Establishment and Characterization of Four Sinclair Swine Cutaneous Malignant Melanoma Cell Lines
Albert Green, Anne Shilkaitis, Laura Bratescu, Max S. Amoss Jr. and Craig W. Beattie
ABSTRACT: Cutaneous malignant melanoma of Sinclair Swine (SSCM) is a heritable, congenital neo- plasm which either proves fatal to the neonatal animal or undergoes spontaneous regression. Four SSCM cell lines, UISO-SSCM-433, UISO-SSCM-438, UISO-SSCM-5052, and UISO-SSCM-8093, were derived from biopsy specimens of primary tumors removed from swine at 26, 8, and 8 weeks of age, and 15 weeks gestation, respectively. Morphologic features, DOPA oxidase staining, and abnormal karyotype were suggestive of malignant melanoma. Each cell line was morphologically heterogeneous in culture with dendritic, spindle- and cuboidal-shaped cells. Pigmented melanosomes and DOPA oxidase activity were present in all cell lines at passages 20-22. UISO-SSCM-433 and UISO-SSCM-5052 contained hypodiploid and hypotetraploid sublines whereas UISO-SSCM-438 and UISO-SSCM-8093 were hypo- diploid and hypotetraploid, respectively. At later passages, all cell lines presented evolutionary, karya- typic changes; the same chromosomes were involved in the alterations, however. Chromosomes 2, 6, 13, and 14 were the most affected, exhibiting numerical and structural alterations in all four cell lines. Despite the presence of multiple chromosomal anomalies in all cell lines, each with a unique set of chromosomal markers, clonal growth was not detected in soft agar, nor were any of the lines tumorigenic following s.c. inoculation in athymic mice. This suggests that the loss of malignant potential in SSCM may be inherent.
INTRODUCTION
Sinclair Swine Cutaneous Melanoma (SSCM) represents a unique opportunity to investigate the development of heritable, congenital neoplasms. Multiple, primary lesions present in Sinclair swine are histopathologically similar to human compound nevi and malignant cutaneous mela- noma, demonstrate significantly different growth rates, have no sex or site preference, and exhibit a high incidence of spontaneous regression, a feature of primary human mel- anoma [1-10]. The high rate of spontaneous transformation and histopathologic similarity to human melanoma suggest the development of stable SSCM cell lines could provide molecular genetic and host environmental markers which may have correlates in the natural history of human cutane- ous melanoma.
Initial work [11, 12] suggested that finite cell lines could be derived from SSCM, but growth terminated at a maxi-
mum of 18-20 passages (75-85 population doublings). As classical melanosomes were only present in primary and passage one cultures, it suggested significant loss of func- tional phenotype during passage of these lines. Further studies suggested that cell lines which exhibited longer term growth could be established, but cytogenetic analysis revealed cell cultures to be diploid (n -- 38) [13], the normal chromosomal number for domestic swine [14].
In human cutaneous melanoma, cytogenetic analysis show chromosomal anomalies at recurring sites [15-20]. Cytogenetic analysis of SSCM cell lines established in our laboratory indicate that SSCM have chromosomal modal and numerical aberrations that involve specific chromo- somes in high frequency. In this paper, we describe four melanotic, karyotypically abnormal cell lines derived from Sinclair swine. In addition, we describe lymphocytic cyto- genetic abnormalities observed in Sinclair Swine that ex- hibited cutaneous melanoma.
From the Specialized Center for Cancer Research and Educa- tion (A.G., A.S., L.B.), University of Illinois at Chicago, Chicago, Illinois; and Department of Veterinary Physiology and Pharmacol- ogy (C.W.B.), Texas A & M University (M.S.A. Jr.), College Station, Texas; and USDA-ARS-MARC, Clay Center, Nebraska.
Address reprint requests to: Craig W. Beattie, Ph.D., USDA- ARS-MARC, P.O. Box 166, State Spur 18 D, Clay Center, NE 68933.
Received September 16, 1991; accepted February 10, 1992.
