Chromosome Abnormalities and Oncogenesis in Cat Leukemias Sara E. Gulino
ABSTRACT: Chromosome abnormalities are found in feline leukemia virus (FeLV)-infected tumor cells as well as in tumor cells free of the virus. Three cell lines derived from tumors in the domestic cat (Felis catus), two of thymic origin and one of multicentric lymphoma origin, were analyzed cytogenetically to determine whether the FeLV virus was associated with chromosomal abnormalities in these tumor cell lines. One thymic tumor and the multicentric lymphoma were FeLV infected. The other thymic tumor cell line was FeLV-free. The normal diploid number in the domestic cat is 36. All three cell lines had numerical chromosome abnormalities with modal numbers of 3Z 38 (pseudodiploid), and 39, respectively and had consistent structural chromosome abnormalities. Three markers in the virus-free cell line (S markers) were shared with one or the other of the virus-positive cell lines. The two FeLV-positive cell lines did not have S markers in common. The finding of chromosome abnormalities in bath the virus-infected and the virus-flee cell lines suggests that these abnormalities may be important in oncogenesis. The FeLV virus could not be considered
the only causative agent of the abnormalities observed.
INTRODUCTION Feline leukemia virus (FeLV) is a retrovirus transmitted, mainly horizontally, in a contagious manner through saliva and other excretions between outbred cats [1-5]. FeLV may have high infectious incidences, reaching 24, 30, and even 60% in multicat households, and causes many feline diseases, including leukemia and l y m p h o m a [1, 6-11]. There is some controversy with respect to the frequency and extent of FeLV virus involvement in causation of disease and death in cats [12], but 70 % of the leukemias and lymphomas occurring in outbred cats are associated with FeLV viremia [2, 13-17]. The other 30% do not have detectable virus or serologically detectable antigens [15, 18-19]. Chromosomal abnormalities are found in tumors infected by retroviruses without onc genes as well as in otherwise indistinguishable uninfected tumors [20-26]. Indeed, chromosome abnormalities are often detected genetic determinants of tumors [22-27]. Investigators have proposed that both virus-positive and virus-negative tumors may be initiated by chromosome abnormalities and that transformation can be a virus-independent event; i.e., it is not the presence of the retrovirus without an onc gene that causes cancer but
From *Department of Molecular Biology, Stanley Hall, University of California, Berkeley, Berkeley, CA 94720. Address reprint requests to Sara E. Gulino, M.S., at Tumor Cytogenetics, Department of Pathology, Harper Hospital, 3990 John R, Detroit, MI 48201. Received May 21, 1992; accepted July 22, 1992.
perhaps the processes initiated by the chromosome abnormalities [22, 28]. The virus infection may induce hyperplasia, which enhances the risk of chromosome aberration, although neither is sufficient for tumorigenesis [22, 28]. On the other hand, the retroviruses with transforming (onc) genes are still the only known genes capable of causing cancer with single-hit kinetics; i.e., every cell infected by the transforming (onc) retrovirus will be transformed [29-31]. These genes may arise by rare truncations and illegitimate recombinations between retroviral and cellular genes [29]. Chromosome rearrangements are also produced by illegitimate recombination [29, 32] and may be considered an alternate pathway. Chromosome abnormalities characterize most malignancies [32], with the exception of some tumors caused by retroviruses with onc genes [21, 26, 33-35]. The FeLV virus lacks the genetic information for transformation; i.e., lacks the onc gene and also has a long latent period before transformation occurs [2, 36-38]. In addition, some healthy FeLV-infected cats excrete high titers of FeLV virus. Furthermore, they usually have higher vires titers than leukemic animals [2, 39]. Therefore, we propose that FeLV virus may not have a direct role in tumorigenesis, but may indirectly influence chromosome abnormalities or virus integration sites that play a role in leukemogenesis, as suggested in a recent study [20]. MATERIALS AND METHODS Lymphoma Cell Lines Line 3272 was established from a FeLV-positive thymic lymphoma in a naturally infected cat (Felts catus). Line 3201 B
149 © 1992 Elsevier Science P u b l i s h i n g Co., Inc. 655 A v e n u e of the Americas, N e w York, NY 10010
Cancer Genet Cytogenet 6 4 : 1 4 9 - 1 5 7 (1992) 0165-4608/92/$05.00
150 was established after in vivo passage through a specific pathogen-free kitten of a naturally occurring FeLV-negative thymic lymphoma and was indistinguishable by allelic isoenzyme signature from the donor tumor [13]. Line FL 74 was established from a renal lymphoma (multicentric lymphoma) after in vivo passage of the original Kawakami-Theilen FeLV virus isolate [13]. All three cell lines were provided by Dr. William Hardy, Jr. of the Sloan-Kettering Institute for Cancer Research (New York, NY).
Cell Culture and Harvest The cell lines were cultured in T-cell m e d i u m made of a 1:1 mix of RPMI 1640 and Leibovitz's L-15, 20% newborn calf serum, 1% penicillin-neomycin-streptomycin cocktail, and 1 × L-glutamine (L-GIn) (200 mM, 100 x ). Stock cultures were split 1:10 every 4 days for 3201 B and FL 74 and once weekly for 3272 cell lines. Cultures for chromosome banding were set up from the stock cultures at high cell concentration of 1.6 x 106 cells/ml in 5 ml T-cell m e d i u m for the FL 74 cell line and 3201 B cell line. The cells were cultured for 4 days at 37°C. Two thirds of the culture m e d i u m was changed, and flesh m e d i u m , to a final volume of 5 ml, plus three drops (51 ~1) L-GIn (200 mM), was a d d e d to 3201 B at 5 hours before harvest, and to FL 74 at 14 to 16 hours before harvest. For 3272, concentrations of 7 x 106 cells/ml were cultured in either 7 or 10 ml of T-cell m e d i u m plus 85 ~1 of L-GIn (200 mM) for 7 days at 37°C. T-cell m e d i u m (4 ml) plus 51 ~1 L-GIn (200 mM) was a d d e d to the culture 78 hours before harvest, with another change of m e d i u m 24 hours before harvest. Because the doubling time of 3272 was greater than that of the other two cell lines, a longer incubation time was required to obtain the same cell turnover and mitotic index.
S.E. Gulino fixative as above for 10 minutes more at room temperature; they were then centrifuged again and r e s u s p e n d e d with enough fixative to yield a translucent suspension.
Banding and Karyotyping Slides were heat-dried at 45°C for 24 h before banding. The Giemsa-trypsin b a n d i n g technique was used as described previously [40]. Metaphases were karyotyped according to the standard of Wurster-Hill and Gray for cat chromosomes (Fig. 1) [41]. The numerical designations used for the markers were unique for each cell line. The chromosome abnormalities designated as S markers were shared by two or more cell lines. Fifty intact metaphases were counted for each line.
A
B
Testing for FeLV: Assure/FeLV Test All cell lines were checked for the presence or absence of the FeLV virus by the Assure/FeLV feline leukemia antigen test kit (according to manufacturer's instructions, Synbiotics Corporation, Kansas City, MO), w h i c h uses an antibody highly specific against FeLV virus antigen p27, one of the proteins c o d e d by the gag gene [1]. Blue development was indicative of the presence of the virus. Lack of color development indicated its absence. All three cell lines tested as expected: FL 74 and 3272 were positive and 3201 B was negative.
'i D
~
~
~i~i~ ~)~
Chromosome Harvest All cell lines were grown in s u s p e n s i o n and harvested without mitotic arrestants such as Colcemid to obtain longer chromosomes. Colcemid, used for 30 minutes before harvest, resulted in short chromosomes. The cells were centrifuged at 800 r p m for 5 minutes. The supernatant was discarded, and the cells were r e s u s p e n d e d in warm hypotonic (0.7% potassium chloride) and incubated in a 37°C water bath for 6 minutes for 3201 B and FL 74, and for 12 minutes for 3272. They were centrifuged again at 800 r p m for 5 minutes, r e s u s p e n d e d in a freshly m a d e fixative of methanol:acetic acid 5:2 and allowed to stand for 30 minutes at room temperature. They were again centrifuged and r e s u s p e n d e d in
I
2
5
4
XX
Figure 1 Normal karyotype of a female cat, Fells catus. The diploid number is 38. A G-banded karyotype shows 18 autosomal chromosomes and two sex chromosomes (picture courtesy of Dr. D. Wurster-Hill).
Chromosome Abnormalities and Oncogenesis in Cat Leukemias
151
A
B
Figure 2 FL 74 mitotic response to addition of extra L-Gln (200 mM). (A) Cells 24 hours after resuspension in 7 ml T-cell medium plus 51 ~1L-Gln(200 mM} with many cell cumuli. (B} Same culture 5 days later after a second 1:10split.
