Karyotypic Analysis of a Hetemseneous Human Transitional Cell Carcinoma of the Bladder
Joanne L. Brown, Robyn Lukeis, Zambela Palividis, Jane Wass, Derek Raghavan, and Pamela J. Russell
ABSTRACT: The UCRU-BL-17 (BL-17} series of xenogmfts, tissue culture sublines, and cloned cell lines [Fig. 1] shows a range of heterogeneous growth characteristics both in vitro and in vivo Crable 1] and represents a model of human bladder cancer heterogeneity. Cytogenetic analysis was undertaken to determine if spe- cific chromosome changes correlated with particular aspects of the heterogeneous phenotypes. The BL-17 sublines and cloned cell lines shared many common chromosome abnormalities. Indeed, the cloned cell lines showed nearly identical karyotypes despite their marked differences in growth characteristics. Karyo- typic evolution with passage through the nude mice was apparent, however. This evolution occurred at the specific chromosome regions of lp12, 3cen-3p21, and 6cen-6q21. Whether the heterogeneity ofkaryo- type between the BL-17 cell lines resulted from the existence of multiple clones in the original patient tumor or from karyotypic instability through passage in nude mice is uncertain, but in either case the specificity of karyotypic evolution observed suggests that lp12, 3cen-3p21, and 6cen-Tq21 are hotspots for rearrungement in the BL-17 tumor. No specific correlations between chromosome abnormalities and biologic character/stics could be made, but several unique karyotypic features arose in the progression to two of the sublines, BL-17/23a and BL-17/O/X2A, coinciding with a loss of anchorage-independent growth by BL-17/23a and a change in growth in vivo from a solid tumor to a fluid-filled tumor by BL-17/O/X2A.
INTRODUCTION
Human bladder cancer progression could result in the exis- tence of several cell subpopulations in one tumor that differ with respect to their ability to invade or metastasize or in their sensitivity to cytotoxic agents or irradiation. This tu- mor heterogeneity may explain the different clinical courses followed by apparently histologically similar human blad- der cancers. An important factor that may contribute to this process is genomic instability of tumor cells [1]. Several cell lines of human bladder cancer have been established from primary or metastatic human tumors to study the biology and heterogeneity of this disease [2-5].
A series of cell lines and xenografts established in our laboratory from a human bladder cancer, UCRU-BL-17 {BL- 17), represents a model of heterogeneity in a single tumor. The original human tumor was a highly invasive, grade III
From Kflnernatsu Laboratories, Royal Prince Alfred Hospital (J. L. B., Z. R, R ]. R.I, the Department of Cytogenetics, St. Vincents Hospital, Victoria (R. L.}, C, enito-Urinary Research Unit, Oncology Research Centre, Prince of Wales Hospital, Bondwick, N.S.W. C/. L. B., Z. P, P./. R.J, Australia, and Divisions of Solid Tumor Oncology and Investigational Therapeutics, Roswell Pork Cancer Institute, Buffalo, New York.
Address reprint requests to : /oanne L. Brawn, Ph.D., Genito- Urinary Research Unit, Oncology Research Centre, Prince of Wales Hospital, High Street, Randwick, NSW 2031, Australia.
Received November 5, 1992; accepted August 3, 1993.
116 Cancer Genet Cytogenet 72:116-125 (1994) 0165-4608/94/$07.00
transitional cell carcinoma of the bladder showing features of glandular and squamous differentiation [6]. When this tumor was grown in vivo as a xenograft line or in vitro as a continuous cell line, a similar histologic heterogeneity was apparent [6, 7]. Four tissue culture sublines established from different xenograft passages and nine cloned cell lines es- tablished by the limit dilution of one of these sublines show heterogeneous phenotypes. Each of the BL-17 sublines is tumorigenic in nude mice but shows heterogeneous growth characteristics; however, three of the nine BL-17 clones are nontumorigenic in nude mice. Of the remaining six tumori- genic clones, three have shown some local invasion from the subcutaneous (s.c.] implantation site in nude mice [8]. One of these, clone B8, is frankly metastatic when implanted in the bladder wall of nude rats [9]. These nine cloned cell lines and the BL-17 sublines thus represent a model of bladder tumor heterogeneity and progression. We performed cyto- genetic analysis of the BL-17 sublines and cloned cell lines to determine whether cytogenetic changes correlate with differences in biologic behavior exhibited by the cells in vitro and in vivo (Table 1).
