specific chromosome aberration in human renal cell carcinoma

8
Itzt. J. Cancer: 40, 1171-178 (1987) 0 1987 Alan R. Liss, Inc. Publication of the International Union Against Cancer Publication de I’Union lnternationale Contre le Cancer SPECIFIC: CHROMOSOME ABERRATION IN HUMAN RENAL CELL CARCINOMA Gyula KOVACS’,~, Sandor szucs‘ , Werner DE RIESE’ and Hermann BAUMGARTEL4 ‘Laboratory of Cytogenetics, Institute o Pathology and ‘Department of Urology, Hanover Medical School, Konstanty- Gutschow-Str 8, 0-3000 Hanover 61; Department of Urology, Siloah Hospital, Roesebeck-Str. 15, 0-3000 Hanover 91, Federal Republic of Germany. l Using G-banding technique, the chromosomes were studied in short-term cultures of 25 primary renal-cell carcinomas (RCC). Phytohaemagglutinin-stimulated peripheral blood lym- phocytes or normal kidney cells of the same patients growing in primary cultures were analysed to define the constitutional karyotype. The modal chromosome number of 23 RCC‘s was found to be pseudo-diploidor near-diploidwith only few struc- tural rearrangements, 22 of the RCC’s showed an aberration of Chromosome 3, deletion of 3p, or translocation of different chromosome segments to the deleted chromosome 3, leading to the loss of variable segments of chromosome 3. The break- points in rearrangements of Chromosome 3 clustered in the region 3p I 1.2-p 13. Shortest-region overlap analysis localized a consistent change to a small area of 3p13-pter. In 8 of the 25 RCCs, the rearrangement of chromosome 3 was the only karyotype change determined, and 4 other tumours had only one chromosomal rearrangement in addition to the aberra- tion of chromosome 3. These results suggest that the aberra- tion of chromosonie 3 is the first cytogenetic event in the clonal evolution of RCCs. Translocation 3;s was preferentially involved in the rearrangements between chromosome 3p and other chromosomes. The breakpoint on chromosome 3 was constant at p13, but the breaks on chromosome 5 varied between bands q II .2 and q22. Monosomy 14 was observed in 10 cases and loss of Y chromosome was detected in 6 of 14 tumours obtained from male patients. Since the normal so- matic cells were free of chromosomal aberrations, one may conclude that the loss of 3p13-pter segment is an acquired, consistent chromosomal aberration which marks human RCCs. In recent years an increasing number of non-random chro- mosome aberrations have been detected in human leukaemias and lymphomas, as well as in solid tumours (for review see Yunis, 1983). Reciprocal (balanced) translocations with pre- cisely defined breakpoints were found in several forms of leukaemia and lymphoma, and the activation and altered func- tion of oncogenes situated close to the chromosomal breakage was demonstrated iusing recombinant DNA technology (Kon- opka and Witte, 1985; Croce and Nowell, 1985; Klein and Klein, 1985). On the oiher hand, chromosomal deletions are associated in a characteristic manner with some solid malignant tumours, primarily with dysontogenetic tumours of childhood. A small interstitial deletion of chromosome 13 involving the Rb-I gene locus at 13q14 was detected in all somatic cells of a number of hereditary retinoblastoma carriers (Knudson et al., 1976) as well as in retinoblastoma tumour tissue (Benedict et al., 1983). An interstitial deletion in the short arm of chromosome 11 at p13 has been found in somatic cells of patients with hereditary Wilms’ tumour (Riccardi et al., 1978) and in tumour cells obtained from sporadic cases (Slater, 1986). Furthermore, a deletion of the short arm of chromosome 1 was found as a non-random chromosomal defect occurring in neuroblastoma (Brodeur et al., MI). In spite of the recent accumulation of cytogenetic data on some types of solid tumours, reports on chromosomal features of human renal-cell carcinomas (RCC) are rare. A constitu- patient showing normal constitutional karyotype, from another family predisposed to renal cancer (Pathak et al., 1982). Re- cently, non-random structural aberrations of chromosome 3 were observed in 8 of 12 RCCs analysed in short-term culture (Yoshida et al., 1986). In another study, the aberration of chromosome 3 was detected in 3 of 8 RCCs (Berger et al., 1986). The question arising from these findings is whether further evidence may be gathered for a specific chromosomal defect in RCC, corresponding to those found in retinoblastoma and nephroblastoma. For a better understanding of the karyol- ogical changes occurring in RCCs, we have carried out cyto- genetic studies on tumour cells from 25 patients and compared them with their constitutional karyotypes. MATERIAL AND METHODS Tumour specimens A total of 34 patients with RCC from the Hanover Medical School and Siloah Hospital was studied without prior selec- tion. Successful banding analysis was obtained in 25 primary human RCCs of histological grades 1 to 3. Tumours from previously untreated patients were obtained by radical ne- phrectomy, and a macroscopically homogeneous area free of fibrosis, necrosis or haemorrhage was excised for preparation. In some cases, larger tumour size permitted separate process- ing of 2 or 3 samples from the same tumour. One portion of the tumour sample was fixed in 4 % formaldehyde and used for histological examination as a “reference slide” for tissue cultures and chromosome analysis. The remainder of the tu- mouc samples were processed for cell culture and karyotyping. Cell cultures Specimens of tumour tissue were cut into pieces of approxi- mately 2-3 mm3 or minced with a surgical scalpel under sterile conditions in a Petri dish containing RPMI 1640 medium, then washed twice in PBS. The small tissue fragments, thus freed from cell debris, were then incubated in 0.1% collagenase (Worthington CLS 111) dissolved in culture medium (RPMI 1640 supplemented with 15% foetal calf serum and an anti- biotic) for 30-60 min at 37°C. The tissue fragments were then gently centrifuged (50g for 3 min) and washed twice in me- dium. Finally, the samples were resuspended in a centrifuge tube containing 2-4 ml culture medium and dispersed vigor- ously with a Pasteur pipette. The suspension was allowed to sediment for 1 min and the supernatant, containing over 90% of single cells with low viability and the cell debris, was decanted. The pellet, consisting of over 90% of small cell clusters of about 5 to 15 cells, was resuspended and seeded in culture. Tumour cells were maintained at 37°C in a humified atmosphere containing 5 % C 0 2 in air, in 25-cm2 Falcon flasks containing 5 ml culture medium. During in vitro growth, the cells were observed with an inverted phase-contrast micro- tional reciprocal translocation 3;8 was found in a family with hereditary RCC (Clohen et al., 1979), and an acquired trans- location 3; I [ was detected in metastatic RCC Cells from a Received: January 7, 1987 and in revised form March 19, 1987 *To whom reprint requests should be addressed.