RESULTS
Cell Growth Cell lines were maintained in the T-cell medium developed by Dr. P. Donahue (Biomedical Research Institute, St. Paul, MN). FL 74 grew in suspension and formed cumuli. The round cells showed a "rough" surface with much cell budding. Cytoplasmic extrusion occurred during cell division, possibly resulting in budding. The 3272 line grew very slowly in suspension without tending to form aggregates. The ameboid or round cells showed a "semirough" surface with cell
budding. The 3201 B line also grew in suspension with most cells dispersed and few dispersable cumuli. The round or ameboid cells had smooth surfaces without budding. The ameboid morphology was most apparent after addition of extra L-Gln (200 raM), which acted as a mitotic agent. Treatment of all cell lines with phytohemagglutinin (PHA} did not increase the mitotic index, but addition of extra L-Gln stimulated mitosis (Fig. 2A and B}. 3201 B cells responded more quickly to the mitotic stimulus than did 3272 cells. The latter also required a greater concentration and longer ex-
152
S.E. Gulino
A
B
II,
It
I¢ i, is
C
Ill
D
I
F 1
2
3
4
XY
i i|.
marl. mar~. mar3. mar4. F i g u r e 3 Representative FL 74 karyotype showing A2, E2, and F 2 monosomies, E1 trisomy, loss of the E 3 and F 1 chromosome pairs, and + X chromosome. Five of the eight consistent markers are shown.
posure of extra L-Gln (200 mM) to obtain the same mitotic index as the other lines. Similarly, none of the lines showed mitotic response to Concanavilin A (ConA). Furthermore, a combination of PHA and ConA failed to increase mitotic index.
Cat Karyotype The normal d i p l o i d (2n) chromosome number in the domestic cat {Felis catus} is 38. The karyotype is c o m p o s e d of 18 autosome pairs grouped from A to F, based on chromosome size and centromere location, and w i t h i n each class is sequentially n u m b e r e d from 1 according to the system of the San Juan (1965} Conference [42]. The A group comprises
three large submetacentric pairs, the B group four large subtelocentric pairs, the C group two large metacentric pairs, the D group four m e d i u m small subtelocentric pairs, the E group three small metacentric pairs, and the F group two small acrocentric pairs. The X chromosome is medium-sized submetacentric, and the Y chromosome is acrocentric [43]. FL 74 Karyotype FL 74 had a m o d a l n u m b e r of 37, with eight consistent markers; i.e., markers appearing in two or more cells. Karyotype analyses showed h y p o d i p l o i d or p s e u d o d i p l o i d cells with monosomies, trisomies, and absence of paired chromosomes (Fig. 3). Groups A, B 1, B 3, D, E, F, and X, and Y chromo-
Chromosome A b n o r m a l i t i e s and Oncogenesis in Cat Leukemias
153
4
FL 74 FeLV(+)
"*
r
marl. mar2. mar3. mar4. marS. mar6.
3272 FeLV(+)
3201 B FeLV(-)
$3
I II
~ & ~ ~ marlb mar2~ mar3b mar4b
~
~
marl~ mar2~ mar3~ mar4~ mar5,
~ mar6~
mar7,
~
*,
marS,
mar9,
S1
$2
S1
$2
$3
Figure 4 C h r o m o s o m e markers in all t h r e e cell lines. All c o n s i s t e n t markers in all t h r e e cell lines are s h o w n . T h e m a r k e r n u m b e r s of t h e c h r o m o s o m e s d e s i g n a t e d m a r are n o t related to t h e n u m b e r s of t h e other cell lines.
somes showed at least one of the anomalies described above. D~ and D 4 monosomies and absence of both F~ were the most c o m m o n occurrence in the karyotypes analyzed. Markers la, 2a, 4a, derivative (der) (Dip), and deletion (del) (D2q) occurred with frequencies of 86, 92, 64, 98 and 66%, respectively. The other markers were found in less than 50% of cells. The only cell w i t h o u t der (D~;?)(p;?) was monosomic for D r In addition, 4% of cells had a double deletion for D 2 (data not shown). Marker $3 [S ("shared") indicates that the marker was found in at least one other of the cell lines analyzed] was also found in 3201 B and in 2% of FL 74 (Fig. 4 and Table 1). Partial identification of markers for FL 74 was possible for marla, mar2a, der (D1), and $3. Marker la was a possible translocation, ?t(E~;F~)(q;?). Marker 2a was interpreted as a possible Robertsonian-translocation, ?t(Fz;F2). Der (D1) was a t(D~;?)(q;?). Marker $3 was a t(A 2 p;F~). No cellular onc genes or i m m u n o g l o b u l i n (Ig) genes have yet been m a p p e d to translocation regions of the identified markers, del (D2q), or der (Dip).
The only identifiable markers for 3272 were $1 and del (B2q). Marker $1 was a t(B2;?)(q;? ). No cellular onc genes or
i m m u n o g l o b u l i n genes have yet been m a p p e d to the breakpoint regions of either marker.
3201 B Karyotype Karyotypes of 3201 B had a modal number of 39, with 13 consistent markers. Analyses showed h y p o d i p l o i d , p s e u d o d i p loid, and h y p e r d i p l o i d cells with monosomy, trisomy, and losses of p a i r e d chromosomes (Fig. 6). All six chromosome groups as well as the sex chromosomes showed at least one of the above-described anomalies. A 1, B2, B 3, D1, El, E 3, and F~ monosomies prevailed, with A1 m o n o s o m y in 100% of cells. Markers 6c, lc, $3, der (Alp), and del (B2o3 were noted in 76, 100, 57, 85, and 64% of cells, respectively. The other markers were found in less than 50% of the cells. W h e n der (A~p) was not found, the cell was either monosomic for A1 or the A~ pair was missing. In addition, 6% of cells had a double deletion for B 2, and 61% of cells were monosomic f o r B 2.
3272 Karyotype The karyotype of 3272 had a m o d a l n u m b e r of 38, with six consistent markers. Analyses showed h y p o d i p l o i d , pseud o d i p l o i d , and h y p e r d i p l o i d cells with monosomy, trisomy, and loss of chromosome pairs (Fig. 5). All six chromosome groups as well as the sex chromosomes showed at least one of the above described anomalies. A 2, A 3, B~, D 2, E 1, E 3, and F 2 monosomies were the prevailing anomalies, with F 1 m o n o s o m y in 80% of cells. Markers lb and $2 occurred w i t h a frequency of 55 and 71%, respectively. The other markers were found in less than 50% of cells. Markers $1 and $2 were shared with 3201 B. The marker frequencies of $1 and $2 in 3272 were 26 and 71%, respectively (Table 1).
Table 1.
Occurrence of S markers and cell origin
Parameter 3272 FeLV-positive FL 74 FeLV-positive 3201 B FeLV-free Relationship
S1 (%)
$2 (%)
26
71
2 13:1
49 1.5:1
$3 (%) 2 57 29:1
Cell origin multicentric thymic thymic
The S ("shared"} marker's number identifies the same chromosome marker in each cell line; 3201 B had the most S markers. The relative relation shows the significance of association for each S marker between cell lines. Percentages were determined in number of cells analyzed, which was different from number of cells counted (> 50 cells).
154
S.E. Gulino
+I'
A
l ++ A der(AO
|
B
iiill
B
II
i,, del(B~)
C
o
¸¸111
C
1!¢
I +ti !
E
++
Q
E
ml
F 1
I+
2
+ marl+ mar4b
D
det(b@)
3
++
4
1
XY
2
3
4
XY
+ ! S1
marl+ mar2~ mar3+ marS+ mar6+
S2
Figure 5
3272 karyotype showing A 2, B1, B4, D2, El, E3, and F1 monosomies and C2, D~ and D 3 trisomy. Five of the six consistent markers are shown.
The only identifiable markers for 3201 B were marlc, $3, der (A1), and del (B2). Marker lc was a t(D~;?)(q;?), marker $3 was a t(A2;F~)(p;?), and der (A1) was a t(A1;?B2)(p;?). Marker B2 had a del (B2q). No cellular onc genes or Ig genes have yet been m a p p e d to the translocation or breakpoint regions of any of the identified markers (44]. However, the distal half of the q arm of the B2 chromosome houses the major histocompatibility complex (MHC) class I and class II antigens [44]. Morphologic observation suggests that the two MHC genes may have been deleted in all cases of del (Bzq), but the B2q deletion may have been terminal or interstitial. Furthermore, the exact band location of the MHC on the q arm is not known (45].
