

3

Ataxia-Telangiectasia Locus

Table 1

Description and source of probes used

Locus

Probe name

Source

STMY CLG FTH DllS85 DllS98 DllS34 DllS35 DllS36 phage 6-8 DllS132 DllS144 CJ52.208 NCAM DRD2 DllS29 APO APO APO HHH172 CD3E CD3G PBGD THY1 cbl-2 CJ52.12 DllS147 DllS138 DllS133 ets-1 DllS83

psp64 pCLLACE1 milL217-1 ph6-3 ph9-11 ph2-11 ph2-22 ph2-14 phage 6-8 L424 pYNB3.12 CJ52.208 PM1.3 hD2G1 L7 APO-A1 APO-A4 APO-3' HHH172 PDJ4

N. Spurr ATCC W. Salser M. Litt M. Litt M. Litt M. Litt M. Litt M. Litt CRI Y. Nakamura C. Julier S. Karathanasis O. Civelli D. Retief J. Breslow J. Breslow S. Karathanasis Y. Nakamura C. Terhorst C. Terhorst ATCC J. Silver H. Morse C. Julier Y. Nakamura CRI CRI ATCC M. Litt

c-cbl CJ52.12 HBI18P1 R548 L451 PHE5.4 ph2-25

Abbreviations: ATCC, American Tissue Culture Collection; CRI, CollaborativeResearch, Inc. in w h i c h it maps a p p r o x i m a t e l y 2 cM telomeric to THY1, 4) the relatively n e w marker, HHH172, localized telomeric to the 4;11 breakpoint, thereby confirming its localization by genetic linkage (C. Julier and G. M. Lathrop, personal communication) and dissociating it from D l l S 1 4 4 , CJ52.208, and thus most likely, from close linkage with the A T A locus; 5) APO localized telomeric to the X;11 breakpoint yet centromeric to the 4;11 breakpoint, thereby distancing it from THY1, in contrast to earlier genetic linkage data that placed THY1 at a p p r o x i m a t e l y I cM from the APO c o m p l e x [8], and in contrast to its telomeric position relative to THY1 in the mouse genome, 6) the oncogene ets-1 localized telomeric to the 4;11 breakpoint, in contrast to its position 9.8 cM centromeric to THY1 in mouse [9]; 7) the new oncogene, cbl-2 [10] maps to c h r o m o s o m e 11q and localizes telomeric to the 4;11 breakpoint; and 8) DRD2 localizes centromeric to the X; 11 breakpoint. This w o u l d place it very near the X; 11 breakpoint, since Gelernter et al. [11] reported a close linkage to D l l S 2 9 . The arbitrary DNA probe D l l S 3 4 (= phage 2-11), described by Maslen et al. [5], presented special problems in localization. In three experiments, we had observed positive h y b r i d i z a t i o n signals to DNA from HY.87Z4, MC-1, and RS4;11 hybrids. This a p p e a r e d to localize D l l S 3 4 centromeric to the X;11 breakpoint. However, because this result conflicted with its localization by Maslen et al. [5] telomeric to the same breakpoint (as defined by the MC-1 hybrid), we continued these experiments. We

4

s. Wei et al.

1. 1 l q d e l (q23-qter) + CHO 2. entire 11 + CHO DNA 3. CliO DNA o n l y 4. h u m a n DNA o n l y

i~ii~i~~

NCAM p r o b e

Figure 1 An autoradiogram from a typical Southern blot used in characterizing breakpoints on chromosome 11q in somatic cell hybrids. Lanes 2 through 4 contain controls, as indicated. A positive human-specific signal in lane 1 indicates the presence of the gene being studied [in this case, neural cell adhesion molecule {NCAM}] and identifies its position as centromeric to the hybrid cell's 11q breakpoint. Because absence of a signal would place a locus telomeric to the breakpoint, autoradiograms were purposely overexposed. Further interpretation is provided in text.

isolated a 2.2-kilobase (kb) SalI/EcoRI insert from the phage vector and confirmed the probe's identity by the pattern of its MspI p o l y m o r p h i s m [5]. We then discovered two $34 bands w h e n h y b r i d DNAs were digested with EcoRI: a strong-intensity band on control DNAs and on 4q + hybrids, and a weaker intensity band present on both 1 1 q - and 4q + h y b r i d s derived from the same patient [3, 6] (and thereby reciprocal}. The latter hybrids (11B and 27B in reference 6) both contained chromosomes 3, 8, 12, 13, 16, 17, 19, 20, 22, and possibly 10. We c o n c l u d e d that the primary D l l S 3 4 locus is on c h r o m o s o m e 11, telomeric to the 4;11 breakpoint, whereas a secondary, crossh y b r i d i z i n g signal is present on one of the above cited chromosomes. The D l l S 3 4 probe d i d not cross-hybridize with either mouse or Chinese hamster DNA.