Upload: gyula-kovacs

Post on 11-Jun-2016

214 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Specific chromosome aberration in human renal cell carcinoma

Itzt. J. Cancer: 40, 1171-178 (1987) 0 1987 Alan R. Liss, Inc.

Publication of the International Union Against Cancer Publication de I’Union lnternationale Contre le Cancer

SPECIFIC: CHROMOSOME ABERRATION IN HUMAN RENAL CELL CARCINOMA Gyula K O V A C S ’ , ~ , Sandor szucs‘ , Werner DE RIESE’ and Hermann BAUMGARTEL4

‘Laboratory of Cytogenetics, Institute o Pathology and ‘Department of Urology, Hanover Medical School, Konstanty- Gutschow-Str 8, 0-3000 Hanover 61; Department of Urology, Siloah Hospital, Roesebeck-Str. 15, 0-3000 Hanover 91, Federal Republic of Germany.

l

Using G-banding technique, the chromosomes were studied in short-term cultures of 25 primary renal-cell carcinomas (RCC). Phytohaemagglutinin-stimulated peripheral blood lym- phocytes or normal kidney cells of the same patients growing in primary cultures were analysed to define the constitutional karyotype. The modal chromosome number of 23 RCC‘s was found to be pseudo-diploid or near-diploid with only few struc- tural rearrangements, 22 of the RCC’s showed an aberration of Chromosome 3, deletion of 3p, or translocation of different chromosome segments to the deleted chromosome 3, leading to the loss of variable segments of chromosome 3. The break- points in rearrangements of Chromosome 3 clustered in the region 3p I 1.2-p 13. Shortest-region overlap analysis localized a consistent change to a small area of 3p13-pter. In 8 of the 25 RCCs, the rearrangement of chromosome 3 was the only karyotype change determined, and 4 other tumours had only one chromosomal rearrangement in addition to the aberra- tion of chromosome 3. These results suggest that the aberra- tion of chromosonie 3 is the first cytogenetic event in the clonal evolution of RCCs. Translocation 3;s was preferentially involved in the rearrangements between chromosome 3p and other chromosomes. The breakpoint on chromosome 3 was constant at p13, but the breaks on chromosome 5 varied between bands q I I .2 and q22. Monosomy 14 was observed in 10 cases and loss of Y chromosome was detected in 6 of 14 tumours obtained from male patients. Since the normal so- matic cells were free of chromosomal aberrations, one may conclude that the loss of 3p13-pter segment is an acquired, consistent chromosomal aberration which marks human RCCs.

In recent years an increasing number of non-random chro- mosome aberrations have been detected in human leukaemias and lymphomas, as well as in solid tumours (for review see Yunis, 1983). Reciprocal (balanced) translocations with pre- cisely defined breakpoints were found in several forms of leukaemia and lymphoma, and the activation and altered func- tion of oncogenes situated close to the chromosomal breakage was demonstrated iusing recombinant DNA technology (Kon- opka and Witte, 1985; Croce and Nowell, 1985; Klein and Klein, 1985).

On the oiher hand, chromosomal deletions are associated in a characteristic manner with some solid malignant tumours, primarily with dysontogenetic tumours of childhood. A small interstitial deletion of chromosome 13 involving the Rb-I gene locus at 13q14 was detected in all somatic cells of a number of hereditary retinoblastoma carriers (Knudson et al., 1976) as well as in retinoblastoma tumour tissue (Benedict et al . , 1983). An interstitial deletion in the short arm of chromosome 11 at p13 has been found in somatic cells of patients with hereditary Wilms’ tumour (Riccardi et al., 1978) and in tumour cells obtained from sporadic cases (Slater, 1986). Furthermore, a deletion of the short arm of chromosome 1 was found as a non-random chromosomal defect occurring in neuroblastoma (Brodeur et al . , M I ) .

In spite of the recent accumulation of cytogenetic data on some types of solid tumours, reports on chromosomal features of human renal-cell carcinomas (RCC) are rare. A constitu-

patient showing normal constitutional karyotype, from another family predisposed to renal cancer (Pathak et a l . , 1982). Re- cently, non-random structural aberrations of chromosome 3 were observed in 8 of 12 RCCs analysed in short-term culture (Yoshida et al . , 1986). In another study, the aberration of chromosome 3 was detected in 3 of 8 RCCs (Berger et al., 1986). The question arising from these findings is whether further evidence may be gathered for a specific chromosomal defect in RCC, corresponding to those found in retinoblastoma and nephroblastoma. For a better understanding of the karyol- ogical changes occurring in RCCs, we have carried out cyto- genetic studies on tumour cells from 25 patients and compared them with their constitutional karyotypes.

MATERIAL AND METHODS

Tumour specimens A total of 34 patients with RCC from the Hanover Medical

School and Siloah Hospital was studied without prior selec- tion. Successful banding analysis was obtained in 25 primary human RCCs of histological grades 1 to 3. Tumours from previously untreated patients were obtained by radical ne- phrectomy, and a macroscopically homogeneous area free of fibrosis, necrosis or haemorrhage was excised for preparation. In some cases, larger tumour size permitted separate process- ing of 2 or 3 samples from the same tumour. One portion of the tumour sample was fixed in 4 % formaldehyde and used for histological examination as a “reference slide” for tissue cultures and chromosome analysis. The remainder of the tu- mouc samples were processed for cell culture and karyotyping. Cell cultures

Specimens of tumour tissue were cut into pieces of approxi- mately 2-3 mm3 or minced with a surgical scalpel under sterile conditions in a Petri dish containing RPMI 1640 medium, then washed twice in PBS. The small tissue fragments, thus freed from cell debris, were then incubated in 0.1% collagenase (Worthington CLS 111) dissolved in culture medium (RPMI 1640 supplemented with 15% foetal calf serum and an anti- biotic) for 30-60 min at 37°C. The tissue fragments were then gently centrifuged (50g for 3 min) and washed twice in me- dium. Finally, the samples were resuspended in a centrifuge tube containing 2-4 ml culture medium and dispersed vigor- ously with a Pasteur pipette. The suspension was allowed to sediment for 1 min and the supernatant, containing over 90% of single cells with low viability and the cell debris, was decanted. The pellet, consisting of over 90% of small cell clusters of about 5 to 15 cells, was resuspended and seeded in culture. Tumour cells were maintained at 37°C in a humified atmosphere containing 5 % C02 in air, in 25-cm2 Falcon flasks containing 5 ml culture medium. During in vitro growth, the cells were observed with an inverted phase-contrast micro-

tional reciprocal translocation 3;8 was found in a family with hereditary RCC (Clohen et al . , 1979), and an acquired trans- location 3; I [ was detected in metastatic RCC Cells from a Received: January 7, 1987 and in revised form March 19, 1987

*To whom reprint requests should be addressed.