Initiation and Propagation of Cell Lines in Culture Blood and biopsy specimens were taken from swine of both sexes with cutaneous melanoma, along with blood samples from Hanford swine housed at Texas A & M University. Blood specimens obtained from clinically normal Hanford swine served as controls for cytogenetic analysis of periph-
77 Cancer Genet Cytogenet 61:77-92 (1992) 0165-4608/92/$05.00
78 A. Green et al.
eral lymphocy te cultures. Blood samples obtained for karyotypic evaluat ions were taken from Sinclair swine that exhibi ted cutaneous melanoma. All Sinclair swine were der ived from stock at the Sinclair Comparat ive Research Farm of the Univers i ty of Missouri and have been main- ta ined essent ia l ly as a c losed colony since 1970 at Texas A & M Universi ty (SMS-TAMU) [1, 3, 9]. The b iopsy sam- ples were t ranspor ted under sterile condi t ions in minimal essential media (MEM) supp lemen ted with 10% heat-inac- t ivated fetal bovine serum, (3 x ) funigizone, penic i l l in 100 /~g/ml, and s t reptomycin 100 tzg/ml (all from GIBCO, Grand Island, NY). The lesions were profusely washed with mini- mal essential media wi th Earle's salts (MEM-E) supple- mented with 2 mM L-glutamine, fungizone, penic i l l in 100 /zg/ml, s t rep tomycin 100/~g/ml (Biologos, Inc., Napervil le , IL), dur ing mechanica l disrupt ion. Specimens, I mm 3, were p laced in 25 cm 3 flasks containing MEM-E, 2mM L-gluta- mine, fungizone, penic i l l in 100 /~g/ml, s t rep tomycin 100 /zg/ml, nonessent ia l amino acids, and 15% heat- inact ivated fetal bovine serum (FBS) (Biologos) and incubated at 37°C, in 5% CO2 in air. Act ively growing cultures were fed weekly and nonadherent cells and debris were d iscarded when the media were changed. Confluent monolayers were passaged by gentle agitation in t rypsin-EDTA [trypsin 0.5 gm/liter with EDTA 0.2 gm/liter in Hanks balanced salt solut ion (HBSS) without Ca + + and Mg + *, [Biologos) resuspended, and transferred to a new 25-cm 3 flask with MEM-E con- taining 10% heat- inact ivated FBS with other media addi- tives kept at the same concentrat ions as in the init ial plat- ing. Cultures were fed once a week and passaged at a ratio of 1:2 , general ly after 15-20 days. Cells were cryopre- served in l iquid ni trogen at various passage levels begin- ning at passage 1, wi th cultures rout inely reestabl ished from frozen stocks.
Determination of Growth In Vitro and In Vivo Exponent ia l ly growing UISO-SSCM-433, UISO-SSCM-438, UISO-SSCM-5052, and UISO-SSCM-8093 cells were plated in t r ipl icate at a densi ty of 1 x 10 4 cells/well in 24- mul t iwel l dishes wi th MEM-E, 10% FBS, and cul tured for 9 days. Culture med ia was renewed on the fourth and seventh day. Cells harvested on days 1, 2, 4, 7, and 9 were counted on a Model ZB1 Coulter Counter, Coulter Electronics, Inc. (Hialeah, FL). UISO-SSCM-433, UISO-SSCM-438, UISO- SSCM-5052, and UISO-SSCM-8093 at passages (p) 21, 18, 22, and 13, respect ively, were harvested with t rypsin EDTA, resuspended in 1 x EBSS (Biologos), and inocu- lated (s.c.) into the flanks of 4- to 5-week-old male a thymic mice (NCI-nu/nu, Freder ick Cancer Resource Center, Fred- erick, MD) in an inocu lum of 1.2-2.1 × 106 cells per mouse suspended in a total of 0.1 ml of 1 x EBSS. Mice were examined weekly for the appearance of tumors over a pe- r iod of 120 days.
Anchorage Independence Assay The abi l i ty of ind iv idua l cell l ines to exhibit anchorage independen t growth was de te rmined according to the method of Kern et al. [21]. Single cells were seeded in a 0.5 ml solid agar (0.525%) feeder layer containing MEM supp lemen ted with 10% heat- inact ivated FBS in 6-well
t issue culture plates at final densi t ies of 3 x 103 to 1 × 104 cell /well . The cells were a l lowed to grow over a 3-week per iod at 37°C and scored according to the method of Sa lmon and Hamburger [22].