MATERIALS AND METHODS
Cell Lines The BL-17 cell lines were established from different ~eno~aft passages of BL-17 [7] (Fig. 1). The original tumor from which
Abbreviations: ND not determined; A, derived from ascitic fluid in the xenogra_~; S, derived from solid tissue surrounding the ascitic fluid.
UCRU-BL-17 was established was obtained from a 69-year- old woman of blood group A who had a grade III, stage T4b transitional cell carcinoma {TCC) of the bladder. She was treated with cisplatin and radiotherapy but had only a tran- sient remission and died 4 months after presentation. The xenograft line was established before treatment. Flow cyto- metric analysis showed a diploid tumor with a tetraploid peak. The BL-17 cloned cell lines were established by limit dilution of the continuous cell line BL-17/2, a cell line estab-
lished from the tumor tissue resulting after the second pas- sage of the BL-17 tumor in nude mice (Fig. 1}, and were characterized previously [8]. Some of the features of the BL- 17 sublines and clones are summarized in Table 1.
Cell Culture Tumor cells were cultured according to previously reported techniques [6]. Tumor epithelial cells were cultured at 37°C
Figure 1 Derivation of UCRU-BL-I? cell lines. * Line failed to take on further serial passage in nude mice. Tumors derived from the BL-17/0CL (cell line) formed fluid-filled cysts f~om passage 2 on when implanted in nude mice. Different sublines were established in culture from cells in the fluid (BL-17/O/X2A) and from cells forming the outer coat of the c y s t (BL-17/0/X2S).
Biopsy 17
Nude Mice Tissue Culture Nude Mice Tissue Culture
BL17 Clones BL17/0/X3
BL17/4 + BL17/O/X4
BL17/23alpha + * BL17/5 f
BL17/1 BL17/O/X1 BL17/O/X2A
BL17/2 ~'~ BL17/2CL BL17/O/X2
BL17/O/X2S B L17/3 from ,o,d
~ t ~ mat of oy~
BL17~/X5
BL17/O ~ BL17/OCL ~ BL17/O/XO BL17CL r r f
1 1 8 J . L . B r o w n e t a l .
A F i g t l r e 2 {A) UCRU-BL-12/0/X1:64-69,<3n.~,XX~IeI(X)(p22), + der(1;3){ql0;ql0)x2,- 2 , - 3 , - 4 , + i(5)(p10)x2~lel{6)(q21), + Z - 8,der{8;11}{q10;q10) ~ler{11)add{11) {pl0), - 13, - 14,add{15)(p11), - 16~lel{1Y){p12), - 18, - 19, - 21, - 22, + M1, + M2, + M3, + M4, + d m i n [ c p 6]. Markers no t n u m b e r e d are nonc lona l . {B} UCRU-BL-17/O/X2A-C2:59-70,<:3n~,XX,- X, + dar{1;5) {?q10-p12 ;q11} [c p 3], - 3, - 4, + i{ 5) {pl0) ,del{ 6) {q21}, + i{7) {plO) ,del{7) {q22) ,der{8 ;11} {ql0;ql0), - lO,der{11) add{11) {pl0),
- 13 [cpS],der{6;13){plo;q10)[cp6] ~ler{14)t{ll;14}(q13;q22}[cp7], - 14[cp4], - 15,add{15){p11}, - 16, + 17[cp5], - 17[cp6], - 18, - 19[cp6],add{19){qll)[cpS],t{2;19){p21;p13)[cpS], - 21 , - 22, + M1 { - M3 BL-I?/O/X1), + M2, + M3 {= M4 BL-17/O/X1), + d m i n [cp11].UCRU-BL-17/O/X2A-C1: i d e m , - der(1;5){?q10-p12;q11), der(1;3}{ql0;qlO)[cp8]. {C) UCRU-BL-17/2 Clone 1 {C1):67-69, <:3n>, X X , - X {or del{X)(p22}), t(1;2}(p12;p24)x2,- 2,addl(3){p21 ), -4,i{5)(p10)x2,del(6}{q21){x1-2},addl(6){q13){xl-2}, + 7,i{7)(p10), - 8,der{8}inv{8)(qllq22)inv(8){q22q24}, - lO,?add{ll)(?p14}, + der{llp}add{11)[plO),add(13){p11), - 13, - 14, - 16, + add{17){p11}, - 18, - 19, - 20, + add[21){q21}x2, - 22, + M[cp 4]. {D) UCRU-BL-17/2 c lone 2 {C2) a n d the BL-17 c loned