S Markers The two FeLV-positive cell lines d i d not share a c o m m o n marker (Fig. 4). Thus, the greatest sharing of chromosomal
$3
$2
Figure 6
3201 B karyotype. This karyotype shows B3, D1, E3, Ft, and F 2 monosomies. Nine of the 13 consistent markers are also observed in this cell, including der (At) and del (B2). The X chromosome is missing.
abnormalities was noted between the virus-free cell line and one of the virus-positive cell lines, 3272. Each cell line had originated independently: the virus-free cell line arose from a thymic l y m p h o m a , whereas one of the FeLV-positive cell lines arose from a multicentric l y m p h o m a (Table 1).
Comparative Karyotypic Analysis All three cell lines showed some monosomies, trisomies, and losses of chromosome pairs. The FL 74 was the only line with some consistent normal pairs of chromosomes; i.e., chromosomes B 2 and B 4. E 3 m o n o s o m y occurred with about 50% frequency in all three cell lines. D 1 m o n o s o m y was greater in FL 74 than in 3201 B: 83 and 69% respectively. F1 monosomy was greater in 3272 (80%) than in FL 74 (16%) and 3201 B (46%). The Ft pair was missing in 83% of cells of FL 74 and missing in only 8% of 3201 B. Even though all chromosome pairs have been n u m e r i c a l l y altered in all three cell
Chromosome Abnormalities and Oncogenesis in Cat Leukemias lines (with the exception of the FL 74 FeLV-positive cell line), the only chromosome pairs affected in both virus-positive cell lines were E3 and F1, with losses less than or equal to 50%. DISCUSSION
We tested whether chromosome abnormalities or virus integration sites could play a role in causation of feline leukemias and whether the FeLV virus may have a direct role in tumorigenesis. Each cell line had a different cell of origin, although all were T lymphocytes. Although each tumor was chromosomally distinct and kals~typically suggestive of origin as a single clone, some common markers were identified. If the FeLV virus were the cause of tumors and chromosomal abnormalities in virus-positive tumors, the lines might be expected to share markers. Furthermore, virus-free cells would not be expected to share chromosome aberrations with the two virus-positive cell lines. The opposite was true. The two FeLV-positive cell lines did not share markers in common, and their karyotypes differed considerably. The FeLV-free cell line had the highest frequency of rearranged chromosomes and the most uniformly clonal patterns of chromosome abnormality. This appears to indicate a lack of chromosomal specificity in relation to viral infection. In contrast, 3272 FeLV-positive cells and 3201 B FeLV-free cells showed the greatest association of chromosome abnormality. Perhaps neither the virus itself nor a viral product was important in tumorigenesis and chromosomal abnormalities, but rather the proviral integration site. If this were true, an integration site within the same locus should cause the same phenotypic tumor, and the chromosome abnormalities, as a consequence of this event, should be common to tumors of the same phenotype. If the virus were integrated and suppressed in the 3201 B FeLV-freecell line, as has been claimed for virus-negative tumors [15], it could have been expected to generate a marker common to FL 74 FeLV, the other thymic tumor, which did not occur. Lacking unique and clonal chromosome abnormalities, either viral specific or tissue specific, the aberrations found may represent late events in tumor development, showing progression in the tumor rather than initiation. A study of 21 naturally occurring feline lymphomas [36] showed only four lymphomas (19%) with FeLV virus integrated in a unique 2.4-kilobase locus. The other tumor integration sites did not map to any of the commonly known integration loci for other proviruses. Seven of the tumors analyzed were in the same phenotypic subgroup and included the four integration sites in the unique locus. Thus, the integration site hypothesis may explain some but not all cases in terms of the malignant potential of retroviruses with no onc genes. The only transforming autonomous genes are the retroviruses with transforming onc genes. They are capable of transforming a permissive host cell with 1:1 kinetics; i.e., each cell infected becomes transformed. Therefore, tumors arising from transforming retroviruses are polyclonal because more than one cell is infected by the retrovirus and its progeny [22]. Usually they do not show chromosome abnormalities, although they may do so later in tumor progression [46]. On the other hand, clonal chromosome abnormalities are commonly recognized in tumors that are not caused by
155
retroviruses with transforming onc genes [21, 27]. This characteristic is also shared with tumors of the same lineage that have not been infected with virus [22]. However, not all tumors show the high degree of specificity of chromosome abnormality demonstrated by the Ph chromosome of chronic myalogenous leukemia (CML). Even in CML, there is not complete uniformity among tumors. Some cases are variant translocations, and a similar Ph chromosome is found in acute lymphoblastic leukemia and in acute myeloblastic leukemia [47]. Polyclonalities of primary tumors may be common in certain tumor types [48-50]. Thus, even in nonviral cases, there are degrees of variability in clonality and chromosome specificity. Based on available genetic mapping of the domestic cat genome, the only chromosome abnormality identified in this study with potential bearing on carcinogenesis was the del{B2q}. The feline leukocyte antigen-1 {MHC class I} and feline leukocyte antigen-2 (MHC, class II) are located on the distal half of the B2q arm [44]. Based on morphology alone, the two MHC genes might have been deleted in all cases of dal(B2q I, but molecular analysis is necessary to verify this possible deletion. Studies of human malignant cancers of the breast, colon, urinary bladder, kidneys, and Lewis lung carcinoma for the expression of human leukocyte antigen (HLA) class I have shown that most primary lesions are HLAimmunoreactive cells whereas most metastatic lesions are HLA-nonreactive [51-52]. Tumors had a distribution of HLAreactive and HLA-nonreactive cells compatible with the suggestion that HLA-nonreactive cells have a selective advantage for growth and metastatic invasion [51]. This thesis is also supported by transfection experiments of H-2Kb histocompatible complex antigen in C57BL/6 genetic mice that showed abrogation of metastasis in H-2K and H-2D compatible mice by inducing alloreactive cytotoxic lymphocytes that kill H-2K-positive cells [52]. Because both virus-infected and virus-free cell lines have chromosomal abnormalities, we can conclude that the FeLV virus may cause chromosome abnormalities but that the aberrations are neither unique nor specific. These abnormalities may nevertheless be important in oncogenesis. Alternatively, as-yet-unknown factors may be causative of both chromosome abnormalities and cancer. If so, chromosome abnormalities may not be necessary for tumor development and progression, as suggested by the normal karyotype found in many human tumors, including breast cancers and meningiomas [46], and murine tumors. Human T-cell leukocyte virus-I (HTLV-I}has been associated with adult T-cell leukemia, but HTLV-I, like the FeLV virus, has no oncogene and the level of viral RNA expression is generally insignificant in infected cells [33]. Its integration site is also random. Another molecular event appears to be necessary for overt ATL to occur. The FeLV-associated tumors in feline may be a good animal model to study the pathophysiology and development of HTLV-I leukemias. This work was supported by Outstanding InvestigatorGrant No. 5R35-CA39915-06from the National Cancer Institute awarded to Peter Duesberg, Ph.D., of the University of California, Berkeley, CA, The author thanks Drs. Peter Duesberg, Sandra Wolman of Wayne State Medical School, Detroit, MI, and Doris Wurster-Hill of Dart-
156
mouth Medical School, Hanover, NH, for constructive criticism of the manuscript, as well as general influence and support. The author also thanks Drs. Kirsten Walen of State Public Health Service, Berkeley, CA, for technical advice and William Hardy, Jr.,of the Sloan-KetteringInstitutefor Cancer Research, N e w York, NY, forthe cell lines.
S.E. G u l i n o
19.
