0165-4608/98/$19.00PII S0165-4608(97)00250-1

Cancer Genet Cytogenet 102:6–11 (1998)

Elsevier Science Inc., 1998655 Avenue of the Americas, New York, NY 10010

Rearrangement of the Long Arm of Chromosome 10 in the Prostate Adenocarcinoma Cell Line LNCaP

Sally Ford, Ian C. Gray, and Nigel K. Spurr

ABSTRACT:

Rearrangement of distal 10q is a common feature of many tumor types and tumor-derivedcell lines. More specifically, loss of 10q23–25 has been demonstrated in a large proportion of prostatetumors, indicative of the presence of a tumor suppressor gene at this location. Using whole-chromosomepaints and human genomic YAC clones as FISH probes, we have performed a detailed cytogenetic anal-ysis of distal 10q rearrangements in the prostate adenocarcinoma cell line LNCaP. Our data revealnonreciprocal translocation of 10q24.1-qter material to two sites on chromosome 5q, giving der(5)t(5;10)(q14–23;q24.1)t(5;10)(q35;q24.2) loss of 10q material at the 10q24.1 breakpoint. Deleted chromatin atthe distal breakpoint includes the cytochrome P450IIC (

CYP2C

) gene cluster, thought to be involved insteroid hormone metabolism and therefore of possible significance to the growth rate of this androgen-dependent cell line. Deleted material at the proximal breakpoint overlaps with a region of deletion atthe 10q23–24 boundary recently identified in a high proportion of prostate tumors, adding to the evi-dence for a tumor suppressor gene in this interval. © Elsevier Science Inc., 1998

INTRODUCTION

Rearrangement of the distal end of the long arm of chromo-some 10 is a common event in prostate tumors and tumor-derived cell lines; 10q deletions and translocations arefrequently observed during cytogenetic analysis [1–3]. Morerefined allelic loss studies have shown that over 60% ofprostate tumors show loss of 10q23–25 [4–7]; strong evi-dence for the presence of a tumor suppressor gene (orgenes) at this location. Several other tumor types also dem-onstrate 10q loss or rearrangement including glioma [8–11],melanoma [12, 13] lymphoma [3, 14] renal cell carcinoma[15], and endometrial adenocarcinoma [16], implying aregion of tumor suppression of general relevance in the gen-esis or progression of a wide range of tumor types.

The most widely studied prostatic adenocarcinoma cellline is the androgen-dependent hypotetraploid cell lineLNCaP, derived from a lymph node metastasis [17]. Previ-ous cytogenetic studies of LNCaP describe a 10q24-qter de-letion [18]. Using the CEPH-Genethon physical map ofchromosome 10 as a scaffold [19, 20], we have recentlyconstructed a detailed yeast artificial chromosome (YAC)

contig spanning 10q23–25 [21]. We have used selectedYACs from this contig as fluorescence in situ hybridization(FISH) probes against metaphase LNCaP spreads. In con-junction with whole chromosome painting, this has al-lowed us to characterize the distal 10q rearrangements inthis cell line in more detail, revealing an event more com-plex than originally described, resulting in loss of 10q23–24material in conjunction with a nonreciprocal translocation.

MATERIALS AND METHODS

The LNCaP cell line was obtained from the AmericanType Culture Collection (ATCC) and grown and maintainedin the recommended media. Exponentially growing cellswere treated with Colcemid at a final concentration of 0.01

m

g/ml for 1 hour, pelleted, washed with phosphate buff-ered saline solution, and resuspended for hypotonicswelling in 0.56% KCl at 37

8

C for 15 minutes. Cells werefixed in methanol:acetic acid (3:1) and metaphase spreadsprepared by dropping the cell suspension onto a micro-scope slide. Following drying, the slides were stored des-iccated at

2

70

8

C and aged for at least 3 days beforehybridization.

Following biotin labeling using a Bionick kit (GIBCO-BRL, USA), YAC DNA and a chromosome 10-specific cen-tromeric probe (Oncor, USA) were hybridized to LNCaPmetaphase spreads as described previously [22]. Biotin la-beled chromosome-specific paints (Cambio, UK) were hy-bridized in accordance with the supplier’s instructions.Hybridization signals were detected with a Zeiss (Germany)

From the Imperial Cancer Research Fund, Clare Hall Labora-tories, Potters Bar, United Kingdom.