DISCUSSION The c h r o m o s o m e 11q23 region is c o m p o s e d of two large Giemsa-negative (light) bands d i v i d e d by a narrow Giemsa-positive (dark} band. A n array of functional genes localizes to this region: stromelysin (STMY), collagenase (CLG), NCAM, d o p a m i n e

Ataxia-Telangiectasia Locus

5

21 22.1

8TMY CLG 8132 (L424)

$85

898

$36

NCAM

Phage 6-8

835 FTH

22.2 22.3

CJ52. 208 8144 (pYNB3.12)

23.1 23.2

DRD2

/

X; 1 1 4;11 ~

23.3

Ewing's 24 25

11q

APO Complex $29 (L7) CD3 Complex

"

8 cM

HHH172 PBGD THY1 Cbl-2 CJ52.12 $147 (HE~18P1) 8133;8138 (L451;R548) Ets-1

834 883 Figure 2 Schematic presentation of distal 11q showing breakpoints and localization of 26 loci in reference to breakpoints X;11 and 4;11. Markers in box are poorly localized on proximal 11q. The Ewing's breakpoint is shown for orientation only [14].

D2 receptor (R2D2), high-density lipoproteins (APOA1-APOC3-APOA4), T cell T3 receptor (CD3E-CD3D-CD3G), THY-1, porphobilinogen deaminase (PBGD), and oncogenes ets-1 and cbl-2. Especially interesting to us is the ataxia-telangiectasia group A (ATA) gene, and in light of the immunodeficiency that exists in this disorder, those genes belonging to the immunoglobulin supergene family [12] (CD3G, CD3D, CD3E, NCAM, and THY1). Before the ATA gene can be further localized within the 11q22-23 region, other markers must be anchored in place in their proper order. Linkage analysis involves finding a polymorphic restriction enzyme digestion site within a specific genomic area and then typing many large families with the polymorphic probe/enzyme combination. By repeating this exercise with other polymorphic markers, one can eventually link markers in a region to one another and estimate the physical distance from the percentage of recombination between them. However, recombination fractions typically have an SD of 5 to 10 cM, depending on the number of meioses that can be characterized with both polymorphisms. Thus, when several markers lie within 5 cM of one another, the amount of linkage data necessary to order them may become prohibitive, especially if these markers are not informative in a major proportion of the population studied. Somatic cell hybrids containing chromosomal breakpoints serve a very useful purpose for physical mapping of markers that lie centromeric or telomeric to specific breakpoints. By combining information derived from hybrids with nearby breakpoints, the physical order of markers can be further refined. Other investigators have used various types of somatic cell hybrids to order loci within the chromosome 11q22-23 region [5, 13-15]. Maslen et al. [5] localized seven markers using a panel of hybrid cells containing 1 1 q - derivatives which were subcloned from human parental cells containing translocations involving chromosome