Page 2: Specific chromosome aberration in human renal cell carcinoma

TABLE I - CYTOGENETIC DATA O N TUMOUR AND NORMAL CELLS FROM 25 PATIENTS WITH NON-FAMILIAL RENAL-CELL CARCINOMA

Ca\e Agc Sex Mode Karyotype of turnour cellb Constitutional karvotvoe

HA1 17

HA 188 HA I94 HA2 10A

B C

HA2IlA

B

HA214

HA2 IS

HA221 HA223 HA225

HA228 HA230

HA245A

B

HA267

HA271

HA273

HA281

HA284

HA290

HA299

HA3 12

HA3 19 HA320 HA343

HA349

56 F 46

32 M 44 53 M 45 48 M 46

68 F 45

75 F 48

71 F 62

SO M 48 66 M 45 74 F 46

79 M 45 82 M 41

74 F 46

75 M 44

67 M 45

73 M 45

72 M 45

79 F 45

63 F 44

48 M 78

49 F 44

75 F 46 SO M 46 61 M 45

80 F 45

sl 46,XX,der(3)t(l;3)(q21; pl3).der( l4)t( 14;?)(q32. I ;?) =30 sdll 46,XX,der(3)t(l;3)(q21; p13)=2 sd12 45,XX,der(3)t( I ;3)(q2 I ; plll),der( 14)t( 14;?)(q32. l;?), - 18 =2 7 cells with additional non-clonal chromosome losses sl sl 45,X,-Y,der(3)t(3;5)(p13;q22),-6,+12.- l4,+16,-l8,+20=17

44,X. - Y .der(3)t(3;5)(p 13; q 1 1.2). - 8, - 9, +mar = 19

sl 46,X,del(Y)(q12),der($t(3;5)(p13;q22),del( 14)(ql3) =2 I sdl I 45,X, -Y ,der(3)t(3;5)(p 13; q22),de1(14)(q 13) =2 sd12 45,X,del(Y)(q 12),der(3)t(3;5)(p13; q22). - 14=2 sl 46,X,del(Y)(q12).der(3)t(3;5)(p13; q22),de1(14)(q13) = I 1 sd13 44,X. - Y ,der(3)t(3;5)(pl3; q22), - 14 = 6 sd14 45,XY,der(3)t(3;5)(~13;q22), - 14=4 sd15 46,XY ,der(3)t(3;5)(pl3; q22). +der(3)t(3;5)(p13; q22), - I4 =3 6 cells with two copies of der(3) missing various other chromosomes sl 45,X, -X,der( I)t(X; I)(pl I .23; p34.3),der( l6)t( I ; 16)(p34.3; q24).

der( 19)t(X; 19)(q 13; p 13. I ) =20 sl 45.X. -X.der( I)t(X;l)(pl I .23;p34.3),der(16)t(l;l6)(p34.3;q24),

der( 19)t(X; 19)(q13; p13.1) = 12 sl 48,XX,der( l)t(1;7)(p36.l;q32),de1(3)(p11.2),der(6)t(3;6)

7 cells with additional non-clonal chromosome losses sl

+der( 16)t(16;'?)(q24;?), + 19, +20, +20.+21, +22=9 5 cells with additional non-clonal chromosome losses

(42 1 ; q2 I ) , +de1(7)(q32) ,der( 17)t( 1 ; 17)(p36.1; q25), + 5, - 13, + 2 1 = 12

62,XX. +2, +5. +5, +7, +7, + 8. + I I . + 1 I , + 12, + 1 S,der( l6)t( 16;?)(q24;?),

sl 48,XY,+7,+17=22 sl 45,XY,der(3)t(3;5)(p13;q15).-14=14 sl 46,XX,r(3)(p25q27j= 14. sdll 92.XXXX.r(3)(p25q27),r(3)(p25q27)=4 sd12 9 I .XXXX,r(3)(p25q27::p25q27) = 3

sl 41,X,-Y.del(l)(p22).-3.-4,der(5)t(5;?)(q35;?),-8,-9,-14~ 16 9 cells with additional non-clonal chromosome losses sl 46,XX,del(3)(p13)=32 sdll 45,XX,de1(3)(p13),- 14=2 7 cells with additional non-clonal chromosome losses sl 46,XX,de1(3)(p13)= I 1 sd12 46,XX,de1(3)(pl3), rcp(6; 12)(p 1 1 ; q 13) = 13 sl 44,X, -Y,der(3)t(3;5)(p13; q22), - 14=27 sdll 43,X,-Y,der(3)t(3;5)(~13; q22),- 14,- 18=3 sl 45.X. -Y,rcp(3;4)(q21 ;q33),der(3)t(3;5)(~13;qIl .2), -6, +7, -8,

sdl1 45,X, -Y,rcp(3;4)(q2 I ;q33),der(3)t(3;5)(pI 3;ql 1.2). -8, +22 =5 sd12 46!X, -Y .rcp(3;4)(q2 1 ;q33),der(3)t(3;5)(p13;ql1.2), +22 =4 7 stemline cells with additional non-clonal chromosome losses sd13 46,X, -Y,der(3)t(3;5)(~13;ql1.2), +22=9 5 sd13 cells with additional non-clonal chromosome losses sl

SI 45,XY,-3=18

+22=13

45.XY .der(3)tdic(3; I8Ko I I .2; a I I .2).del( I 5)(q I3),der( 16) t(is;16)(qi3;pi 1.2),-i8=22 '

sdll 45.XY.der(3)tdic(3;18)(p11.2;sl 1.2).-18=4 sd12 50,XY,der(3)tdic(3;18j(pI 1.2;q11.2).+5, +7,+ 11,+ 12,- 18,+20=2 3 sdl1 cells with additional non-clonal chromosome losses s l 45 ,X, - Y ,der(3)tdic(Y ;3)(p I 1.2; p 1 1.2), inv(9)(p 12; q 13) = I7 sdll 44,X,-Y,der(3)tdic(Y;3)(pll.2;pl1.2),inv(9)(p12;q13),- 14=13 2 cells with additional structural aberrations sl

7 cells with additional non-clonal chromosome losses sl 44,X, -X.del(3)(pl2.2),der(4)t(4;8)(p16. I ;qlI .2),der(6)t(1;6)

9 cells with additional non-clonal chromosome losses sl 78,XXY,+1,+2,+2,+der(3)t(3;5)(p13;q22),+4,+5,+5,+6,+7,+7,

45,X, -X.del(l)(q2l),der(3)t(1;3)(ql l ;pl3), +der(3)t( 1:3)(qlI ;p13). der(4)t(3;4)(q21;q33),+7,-8,-9,- 14,+ 16= 14