Morphology and Ultrastructure SSCM cells were seeded on plast ic covers l ips p laced in 35- × 10-mm Permanox petri dishes (Nunc, Inc., Napervi l le , IL) in MEM-E, 10% FBS and a l lowed to grow for 24-48 hours at 37°C. The seeded covers l ips were then fixed in formaldehyde, dehydra ted in ethanol , and s ta ined with hematoxyl in and eosin. For electron microscopy, cells were grown to confluency and processed in Permanox 35- x 10- mm tissue cul ture dishes. Cells were fixed for 2 hours in 4% phosphate-buffered g lu tera ldehyde , post-f ixed 1 hour in osmium tetroxide, dehydra ted in a graded series of etha- nol, infi l trated with a series of increasing epoxy resin to 100% ethanol mixtures, embedded in epoxy resin, and po lymer ized in a 60°C oven overnight. Thin sections were cut using a LKB Ultra Nova ul t ramicrotome, s ta ined with urany] acetate and lead citrate, and examined on a Joel 100 S electron microscope.
Isoenzyme Determination and Mycoplasma Testing Species origin was confirmed e lec t rophore t ica l ly by a lac- tate dehydrogenase [LDH) i soenzyme banding pat tern in an agarose 1% (w/v) gel according to the manufacturers instruct ions (Corning, Inc., Corning, NY). Cells from mono- layer culture were rout ine ly screened for Mycoplasmal con- taminat ion by a direct method using Microt r im TC flasks obtained from Hana Media (Berkeley, CA). Al l l ines were mycoplasma negative.
DOPA-Oxidase Histochemistry The presence of DOPA oxidase act ivi ty in ind iv idua l cell l ines was de te rmined by the method of Sheenan and Hra- phak [23]. A melan in synthes iz ing human melanoma, UISO-MEL-21 (p 41) served as a control. Posi t ive control staining was assigned an arbi trary value and the percentage staining of SSCM cell l ines de te rmined subjectively.
Karyotype Analysis A m i n i m u m of 100 cells per cell l ine were analyzed over mul t ip le passages. UISO-SSCM-433 cells were karyotyped at passages 21, and 28, UISO-SSCM-438 at p 11 and 29, UISO-SSCM-5052 at p 12 and 27, and UISO-SSCM-8093 at p 9 and 23. Cells adherent to t issue cul ture flasks were treated with 0.05 ml of co lcemid (10 tzg/ml, Gibco) in 2 ml of t rypsin-EDTA at 37°C for 25 rain, harvested, and centrifuged at 1000 × g for 10 rain, and resuspended in 0.47% KC1 at 37°C for 20 min. Acetic ac id /methanol fixative (1 : 3) was added to these suspens ions and the suspens ion centrifuged 1000 × g for 10 minutes. The supernatant was removed and fresh fixative added to the pellet . Sl ides were prepared using s tandard techniques [24]. The his topatho- logic characterist ics of each tumor, passage number, and number of cells counted are shown in Table 1.
All blood spec imens from Sinclair and Hartford animals underwent s tandard 3-day cul ture in RPMI 1640 media supp lemented with 20% heat- inact ivated FBS. Metaphase
No. counted/ Histology; primary; ° Tumor Doubling time c No. karyotyped
Cell line age stage b (hours) at passage no. ~ Modal no./range ~
SSCM°433 4 Cutaneous SSCM; IV P 19 51.6 P 21 100/5 40% tumor cells; P 29 98.5 26 wks
SSCM°438 Cutaneous SSCM; IV P 23 65.6 30% tumor cells; P 29 98.4 8 wks
SSCM-5052 Cutaneous SSCM; III/IV P 11 106.4 40% tumor ceils; P 29 76.8 8 wks
66/31-94
P 28 21/3 64/31-75 P 11 75/5 35/29-44 P 24 25/3
P 29 10/3 36/33-39 P 10 50/3 37/33-77 P 12 50/3
P 29 21/6 37/35-80 SSCM-8093 Cutaneous SSCM; II P 10 102.0 P 1 12/2 66/34-72
100% tumor cells; P 23 49.9 P 8,9 38/2 15 wks gestation P 11 50/3
P 23 18/3 63/52-71
,, Primary histology and swine age at the time of the biopsy. b Tumor stage at biopsy.
Doubling times observed in earlier and later passages. d Passage number of cell lines at the times of chromosomal analysis; number of cells counted and number of cells karyotyped.