cell l ines: 44-48,<:2nY,XX, + del{X}{p22),add(1){p12)~lel{2){q14q31), - 4, + i{5){p10),del{6}{q21), + Z - 8,der{8;11){q10;q10}, der{ll)add{11){plO},-13,-16,add{19){q13),-22 [cp4]. {E} UCRU-BL-1Y/23a ear ly passage : 57-66,<:3n>,XXX,?add(X} [p22),add{1)(p12}x2, - 2,add2(3){p21)x2, - 4,i(5){p10}x2,del{5}{q21},add{6)(q27}, - 6,der{7;9}{pl0;qlO), + 7, - 8,add{9){p22}, - 10,?add{11)(?p14), - 15, - 16, - 16, - 16, + add{17){q21), - 18, - 19, - 21, - 22, + MI{ = M3 BL17/O/X1), + M2 ( = M2 BL-17/0/X2}, + M3 [cp8]. Markers no t n u m b e r e d are nonc lona l . {F) UCRU-BL-17/23a late passage: 57-66,<~nY,XXX,?add{X){p22), add(1}{p12)x2, - 2,addz[3}{p21)x2, - 4,i{5){p10)x2,del(6){q21),add2{6){q13), - 6~ler {~ 9}(p10;qlO) ,inv{7)(q22q36), - 8,add{9} {p22), - lo,?add(11){?p14), - lS, - 16, - 16, - 16, + add(ly){q21), - 18, - 19, - 21, - 22, + MI{ ffi BM3 L17/O/X1), + M2 { = M2 BL- 17/0/X2), +M3[cp5]. Arrows indicate n o n c l o n a l changes .
i n Z 5 % COz i n a i r i n R o s w a l l P a r k M e m o r i a l I n s t i t u t e t i s -
s u e c u l t u r e m e d i u m , R P M I 1640 ( F l o w L a b o r a t o r i e s , N o r t h R y d e , N.S .W. , A u s t r a l i a } , s u p p l e m e n t e d w i t h 0 . 2 1 % s o d i u m
b i c a r b o n a t e , 4 m M lrglotAmine (Flow Labora to r i e s} , a n d 10%
fe t a l c a l f s e r u m { C o m m o n w a a l t h S e r u m L a b o r a t o r i e s , Vic to -
r i a , A u s t r a l i a ) , h e a t e d to 5 6 ° C for 30 m i n u t e s . C o n f l u e n t c e l l s
w a r e h a r v e s t e d b y t r y p s i n i z a t i o n w i t h 0 . 2 5 % t r y p s i n {F low
L a b o r a t o r i e s } .
C y ~ e n e t i c A n a l ~ i s
S u b c o n f l u e n t m o n o l a y e r s w e r e e x p o s e d to 0 .8 ~ g / m l Co l - cemid for 30 minutes, trypsinized and harvested by standard techniques. Metaplmse spreads ware G-banded by a modi- fied method of Seabright [10], and cytogenetic analysis was performed according to the Paris convention [11] on at least 6 cells. In general abnormalities discussed are those pres- ent in the majority {>80%} of cells.
IP
S t
9f
4-,
-"
lid
Im
I--,
--
Id ]
~ 4.
. 41
'-"
Ib
II
d~
t~
C~
L"I
12 2 J.L. Brown et al.
RESULTS
Complex karyotypes were obtained from all sublines and cloned cell lines except BL-17/0/X2S and Bll. Complete keryo- types shown in Figure 2 are representative of most cells but also include nonclonal markers. We previously described the karyotype of the subline BL-17CL [6] as near-triploid and con- raining a translocation between chromosomes 11 and 13. Sub- sequent cytogenetic analysis of this cell line, other BL-lY sub- lines, and the BL-17 clones suggested that this translocation was actually between chromosomes 8 and 11. The full de- scription of the resulting chromosome is der{8;11}(8qter-* 8q10: :11q10~llqter).