20. REFERENCES
1. Lutz H (1990): Feline retrovirnses: A brief review. Vet Micrebiol 23:131-146. 2. Essex ME (1982): Feline leukemia: A naturally occurring cancer of infectious origin. Epidemiol Rev 4:189-203. 3. Hardy WD Jr, McClelland AJ, DPhil E, MacEwen G, Hess PW, Hayes AA, Zuckerman EE (1977): The epidemiology of the feline leukemia virus (FeLV}. Cancer 39:1850-1855. 4. Hardy WD, Old LJ, Hess PW, Essex M, Cotter S [1973): Horizontal transmission of feline leukaemia virus. Nature 244:266-269. 5. Jarrett WFH, Jarrett O, Mackey L, Laird W, Hardy WD, Essex M [1973): Horizontal transmission of leukemia virus and leukemia in the cat. J Natl Cancer Inst 51;833-841. 6. RomatowskiJ(1989): Interpreting feline leukemia test results. Am Vet Med Assoc 195:928-930. 7. Donahue PR, Hoover EA, Beltz GA, Riedel N, Hirsch VM, Overbaugh J, Mullins JI (1988): Strong sequence conservation among horizontally transmissible, minimally pathogenic feline leukemia viruses. J Virel 62:722-731. 8. Reinacher M, Theilen G (1987): Frequency and significance of feline leukemia virus infection in necropsied cats. Am J Vet Res 48:939-945. 9. Jarrett O (1984): Pathogenesis of feline leukemia virus-related diseases. In: Mechanisms of Viral Leukaemogenesis, JM Goldman, O Jarrett, eds. Churchill-Livingstone, Edinburgh, pp. 135-154. 10. Lutz H, Pedersen NC, Theilen GH (1983): Course of feline leukemia virus infection and its detection by enzyme-linked immunosorbent assay and monoclonal antibodies. Am J Vet Res 44:2054-2059. 11. Dorn CR, Taylor DON, Schneider R, Hibbard HH, Klanber MR (1968): Survey of animal neoplasms in Alameda and Centre Costa Counties, California. II. Cancer morbidity in dogs and cats from Alameda County. JNCI 40:307-318. 12. Reinacher M (1989): Diseases associated with spontaneous feline leukemia virus (FeLV) infection in cats. Vet Immunol Immunopathol 21:85-95. 13. Rojko JL, Kociba GJ, Abkowitz JL, Hamilton KL, Hardy WD Jr, Ihle JN, O'Brien SJ (1989): Feline lymphomas: Immunological and cytochemical characterization. Cancer Res 49:345-351. 14. Hardy WD Jr (1980): Feline leukemia virus diseases. In: Feline Leukemia Virus, WD Hardy Jr, M Essex, AJ McClelland, eds. Elseviar/North Holland Press, New York, pp. 3-32. 15. Koshy R, Gallo C, Wong-Staal F (1980): Characterization of the endogenous feline leukemia virus-related DNA sequences in cats and attempts to identify exogenous viral sequences in tissues of virus-negative leukemic animals. Virology 103:434-445. 16. Hardy WD Jr, McClelland AJ, Zuckerman EE, Snydar HW Jr, MacEwen EG, Francis D, Essex M (1980): Development of virus nonproducer lymphosarcomas in pet cats exposed to FeLV. Nature 288:90-92. 17. Francis DP, Cotter SM, Hardy WD Jr, Essex M (1979): Comparison of virns-positive and virus-negative cases of feline leukemia and lymphoma. Cancer Res 39:3866-3870. 18. Essex M, Grant CK, Cotter SM, Sliski AH, Hardy WD (1979): Leukemia specific antigens: FOCMA and immunosurveillance. In: Modern Trends in Human Leukemia, vol. III, R Neth, RC
21.
22. 23.
24.
25. 26.
27. 28. 29.
30. 31. 32.
33.
34.
35. 36.
37.
38.
39.
40. 41.
Gallo, PH Hofschneider, K Mannweiler, eds. Springer-Verlag, New York, pp. 453-467. Hardy WD Jr, Geering G, Old LG, De Harven E, Brody RS, McDonough S (1969): Feline leukemia virus; Occurrence of viral antigen in the tissues of cats with lymphosarcoma and other diseases. Science 166:1019-1021. Grindem CB, Buoen LC (1989): Cytogenetic analysis in nine leukaemic cats. J Comp Pathol 101:21-30. Duesbarg PH (1987): Cancer genes: Rare recombinants instead of activated oncogenes Ca review). Proc Natl Acad Sci USA 84:2117-2124. Duasberg PH (1987): Retrovirnses as carcinogens and pathogens: Expectations and reality. Cancer Res 47;1199-1220. Rowley JD (1984): Consistent chromosomal aberrations and oncogenas in human t u m o u r s - a n overview. Cancer Surv 3:355-357. Trent JM (1984): Chromosomal alterations in human solid tumours: Implications of the stem model to cancer cytogenetics. Cancer Surv 3:395-422. Levan A (1956): Chromosomes in cancer tissue. Ann NY Acad Sci 63:774-792. Sandberg AA (1990): The Chromosomes in Human Cancer and Leukemia, 2nd Ed. Elsevier Science Publishing Co., New York, pp. 753-760. Wolman SR (1983): Karyotypic progression in human tumors. Cancer Metastasis Rev 2:257-293. Gough N (1985): Chromosomal translocations and the c-myc gene: Paradigm lost? Trends Genet 1:63-64. Duasberg PH (1987): Cancer genes generated by rare chromosomal rearrangements rather than activation of oncogenes. Haematol Blood Transfusion 31:496-510. Duasberg PH (1985): Activated proto-onc genes: Sufficient or necessary for cancer? Science 228:669-677. Duesherg PH (1983): Retroviral transforming genes in normal cells? Nature 304:219-226. Zhou R-P, Duesberg PH (1988): Myc protooncogene linked to retroviral promoter, but not to enhancer, transforms embryo cells. Proc Natl Acad Sci USA 85:2924-2928. Sadamori N (1991): Cytogenetic implication in adult T-cell leukemia- a hypothesis of leukemogenesis. Cancer Genet Cytogenet 51:131-136. Weiss R, Teich N, Varmus H, Coffin J (1985): RNA Tumor Viruses, 2nd Ed., Cold Spring Harbor Press, Cold Spring Harbor, NY. Weiss R, Teich N, Varmus H, Coffin J (1982): RNA Tumor Viruses. Cold Spring Harbor Press, Cold Spring Harbor, NY. Levesque KS, Bonham L, Levy LS (1990): flvi-1, A common integration domain of feline leukemia virus in naturally occurring lymphomas of a particular type. J Virol 64:3455-3462. Miura T, Tsujimoto H, Fukasawa M, Kodama T, Shibuya M, Hasegawa A, Hayami M (1987): Structural abnormality and over-expression of the myc gene in feline leukemias. Int J Cancer 40:564-569. Neel BG, Hayward WS, Robinson HL, Fang J, Astrin S (1981): Avian leukosis virus-induced tumors have common proviral integration sites and synthesize discrete new RNAs: Oncogenesis by promoter insertion. Cell 23:323-334. Francis DP, Essex M, Cotter SM, Jakowski RM, Hardy WD Jr. (1979b): Feline leukemia virus infections: The significance of chronic viremia. Leukemia Res 3:435-441. Seabright M [1971): A rapid banding technique for human chromosomes. Lancet 2:971-972. Wurstar-Hill DH, Gray CW (1973): Giemsa banding patterns in the chromosomes of twelve species of cats (Felidoe). Cytogenet Cell Genet 12:377-397.
C h r o m o s o m e Abnormalities and Oncogenesis in Cat Leukemias
42. Jones TC (1965): San Juan conference on karyotypes of Felldoe. Special report. Mammal Chromosomes Newslett 15:121122. 43. Hare WCD, Weber WT, McFeely RA, Yang Tsu-Ju (1966): Cytogenetics in the dog and cat. J Small Anim Pract 7:575-592. 44. O'BrienSJ, ed. (1990): Genetic Maps-Locus Maps of Complex Genomes, 5th Ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 45. Yuhki NY, O'Brien SJ (1988): Molecular characterization and genetic mapping of class I and class 1I MHC genes of the domestic cat. Immunogenetics 27:414-425. 46. Wolman SR (1986): Cytogenetic heterogeneity: Its role in tumor evolution. Cancer Genet Cytogenet 19:129-140. 47. Larson RA, Golomb HM, Rowley JD (1981): Chromosome changes in hematologic malignancies. CA-A Cancer J Clinicians 31:222-238.
157
48. Wolman SR (1991): Clonality. In: Biochemical and Molecular aspects of Selected Cancers, vol. 1, TG Pretlow II, T Pretlow, eds. Harcourt Brace Jovanovich, San Diego, pp. 409-410. 49. Heim S, Mandahl N, Mitelman F (1988): Genetic convergence and divergence in tumor progression. Cancer Res 48:5911-5916. 50. Aldaz CM, Conti CJ, Yuspa SH, Slaga TJ (1986): Cytogenetic profile of mouse skin tumors induced by the viral Harvey-ras gone. Carcinogenesis 9:1503-1505. 51. Cordon-Cardo C, Fuks Z, Drobnjak M, Moreno C, Eisenbach L, Feldman M (1991): Expression of HLA-A,B,C antigens on primary and metastatic tumor cell populations of human carcinomas. Cancer Res 51:6372-6380. 52. Gelber C, Plaksin D, Vadai E, Feldman M, Eisenbach L (1989): Abolishment of metastasis formation by murine tumor cells transfected with "foreign" H-2K genes. Cancer Res 49:23662373.