(ql2;q21), -8= 13

+8,+8,+10,+10,+l1,+l1,+12,+13,+l3,+l4,+15,+15,+16,+l7. +18,+19,+19,+20,+20.+22,+22,=16

9 sl cells with non-clonal chromosome losses sl 44.XX.-3.der(12~t(3;12)(pl1.2;q24.3).-8.der(15)t(8;15)

(q11.2;pl I 2)= 16 sdll 43,XX. -3,der(12)t(3;12)(plI .2;q24.3), -8, - 14,der(15)t(8;15)

(q11.2;p11.2)=3

(91 1.2;pl I .2)=5 sd12 44,XX,inv(2)(p23ql I) , -3. - 14,der(IS)t(8; 15)

sl 46,XX,de1(3)(plI .2)=25 sl 46,XY.der(3)t(3;5)(p13;q22), + 13.- l4=2l sl 45,XY,der(3)t(3;8)(pI 1.2;ql1.2),-8=19 sdll 48,XY,der(3)t(3;8)(~11.2;qI 1.2), +5,+7, -8, + 12= 1 I

ND'

46,XY 46,XY 46,XY

ND

ND

46,XX

46,XY 46, XY 46,XX

46,XY 46,XY

ND

46,XY

46,XY

46,XY

46,XY ,inv(9) (p 12; q 13)

46,XX

ND

46.XY

46,XX

46,XX 46,XY 46,XY

sl sdll 44,XX,der(3)t(3;7)(p12.2; ql 1.2), -6, -7 =8

4S,XX,der(3)t(3;7)612.2:ql I .2), -6,-7. +mar=21

' N U - not determined

46,XX

Page 3: Specific chromosome aberration in human renal cell carcinoma

CHROMOSOME ABERRATION IN RCC

scope. The small cell clumps attached to the flasks by 24 hr and grew out to a near-confluent monolayer, in most cases within one week after seeding. The morphology of the tumour cell in vim) in the lximary cultures and in the early passages differed very little from the cell type of RCCs observed in the reference slides. Growth of fibroblast-like cells was very rare in the cultures. Chromosome ana1y:ris

Cytogenetil; analysis was carried out on the days 3-7 of primary culture, depending on the proliferation of tumour cells. by adding 0.1 pg/ml colchicine for 30-60 min or 0.02 pg/ml Colcernid for 2-3 hr. The cells were detached by treat- ment with 0.025 % trypsin-EDTA solution, then treated with 0.075 molar potassium chloride solution for 15 to 30 min. This step of the preparation seems to be very important to obtain intact metaphases. In some cases a hypotonic treatment exceeding 20 min resulted in metaphases showing extreme chromosome loss, but a repeated preparation with briefer hypotonic treatment, on the same day of culture, resulted in metaphases :showirlg a well-defined chromosomal mode. It may be that over-long hypotonic treatment was instrumental in obtaining metaphases defying analysis in 9 of the cases. After the hypotonic treatment the cells were fixed 2 or 3 times with cold methanol-acetic acid (3: 1). Chromosome prepara- tions were made by the air-drying technique and the slides were kept at room temperature for 3 to 7 days before banding. Karyotype analysis was routinely performed using GTG (Sea- bright, 1971:i and CBG techniques (Sumner et al., 1971), and the QFQ technique (Caspersson et al., 1970) for identification of the Y chromosome. Each well-banded metaphase was karyotyped ,and thle stemline and sideline were determined according to the ISCN (1985).

Analysis of the con,rtitutional karyotype To define the constitutional karyotype of the patients, phy-

tohaemagglutinin-stimulated peripheral blood lymphocytes, or normal kidney cells growing in v i m , were analysed. Tissue specimens obtained from normal kidney were prepared for in virro cultures in the manner described above for tumour sam- ples. Metaphases were analysed using GTG, CBG and occa- sionally QFQ techniques, and more than 15 metaphases were karyotyped in each1 case.

RESULTS

The chromosome findings in the 25 RCCs reported are summarized in Table I . The cytogenetic analysis of most tu- mows revealed a well-defined stemline karyotype with occa- sional sidelines and some cells with non-clonal structural aberrations or chromosomal loss.

The modal chromosome number usually lay in the diploid range: in 6 tumours of pseudodiploid karyotype was noted, 10 tumours showed a. chromosomal mode of 45. In 4 cases the chromosome numlber was 44, one tumour was characterized by a mode of 41 and 2 others by a mode of 48. Only 2 of the 25 RCCs showed a modal number in polyploid region, namely at 62 and 78.

by minimal numerical deviation from the diploid karyotype, but also showed only a few structural rearrangements. Upon statistical analysis certain chromosomes seem preferentially involved in karyotype changes. We analysed the gain and loss The preferential involvement of chromosomes in the rear- of whole chromosomes and the short and long arm of each rangements occurring in RCCs shows greater significance chromosome for structural aberrations (Fig. 1). A gain of when structural aberrations are considered (Fig. 1). The short chromosome 7 was found in 9 cases, while trisomy of chro- arm of chromosome 3 was involved in aberrations resulting in mosomes 5 and 20 was seen 7 and 6 times, respectively. 22 marker chromosomes, and the long arm of the chromosome Monosoiny 14 was detected in 10 tumours and the loss of Y 5 was also affected in 10 cases. When all numerical and chromosome was shown in 6 of the 14 male patients. structural aberrations are taken into account, chromosome 3

Ion g

FIGURE 1 - Distribution of gains and losses o f chromosomes (above) The RCC!~ analysed in this study are not only characterized and invohement of chromosome arms in structural rearrangements (be-

low) in 25 RCCs. Note the high frequency of rearrangement of the short arm of chromosome 3 ,

short

173

Page 4: Specific chromosome aberration in human renal cell carcinoma

I74 KOVACS ET AL.

2 2 I I q 23 I I

I I 25 I I

I I I I

2 1

26 1 26 2 26 3 27 20 29

3 FIGURE 2 - Diagrammatic representation of a normal chromosome 3.

Dotted lines represent losses, solid lines gains of chromosome segments or entire chromosomes. Breakpoints are marked with an asterisk. Analysis of the shortest overlap showed a loss of segment distal from 3p13 In all but one RCCs marked with chromosome 3 abnormality.

was involved 29 times, and chromosome 5 was affected 17 times.