Address correspondence to: Dr. Ian C. Gray, Smithkline Bee-

cham Pharmaceuticals, Biopharmaceutical R&D,

New FrontiersScience Park, Harlow, Essex CM10 5AW, United Kingdom.

Present address for I. C. G., N. K. S.: Smithkline BeechamPharmaceuticals, Biopharmaceutical R&D, New Frontiers Sci-ence Park, Harlow, Essex CM10 5AW, United Kingdom.

Received December 11, 1996; accepted June 10, 1997.

10q Rearrangements in the LNCaP Cell Line

7

Axiophot microscope and images captured using aHamamatsu (Japan) cooled CCD camera.

RESULTS

The LNCaP cell line is hypotetraploid and has four copiesof chromosome 10 [17, 18]. To identify any large-scaleloss, gain, or translocation of chromosome 10 material, awhole chromosome 10 paint was applied to LNCaPmetaphase spreads (Fig. 1A). The resulting fluorescencepattern clearly shows two truncated chromosome 10homologues from which 10q material appears to havetranslocated to an A group chromosome. The two remain-

ing chromosome 10 homologues appear normal. Closeinspection of the fluorescein isothiocyanate (FITC) signalreveals two sites of incorporation of chromosome 10 mate-rial on the recipient chromosome. The majority of 10-derived chromatin is fused to the tip of the q arm, with asmaller insertion at proximal q (Fig. 1A). Application of asuccession of A group whole chromosome paints revealedthe recipient to be chromosome 5, which gives reverseFITC signals with chromosome 10 and 5 paints on two

Figure 1 Chromosome-specific paints applied to hypotetraploid LNCaP metaphase spreads. (A) Chromosome 10specific paint. Two copies of chromosome 10 appear truncated (white arrows) as a result of translocation of chro-matin to an A-group chromosome. This derivative chromosome has two sites of incorporation of chromosome 10material; the majority is fused to the tip of the q-arm (yellow arrows) with a further signal at proximal q (yellowarrowheads). The remaining chromosome 10 copies appear normal (white arrowheads). (B) Chromosome 5 specificpaint. A reverse of the chromosome 10 paint pattern is obtained, with unstained additional material at 5qter(arrow) and proximal 5q (arrowhead). No signals are evident on any other chromosomes, suggesting a nonrecipro-cal 10q;5q translocation.

Figure 2

FISH signals from 10q24 YAC clones applied toLNCaP metaphase spreads in conjunction with a chromosome 10centromeric probe. (A) 10q24–25 YAC clone 878-F-11. Two cop-ies of chromosome 10 retain 878-F-11 (white arrowheads),whereas the remaining copies give a centromeric signal only(white arrows), 878-F-11 having translocated to distal 5q (yellowarrows). (B) 10q23–24 YAC clone 958-A-7. Two copies of chromo-some 10 retain 958-A-7 (white arrowheads), whereas the remain-ing copies give a centromeric signal only (white arrows), 958-A-7having translocated to proximal 5q (yellow arrows). (C) 10q24YAC clone 912-C-7. A clear signal can be seen on two copies ofchromosome 10 (arrowheads), but no signal is present on theremaining copies (arrows). No signal is evident on any otherchromosome, suggesting a deletion spanning clone 912-C-7. (D)YAC clone 796-D-5 from the 10q23–24 boundary. An obvious sig-nal can be seen on two copies of chromosome 10 (arrowheads),but no signal is present on the remaining copies (arrows). No sig-nal is evident on any other chromosome, indicative of a seconddeletion spanning clone 796-D-5.