6

S. Wei et al. 11. Glaser [13] established HAT-sensitive hybrids with 11;22 breakpoints in an attempt to fine map the region around the Ewing's sarcoma breakpoint at 11q23. Budarf et al. [14] were similarly interested in localizing markers around the constitutional and tumor-associated 11;22 translocation sites. [Although t(11;22)(q23;q11) is the most common constitutional translocation, it is not generally associated with disease or cancer risk.] Savage et al. [15] were interested in mapping the ets-1 oncogene relative to breakpoints associated with the 9;11 and 4;11 translocations that occur in certain leukemias [16]. From these studies a general physical map of the region has developed. Budarf et al. [14] localized [D11S35-D11S85-D11S144] centromeric to the constitutional breakpoint, [D11S29-D11S34-D11S147-APO-CD3-ETS1-PBGD-THY1] in between the constitutional and Ewing's breakpoints, and [D11S83-D11S 129] telomeric to the Ewing's breakpoint. Maslen et al. [5] localized [D11S35-D11S36-D11S84] and [D11S85DllS98] centromeric to the MC-1 breakpoint, and [D11S34-D11S83] telomeric to it. Glaser [13] localized [APO-CD3-PBGD-ETS1-THY1-ESA4] to the telomeric portion of 11q23. When the information derived from studying both of our hybrids is considered together, the two breakpoints define a region of approximately 8 cM containing the APO and CD3 gene complexes, as well as the polymorphic marker DllS29. Our studies, however, do not resolve the order of these markers. Because either of the breakpoints could have intersected the A1-C3--A4 complex or the CD3E--CD3D-CD3G complex, we tested probes at both ends of these complexes. The breakpoints did not appear to fall within either gene complex. Data derived from studies with radiation-induced or microcell hybrids, such as MC-1 [4], cannot be interpreted as confidently as data using hybrids subcloned from human cells containing reciprocal translocations because radiation-induced and microcell hybrids may contain two distinct segments of DNA within the same hybrid cell. Such data would misleadingly appear to localize a group of markers to a single contiguous segment. We used a large array of markers in our studies so that these localizations could be integrated with other partial maps into a preliminary consortium map of 11q23. Our preliminary linkage analyses of loci in this region [17] are in agreement with the physical order we now report. Based on these preliminary localizations, we were able to confirm our previous mapping of the ATA gene to a region centromeric to DllS144 (pYNB3.12). The region of approximately 10 cM between DllS144 and THY1 [17] now appears to be excluded by three-point mapping of three contiguous intervals between DllS144 and THY1 (data not shown: ref 18). Furthermore, linkage analyses of AT with markers telomeric to THY1 give decreasingly significant LOD scores [18]. The murine chromosome 9 region includes a number of genes that are also contained in the human 11q23 region, eg, NCAM, the CD3 and APO complexes, PBGD, THY1, ets-1, and cbl-2. Several of these are members of the immunoglobulin supergene family [12], and this region is a good candidate for the location of the primordial gene from which this supergene family evolved. Our physical and linkage mapping data suggest that the order of functional genes such as NCAM, APO, and ets-1 in the homologous syntenic region on mouse chromosome 9 may be inverted in comparison to the human region on chromosome 11q23. Despite this, the ATA locus may be included in this syntenic group containing several candidate genes. Both THY-1 and CD3-epsilon are expressed on thymocytes, lymphocytes and on cerebellar Purkinje cells. NCAM is believed to play a major role in neural tissue morphogenesis and nerve fiber pattern formation. The immunoglobulin supergene family has been postulated to have evolved from a cellular adhesion-type molecule [19]. Although the function of the oncogene ets-1 is unknown, it is highly expressed in thymocytes and T lymphocytes [20] and may also be a member of the immunoglobulin supergene family. Thus,

Ataxia-Telangiectasia Locus

7

in keeping with the suggestion that the 11q23 region contains genes that play a major role in the differentiation and development of the i m m u n e and neurological systems, it is noteworthy that both thymus and cerebellum are affected in AT homozygotes, suggesting that the ATA gene on chromosome 11q23 may also belong to this neuroimm u n e supergene family.

The authors thank the investigators listed in Table 1 for contributing the DNA probes used in this work. They also thank P. Charmley for assistance in manuscript preparation and A. Geurts van Kessel for furnishing hybrid cell lines carrying the 4;11 breakpoint. The information contained in this report was incorporated into the HGMIO maps of chromosome 11q. R. A. G. is a member of the Jonsson Comprehensive Cancer Center. This work was supported by grants from the US Department of Energy, The American Cancer Society, the Ataxia-TelangiectasiaMedical Research Foundation, the Joseph Drown Foundation, and the ELM fund.