Distribution of breakpoints We analysed the distribution of chromosomal breakpoints

throughout the genome, to assess which chromosomal regions are involved preferentially in the structural rearrangements. All breakpoints observed in the stemlines and sidelines of the 25 RCCs were presented in a schematic karyogram. Chromo- somal breaks involved in complex translocations were sub- summated as single aberrations. A total of 74 breakpoints were detected, 26 of which were localized to chromosome 3. Thus, more than 35% of the total breaks observed were demonstra- bly localized to chromosome 3, mostly between bands p11.2 and p13. The breakpoint leading to the deletion of 3p or translocation between 3p and another chromosome occurred at p13 in 13 marker chromosomes, at 3p12.2 in 2 cases and at 3~11.2 in 6 cases (Fig. 2). Furthermore, band 3q21 was involved in structural rearrangements 3 times and the break- points at 3p25 and q27 were noted in a case with ring chro- mosome 3. Although the breakpoints on chromosome 3 varied, 80% of the breaks clustered to the short arm between p13 and

Chromosome 5 was also often involved in structural aber- rations. Breakpoints were clustered to band q22 on chromo- some 5, and were always involved in translocation occuring between chromosome 3 and 5 (Fig. 3). Chromosome 3p aberration

We examined the specific types and sites of the abnormali- ties of chromosomes 3 in RCCS. Three types of chromosome 3 aberrations were detected:

1. Monosomy of chromosome 3. The loss of an entire chromosome 3 occurred in 2 tumours. Tumour HA228 showed monosomy 3 as the only karyotype change (Fig. 4), and RCC HA230 was characterized by additional numerical and struc- tural aberrations, beyond the loss of one of the 2 homologous chromosomes 3.

p11.2.

2 . Deletion of the short arm of chromosome 3. This form of rearrangement involving chromosome 3 was found in 5 RCCs. In tumour HA214 and HA319 the breakpoint was local- ized to 3~11.2, in tumour HA290 to 3~12 .2 and in HA245 to 3p13. The renal cancer HA225 showed a ring chromosome 3, with breakpoints at p25 and q27, as the only karyotype devia- tion (Fig. 5). In the stemline cells of tumour HA245 (Fig. 6), and also in all the cells of renal cancer HA319, the deletion 3p was the only chromosome change.

3. Translocation between chromosome 3 and other chro- mosomes. This form of chromosomal rearrangement was found in 15 renal carcinomas. In some tumours 2 copies of the rearranged chromosome were noted. The long arm of chromo- some 1 was detected 3 times in this type of rearrangement (HA117 and twice in HA284), chromosome 4 was affected in 2 cases (HA271 and HA284). An unbalanced translocation of the 7q, 8q, 12q segment, deleted chromosome 18 and the Y chromosome, to the short arm of chromosome 3 were each observed only in single cases (HA349, HA343, HA273 and HA281). Chromosome 5 was more frequently involved in this type of rearrangement, namely in tumours HA188, HA194, HA210 (Fig. 7), HA223, HA267 (Fig. 8), HA271, HA299 and HA320.

LAXS of segment 3p13-pter Chromosome 3 was involved in numerical or structural

aberrations in 22 of the 25 RCCs. Irrespective of the type of aberrations, such as monosomy, deletion or translocation, the loss of different segments of chromosome 3 was encountered in all these cases (Fig. 2). The largest group of tumours showed deletion of the short arm of chromosome 3. In a sideline of tumour HA210, the translocation 3;5, leading to

15 3 15 2

* 12

13

14

15

q 21

22

23

31

32

33

34

35

i *

*

5 FIGURE 3 - Diagrammatic representation of a normal chromosome 5 .

Solid lines represent gains of chromosome segments or entire chromo- somes. Breakpoints are marked with an asterisk. The shortest region overlap analysis showed a trisomy for the 5q22-qter segment.

Page 5: Specific chromosome aberration in human renal cell carcinoma

CHROMOSOME ABERRATION IN RCC 175

occurring in the karyotype. In case HA225, the formation of a ring chromosome 3 was the only rearrangement. In 4 cases (HA343, HA223, HA267 and HA281), a translocation be- tween chromosome 3p and chromosomes 5q, 5q, 8q and Yp, leading to the loss of 3p segment, was noted as the only rearrangement. Furthermore, 4 other tumours were character- ized by a single chromosomal change in addition to the rear- rangement of chromosome 3. HA188 showed a translocation 1;3 and one additional change in the form of 14q+. The stemline cells of tumour HA2 10 manifested a translocation 3;5 and deletion of the Y-chromosome. Tumour HA312 was char- acterized by 2 unbalanced translocations, namely t(3; 12) and t(8;15). Trisomy 13, in addition to translocation 3;5 was ob- served in all cells of tumour HA320. Rearrangement of chromosome 5

The chromosome 5 was involved 17 times in karyological abnormalities observed in 13 of the 25 tumours. Trisomy or tetrasomy of chromosome 5 was found in 5 cases (HA214, HA215, HA273, HA299 and HA343), while partial trisomy was found in 6 cases (HA1887 HA1949 HA223, HA267, HA27 1 and HA320) and partial tetrasomy and pentasomy each in one case, in HA299 and HA210, respectively. All but one (HA230) structural aberration of the chromosome 5 involved a translocation of 5q segment to the short arm of chromosome 3. The breakpoint in this type of rearrangement was always located at p13 on chromosome 3, but varied between bands q11.2 and q22 on chromosome 5. The analysis of shortest region overlap revealed a tri- or polysomy of the 5q22-qter region in 12 of the 25 RCCs. A duplication of a marker chromosome t(3;5) was found in a sideline of tumour HA210. In HA299, a tetrasomy of chromosome 5 with several other chromosomal gains was noted in addition to t(3;5). It is re- markable that loss of chromosome 5 or constituent segments was not observed.

FIGURF 4 - Gbanded karyotype of a tumour cell from case HA228 showing monosomy 3 BS the only chromosome change.

DISCUSSION

The aberration of chromosome 3 was the most common cytogenetic finding in the 25 non-familial RCCs presented here. Structural or numerical changes of chromosome 3 were detected in 22 of the 25 tumours. Three forms of aberration of

FIGURE 5 - Ci-banded karyotype of the tumour HA225 showing a ring chromosome 3 :as the only karyotype change.

the loss of 3p, occurred twice; the RCC HA214 showed a chromosomal defect for 3p, as well as trisomy for 3q21-qter segment. A dual aberration of chromosome 3, leading to the loss of the short arrn and trisomy for 3q, was found in tumour HA284. Disregarding case HA225, with ring chromosome 3, shortest region overlap analysis consistently localized changes to an area of 3p13-pter (Fig. 2). This chromosomal defect was observed in all cells in 21 of 25 RCCs. A very careful retro- spective review of well-banded metaphases failed to identify this deleted segment as translocated to another site. One tu- mow, HA225, showed only terminal deletion of both arms of chromosome 3, and only three of 25 RCCs, namely HA2 11, HA215 and HA221 (Fig. 9), showed a “normal” chromosome 3 pair without micnoscopically demonstrable rearrangement. Rearrangement of chromosome 3 as the sole karyotype change

Karyotyping revealed minimal chromosomal changes in 12 of 25 RCCs. In 8 cases the rearrangement of chromosome 3 was the only karyotype change. Tumour HA228 was charac- terized by monosonny 3 and in 2 cases. HA245 and HA3197 a deletion 3p was observed as the sole chromosomal defect Note the deletion of the short arm of chromosome 3 at p13.