ABSTRACT: Chromosome abnormalities are found in feline leukemia virus (FeLV)-infected tumor cells as well as in tumor cells free of the virus. Three cell lines derived from tumors in the domestic cat (Felis catus), two of thymic origin and one of multicentric lymphoma origin, were analyzed cytogenetically to determine whether the FeLV virus was associated with chromosomal abnormalities in these tumor cell lines. One thymic tumor and the multicentric lymphoma were FeLV infected. The other thymic tumor cell line was FeLV-free. The normal diploid number in the domestic cat is 36. All three cell lines had numerical chromosome abnormalities with modal numbers of 3Z 38 (pseudodiploid), and 39, respectively and had consistent structural chromosome abnormalities. Three markers in the virus-free cell line (S markers) were shared with one or the other of the virus-positive cell lines. The two FeLV-positive cell lines did not have S markers in common. The finding of chromosome abnormalities in bath the virus-infected and the virus-flee cell lines suggests that these abnormalities may be important in oncogenesis. The FeLV virus could not be considered
the only causative agent of the abnormalities observed.
INTRODUCTION Feline leukemia virus (FeLV) is a retrovirus transmitted, mainly horizontally, in a contagious manner through saliva and other excretions between outbred cats [1-5]. FeLV may have high infectious incidences, reaching 24, 30, and even 60% in multicat households, and causes many feline diseases, including leukemia and l y m p h o m a [1, 6-11]. There is some controversy with respect to the frequency and extent of FeLV virus involvement in causation of disease and death in cats [12], but 70 % of the leukemias and lymphomas occurring in outbred cats are associated with FeLV viremia [2, 13-17]. The other 30% do not have detectable virus or serologically detectable antigens [15, 18-19]. Chromosomal abnormalities are found in tumors infected by retroviruses without onc genes as well as in otherwise indistinguishable uninfected tumors [20-26]. Indeed, chromosome abnormalities are often detected genetic determinants of tumors [22-27]. Investigators have proposed that both virus-positive and virus-negative tumors may be initiated by chromosome abnormalities and that transformation can be a virus-independent event; i.e., it is not the presence of the retrovirus without an onc gene that causes cancer but
From *Department of Molecular Biology, Stanley Hall, University of California, Berkeley, Berkeley, CA 94720. Address reprint requests to Sara E. Gulino, M.S., at Tumor Cytogenetics, Department of Pathology, Harper Hospital, 3990 John R, Detroit, MI 48201. Received May 21, 1992; accepted July 22, 1992.
perhaps the processes initiated by the chromosome abnormalities [22, 28]. The virus infection may induce hyperplasia, which enhances the risk of chromosome aberration, although neither is sufficient for tumorigenesis [22, 28]. On the other hand, the retroviruses with transforming (onc) genes are still the only known genes capable of causing cancer with single-hit kinetics; i.e., every cell infected by the transforming (onc) retrovirus will be transformed [29-31]. These genes may arise by rare truncations and illegitimate recombinations between retroviral and cellular genes [29]. Chromosome rearrangements are also produced by illegitimate recombination [29, 32] and may be considered an alternate pathway. Chromosome abnormalities characterize most malignancies [32], with the exception of some tumors caused by retroviruses with onc genes [21, 26, 33-35]. The FeLV virus lacks the genetic information for transformation; i.e., lacks the onc gene and also has a long latent period before transformation occurs [2, 36-38]. In addition, some healthy FeLV-infected cats excrete high titers of FeLV virus. Furthermore, they usually have higher vires titers than leukemic animals [2, 39]. Therefore, we propose that FeLV virus may not have a direct role in tumorigenesis, but may indirectly influence chromosome abnormalities or virus integration sites that play a role in leukemogenesis, as suggested in a recent study [20]. MATERIALS AND METHODS Lymphoma Cell Lines Line 3272 was established from a FeLV-positive thymic lymphoma in a naturally infected cat (Felts catus). Line 3201 B
149 © 1992 Elsevier Science P u b l i s h i n g Co., Inc. 655 A v e n u e of the Americas, N e w York, NY 10010
Cancer Genet Cytogenet 6 4 : 1 4 9 - 1 5 7 (1992) 0165-4608/92/$05.00
150 was established after in vivo passage through a specific pathogen-free kitten of a naturally occurring FeLV-negative thymic lymphoma and was indistinguishable by allelic isoenzyme signature from the donor tumor [13]. Line FL 74 was established from a renal lymphoma (multicentric lymphoma) after in vivo passage of the original Kawakami-Theilen FeLV virus isolate [13]. All three cell lines were provided by Dr. William Hardy, Jr. of the Sloan-Kettering Institute for Cancer Research (New York, NY).
Cell Culture and Harvest The cell lines were cultured in T-cell m e d i u m made of a 1:1 mix of RPMI 1640 and Leibovitz's L-15, 20% newborn calf serum, 1% penicillin-neomycin-streptomycin cocktail, and 1 × L-glutamine (L-GIn) (200 mM, 100 x ). Stock cultures were split 1:10 every 4 days for 3201 B and FL 74 and once weekly for 3272 cell lines. Cultures for chromosome banding were set up from the stock cultures at high cell concentration of 1.6 x 106 cells/ml in 5 ml T-cell m e d i u m for the FL 74 cell line and 3201 B cell line. The cells were cultured for 4 days at 37°C. Two thirds of the culture m e d i u m was changed, and flesh m e d i u m , to a final volume of 5 ml, plus three drops (51 ~1) L-GIn (200 mM), was a d d e d to 3201 B at 5 hours before harvest, and to FL 74 at 14 to 16 hours before harvest. For 3272, concentrations of 7 x 106 cells/ml were cultured in either 7 or 10 ml of T-cell m e d i u m plus 85 ~1 of L-GIn (200 mM) for 7 days at 37°C. T-cell m e d i u m (4 ml) plus 51 ~1 L-GIn (200 mM) was a d d e d to the culture 78 hours before harvest, with another change of m e d i u m 24 hours before harvest. Because the doubling time of 3272 was greater than that of the other two cell lines, a longer incubation time was required to obtain the same cell turnover and mitotic index.
S.E. Gulino fixative as above for 10 minutes more at room temperature; they were then centrifuged again and r e s u s p e n d e d with enough fixative to yield a translucent suspension.
Banding and Karyotyping Slides were heat-dried at 45°C for 24 h before banding. The Giemsa-trypsin b a n d i n g technique was used as described previously [40]. Metaphases were karyotyped according to the standard of Wurster-Hill and Gray for cat chromosomes (Fig. 1) [41]. The numerical designations used for the markers were unique for each cell line. The chromosome abnormalities designated as S markers were shared by two or more cell lines. Fifty intact metaphases were counted for each line.
A
B
Testing for FeLV: Assure/FeLV Test All cell lines were checked for the presence or absence of the FeLV virus by the Assure/FeLV feline leukemia antigen test kit (according to manufacturer's instructions, Synbiotics Corporation, Kansas City, MO), w h i c h uses an antibody highly specific against FeLV virus antigen p27, one of the proteins c o d e d by the gag gene [1]. Blue development was indicative of the presence of the virus. Lack of color development indicated its absence. All three cell lines tested as expected: FL 74 and 3272 were positive and 3201 B was negative.
'i D
~
~
~i~i~ ~)~
Chromosome Harvest All cell lines were grown in s u s p e n s i o n and harvested without mitotic arrestants such as Colcemid to obtain longer chromosomes. Colcemid, used for 30 minutes before harvest, resulted in short chromosomes. The cells were centrifuged at 800 r p m for 5 minutes. The supernatant was discarded, and the cells were r e s u s p e n d e d in warm hypotonic (0.7% potassium chloride) and incubated in a 37°C water bath for 6 minutes for 3201 B and FL 74, and for 12 minutes for 3272. They were centrifuged again at 800 r p m for 5 minutes, r e s u s p e n d e d in a freshly m a d e fixative of methanol:acetic acid 5:2 and allowed to stand for 30 minutes at room temperature. They were again centrifuged and r e s u s p e n d e d in
I
2
5
4
XX
Figure 1 Normal karyotype of a female cat, Fells catus. The diploid number is 38. A G-banded karyotype shows 18 autosomal chromosomes and two sex chromosomes (picture courtesy of Dr. D. Wurster-Hill).
Chromosome Abnormalities and Oncogenesis in Cat Leukemias
151
A
B
Figure 2 FL 74 mitotic response to addition of extra L-Gln (200 mM). (A) Cells 24 hours after resuspension in 7 ml T-cell medium plus 51 ~1L-Gln(200 mM} with many cell cumuli. (B} Same culture 5 days later after a second 1:10split.