Chromosome changes common to all BL-17 cell lines were: - 4, + 7 {in all but the late passage of BL-17/23a), - 16, - 22, i{5}{p10), and del(6)(q21}. Identical abnormalities common to some but not all BL-17 cell lines are shown in Table 2 and include numerical and/or structural alterations in chromo- somes X, 1, 2, Z 8, 11, 15, 18, and 19, and double minute (drain) chromosomes. Different chromosome abnormalities involving the same chromosomes, often with breakpoints in the same chromosome region, were also apparent in the cell lines {Table 3; Fig. 3), in particular including abnormalities of chromosomes lp, 3p, and 6q, but also of 2p, 7q22, 11q, 13p, 17p, 19q, and Xp. The abnormalities of chromosome 1
included t(1;3), t{1;5), and add{lp} in different sublines. Different abnormalities also involved 3p It{l;3}, addl{3p), add2(3p}] and 6q [del{e)q21, addl{6}(q13), add2(6}(q13}, and der{6p;13q)]. Chromosomes unique to particular BL-17 cell lines included add{21}(q21} in karyotypic clone 1 of BL-17/2, der{7;9){pl0;ql0), add(9)(p22), and no normal chromosomes 16 in BL-17/23a, and add(X)(q28} in one of the cloned cell lines, C3.
The parent of the BL-17 cloned cell lines, BL-17/2, con- tained two distinct populations by cytogenetic analysis, a near-diploid clone and a near-triploid clone. The triploid clone did not contain the der(8;ll} described earlier and had several different cloned chromosome abnormalities (Fig. 2}. The diploid clone did not contain this marker. All BL-17 cloned cell lines contained the der{8;ll} and karyotypically were the same as the diploid clone of BL-17/2, indicating that only the diploid clone was selected in the initial limit dilu- tion of this cell line.
All BL-17 cloned cell lines shared very similar karyotypes {Fig. 2), and each had additional noncloned abnormalities. Clone C3 had the same karyotype as the other cloned cell lines plus the add(X)(q28}.
The BL-lY/23a subline was derived from the BL-17/5 xeno- graft (Fig. 1). The karyotype appears to be similar to that of the triploid karyotypic clone of BL-17/2, with some notable
Table 2 Abnormalities common to multiple cell lines of UCRU-BL-17
Cel l l i n e - X del{Xp} a d d { 1 ) ( p 1 2 ) - 2 i{7)(p lO) - 8 d e r ( 8 ; 1 1 ) - 1 0 ? a d d ( 1 1 ) ( p 1 4 } a d d ( 1 5 ) { p 1 1 ) - 1 8 - 1 9 d r a i n
BL-17 /O/X1 _ a o _ a _ a _ a _ a _ a a
B L - 1 7 / 0 / X 2 A _ a _ a _ a _ a _ a _ a
B L - 1 7 / 2 C 2 a n d
clones _ a _ o _ a B L - 1 7 / 2 C 1 b _ b _ a _ a _ a _ c _ a _ a _ a _ o
B L 1 7 / 2 3 a a a a _ a _ a a _ a _ a
Changes c o m m o n to all l ines inc lude - 4,i(5)(plO},del(6}(q21), + 7 (except in late passage of BL-17/23n), - 16, and - 22.
a Abnormal i ty present in cell line.
b Either - X or del(Xp} was present in BL-17/2-C1 cells.
c The karyo typ ic clone of UCRU-BL-17/2 conta ins a der(11p).