Chromosome modal number and range.
harvesting, staining, and banding techniques were carried out using standard techniques [24]. A m i n i m u m of 25 cells per sample specimen were screened and analyzed using criteria defined by the Committee for the Standardization of the Domestic Pig [14].
RESULTS
Morphology and Ultrastructure Early in culture, SSCM lines exhibited a cont inuous layer of either predominant ly cubodial, epitheloid, dendritic, or spindle-shaped cells. Bi- and mult inucleated cells with 1 to 2 prominent nucleoli were seen. Later passages were phenotypical ly similar (Figs. 1-4). Mult inucleated giant cells were detected in less than 1% of UISO-SSCM-8093 at passage 24. All cell lines grew as monolayers and expressed the tendency to dislodge from monolayer if allowed to cont inue growing past confluence.
Electron microscopy of SSCM cell lines revealed irregu- larly shaped nuclei with i to 2 prominent nucleoli. Smooth and rough endoplasmic ret iculum were increased and cross-linked filaments were seen throughout the cell, as were irregularly shaped (sickle, bilobed, and clawed forms) mitochondria. Lightly pigmented, whorled, or cristae-like inclusions identified as stage 2 premelanosomes, as well as more heavily pigmented stage 3, 4 melanosomes, were prominant in each cell line. Premelanosomes and melano- somes were irregularly shaped with heavy pigmentation occurring along the edges of whorled inclusions. A variable degree of pigmentat ion was present in all cell lines at least through passage 27. Addit ional characteristics typical of a mal ignant phenotype (irregular nuclei, poor nuclear-to-
cytoplasmic ratio, etc.) present in init ial cultures were also essentially mainta ined in later passages (Figs. 1-4).
Growth Characteristics Doubling times (D.T.) varied significantly, with U/SO- SSCM-433 (p 19), and UISO-SSCM-438 (p 23), having faster initial growth rates (D.T. 51.6 and 65.5 hours) compared to UISO-SSCM-5052 (p 11), and UISO-SSCM-8093 (p 10) (D.T. 106.4 and 102 hours). Doubling times for all four UISO- SSCM cell lines changed with increased passages. UISO- SSCM-433 (p 29) and UISO-SSCM-438 (p 30) grew at a slower rate (D.T. 98.4 and 117.6 hours, respectively) com- pared to UISO-SSCM-5052 (p 29) and UISO-SSCM-8093 (p 23) where growth rate increased (D.T. 76.8 and 49.9 hours, respectively) (Table 1). Plateau phase cells cont inued to grow as long as nutr ients were provided, eventual ly reach- ing saturation densities (~ 2 x 104 cells/cm 2) when cells began to dislodge from monolayer. UISO-SSCM, 433,438, 5052, and 8093 are currently in passage 34, 33, 37, and 35, respectively, with no addit ional change in growth rate or phenotype.
Single cells suspended in soft nut r ien t agar did not form colonies over a period of 21 days. Subcutaneous inocula- tion of UISO-SSCM-433 (p 21), UISO-SSCM-438 (p 21), UISO-SSCM-5052 (p 10), and UISO-SSCM-8093 (p 9), into groups of 5 male, 4 -5 week-old athymic mice did not prove tumorigenic over a period of 120 days.
DOPA Oxidase Histochemistry Results indicate that all SSCM cell lines had DOPA oxidase activity comparable to the positive huma n melanoma con- trol cell line, where 60% of all ceils stained positively.
80 A. Green et al.
A B
C
T~
D
Figure 1 Phase contrast and TEM of cell line UISO-SSCM-433 at early and late passage. (A) UISO-SSCM-433 passage 17 ( x 25). (B) UISO-SSCM-433 passage 30 ( x 25). (C) UISO-SSCM-433 passage 20 ( x 5000). (D) UISO-SSCM- 433 passage 28 (x 6000).
Staining in UISO-SSCM lines ranged from 24 (UISO-SSCM- 433) to 39% (UISO-SSCM-8093) of control. DOPA oxidase staining was main ta ined over 25 passages (125 popula t ion doublings) in UISO-SSCM-433 cul tured from a pr imary tumor removed from an animal at 26 weeks of age (data not shown).