RESULTS
Cell Growth Cell lines were maintained in the T-cell medium developed by Dr. P. Donahue (Biomedical Research Institute, St. Paul, MN). FL 74 grew in suspension and formed cumuli. The round cells showed a "rough" surface with much cell budding. Cytoplasmic extrusion occurred during cell division, possibly resulting in budding. The 3272 line grew very slowly in suspension without tending to form aggregates. The ameboid or round cells showed a "semirough" surface with cell
budding. The 3201 B line also grew in suspension with most cells dispersed and few dispersable cumuli. The round or ameboid cells had smooth surfaces without budding. The ameboid morphology was most apparent after addition of extra L-Gln (200 raM), which acted as a mitotic agent. Treatment of all cell lines with phytohemagglutinin (PHA} did not increase the mitotic index, but addition of extra L-Gln stimulated mitosis (Fig. 2A and B}. 3201 B cells responded more quickly to the mitotic stimulus than did 3272 cells. The latter also required a greater concentration and longer ex-
152
S.E. Gulino
A
B
II,
It
I¢ i, is
C
Ill
D
I
F 1
2
3
4
XY
i i|.
marl. mar~. mar3. mar4. F i g u r e 3 Representative FL 74 karyotype showing A2, E2, and F 2 monosomies, E1 trisomy, loss of the E 3 and F 1 chromosome pairs, and + X chromosome. Five of the eight consistent markers are shown.
posure of extra L-Gln (200 mM) to obtain the same mitotic index as the other lines. Similarly, none of the lines showed mitotic response to Concanavilin A (ConA). Furthermore, a combination of PHA and ConA failed to increase mitotic index.
Cat Karyotype The normal d i p l o i d (2n) chromosome number in the domestic cat {Felis catus} is 38. The karyotype is c o m p o s e d of 18 autosome pairs grouped from A to F, based on chromosome size and centromere location, and w i t h i n each class is sequentially n u m b e r e d from 1 according to the system of the San Juan (1965} Conference [42]. The A group comprises
three large submetacentric pairs, the B group four large subtelocentric pairs, the C group two large metacentric pairs, the D group four m e d i u m small subtelocentric pairs, the E group three small metacentric pairs, and the F group two small acrocentric pairs. The X chromosome is medium-sized submetacentric, and the Y chromosome is acrocentric [43]. FL 74 Karyotype FL 74 had a m o d a l n u m b e r of 37, with eight consistent markers; i.e., markers appearing in two or more cells. Karyotype analyses showed h y p o d i p l o i d or p s e u d o d i p l o i d cells with monosomies, trisomies, and absence of paired chromosomes (Fig. 3). Groups A, B 1, B 3, D, E, F, and X, and Y chromo-
Chromosome A b n o r m a l i t i e s and Oncogenesis in Cat Leukemias
153
4
FL 74 FeLV(+)
"*
r
marl. mar2. mar3. mar4. marS. mar6.
3272 FeLV(+)
3201 B FeLV(-)
$3
I II
~ & ~ ~ marlb mar2~ mar3b mar4b
~
~
marl~ mar2~ mar3~ mar4~ mar5,
~ mar6~
mar7,
~
*,
marS,
mar9,
S1
$2
S1
$2
$3
Figure 4 C h r o m o s o m e markers in all t h r e e cell lines. All c o n s i s t e n t markers in all t h r e e cell lines are s h o w n . T h e m a r k e r n u m b e r s of t h e c h r o m o s o m e s d e s i g n a t e d m a r are n o t related to t h e n u m b e r s of t h e other cell lines.
somes showed at least one of the anomalies described above. D~ and D 4 monosomies and absence of both F~ were the most c o m m o n occurrence in the karyotypes analyzed. Markers la, 2a, 4a, derivative (der) (Dip), and deletion (del) (D2q) occurred with frequencies of 86, 92, 64, 98 and 66%, respectively. The other markers were found in less than 50% of cells. The only cell w i t h o u t der (D~;?)(p;?) was monosomic for D r In addition, 4% of cells had a double deletion for D 2 (data not shown). Marker $3 [S ("shared") indicates that the marker was found in at least one other of the cell lines analyzed] was also found in 3201 B and in 2% of FL 74 (Fig. 4 and Table 1). Partial identification of markers for FL 74 was possible for marla, mar2a, der (D1), and $3. Marker la was a possible translocation, ?t(E~;F~)(q;?). Marker 2a was interpreted as a possible Robertsonian-translocation, ?t(Fz;F2). Der (D1) was a t(D~;?)(q;?). Marker $3 was a t(A 2 p;F~). No cellular onc genes or i m m u n o g l o b u l i n (Ig) genes have yet been m a p p e d to translocation regions of the identified markers, del (D2q), or der (Dip).
The only identifiable markers for 3272 were $1 and del (B2q). Marker $1 was a t(B2;?)(q;? ). No cellular onc genes or
i m m u n o g l o b u l i n genes have yet been m a p p e d to the breakpoint regions of either marker.
3201 B Karyotype Karyotypes of 3201 B had a modal number of 39, with 13 consistent markers. Analyses showed h y p o d i p l o i d , p s e u d o d i p loid, and h y p e r d i p l o i d cells with monosomy, trisomy, and losses of p a i r e d chromosomes (Fig. 6). All six chromosome groups as well as the sex chromosomes showed at least one of the above-described anomalies. A 1, B2, B 3, D1, El, E 3, and F~ monosomies prevailed, with A1 m o n o s o m y in 100% of cells. Markers 6c, lc, $3, der (Alp), and del (B2o3 were noted in 76, 100, 57, 85, and 64% of cells, respectively. The other markers were found in less than 50% of the cells. W h e n der (A~p) was not found, the cell was either monosomic for A1 or the A~ pair was missing. In addition, 6% of cells had a double deletion for B 2, and 61% of cells were monosomic f o r B 2.
3272 Karyotype The karyotype of 3272 had a m o d a l n u m b e r of 38, with six consistent markers. Analyses showed h y p o d i p l o i d , pseud o d i p l o i d , and h y p e r d i p l o i d cells with monosomy, trisomy, and loss of chromosome pairs (Fig. 5). All six chromosome groups as well as the sex chromosomes showed at least one of the above described anomalies. A 2, A 3, B~, D 2, E 1, E 3, and F 2 monosomies were the prevailing anomalies, with F 1 m o n o s o m y in 80% of cells. Markers lb and $2 occurred w i t h a frequency of 55 and 71%, respectively. The other markers were found in less than 50% of cells. Markers $1 and $2 were shared with 3201 B. The marker frequencies of $1 and $2 in 3272 were 26 and 71%, respectively (Table 1).
Table 1.
Occurrence of S markers and cell origin
Parameter 3272 FeLV-positive FL 74 FeLV-positive 3201 B FeLV-free Relationship
S1 (%)
$2 (%)
26
71
2 13:1
49 1.5:1
$3 (%) 2 57 29:1
Cell origin multicentric thymic thymic
The S ("shared"} marker's number identifies the same chromosome marker in each cell line; 3201 B had the most S markers. The relative relation shows the significance of association for each S marker between cell lines. Percentages were determined in number of cells analyzed, which was different from number of cells counted (> 50 cells).
154
S.E. Gulino
+I'
A
l ++ A der(AO
|
B
iiill
B
II
i,, del(B~)
C
o
¸¸111
C
1!¢
I +ti !
E
++
Q
E
ml
F 1
I+
2
+ marl+ mar4b
D
det(b@)
3
++
4
1
XY
2
3
4
XY
+ ! S1
marl+ mar2~ mar3+ marS+ mar6+
S2
Figure 5
3272 karyotype showing A 2, B1, B4, D2, El, E3, and F1 monosomies and C2, D~ and D 3 trisomy. Five of the six consistent markers are shown.
The only identifiable markers for 3201 B were marlc, $3, der (A1), and del (B2). Marker lc was a t(D~;?)(q;?), marker $3 was a t(A2;F~)(p;?), and der (A1) was a t(A1;?B2)(p;?). Marker B2 had a del (B2q). No cellular onc genes or Ig genes have yet been m a p p e d to the translocation or breakpoint regions of any of the identified markers (44]. However, the distal half of the q arm of the B2 chromosome houses the major histocompatibility complex (MHC) class I and class II antigens [44]. Morphologic observation suggests that the two MHC genes may have been deleted in all cases of del (Bzq), but the B2q deletion may have been terminal or interstitial. Furthermore, the exact band location of the MHC on the q arm is not known (45].