Table 3 Different abnormalities of the same chromosomes in the UCRU-BL-17 cell lines
B L - 1 7 / 0 f X 2 A B L - 1 7 / 2 C 2
Chromosome BL-17/O/X1 { - C1 and - C2) and clones BL-17/2C1 BL-17/23a
1 + d e r ( 1 ; 3 ) { ? q l O ; q l 0 } + d e r ( 1 ; 5 ) ( ? q 1 0 - p 1 2 ; q 1 1 ) { C 2 } _ a t{1 ;Z}{p12;q24) _ a
+ der{1;3)(q10;qlO)(C1) 2 a t{2;19}{p21;p13}a del(2){q14q31)
3 der(l;3){ql0;ql0} - 3, a der{1;3}[qlO;q10} - 3
fi _a ?der{6;13){plO;qlO) a _a
7 _o del(7){q22)a _a 8 a Q a
11 _ a _ o
13 _ a _ a 17 del(17}{p12) 19 a a d d ( 1 9 ) { q 1 3 ) X c I a
t(1;2}{p12;p24}a _a add(3}{p2 I} add(3)z{p2 I) add{6h(q13} a L add{6}2(q13)a;
E add{6}(q27) a _ a inv(7}{q22q36)
der(S}ainv(8}{qllq2 2) _ a inv{8}{q22q24) a
t{11;14){q22;q13} a ?add{11}{?p14} a ?add(11)(?p14) der{6;13){plO;q10)a add{13){p11)a a + or - 17 + add{17}{p11} add(17){q21) t{2;19)(p21;p13)add(Ig)(q11} . . . .
. . . . ?add(X)(p22}
Abbreviations: E and L, early and late passage.
a See Table 2 for abnormalities relating to these chromosomes. Subscripts indicate derivative chromosomes with the same breakpoint but involving trans- location of different unknown chromosome material.
123
|
a
Figure 3 Partial karyotypes representing the different abnormalities involving the same chromosomes observed in the BL-17 cell lines shown for chromosomes X, 1, 3, 6, 7, 11, 17, and 19. A normal homolos, when present, is at left; arrows indicate breakpoints. Subscripts havebeen used to differentiate derivative chromosomes with the same breakpoint but involving translocation of different unknown chromosome material. Abbreviated nomenclature is used. Full details of karyotypes are shown in Table 2.
124 J.L. Brown et al.
differences. BL-17/2 markers not present after the progres- sion to BL-17/23a included the translocation between 1 and 2, i(7)(pl0), der(llp), and derivatives of 13 and 21. Additional markers were also noted: a translocstion between 7 and 9, a derivative of chromosome 9 and, in later passages of BL- 17/23a, an inversion in 7q. Additional unknown markers were also evident. The BL-17/0/X2A subline was derived from one xenograft passage later than BL-17CL. Additional chromo- some alterations noted in BL-17/0/X2A included transloca- tions involving I and 5, 2 and 19, 11 and 14, as well as a der(19), an i(7)(p10), and deletions of 7(q22).
DISCUSSION
The BL-17 series of sublines and cloned cell lines represents a model of the level of tumor heterogeneity that can exist in TCC of the bladder. The keryetypes of the BL-17 cell lines were analyzed to determine ff the differences in growth char- actaristics noted both in vitro and in vivo between the lines (Table 1) correlated with cytogenatic differences between them. All BL-17 sublines are tumorigenic in nude mice, form- ing an s.c. tumor with no evidence of invasion or metastasis. In contrast, only six of the nine BL-17 cloned cell lines are tumorigenic and three of these have shown some invasion into local tissues. This did not correlate with differences ob- served in the karyotype. The ability of the cell lines to grow in an anchorage-independent manner also shows some het- erogeneity. All the sublines and cloned cell lines form colo- nies in methylcellulose except for the three nontumorigenic clones and BL-17/23a. Although the keryotypes of the non- tumorigenic clones were not different from the tumorigenic clones, BL-17/23a had unique karyotypic features (add2(3), add2(6), add(9), dar(7;9), and no normal chromosomes 6 or 16) that could be postulated to be related to this phenotypic change through a mechanism different from that of the non- tumorigenic clones. It is of interest that the colony-forming efficiency of the tumorigenic clones that contain only one population of BL-17/2 cells, is consistently higher than that of the other sublines, indicating this population may have different characteristics to the BL-17/2-CA/BL-17/23a popula- tion. Although each of the sublines and cloned cell lines was derived from a single human bladder tumor, they showed variations in tumorigenicity, invasiveness in vivo, anchorage- independent growth, and DNA ploidy (Table 1).
Several additional chromosome changes occurred in the passage from BL-17CL to BL-17/O/X2A, coinciding with, and perhaps involved in, a change in the nature of the tumors formed. BL-17CL was established from a solid tumor, whereas one passage later the tumor cells formed fluid-filled cysts [7]. BL-17/O/X2A was established from tumor cells in the tumor fluid [7].