Karyo type Ana lys i s
Cytogenetic aberrat ions of UISO-SSCM-438, UISO-SSCM- 5052, and UISO-SSCM-8093 cell l ines observed during ini-
t ial passages (p < 11) suggest that the anomal ies descr ibed were also present at the lower passage numbers . Analys is of 100 Giemsa-banded metaphases of UISO-SSCM-433 at p 21 (115 popula t ion doublings) showed a numer ica l range of 31 to 94 chromosomes per cell (Fig. 5A, B; Tables 1, 2). Clonal chromosome aberrat ions of der(6p) and t(2q;13q) were detected in cells of this line. Several r inged chromo- somes of various sizes were also present , as wel l as mul t ip le homologs of chromosomes 1, 2, 4, 7, 13, and 18. Karyotypi- cally, there appeared to be two subl ines present , wi th each
Swine Melanoma Cytogenetics 81
A
C D
Figure 2 Phase contrast and TEM cell line of UISO-SSCM-438 at early and late passage. (A) UISO-SSCM-438 passage 8 (x 25). (B) UISO-SSCM-438 passage 35 (x 25). (C) UISO-SSCM-438 passage 22 (x 2000); insert, cristae structure of melanosome ( x 13,000). (D) UISO-SSCM-438 passage 35 ( x 2000); insert, cristae structure of me|anosome (x 10,000).
subl ine exhibi t ing unique markers. The first was shown to be hypod ip lo id , ranging from 32 to 39 chromosomes per mitosis. This subl ine presented the der(6p), t(2q;13q) and i(18q) as clonal chromosome anomalies. The second sub- l ine was shown to be hypote t rap lo id , wi th a chromosomal count that ranged from 64 to 77. Structural anomalies char- acteristic for the subl ine were der(2p), t(lq;2q) and the t(2q;13q), the one marker unique to both sublines.
At p 28 (168 popula t ion doublings) (Fig. 5B; Tables 1, 2), the cell l ine still presented both h y p o d i p l o i d and hypo- te t raploid sublines. However, karyotypic changes were ob- served. The t(6q;14q) and t(4q;13q) were representa t ive clonal markers for this later passage.
UISO-SSCM-438 cells at p 11 (66 popu la t ion doublings) had a hypod ip lo id modal chromosomal number varying between 29 and 44 chromosomes (Fig. 6A, B; Tables 1, 2).
82 A. Green et al.
A
D
Figure 3 Phase contrast and TEM of cell line UISO-SSCM-5052 at early and late passage. (A) UISO-SSCM-5052 passage 9 (x25). (B) UISO-SSCM-5052 passage 30 (x25). (C) UISO-SSCM-5052 passage passage 15 (x4000); insert, pigmented melanosome (x 13,000). (D) UISO-SSCM-5052 passage passage 30 (x 5000); insert, pigmented melanosome ( x 13,000/,
Although the incidence of structural anomalies was limited in the 100 mitosis screened, chromosomes 9, 10, 11, and 14 were lost. Metaphases that counted 38,XY, were not karyotypically normal, with either addit ional chromo- somes 2 or 14, with the absence of chromosomes 9, 10, or 11.
In contrast to the earlier passage at p 29 (174 populat ion doublings) (Fig. 6B; Tables 1, 2), mult iple structural alter-
ations were observed. Translocations involving chromo- somes 6, 7, 12, 14, 16, and 17 were detected with der(7q) and i(14q) as the clonal markers representative of this pas- sage. The cell line remained hypodiploid.
UISO-SSCM-5052 cells screened at passages 10 and 12 (60 and 72 populat ion doublings) also contained both hypo- diploid and hypotetraploid subl ines [modal range 33-77 (Figure 7A, B; Tables 1, 2)]. The modal range of the hypo-
Swine Melanoma Cytogenetics 83
A
D C
Figure 4 Phase contrast and TEM of cell line UISO-SSCM8093 at early and late passage. (A) UISO-SSCM- 8093 passage 6 (x 25). (B) UISO-SSCM-8093 passage 28 (x 25/. (C) UISO-SSCM-SO93 passage 11 (x 2600); insert, pigmented melanosome (x9000). (D) UISO-SSCM-8093 passage 26 (x 3000); insert, pigmented melanosome ( x 9000).
d ip lo id subl ine was 33 to 40 chromosomes per cell. Clonal markers t(2q;14q) and i(2q) represented consistent identi- fiable chromosomal abnormal i t ies in this subline. The chro- mosomal number ranged from 62 to 71 of the hypotetra- p lo id subline. Mul t ip le homologs of chromosomes 2, 5, 6, 7, 9, 10, 13, and 14 with mul t ip le unknown markers were also present. The one common marker present in both sub- sets was i(2q). Derived t ranslocat ions such as a der(2p),
der(2q), and t(13q;18q) were character is t ic markers not present in the hypod ip lo id subline.