FIGURE 6 - G-banded karyotype of a stem-line cell from RCC HA245.

Page 6: Specific chromosome aberration in human renal cell carcinoma

176 KOVACS ET AL.

Renal cancer occurs in sporadic and hereditary forms. Cohen er al. (1979) reported a constitutional reciprocal translocation 3;8 in 10 members of a family predisposed to renal cancer, and suggested that this inherited chromosomal rearrangement predisposes carriers to development of RCC. Despite this highly intriguing finding in rare familial RCC, reports con- cerning the cytogenetic features of non-familial renal cancers, which occur far more frequently, are scanty. In cell lines established from RCC, several structural changes, including the rearrangement of chromosome 3, were described (Hage- meijer et al., 1979; Sytkowski et al., 1983; Wang er al., 1983). In direct preparation of a renal cancer rearrangement of chro- mosomes 1, 3, 6, 7 and 17 were found (Ferti-Passantonopou- IOU et al., 1984). Yoshida et al. (1986) have reported an aberration of chromosome 3 in 8 of 12 primary RCCs analysed in short-term culture. Although the detailed karyotype was not reported for each case, several clonal and non-clonal chromo- some aberrations were seen to include rearrangement of chro- mosome 3. Yoshida et al. suggested that RCC may be cytogenetically classified into 3 groups, namely tumours with aberration of chromosome 3, tumours with other clonal changes and tumours without clonal changes. In another study, Berger et al. (1986) analvsed 8 RCCs using a short-term

FIGURE 7 - A representative, G-banded karyotype of a stem-line cell cultke method.' The'compiexity of the chrom&omal changes

why their report offers no evidence of a common chromosomal from tumour HA210: 46,X,del(Y)(q12),der(3)t(3;5)(pI3;q22). and the low number Of metaphases Observed Seems to explain del( 14)(q 13).

abnormality. In contrast to these studies carried out on non-familial RCCs,

we found that each RCC was derived from a single progenitor cell and was characterized by constant chromosomal aberra- tions, a well-defined stem-line karyotype and sometimes by one or more sidelines with additional karyotype changes, marking the clonal evolution of such tumours. We karyotyped at least 14, commonly 20 to 30 metaphases in each case; thus our results may be regarded as characteristic of the entire tumour cell population occurring in the RCCs studied. Fur- thermore, the aberration of chromosome 3 was a very consis- tent karyotypic change in the stemline as well as in the sideline karyotypes of 22 of 25 RCCs. Therefore, our results present strong evidence for the rearrangement of chromosome 3 as a specific chromosomal defect in human RCCs.

Aberration of chromosome 3 may occur in various human tumours (Mitelman, 1986). Involvement of chromosome 3 in karyotype changes, in the form of t(3;8) was found frequently in benign pleomorphic adenomas of the salivary gland (Mark

FIGURE 8 - G-banded karyotype of a side-line cell from RCC HA267 43.X ,-Y,der(3)t(3;5)(pl3;q22),-14,-18.

chromosome 3 were found, namely loss of the entire chromo- some 3, deletion of the short arm without translocation, and deletion followed by translocation of another chromosome to the deleted 3p. All 3 types of karyotype change led to the loss of variable segments of chromosome 3, but shortest-region overlap analysis revealed that segment 3p13-pter was lost in 21 of the 25 RCCs. Since the karyotype of most RCCs ana- lysed in this study was simple, with a single or only very few marker chromosomes, and because the rearrangement of chro- mosome 3 was the only karyotype change in 8 of the RCCs, it is very tempting to assume that this chromosomal defect is the initial cytogenetic change during the development and progres- sion of human RCCs. Chromosomal aberration was absent in the peripheral blood lymphocytes or normal kidney cells, therefore the deletion 3p must be an acquired, specific chro- mosomal defect in non-familial RCCs.

FIGURE 9 - G-banded kayotype of a tumour cell from case HA221 :48,XY, +7. + 17. Note the 2 normal chromosomes 3.

Page 7: Specific chromosome aberration in human renal cell carcinoma

CHROMOSOME ABERRATION IN RCC 177

and Dahlenfors, 1986). A chromosomal defect with deletion 3p( 14-23) was believed to be specific for small-cell carcinoma of the lung (Whang-Peng et al., 1982), but these findings have recently been strongly questioned by other investigators (Wurster-Hill et al., 1984; Zech et af., 1985). Most solid malignant tumours at the time of cytogenetic analysis are characterized by complex chromosomal rearrangements. Therefore, the initial “tumour-specific” chromosomal changes are impossible to determine at this point of the tumour devel- opment. Some chromosomes, including chromosome 3, are known to be preferentially involved in the secondary changes during the clonal evolution of human neoplasms, without evi- dencing specificity for a distinct tumour type (Mitelman and Levan, 1981). In most carcinomas, random or frequent chro- mosomal aberrations, secondary to the primary changes, can- not be differentiated from the tumour-specific, initial chro- niosomal rearrangements. Therefore, tumours showing only single or minimal karyotypic changes are very important in the determination of chromosomal aberrations of prime signif- icance. Among 5,345 human cancers and leukaemias, only 2 carcinomas, showing rearrangement of chromosome 3 as the sole karyotypic change, were detected (Mitelman, 1986), and cancer with monosoiny 3 as a single aberration was not found (Heim and Mitelman, 1986). In the present study, 8 of 25 RCCs were characterized by numerical or structural aberration of chromosome 3 as the sole karyotype deviation. We believe that RCC, as a type of solid malignant tumour marked by minimal, consistent karyotype changes, is therefore a highly suitable model for the analysis of cytogenetic, as well as molecular biological events, corresponding to the initiation of malignant proliferation from normal cells, in this case from normal tubular cells of the kidney. The loss of 3p segment is, in all probability, the first step in the clonal evolution of RCCs, to be followed by other non-random and accidental chromo- some changes.