S Markers The two FeLV-positive cell lines d i d not share a c o m m o n marker (Fig. 4). Thus, the greatest sharing of chromosomal
$3
$2
Figure 6
3201 B karyotype. This karyotype shows B3, D1, E3, Ft, and F 2 monosomies. Nine of the 13 consistent markers are also observed in this cell, including der (At) and del (B2). The X chromosome is missing.
abnormalities was noted between the virus-free cell line and one of the virus-positive cell lines, 3272. Each cell line had originated independently: the virus-free cell line arose from a thymic l y m p h o m a , whereas one of the FeLV-positive cell lines arose from a multicentric l y m p h o m a (Table 1).
Comparative Karyotypic Analysis All three cell lines showed some monosomies, trisomies, and losses of chromosome pairs. The FL 74 was the only line with some consistent normal pairs of chromosomes; i.e., chromosomes B 2 and B 4. E 3 m o n o s o m y occurred with about 50% frequency in all three cell lines. D 1 m o n o s o m y was greater in FL 74 than in 3201 B: 83 and 69% respectively. F1 monosomy was greater in 3272 (80%) than in FL 74 (16%) and 3201 B (46%). The Ft pair was missing in 83% of cells of FL 74 and missing in only 8% of 3201 B. Even though all chromosome pairs have been n u m e r i c a l l y altered in all three cell
Chromosome Abnormalities and Oncogenesis in Cat Leukemias lines (with the exception of the FL 74 FeLV-positive cell line), the only chromosome pairs affected in both virus-positive cell lines were E3 and F1, with losses less than or equal to 50%. DISCUSSION
We tested whether chromosome abnormalities or virus integration sites could play a role in causation of feline leukemias and whether the FeLV virus may have a direct role in tumorigenesis. Each cell line had a different cell of origin, although all were T lymphocytes. Although each tumor was chromosomally distinct and kals~typically suggestive of origin as a single clone, some common markers were identified. If the FeLV virus were the cause of tumors and chromosomal abnormalities in virus-positive tumors, the lines might be expected to share markers. Furthermore, virus-free cells would not be expected to share chromosome aberrations with the two virus-positive cell lines. The opposite was true. The two FeLV-positive cell lines did not share markers in common, and their karyotypes differed considerably. The FeLV-free cell line had the highest frequency of rearranged chromosomes and the most uniformly clonal patterns of chromosome abnormality. This appears to indicate a lack of chromosomal specificity in relation to viral infection. In contrast, 3272 FeLV-positive cells and 3201 B FeLV-free cells showed the greatest association of chromosome abnormality. Perhaps neither the virus itself nor a viral product was important in tumorigenesis and chromosomal abnormalities, but rather the proviral integration site. If this were true, an integration site within the same locus should cause the same phenotypic tumor, and the chromosome abnormalities, as a consequence of this event, should be common to tumors of the same phenotype. If the virus were integrated and suppressed in the 3201 B FeLV-freecell line, as has been claimed for virus-negative tumors [15], it could have been expected to generate a marker common to FL 74 FeLV, the other thymic tumor, which did not occur. Lacking unique and clonal chromosome abnormalities, either viral specific or tissue specific, the aberrations found may represent late events in tumor development, showing progression in the tumor rather than initiation. A study of 21 naturally occurring feline lymphomas [36] showed only four lymphomas (19%) with FeLV virus integrated in a unique 2.4-kilobase locus. The other tumor integration sites did not map to any of the commonly known integration loci for other proviruses. Seven of the tumors analyzed were in the same phenotypic subgroup and included the four integration sites in the unique locus. Thus, the integration site hypothesis may explain some but not all cases in terms of the malignant potential of retroviruses with no onc genes. The only transforming autonomous genes are the retroviruses with transforming onc genes. They are capable of transforming a permissive host cell with 1:1 kinetics; i.e., each cell infected becomes transformed. Therefore, tumors arising from transforming retroviruses are polyclonal because more than one cell is infected by the retrovirus and its progeny [22]. Usually they do not show chromosome abnormalities, although they may do so later in tumor progression [46]. On the other hand, clonal chromosome abnormalities are commonly recognized in tumors that are not caused by
155
retroviruses with transforming onc genes [21, 27]. This characteristic is also shared with tumors of the same lineage that have not been infected with virus [22]. However, not all tumors show the high degree of specificity of chromosome abnormality demonstrated by the Ph chromosome of chronic myalogenous leukemia (CML). Even in CML, there is not complete uniformity among tumors. Some cases are variant translocations, and a similar Ph chromosome is found in acute lymphoblastic leukemia and in acute myeloblastic leukemia [47]. Polyclonalities of primary tumors may be common in certain tumor types [48-50]. Thus, even in nonviral cases, there are degrees of variability in clonality and chromosome specificity. Based on available genetic mapping of the domestic cat genome, the only chromosome abnormality identified in this study with potential bearing on carcinogenesis was the del{B2q}. The feline leukocyte antigen-1 {MHC class I} and feline leukocyte antigen-2 (MHC, class II) are located on the distal half of the B2q arm [44]. Based on morphology alone, the two MHC genes might have been deleted in all cases of dal(B2q I, but molecular analysis is necessary to verify this possible deletion. Studies of human malignant cancers of the breast, colon, urinary bladder, kidneys, and Lewis lung carcinoma for the expression of human leukocyte antigen (HLA) class I have shown that most primary lesions are HLAimmunoreactive cells whereas most metastatic lesions are HLA-nonreactive [51-52]. Tumors had a distribution of HLAreactive and HLA-nonreactive cells compatible with the suggestion that HLA-nonreactive cells have a selective advantage for growth and metastatic invasion [51]. This thesis is also supported by transfection experiments of H-2Kb histocompatible complex antigen in C57BL/6 genetic mice that showed abrogation of metastasis in H-2K and H-2D compatible mice by inducing alloreactive cytotoxic lymphocytes that kill H-2K-positive cells [52]. Because both virus-infected and virus-free cell lines have chromosomal abnormalities, we can conclude that the FeLV virus may cause chromosome abnormalities but that the aberrations are neither unique nor specific. These abnormalities may nevertheless be important in oncogenesis. Alternatively, as-yet-unknown factors may be causative of both chromosome abnormalities and cancer. If so, chromosome abnormalities may not be necessary for tumor development and progression, as suggested by the normal karyotype found in many human tumors, including breast cancers and meningiomas [46], and murine tumors. Human T-cell leukocyte virus-I (HTLV-I}has been associated with adult T-cell leukemia, but HTLV-I, like the FeLV virus, has no oncogene and the level of viral RNA expression is generally insignificant in infected cells [33]. Its integration site is also random. Another molecular event appears to be necessary for overt ATL to occur. The FeLV-associated tumors in feline may be a good animal model to study the pathophysiology and development of HTLV-I leukemias. This work was supported by Outstanding InvestigatorGrant No. 5R35-CA39915-06from the National Cancer Institute awarded to Peter Duesberg, Ph.D., of the University of California, Berkeley, CA, The author thanks Drs. Peter Duesberg, Sandra Wolman of Wayne State Medical School, Detroit, MI, and Doris Wurster-Hill of Dart-
156
mouth Medical School, Hanover, NH, for constructive criticism of the manuscript, as well as general influence and support. The author also thanks Drs. Kirsten Walen of State Public Health Service, Berkeley, CA, for technical advice and William Hardy, Jr.,of the Sloan-KetteringInstitutefor Cancer Research, N e w York, NY, forthe cell lines.
S.E. G u l i n o
19.