The i(2q) and t(2q;14q) of earl ier passages were lost, to be replaced by t(2q;6q;6q) (Fig. 7B; Tables 1, 2), the repre- sentative marker at p 27 (162 popula t ion doublings). This cell l ine also re ta ined both hypod ip lo id and hypote t rap to id sublines.
Cell l ine UISO-SSCM-8093, der ived from a lesion re-
e"
u m
3
L 2
~ ..l
J-.,.
3
4 S
1 2
8uaz
6 7
6 ?
8 9
10
1
1
12
&w
8
9
i0
ii
12
A
L&
15
1
6
1";
18
s
ex
Markierm
B
Markers
Fig
ure
5
(A)
(p21
), (B
) (p
28).
G-b
ande
d ka
ryot
ypes
of
UIS
O-S
SCM
-433
. See
Tab
le 2
for
int
erpr
etat
ion.
17
18
sex
|I
11}
~:
; .
., jl
.~:
1 :3
3
4 5
1 :3
Vtl
l ~a
4
.¢0
II
o.
I:
i ,|
~. 6
7 6
7 j~
"l
i'-"
..
..
.
o,,
S ~,
.... !
e
w
8 9
10
11
12
II 9
tm
,,.
J'..
gO
11
list
! ,,
~
. ~
. s.
il
13
1
4
ls
16
1
7
18
m
q~c
13
| 7"
I4
15
1
6
11
18
m
Mar
ker
s A
M
arke
~s
B
Fig
ure
6
(A)
(p11
), (
B)
{p29
). G
-ban
ded
kary
otyp
es o
f U
ISO
oSSC
M-4
38.
See
Tab
le 2
for
int
erpr
etat
ion.
GO
¢.
21
%(
1
1
4
5
2 3
4 5
""
|I
6
?
il
....
....
....
. m
A
~
; ! •
"s
9 I0
11
12
8
a
..
I[
! II
10
ii
12
A
me
ww
"
13
14
15
16
17
18
sex
O
B
|A
¢Q,,
,.
- I"
13
14
15
16
17
18
sex
0 m
? :
Fig
ure
7
(A)
(p12
), (
B)
(p27
). G
-ban
ded
kary
otyp
es o
f U
ISO
-SSC
M-5
052.
See
Tab
le 2
for
int
erpr
etat
ion.
__~1
~,4
~_
- 2
3
&i
~ P 6
3'
lP
~"
--
jM~
a~
4 5
1 2
3 4
5
6 7
,llm
m
em~
w~
8 9
I0
13
Ii
12
17
18
lli
8~'
8 9
10
11
12
sex
....
..
4--
" 1,
6
,,,,
o,
~,b
, .....
SA
1
7
18
so
x
A
~L
~ ~
~ ~
• ....
~ ~
~e
Fs
]3
Ma
rke
rs
Fig
ure
8
(A)
(p9)
, (B
) (p
23).
G-b
ande
d ka
ryot
ypes
of
UIS
O-S
SCM
-809
3. S
ee T
able
2 f
or i
nter
pret
atio
n.
O0
"-.1
88
"-F-~ +' "'i - , ~ 2 ' - •
F 4 q ; " " I, = ~ 12~ 23 I ' ~ 21 45 ., :1
ll+p, , • • 1 - ; 7
2 3
3 2 I I
+:i 4 5
~ o ~ , ..... q--~ I I ~'mtl2 .J__~)l
q ~, ,-17~.;.'
• I ' e . , . . . .
6 7
~ -r~+ ~!I~ ,_, ! "-~, T2 ~ i - - ~ ' ~ I ~-#, +~
I L"~ I ~-#~+ ~+~ 8 9 10 11 12
- + 2 . + +
I ' = ; : 14 s
13
.17~i 17 ;8 16 !