8

S. Ford et al.

Tab

le 1

A s

um

mar

y of

dis

tal

10q

rear

ran

gem

ents

in

LN

CaP

der

ived

usi

ng

10q2

3–25

YA

Cs

as F

ISH

pro

bes

agai

nst

met

aph

ase

spre

ads

Cen

trom

ere

Mar

ker

a

Tel

omer

e

Nor

mal

cyto

loca

tion

c

LN

CaP

D10

S57

9D

10S

215

D10

S54

1D

10S

1442

D10

S15

71D

10S

1753

D10

S56

4D

10S

1755

D10

S58

3D

10S

185

D10

S20

0P

DE

6CR

BP

4C

YP

2CD

10S

571

D10

S16

80D

10S

208

D10

S57

4D

10S

2172

D10

S19

8D

10S

603

D10

S19

2D

10S

1268

D10

S56

6D

10S

530

D10

S54

0S

ize

(KB

)

b

YA

C––

–––—

D —

––––

–––

––––

— T

ran

sloc

ated

to P

5q

—––

––––

––––

––––

––––

––––

D –

––––

––––

––––

–––

––––

––––

Ret

ain

ed o

n 1

0q24

–––

––––

–––

–– T

ran

sloc

ated

to

5qte

r ––

––

746H

8

11

1

1200

10q2

3–24

D82

1D2

11

1

1150

10q2

3–24

D83

1E5

11

1

1110

10q2

3–24

D79

6D5

1

800

10q2

3–24

D90

6D1

11

1

1010

10q2

3–24

P 5

q75

9C9

1

NK

10q2

3–24

P 5

q88

5H11

1

1410

10q2

3–24

P 5

q76

1B1

11

980

10q2

3–24

P 5

q95

8A7

11

1730

10q2

3–24

P 5

q74

5D8

1

850

10q2

3–24

P 5

q79

1C3

11

1360

10q2

3–24

P 5

q91

2C4

11

11

11

2100

10q2

4D

912C

7

11

11

11

2100

10q2

4D

853A

4

11

920

10q2

410

q24

744D

4

11

1

1360

10q2

410

q24

857E

10

11

810

10q2

410

q24

941F

5

11

1400

10q2

410

q24

845H

7

11

420

10q2

410

q24

926D

8

11

1780

10q2

4–25

5qte

r95

4E12

1

1580

10q2

4–25

5qte

r87

8F11

11

1230

10q2

4–25

5qte

r95

4D6

11

1790

10q2

4–25

5qte

r79

0A8

1

1760

10q2

55q

ter

Abb

revi

atio

ns

: D, d

elet

ed; P

, pro

xim

al; N

K, n

ot k

now

n.

a

Gen

etic

mar

ker

assi

gnm

ents

to

YA

Cs

are

den

oted

by

cros

ses

and

are

tak

en f

rom

[19

] an

d [

21].

b

YA

C s

izes

are

tak

en f

rom

th

e G

enet

hon

Qu

ickm

ap d

atab

ase

[19]

, oth

er t

han

siz

es f

or 9

12-C

-4 a

nd

912

-C-7

, wh

ich

are

fro

m [

25].

c

Th

e n

orm

al c

ytog

enet

ic l

ocat

ion

of

each

YA

C i

s ta

ken

fro

m [

21].

10q Rearrangements in the LNCaP Cell Line

9

homologues (Fig. 1A and B). The remaining chromosome5 homologues appear normal. No chromosome 5 signal isvisible on any of the chromosome 10 copies, suggestingthe translocation to be nonreciprocal.

A 10q24-qter deletion, observed in G-banding patterns,has previously been reported for LNCaP [18]. It thereforeseemed likely that the material translocated to 5q consistsof all or part of 10q24-qter. To confirm this and to charac-terize the rearrangement in more detail, YAC clones map-ping to 10q23–25 were used as FISH probes against LNCaP

metaphase spreads. YACs were selected from the CEPH-Genethon mega-YAC physical map of chromosome 10 [19,20] and their cyto-location confirmed by FISH mapping tometaphase spreads of normal blood lymphocytes [21].Chimaeric YACs, i.e., those mapping to other cytogeneticregions in addition to 10q23–25, were excluded.

A total of 23 YACs mapping between the genetic mark-ers

D10S579

at the 10q23–24 boundary and

D10S540

at10q25 [19–21] were used to probe LNCaP metaphasespreads. The FISH images generated by these probes re-

Figure 3 Ideogram summarizing the10q;5q translocation in LNCaP. 10q24.2-qter is translocated to 5qter, whereas chro-matin from 10q24.1 is translocated to5q14–23 (with some loss of material at the10q24.1 breakpoints). A small amount ofchromatin between the two translocatedregions is retained at 10q24. It is notknown if loss of 5q material has beenincurred during this event.