20. REFERENCES
1. Lutz H (1990): Feline retrovirnses: A brief review. Vet Micrebiol 23:131-146. 2. Essex ME (1982): Feline leukemia: A naturally occurring cancer of infectious origin. Epidemiol Rev 4:189-203. 3. Hardy WD Jr, McClelland AJ, DPhil E, MacEwen G, Hess PW, Hayes AA, Zuckerman EE (1977): The epidemiology of the feline leukemia virus (FeLV}. Cancer 39:1850-1855. 4. Hardy WD, Old LJ, Hess PW, Essex M, Cotter S [1973): Horizontal transmission of feline leukaemia virus. Nature 244:266-269. 5. Jarrett WFH, Jarrett O, Mackey L, Laird W, Hardy WD, Essex M [1973): Horizontal transmission of leukemia virus and leukemia in the cat. J Natl Cancer Inst 51;833-841. 6. RomatowskiJ(1989): Interpreting feline leukemia test results. Am Vet Med Assoc 195:928-930. 7. Donahue PR, Hoover EA, Beltz GA, Riedel N, Hirsch VM, Overbaugh J, Mullins JI (1988): Strong sequence conservation among horizontally transmissible, minimally pathogenic feline leukemia viruses. J Virel 62:722-731. 8. Reinacher M, Theilen G (1987): Frequency and significance of feline leukemia virus infection in necropsied cats. Am J Vet Res 48:939-945. 9. Jarrett O (1984): Pathogenesis of feline leukemia virus-related diseases. In: Mechanisms of Viral Leukaemogenesis, JM Goldman, O Jarrett, eds. Churchill-Livingstone, Edinburgh, pp. 135-154. 10. Lutz H, Pedersen NC, Theilen GH (1983): Course of feline leukemia virus infection and its detection by enzyme-linked immunosorbent assay and monoclonal antibodies. Am J Vet Res 44:2054-2059. 11. Dorn CR, Taylor DON, Schneider R, Hibbard HH, Klanber MR (1968): Survey of animal neoplasms in Alameda and Centre Costa Counties, California. II. Cancer morbidity in dogs and cats from Alameda County. JNCI 40:307-318. 12. Reinacher M (1989): Diseases associated with spontaneous feline leukemia virus (FeLV) infection in cats. Vet Immunol Immunopathol 21:85-95. 13. Rojko JL, Kociba GJ, Abkowitz JL, Hamilton KL, Hardy WD Jr, Ihle JN, O'Brien SJ (1989): Feline lymphomas: Immunological and cytochemical characterization. Cancer Res 49:345-351. 14. Hardy WD Jr (1980): Feline leukemia virus diseases. In: Feline Leukemia Virus, WD Hardy Jr, M Essex, AJ McClelland, eds. Elseviar/North Holland Press, New York, pp. 3-32. 15. Koshy R, Gallo C, Wong-Staal F (1980): Characterization of the endogenous feline leukemia virus-related DNA sequences in cats and attempts to identify exogenous viral sequences in tissues of virus-negative leukemic animals. Virology 103:434-445. 16. Hardy WD Jr, McClelland AJ, Zuckerman EE, Snydar HW Jr, MacEwen EG, Francis D, Essex M (1980): Development of virus nonproducer lymphosarcomas in pet cats exposed to FeLV. Nature 288:90-92. 17. Francis DP, Cotter SM, Hardy WD Jr, Essex M (1979): Comparison of virns-positive and virus-negative cases of feline leukemia and lymphoma. Cancer Res 39:3866-3870. 18. Essex M, Grant CK, Cotter SM, Sliski AH, Hardy WD (1979): Leukemia specific antigens: FOCMA and immunosurveillance. In: Modern Trends in Human Leukemia, vol. III, R Neth, RC
21.
22. 23.
24.
25. 26.
27. 28. 29.
30. 31. 32.
33.
34.
35. 36.
37.
38.
39.
40. 41.
Gallo, PH Hofschneider, K Mannweiler, eds. Springer-Verlag, New York, pp. 453-467. Hardy WD Jr, Geering G, Old LG, De Harven E, Brody RS, McDonough S (1969): Feline leukemia virus; Occurrence of viral antigen in the tissues of cats with lymphosarcoma and other diseases. Science 166:1019-1021. Grindem CB, Buoen LC (1989): Cytogenetic analysis in nine leukaemic cats. J Comp Pathol 101:21-30. Duesbarg PH (1987): Cancer genes: Rare recombinants instead of activated oncogenes Ca review). Proc Natl Acad Sci USA 84:2117-2124. Duasberg PH (1987): Retrovirnses as carcinogens and pathogens: Expectations and reality. Cancer Res 47;1199-1220. Rowley JD (1984): Consistent chromosomal aberrations and oncogenas in human t u m o u r s - a n overview. Cancer Surv 3:355-357. Trent JM (1984): Chromosomal alterations in human solid tumours: Implications of the stem model to cancer cytogenetics. Cancer Surv 3:395-422. Levan A (1956): Chromosomes in cancer tissue. Ann NY Acad Sci 63:774-792. Sandberg AA (1990): The Chromosomes in Human Cancer and Leukemia, 2nd Ed. Elsevier Science Publishing Co., New York, pp. 753-760. Wolman SR (1983): Karyotypic progression in human tumors. Cancer Metastasis Rev 2:257-293. Gough N (1985): Chromosomal translocations and the c-myc gene: Paradigm lost? Trends Genet 1:63-64. Duasberg PH (1987): Cancer genes generated by rare chromosomal rearrangements rather than activation of oncogenes. Haematol Blood Transfusion 31:496-510. Duasberg PH (1985): Activated proto-onc genes: Sufficient or necessary for cancer? Science 228:669-677. Duesherg PH (1983): Retroviral transforming genes in normal cells? Nature 304:219-226. Zhou R-P, Duesberg PH (1988): Myc protooncogene linked to retroviral promoter, but not to enhancer, transforms embryo cells. Proc Natl Acad Sci USA 85:2924-2928. Sadamori N (1991): Cytogenetic implication in adult T-cell leukemia- a hypothesis of leukemogenesis. Cancer Genet Cytogenet 51:131-136. Weiss R, Teich N, Varmus H, Coffin J (1985): RNA Tumor Viruses, 2nd Ed., Cold Spring Harbor Press, Cold Spring Harbor, NY. Weiss R, Teich N, Varmus H, Coffin J (1982): RNA Tumor Viruses. Cold Spring Harbor Press, Cold Spring Harbor, NY. Levesque KS, Bonham L, Levy LS (1990): flvi-1, A common integration domain of feline leukemia virus in naturally occurring lymphomas of a particular type. J Virol 64:3455-3462. Miura T, Tsujimoto H, Fukasawa M, Kodama T, Shibuya M, Hasegawa A, Hayami M (1987): Structural abnormality and over-expression of the myc gene in feline leukemias. Int J Cancer 40:564-569. Neel BG, Hayward WS, Robinson HL, Fang J, Astrin S (1981): Avian leukosis virus-induced tumors have common proviral integration sites and synthesize discrete new RNAs: Oncogenesis by promoter insertion. Cell 23:323-334. Francis DP, Essex M, Cotter SM, Jakowski RM, Hardy WD Jr. (1979b): Feline leukemia virus infections: The significance of chronic viremia. Leukemia Res 3:435-441. Seabright M [1971): A rapid banding technique for human chromosomes. Lancet 2:971-972. Wurstar-Hill DH, Gray CW (1973): Giemsa banding patterns in the chromosomes of twelve species of cats (Felidoe). Cytogenet Cell Genet 12:377-397.
C h r o m o s o m e Abnormalities and Oncogenesis in Cat Leukemias
42. Jones TC (1965): San Juan conference on karyotypes of Felldoe. Special report. Mammal Chromosomes Newslett 15:121122. 43. Hare WCD, Weber WT, McFeely RA, Yang Tsu-Ju (1966): Cytogenetics in the dog and cat. J Small Anim Pract 7:575-592. 44. O'BrienSJ, ed. (1990): Genetic Maps-Locus Maps of Complex Genomes, 5th Ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 45. Yuhki NY, O'Brien SJ (1988): Molecular characterization and genetic mapping of class I and class 1I MHC genes of the domestic cat. Immunogenetics 27:414-425. 46. Wolman SR (1986): Cytogenetic heterogeneity: Its role in tumor evolution. Cancer Genet Cytogenet 19:129-140. 47. Larson RA, Golomb HM, Rowley JD (1981): Chromosome changes in hematologic malignancies. CA-A Cancer J Clinicians 31:222-238.
157
48. Wolman SR (1991): Clonality. In: Biochemical and Molecular aspects of Selected Cancers, vol. 1, TG Pretlow II, T Pretlow, eds. Harcourt Brace Jovanovich, San Diego, pp. 409-410. 49. Heim S, Mandahl N, Mitelman F (1988): Genetic convergence and divergence in tumor progression. Cancer Res 48:5911-5916. 50. Aldaz CM, Conti CJ, Yuspa SH, Slaga TJ (1986): Cytogenetic profile of mouse skin tumors induced by the viral Harvey-ras gone. Carcinogenesis 9:1503-1505. 51. Cordon-Cardo C, Fuks Z, Drobnjak M, Moreno C, Eisenbach L, Feldman M (1991): Expression of HLA-A,B,C antigens on primary and metastatic tumor cell populations of human carcinomas. Cancer Res 51:6372-6380. 52. Gelber C, Plaksin D, Vadai E, Feldman M, Eisenbach L (1989): Abolishment of metastasis formation by murine tumor cells transfected with "foreign" H-2K genes. Cancer Res 49:23662373.