The BL-17 cloned cell lines were established by the limit dilution of BL-17/2. In contrast with their heterogeneous growth characteristics in vitro and in vivo, the karyotype of all the clones was remarkably constant. These results con- trast with those described by Hastings and Franks [12], who suggested that in the case of the bladder cancer cell lines RT4, RTl12, T24, and EJ, deletions of chromosomes 8 and 9 [del(8){p21};dal(9){p13)] and a translocstion between chro- mosomes 15 and 18 [t{15;18){15qter--~15q21::lgpll~lgqter}]
were linked to the ability of the cells to undergo anchorage- independent growth in soft agar.
Karyotypically, the sublines and cloned cell lines also showed some degree of heterogeneity. A progression of changes appears to have occurred in the genome that may haw been due to either progression inherent in the tumor or to passaging through the artificial systems used for tissue amplification. All the cell lines share many common chro- mosome abnormalities, indicating their derivation from the same tumor. The BL-17 cloned cell lines shared a common karyotype with one of the metaphase clones of BL-17/2, whereas BL-17/23a shared more features with the triploid metaphase clone of BL-I?/2. Similarly, the karyotype of BL- 17/0/X2 shared common features with BL-17CL.
Some chromosome abnormalities were not common be- tween the sublines, but these changes often occur in the same chromosomes, suggesting that programmed changes are in- volved in the progression of the BL-17 tumor. At least some of the changes may not have been caused by the artificial growth environments. It is of interest that the particular chro- mosome regions of lp, 3p, and 6q were involved in different abnormalities, indicating karyotypic evolution at specific chromosome sites. One explanation for the heterogeneity of karyotype of this human bladder cancer model is that mul- tiple clones existed in the initial xenograft, as in the patient. Successive passage in vivo and in vitro of these multiple populations has selected related but distinct lines with shared and unique keryotypic abnormalities. An alternative hypoth- esis is that the karyotypic evolution occurred with serial pas- sage. In either case, this would assume that changes occurred at the same breakpoints, with progression; e,g., der(1;3] (BL- 17/0/X1 and X2A] progressing to t(1;2) plus add1(3} (BL-17/2- C1} and then to add(lp) and addz(3)(p21) (BL-17/23a), and addl(6)(q13) (BL-17/2-CQ to add2(6}(q13) (BL-12/23a). This in- dicates that the regions lcen-lp12, 3cen-3p21, and 6cen-6q21 are hotspots for rearrangement in this tumor. The cytogenetic differences detected between the BL-17 cell lines might have consequences as yet undetected in the growth assays used to characterize the cell lines. Clonal markers observed in the BL-17 cell lines involved chromosomes also shown to be in- volved in apparent nonrandom abnormalities in other blad- der tumors [13, 14].
Cytogenefic analysis of a heterogeneous tumor has demon- strated the presence of different keryetypic clones, implicat- ing it as a genetically labile tumor. The BL-17 sublines and cloned cell lines provide a unique model for analysis of the contribution made by merkers to the phenotype of each line. Specific keryotypic differences between the sublines and cloned cell lines did not correlate with other known biologic characteristics, however, although the keryotypic features unique to BL-17/23a may contribute to its inability to undergo anchorage-independent growth, and differences between karyotypes of BL17CL and BL-17/O/X2 may be involved in a change from solid tumor to fluid-filled tumor growth in vivo. Therefore, analyses at the molecular level, such as the re- striction fragment length polymorphism analyses described concerning chromosomes 9 and 17 [15, 16], are required to determine the genetic changes responsible for these differ- ent characteristics.
The UCRU-BL-17 sublines and cloned cell lines represent
Cytogenetics of a Human Bladder Cancer Model 125
only one model of b ladder tumor heterogeneity. Recently, a human b ladder cancer cell l ine was es tabl ished wi th fea- tures similar to those of UCRU-BL-17 [17]. The cell l ine HOK-1 was established from a transitional cell carcinoma of the blad- der that contained some g landula r and squamous differen- t iation. But the karyotype of HOK-1 showed only structural alterations of chromosome 1, suggesting that the hetero- geneous histologic characterist ics of this cell l ine may not be correlated wi th cytogenetic markers. Further s tudies of the cytogenetics and molecular biology of b ladder tumor cell subpopulat ions , inc luding the UCRU-BL-17 cell l ines and the HOK-1 model , are required to determine the relevance of specific chromosome and molecu la r markers to diagno- sis and prognosis of b ladder cancer.