10

S. Ford et al.

veal a complex pattern (summarized in Table 1): 10q24-qter distal to marker

D10S603

translocated to 5qter (Fig.2A), plus a second interstitial translocation of approxi-mately 8 cM of proximal 10q24 material from betweenmarkers

D10S541

and

D10S185

[23] to proximal 5q (Fig.2B), with loss of some 10q material flanking the breakpoints. Approximately 2 Mb of DNA has been lost distal to

D10S583

, including the cytochrome P450IIC gene cluster(

CYP2C

) and the serum retinol-binding protein and phos-phodiesterase 6-C genes (

RBP4

and

PDE6C

; Fig. 2C; seeTable 1). More than 1 Mb has also been lost proximal to

D10S1442

(Fig. 2D; see Table 1). It was not possible to de-fine the proximal boundary of this deletion as suitableYACs were not available. However, a more centromericYAC, 928-E-1 at 10q22–23 [19], is retained on chromosome10 (data not shown). This region of deletion at the 10q23–24boundary overlaps with a commonly deleted region betweenmarkers

D10S1644

(AFMa124ye1) and

D10S583

recentlyidentified in prostate tumors [4]. An ideogram summariz-ing the 10q rearrangements in LNCaP is given in Figure 3.

DISCUSSION

Rearrangement of 10q23–25 is a common event in prostateadenocarcinoma and other tumor types, suggesting thepresence of a gene (or genes) within this cytogenetic inter-val of relevance to tumor progression. As 10q23–25 is afrequent target for deletion events in tumors, a likely rolefor such a gene is tumor suppression. A characteristic ofthe prostate tumor cell line LNCaP is a 10q24 rearrange-ment, described previously as a 10q24-qter deletion [18].As 10q23–25 aberrations are frequently observed in uncul-tured tumor cells it is likely that the 10q anomaly inLNCaP occurred during tumor genesis or progression,rather than being an artifact of cell culture. To gain furtherinsight into the nature and consequences of this event andwith the eventual aim of identifying those genetic ele-ments involved in tumor progression, we have used wholechromosome painting in conjunction with large genomicYAC clone FISH probes to give a detailed picture of distal10q rearrangements in LNCaP.

Our data reveal an event more complex than previouslydescribed: nonreciprocal translocation of 10q24.1-qtermaterial to two sites on 5q, giving der(5)t(5;10)(q14–23;q24.1)t(5;10)(q35;q24.2). Some material between the twotranslocated regions is retained at 10q24 (Table 1; Fig. 3),but more significantly there is loss of 10q material at the10q24.1 breakpoints. Approximately 2 Mb of DNA at thedistal breakpoint has been lost, including the cytochromeP-450IIC (

CYP2C

) gene cluster. The cytochrome P450IICenzymes are monooxygenases thought to be involved insteroid hormone metabolism [24]; consequently loss ofthis gene cluster may confer a growth advantage on the an-drogen-dependent LNCaP by hindering androgen break-down. Mutation analyses of the remaining

CYP2C

genecopies on the cytogenetically normal chromosome 10 ho-mologues, to assess complete loss of enzyme function,may therefore prove fruitful. However, the genetic intervalspanning the

CYP2C

gene cluster has recently been ex-

cluded as a major prostate tumor suppressor locus by al-lele loss studies [4].

The second site of lost chromatin, proximal to the10q24.1 translocation at the 10q23–24 boundary (Table 1),may have more significance. This region of loss overlapswith an area of deletion common to up to 60% of prostatetumors [4, 5, 7]; striking evidence for the presence of a tu-mor suppressor gene in this interval. With the density ofhighly polymorphic genetic markers currently available[23] it should be feasible to narrow this region to a man-ageable size for physical map construction and candidatetumor suppressor gene isolation in the near future. Giventhe broad spectrum of tumor types showing 10q23–25 rear-rangement, the 10q23–24 boundary may yield a gene of gen-eral relevance to tumorigenesis and/or tumor progression.

ADDENDUM

We and others have recently identified a candidate tumorsuppressor gene, designated

PTEN

or

MMAC1

, at the10q23–24 boundary [26–28]. This gene resides in theinterval spanned by YAC 796-D-5, which is deleted fromone chromosome 10 homologue in LNCaP (Fig. 2D). Muta-tions in

PTEN/MMAC1

have been found in a variety oftumor types, including prostate. A 2bp deletion has beenidentified in the retained copy in LNCaP [26, 27].