Although the breakpoints on chromosome 3 were not always precisely determined in the RCCs reported previously, they were localized between bands p l l and p21, mostly at p14. This is interesting, because the breakpoint at 3p14 represents the most common fragile site within the entire human genome (Smeets et al, 1986), and the majority of the proven fragile sites is associated with a higher rate of breakage as well as with specific chrornosome rearrangements in cancer (Hecht and Sutherland, 1984; LeBeau and Rowley, 1984; De Braeke- leer et al., 1985). Our results are in partial contradiction to those previously re:ported. In the RCCs presented here, the breakpoints occurred at bands 3p13, p12.2 and p11.2; thus the more distal break on the short arm of chromosome 3 occurred at p13. One rnay co’nclude that the breakpoint on chromosome 3 is far less significant than the following karyotypic aberra- tion, namely the lO!iS of the deleted 3p segment. However, we believe that definite localization of the more distal breakpoint on chromosome 3p is of great importance. In the constitutional karyotype of family members developing RCC (Cohen et al. , 1979) a reciprocal translocation with breakpoint at p14.2 on chromosome 3 was determined (Wang and Perkins, 1984). If the rearrangement within the band 3p14 or a submicroscopic deletion predisposes to the development of RCC in this family, one may suspect that the rearrangement or loss of such chro- mosomal region is also indicative of the development of spo- radic RCCs. If the band 4p14 is the site of a recessive “cancer gene”, the development of RCCs requires the deletion of this gene either by chromosomal aberration or by submicroscopi- cal loss of DNA. The band 3p14 was deleted on one of the homologous chroniosomes 3 in all cases showing rearrange-

ment of chromosome 3 in this study, and more distal breaks occurred at greater frequency at 3p13. These data support the hypothesis that the rearrangement or loss of the band 3p14 is the first detectable chromosomal change both in familial and in sporadic RCCs. Renal cancer without visible rearrangement of chromosome 3, for example cases HA211, HA215 and HA221 in our series, perhaps acquired the loss of that gene by other mechanisms, such as by non-disjunction followed by reduplication of the chromosome 3, carrying a mutation or submicroscopical loss of suppressor gene, or by mitotic re- combination as proposed for retinoblastoma (Cavenee et d.. 1983).

Chromosome 5 was very often involved in the translocation between chromosome 3p and other chromosomes. The break- point was very constant at p13 on chromosome 3 but varied between bands q11.2 and q22 on chromosome 5. This type of chromosomal rearrangement, in addition to the gains of chro- mosome 5 observed, led to a polysomic state of the segment 5q22-qter. The cellular oncogene fms has been localized to 5q34 (Groffen et a l . , 1984), which is distal to the breakpoints found in the RCCs presented here. One could speculate that increase in the copy number of 5q and thus of the genomic locus of the c-fms is a mechanism leading to higher levels of expression of this oncogene. However, the significance of preferential involvement of chromosome 5 in the clonal rear- rangement of RCCs is not yet clear. Similarly, the role of 2 other non-random karyotype changes, the loss of chromosome 14 and the gain of chromosome 7, in the clonal evolution and progression of RCCs, remain to be elucidated.

We propose that the loss of variable segments of the short arm of chromosome 3 is a very consistent chromosomal ab- normality, which marks human RCCs and may be the initiating cytogenetic event in their development. The presence of this chromosome rearrangement in the great majority of the RCCs we have analyzed suggests that a causal relationship may exist between the loss of a specific chromosomal segment and this neoplasm. The present data on RCCs, as well as previous reports on constitutional and acquired rearrangements of chro- mosome 3 associated with RCC, provide strong evidence that a particular location of 3p, probably at 3p14, should be re- garded as the site of a recessive “cancer gene” responsible for the regulation of growth and differentiation of normal tubular cells of the kidney. Because the loss of this region was not microscopically demonstrable in 3 of the 25 tumours, and when recognizable, involved only one of the 2 homologous chromosomes, it seems possible that a mutation or loss of DNA may take place at a submicroscopic level, analogous to that demonstrated for retinoblastoma (Cavenee et al., 1983; Dryja et al., 1984) and Wilms’ tumour (Koufos et al., 1984; Orkin et al., 1984; Reeve et a/. , 1984; Fearon et al., 1984). A cellular oncogene c-rafl has been localized to the region 3p24-25 (Bonner et al., 1984). Whether the c-raf oncogene is involved in the initiation or progression of RCCs remains to be elucidated. Further studies are required to evaluate the molecular biological events corresponding to this apparently specific chromosomal defect, which seems responsible for the development of RCCs.

ACKNOWLEDGEMENTS

The authors thank Ms. M. Cassens and Ms. A. Emanuel for expert technical assistance. This work was supported by a grant from the Deutsche Forschungsgemeinschaft (DFG KO 841 /3- I ) .

Page 8: Specific chromosome aberration in human renal cell carcinoma

178 KOVACS ET AL

REFERENCES

BENEDICT, W.F., BANERIEE, A., MARK, C. , and MURPHREE, A.L., Non- random chromosomal changes in untreated retinoblastomas. Cancer Ge- ner. Cvrogener., 10, 311-333 (1983). BERGER, C.S., SANDBERG, A.A., TODD, I.A.D., PENNINGTON, R.D., HADDAD, F.S., HECHT, B.K., and HECHT, F., Chromosomes in kidney, ureter, and bladder cancer. Cancer Gener. Cytogenet., 23, 1-24 (1986). BONNER, T., O’BRIEN, S.J., NASH. W.G.. and RAPP, U.R., The human homologs of the raf (mil) oncogene are located on human chromosomes 3 and 4. Science, 223,71-74 (1984). BRODEUR. G.M., GREEN, A.A., HAYES, F.A., WILLIAMS, K.J., WILLIAMS, D.L., and TSIATIS, A.A., Cytogenetic features of human neuroblastomas and cell lines. Cancer Rex, 41,4678-4686 (1981). CASPERSSON, T., ZECH, L., JOHANNSSON, C. , and MODEST. E.J., Identifi- cation of human chromosomes by DNA-binding fluorescent agents. Chro- niosoma (Bed.), 30, 215-227 (1970). CAVENEE, W.K., GRYJA, T.P., PHILIPS, R.A., BENEDICT, W.F., GODBOUT, R., GALLIE, B.L., MURPHREE, A.L., STRONG, L.C., and WHITE, R.L., Expression of recessive alleles by chromosomal mechanisms in retinoblas- toma. Nurure (Lond.), 305,779-784 (1983). COHEN, A.J.. LI, F.P.. BERG, S . , MARCHETTO, D.J., TSAI, S. . JACOBS, S.C. , and BROWN. R.S., Hereditary renal-cell carcinoma associated with a chromosomal translocation. N . Engl. J. Med., 301, 592-595 (1979). CROCE, C.M., and NOWELL, P.C., Molecular basis of human B cell neoplasia. Blood, 65, 1-7 (1985). DE BRAEKELEER. M., SMITH. B., and LIN. C.C., Fragile sites and struc- tural rearrangements in cancer. Hum. Genet., 69, 112-116 (1985). DRYIA. T.P., CAVENEE, W., WHITE, R., RAPAPORT, J.M., PETERSEN, R., ALBERT, D.M., and BRUNS, G.A.P.. Homozygosity ofchromosome 13 in retinoblastoma. N . Engl. J . Med., 310, 550-553 (1984). FEARON, E.R., VOGELSTEIN, B., and FEINBERG, A.P., Somatic deletion and duplication of genes on chromosome 11 in Wilms’ tumours. Narure (Lond.), 309, 176-178 (1984). FERTI-PASSANTONOPOULOU, A, , PANANI, A., and RAPTSI, S . , G-banded karyotype of a primary renal-cell carcinoma. Cancer Genet. Cyrogenet.,