REFERENCES
1. Heppner GH (1984): Tumor heterogeneity. Cancer Res 44: 2259-2265.
2. Christensen B, Kieler ], Vilien M, Don P, Wang CY, Wolf H (1964): A classification of human urothelial cells propagated in vitro. Anticancer Bes 4:319-337,
3. Hepburn P, Masters J (1983): The biological characteristics of continuous cell lines derived from human bladder. In: The Pathology of Bladder Cancer, Vol. II., G Bryan, S Cohen, eds. Academic Press, Boca Raton, p. 241.
4. Masters JRW, Hepburn PJ, Walker L, Highman WJ, Trejdosie- wicz L, Povey P, Parker M, Hill BT, Riddle PR, Franks LM (1986): Tissue culture model of transitional cell carcinoma: Charac- terizetion of twenty-two b-mun urothelial cell lines. Cancer Res 46:3630-3636.
5. Russell PJ, Raghavan D, Philips J, Wills EJ (1968): The biology of urothelial cancer. In: Management of Bladder Cancer. D Raghavan, ed. Edward Arnold, London, pp. 1-41.
6. Russell PJ, Jelbart M, Wills E, Singh S, Wass J, Wotherspoon J, Raghavan D (1988}: Establishment and characterization of a new human bladder cancer cell line showing features of squamous and glandular differentiation. Int J Cancer 41:74-82.
7. Russell P], Wills El, Philips J, Jelbart M, Gregory P, Raghavan D (1988): Features of squumous and adenocarcinoma in the same cell in a xenografted human transitional cell carcinoma: Evi- dence of a common histogenesis? Urol Res 16:79-85.
8. Brown JL, Russell P], Philips ], Wotherspoon J, Raghavan D (1990}: Clonal analysis of a bladder cancer cell line: An ex- perimental model of tumour heterogeneity. Br J Cancer 61:369-376.
9. Russell P], Ho Shon I, Boniface GR, Izard ME, Phih'ps J, Raghavan D, Walker KZ (1991): Growth and metastasis of human bladder cancer xvnografts in the bladder of nude mrs: A model for in- travesical mdioimmunotherapy. Urol Res 19:207-213.
10. Seabright M (1971): A mpid banding technique for human chro- mosomes. Lancet 2:971-972.
11. ISCN (1965): An International System for Human Cytogenetic Nomenclature. Harnden DG, Klingar HP (eds.); Published in collaboration with Cytogenet Cell Genet (Karger, Basel, 1965); also in Birth Defects: Original Article Series, Vol. 21, No. 1 (March of Dimes Birth Defects Foundation, New York, 1985).
12. Hastings RJ, Franks LM (1981): Chromosome pattern, growth in agar and tumorigenicity in nude mice of four human bladder carcinoma cell lines. Int J Cancer 27:15-21.
13. Perucca D, Szepetowski P, Simon M-P, Gandray P (1990): Mo- lecular genetics of human bladder cancer. Cancer Genet Cytoge- net 49:143-156.
14. Teyssiar JR (1987): Nonrandom chromosomal changes in human solid tumors: Application of an improved culture method. J Nat[ Cancer Inst 79:1189-1198.
15. Olumi AF, Tsai YC, Nichols PW, Skinner DG, Cain DR, Bender LI, Jones PA (1990): Allelic loss of chromosome 17p distinguishes high grade from low grade transitional cell carcinomas of the bladder. Cancer Res 50:7081-7083.
16. Tsai YC, Nichols PW, Hiti AL, Williams Z, Skinner DG, Jones PA (1990): Allelic losses of chromosomes 9, 11, and 17 in hu- man bladder cancer. Cancer Res 50:44-47.
17. Offner FA, Ott G, Povey S, Knuechel R, Preislar V, Fuezesi L, Klosterhalfen B, Ruebben H, Hof~taedter F, Kirkpatrick CJ (1991): Characterization of the new bladder cancer cell line HOK-I: Ex- pression of transitional, squamous and glandular differentia- tion patterns. Int J Cancer 49:122-126.