Yeast artificial chromosome clones were supplied by the UKHuman Genome Mapping Project Resource Centre. This workwas supported by the Imperial Cancer Research Fund.

REFERENCES

1. Arps S, Rodewald A, Schmalenberger B, Carl P, Bressel M,Kastendieck H (1993): Cytogenetic survey of 32 cancers ofthe prostate. Cancer Genet Cytogenet 66:93–99.

2. Lundgren R, Mandahl N, Heim S, Limon J, Henrikson H,Mitelman F (1992): Cytogenetic analysis of 57 primary pros-tatic adenocarcinomas. Gene Chromosom Cancer 4:16–24.

3. Mitelman F, Kaneko Y, Trent JM (1990): Report of the com-mittee on chromosome changes in neoplasia. Cytogenet CellGenet 55:358–386.

4. Gray IC, Phillips SMA, Lee SJ, Neoptolemos JP, WeissenbachJ, Spurr NK (1995): Loss of the chromosomal region 10q23-25in prostate cancer. Cancer Res 55:4800–4803.

5. Ittmann M (1996): Allelic loss on chromosome 10 in prostateadenocarcinoma. Cancer Res 56:2143–2147.

6. Lacombe L, Orlow I, Reuter VE, Fair WR, Dalbagni G, ZhangZF, Cordoncardo C (1996): Microsatellite instability anddeletion analysis of chromosome-10 in human prostate-can-cer. Int J Cancer 69:110–113.

7. Trybus TM, Burgess AC, Wojno KJ, Glover TW, Macoska JA(1996): Distinct areas of allelic loss on chromosomal regions10p and 10q in human prostate cancer. Cancer Res 56:2263–2267.

8. Karlbom AE, James CD, Boethius J, Cavenee WK, Collins VP,Nordenskjold M, Larsson C (1993): Loss of heterozygosity inmalignant gliomas involves at least three distinct regions onchromosome 10. Hum Genet 92:169–174.

9. Fults D, Pedone C (1993): Deletion mapping of the long armof chromosome 10 in glioblastoma multiforme. Genes Chro-mosom Cancer 7:173–177.

10q Rearrangements in the LNCaP Cell Line

11

10. Ransom DT, Ritland SR, Moertel CA, Dahl RJ, O’Fallon JR,Scheithauer BW, Kimmel DW, Kelly PJ, Olopade OI, DiazMO, Jenkins RB (1992): Correlation of cytogenetic analysisand loss of heterozygosity studies in human diffuse astrocy-tomas and mixed oligo-astrocytomas. Genes ChromosomCancer 5:357–374.

11. Rasheed BKA, Fuller GN, Friedman AH, Bigner DD, BignerSH (1992): Loss of heterozygosity for 10q loci in human glio-mas. Genes Chromosom Cancer 5:75–82.

12. Parmiter AH, Balaban G, Clark WHJ, Nowell PC (1988): Pos-sible involvement of the chromosome region 10q24-q26 inearly stages of melanocytic neoplasia. Cancer Genet Cytoge-net 30:313–317.

13. Herbst RA, Weiss J, Ehnis A, Cavanee WK, Arden KC (1994):Loss of heterozygosity for 10q22-10qter in malignant mela-noma progression. Cancer Res 54:3111–3114.

14. Speaks SL, Sanger WG, Masih AS, Harrington DS, Hess M,Armitage JO (1992): Recurrent abnormalities of chromosomebands 10q23-q25 in non-Hodgkins lymphoma. Genes Chro-mosome Cancer 5:239–243.

15. Morita R, Saito S, Ishikawa J, Ogawa O, Yoshida O,Yamakawa K (1991): Common regions of deletion on chro-mosomes 5q, 6q and 10q in renal cell carcinoma. Cancer Res51:5817–5820.

16. Simon D, Heyner S, Satyaswaroop PG, Farber M, Noumoff JS(1990): Is chromosome 10 a primary chromosomal abnormal-ity in endometrial adenocarcinoma? Cancer Genet Cytogenet47:155–162.