GROFFEN. J. , HEISTERKAMP, N., SPURR. N.K.. DANA, S.L. , WASMUTH, J.J., and STEPHENSON, J.R., Regional assignment of the human c-fils oncogene to band q34 of chromosome 5 (HGM7). Cytogenet. Cell Genet., 37,484 (1984). HAGEMEIJER, A,. HOEHN, W., and SMIT, E.M.E., Cytogenetic analysis of human renal carcinoma cell lines of common origin (NC 65). Cancer Res.. 39,4662-4667 (1979). HECHT, F., and SUTHERLAND, G.R., Fragile sites and cancer breakpoints. Cancer Gener. Cytogenet. , 12, 179- I8 1 ( 1984). HEIM, S. . and MITELMAN, F., Numerical chromosome aberrations in human neoplasia. Cancer Gener. Cytogenet., 22, 99-108 (1986).

ISCN 1985: An International Sysrem for Human Cytogenetic Nomencla- ture. Karger, Base1 (1985). KLEIN, G., and KLEIN, E., myc/Ig juxtaposition by chromosomal translo- cations: some new insights, puzzles and paradoxes. Itnmunol. Today, 6, 208-215 (1985). KNUDSON. A.G., MEADOWS, A.T.. NICHOLS, W.W., and HILL, R., Chro- mosomal deletion and retinoblastoma. N . Engl. J . Med., 292, 1120-1123 (1976). KONOPKA. J.B.., and WITTE, O.N.. Activation of the abl oncogene in murine and human leukaemias. Biochim. biophys. A m , 823, 1-17 (1985).

LAND, N.G.. JENKINS, N.A., and CAVENEE, W.K., Loss of alleles at loci on human chromosome 11 during genesis of Wilms’ tumour. Nature

11, 227-232 (1984).

KOUFOS. A,. HANSEN, M.F., LAMKPIN, B.C., WORKMAN. M.L., COPE-

(Lurid.). 309, 170-172 (1984). LE BEAU, M.M. and ROWLEY, J.D., Heritable fragile sites in cancer. Nature (Lond. ) , 308,607-608 (1984). MARK, J., and DAHLENFORS, R. , Cytogenetical observations in 100 human benign pleomorphic adenomas: specificity of the chromosomal aberrations and their relationship to sites of localised oncogenes. Anticancer Rex, 6, 299-308 (1986). MITELMAN, F., Clustering of breakpoints to specific chromosomal regions in human neoplasia. A survey of 5,345 cases. Hereditas. 104, 113-119 (1986). MITELMAN. F., and LEVAN, G., Clustering of aberrations to specific chromosomes in human neoplasms. IV. A survey of 1,871 cases. Heredi-

ORKIN. S.H., GOLDMAN. D.S . , and SALLAN, S.E., Development of homo- zygosity for chromosome I lp markers in Wilms’ tumour. Nature (Lond.).

PATHAK, S. , STRONG, L.C., FERREL, R.E., and TRINDADE. A, , Familial renal cell carcinoma with a 3;11 chromosome translocation limited to tumor cells. Science, 217,939-941 (1982). REEVE, A.E., HOUSIAUX. P.J., GARDNER, R.J.M., CHEWINGS, W.E., GRINDLEY, R.M., and MILLOW. L.J., Loss of a Harvey ras allele in sporadic Wilms’ tumour. Nature (Lond.). 309, 174-176 (1984). RICCARDI, V.M., SUJANSKY, E.. SMITH, A.C., and FRANCKE, U . , Chro- mosomal imbalance in the aniridia-Wilms’ tumor association: 1 Ip inter- stitial deletion. Pediatrics, 61, 604-610 (1978). SEABRIGHT, M., A rapid banding technique for human chromosomes. Lancet, 11, 971-972 (1971). SLATER, R.M.. The cytogenetics of Wilms’ tumor. Cancer Genet. Cyto- genet., 19,37-41 (1986). SMEETS, D.F.C.M., SCHERES, J.M.J.C., and HUSTINX, T.W.J.. The most common fragile site in man is 3p14. Hum. Genet., 72, 215-220 (1986). SUMNER, A.T., EVANS, H.J., and BUCKLAND, R.A., New technique distin- guishing between human chromosomes. Nature (Lond.). 232, 3 1-32 (1971). SYTKOWSKI, A.J., RICHIE, J.P. , and BICKNELL, K.A.. New human renal carcinoma cell line established from a patient with erythrocytosis. Cancer Res., 43, 1415-1429 (1983). WANG, N., SOLDAT. L., FAN, S . , FIGEMSHAU, S. , CLAYMAN, R., and FRALEY, E., The consistent involvement of chromosome 3 and 6 aberra- tions in renal cell carcinoma. Amer. J . hum. Genet., 35, 73A (1983). WANG, N., and PERKINS, K.L., Involvement of band 3p14 i n t(3;8) hereditary renal carcinoma. Cancer Genet Cytogener., 11, 479-481 ( 1984). WHANG-PENC, I . , BUNN, P.A., KAO-SHAN, C.S., LEE, E.C., CARNEY, D.N., GAZDAR, A., and MINNA, J.D., A nonrandom chromosomal abnor- mality, del 3p( 14-23), in human small cell lung cancer (SCLC), Cancer Gener. Cytogenet., 6, 119-134 (1982). WURSTER-HILL, D.H., CANNIZZARO, L.A., PETTENGILL, 0,s.. SORENSEN, G.D., CATE, C.C.. and MAURER, L.H., Cytogenetics of small cell carci- noma of the lung. Cancer Genet. Cyrogenet., 13, 303-330 (1984). YOSHIDA. M.A., OHYASHIKI, K. , OCHI, H. , GIBAS, Z . , PONTES. J.E., PROUT, G.R.. and SANDBERG, A.A., Cytogenetic studies of tumour tissue from patients with nonfamilial renal cell carcinoma. Cancer Rex, 46, 2139-2147 (1986). YLJNIS, J.J., The chromosomal basis of human neoplasia. Science, 221, 227-236 (1983). ZECH, L., BERGH, J . , and NrLssON, K., Karyotypic characterisation of established cell lines and short term cultures of human lung cancers. Cancer Genet. Cytogenet., 15, 335-347 (1985).

IUS, 95, 79-139 (1981).

309, 172-174 (1984).