REFERENCES 1. Gatti RA, Berkel I, Boder E, Braedt G, Charmley P, Concannon P, Ersoy F, Foroud T, Jaspers NGJ, Lange K, Lathrop GM, Leppert M, Nakamura Y, O'Connell P, Paterson M, Salser W, Sanal O, Silver J, Sparkes RS, Susi E, Weeks DE, Wei S, White R, Yoder F (1988): Localization of an ataxia-telangiectasia gene to chromosome 11q22-23. Nature 336:577-580. 2. Rocchi M, Roncuzzi L, Santamaria R, Archidiacono N, Denter L, Romeo G (1986): Mapping through somatic cell hybrids and cDNA probes of protein C to chromosome 2, factor X to chromosome 13, and alpha-a-glycoacidprotein to chromosome 9. Hum Genet 74:30-33. 3. Strong RC, Korsmeyer SJ, Parkin JL, Arthur DC, Kersey JH (1986): Human acute leukemia cell line with the t(4;11) chromosomal rearrangement exhibits B lineages and monocytic characteristics. Blood 65:21-31. 4. Scott A, Phillips JA, Migeon BR (1979): DNA restriction endonuclease analysis for localization of human chromosome beta and gamma globlin genes on chromosome 11. Proc Natl Acad Sci USA 76:4563-4565. 5. Maslen CL, Jones C, Glaser T, Magenis RE, Sheehy R, Kellogg J, Litt M (1988): Seven polymorphic loci mapping to human chromosomal region 11q22-qter. Genomics 2:66-75. 6. Sacchi N, Watson DK, Guerts van Kessel AHM, Hagemeijer A, Kersey J, Drabkin HD, Patterson D, Papas TS (1986): Hu-ets-1 and Hu-ets-2 genes are transposed in acute leukemias with (4;11) and (8;21) translocations. Science 231:379-382. 7. Gatti RA, Shaked R, Mohandas TK, Salser W (1987): Human ferritin genes: chromosomal assignments and polymorphisms. Am J Hum Genet 41:654-667. 8. Gatti RA, Lathrop GM, Salser W, Silver J, Lalouel JM, White R (1987): Location of thy-1 with respect to a primary linkage map of chromosome 11q. Human Gene Mapping 9. Cytogenet Cell Genet 46:737A. 9. Krissansen GW, Gorman PA, Kozak CA, Sheer NKD, Goodfellow PN, Crumpton MJ (1987): Chromosomal locations of the gene coding for the CD3 (T3) gamma subunit of the human and mouse CD3/T-cell receptor complexes. Immunogenetics 26:258-266. 10. Regnier DC, Kozak CA,Kingsley DM, Jenkins NA, Copeland NG, Langdon WY, Morse HC: Identification of two murine loci homologous to the v-cbl oncogene. J Virol (in press). 11. Gelernter J, Grandy DK, Bunzow J, Civelli O, Retief AE, Litt M, Kidd KK (1989): D2 dopamine receptor locus (hD2G1) maps close to DllS29 (probe L7) on 11q using non-CEPH families Cytogenet Cell Genet 51:1002A. 12. Williams AF, Barclay AN (1988): The immunoglobulinsupergene family: Domains for cell surface recognition. Annu Rev Immunol 6:381-405. 13. Glaser TM (1988): Fine structure and evolution of the eleventh human chromosome. Ph.D. thesis, Massachusetts Institute of Technology. 14. Budarf M, Sellinger B, Griffin C, Emanuel BS (1989): Comparative mapping of the constitutional and tumor associated 11;22 translocations. Am J Hum Genet 45:128-139. 15. Savage PD, Jones C, Silver J, Geurts van Kessel AHM, Gonzalez-Sarmiento R, Palm L, Hanson CA, Kersey JH (1988): Mapping studies and expression of genes located on human chromosome 11, band q23. Cytbgenet Cell Genet 49:289-292.

8

S. Wei et al,

16. Koeffler HP (1987): Syndromes of acute nenlymphocytic leukemia. Ann Intern Med 107:748-758. 17. Charmley P, Foroud T, Malholtra U, Concannon P, Wei S, Lange K, Gatti RA (1989): Genetic map of the human chromosome 11q22-23 region: Implications for further localization of a gene for ataxia-telangiectasia Cytogenet Cell Genet 51:976A. 18. Sanal O, Wei S, Charmley P, Concannon P, Foroud T, Lange K, Gatti RA (in press): Fine mapping the ataxia-telangiectasia locus within the chromosome 11q23 region. Am J Hum Genet. 19. Edelman G (1989): Topobiology. Scientific American May:76-88. 20. Sacchi N, de Klein A, Showalter SD, Bigi G, Papas TS (1988): High expression of ets-1 gene in human thymocytes and immature T leukemic cells. Leukemia 2:12-18.