17. Horoszewicz JS, Leong SS, Ming Chu T, Wajsman ZL, Fried-man M, Papsidero L, Kim U, Chai LS, Kakati S, Arya SK,Sandberg AA (1980): The LNCaP cell line—a new model forstudies on human prostatic carcinoma. Prog Clin Biol Res37:115–132.

18. Gibas Z, Becher R, Kawinski E, Horoszewicz J, Sandberg AA(1984): A high-resolution study of chromosome changes in ahuman prostatic carcinoma cell line (LNCaP). Cancer GenetCytogenet 11:399–404.

19. Cohen D, Chumakov I, Weissenbach J (1993): A first genera-tion physical map of the human genome. Nature 366:698–701.

20. Chumakov IM, Rigault P, Le Gall I, Bellanne-Chantelot C, Bil-lault A, Guillou S, Soularue P, Guasconi G, Poullier E, Gros I,Belova M, Sambucy JL, Susini L, Gervy P, Glibert F, BeaufilsS, Bui H, Massart C, Detand MF, Dukasz F, Lecoulant S,Ougen P, Perrot V, Saumler M, Soravito C, Bahouayila R,

Cohenakenine A, Barillot E, Bertrand S, Codani JJ, CaterinaD, Georges I, Lacroix B, Lucotte G, Sahbatou M, Schmit C,Sangouard M, Tubacher E, Dib C, Faure C, Fizames C, Gya-pay G, Millasseau P, NGugen S, Muselet D, Vignal A, Moris-sette J, Menninger J, Lieman J, Desai T, Banks A, Bray-WardP, Ward D, Hudson T, Gerety S, Foote S, Stein L, Page DC,Lander LS, Weissenbach J, Le Paslier D, Cohen D (1995): A yaccontig map of the human genome. Nature 377(Suppl):175–297.

21. Gray IC, Fallowfield J, Ford S, Nobile C, Spurr NK (1997): Anintegrated physical and genetic map spanning chromosomeband 10q24. Genomics 28:328–322.

22. Ragoussis J, Monaco A, Mockridge I, Kendall E, CampbellRD, Trowsdale J (1991): Cloning of the HLA class II region inyeast artificial chromosomes. Proc Natl Acad Sci USA88:3753–3757.

23. Dib C, Faure S, Fizames C, Samson D, Drouot N, Vignal A,Millasseau P, Marc S, Hazan J, Seboun E, Lathrop M, GyapayG, Morissette J, Weissenbach J (1996): A comprehensivegenetic-map of the human genome based on 5,264 microsat-ellites. Nature 380:152–154.

24. Meehan RR, Speed M, Gosden JR, Rout D, Hutton J, TaylorBA, Hilkens J, Hastie ND, Wolf CR (1988): Chromosomalorganisation of the cytochrome P4502C gene family in themouse: A locus associated with constitutive aryl hydrocar-bon hydroxylase. Proc Natl Acad Sci USA 85:2662–2666.

25. Gray IC, Nobile C, Moresu R, Ford S, Spurr NK (1995): A 2.4megabase physical map spanning the

CYP2C

cluster on chro-mosome 10q24. Genomics 28:328–332.

26. Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, Puc J,Miliaresis C, Rodgers L, McCombie R, Bigner SH, GiovanellaBC, Ittmann M, Tycko B, Hibshoosh H, Wigler MH, Parsons R(1997):

PTEN

, a putative protein tyrosine phosphatase genemutated in human brain, breast, and prostate cancer. Science275:1943–1947.

27. Steck PA, Pershouse MA, Jasser SA, Yung WKA, Lin H,Ligon AH, Langford LA, Baurngard ML, Hattier T, Davis T,Frye C, Hu R, Swedlund B, Teng DHF, Tavtigian SV (1997):Identification of a candidate tumour suppressor gene,

MMAC1

, at chromosome 10q23.3 that is mutated in multipleadvanced cancers. Nature Genet 15:356–362.

28. Gray IC, Stewart LMD, Phillips SMA, Hamilton JA, Gray NE,Watson GJ, Spurr NK, Snary D (submitted): Mutation andexpression analysis of the putative prostate tumour suppres-sor gene

PTEN

. Br J Cancer.