2
X
Figure 9 Ideogram of G-banded swine chromosomes documenting the band regions presenting structural chromo- some alterations. The solid circles show the sites and frequency at which chromosomal band regions were involved in structural alterations from the four cases of swine melanoma described; ~" indicates the regions of those chromo- somes most frequently altered in SSCM.
89
1 ? ] 4 5
6
} l 0 11 12
]~ 2~ 15 16 ]7 18 sex
S Markers
F i g u r e 10 G-banded karyotype of Sinclair swine 0007. See Table 3 for in terpre ta t ion .
Table 2 Consensus karyotypes of SSCM cell lines
UISO-SSCM-433 P 21
UISO-SSCMo433 P28
UISOoSSCM-438 P 11 UISO-SSCM-438 P 29
UISO-SSCM-5052 P 12 UISO-SSCM-5052 P 27 UISO-SSCM-8093 P 9
33, XY, - 3, - 7, - 9, - 11, - 15. Occas iona l un iden t i f i ed r ing c h r o m o s o m e s observed . 36, XYY, + 2, - 7 , der(7)t(7;?)(q26;?), - 9 , der(12)t(12;16)(q15;q11) - 1 4 , der(14) t(14;14)(q11;q11), - 1 6 ,
The consensus karyotype is the most consistent karyotype for each cell line. In the consensus karyotype, the number of chromosomes may differ from the modal number due to random chromosome gains and losses. Changes present in individual cells, representing random loss, gain, or rearrangement, are discounted.
90 A. Green et el.
Table 3 Representat ive karyotyping of Sinclair and Hanford lymphocyte cultures
Number of Specimen normal cells Karyotype Abnormal cells identification Specimen detected interpretation detected
Karyotype Total interpretation count
0005 Sinclair 24 38, XX, normal 1 female
0007 Sinclair 23 38, XX, normal 2 female
0012 Sinclair 24 38, XY, normal 1 male
0032 Sinclair 24 38, XY, normal 1 male 2 cells 36, XX, with
different random losses H06 Hanford 25 38, XY
male H09 Hanford 25
female 38, XX
38, XX, der(13q), 25 der(16q)
1) 47, XX, +9mar. 25 2) 37, XX, -3 , -14 , + mar.
38, XY, +2mar. 25
37, XY, -15 t(15q;17q) 25
25
25
moved at 15 weeks gestat ional age and screened over mult i - ple passages (p8-p11) , ranged from 59 to 68 chromosomes (Fig. 8A, B; Tables 1,2). This cell l ine also presented numer- ical and structural aberrat ions that differed from the previ- ous three cells lines. A complex three-way t(6q;7q;14q) and t(2q;13q) were the clonal aberrations observed. Between two and six unident i f ied markers were observed in each cell examined and at least two of these marker chromo- somes were present throughout most of the mitosis screened.
At passage 23 (138 popula t ion doublings), the complex t(6q;7q;14q;) seen at earlier passage was lost. The t(2q;13q) was only observed infrequently. The one consis tent chro- mosomal al terat ion that remained was the der(7q). The new clonal marker der(2p) was observed in the majori ty of mitosis at this passage. An idiogram of the structural band regions most affected in all four cell l ines is presented in Figure 9.
Fifty cells screened from the lymphocyte cultures of male and female Hanford miniature swine (unaffected breed) presented all normal d ip lo id (n = 38) metaphases. Analys is of male and female Sinclair blood specimens, (25 cells analyzed/sample) demons t ra ted a majori ty of cells count ing 38,XX or 38,XY. However, in each spec imen screened, there were at least one or more numerica l or s tructural anomal ies detected in at least one metaphase. These inc luded delet ions, addi t ions, and translocat ions which pr imar i ly involved chromosomes 13, 14, and 15, which are also affected in SSCM cells (Figure 10; Table 3).
DISCUSSION
Our results suggest that cell l ines UISOoSSCM-433, 438, 5052, and 8093 are composed of phenotypica l ly melanot ic and karyotypica l ly abnormal cells. However, they do not exhibit anchorage- independent growth in vitro and are not tumorigenic in a thymic mice. Previous reports of tumorige- nici ty of SSCM tumor biopsies t ransplanted directly to the
cheek pouch of cor t i sone-suppressed hamsters also noted that tumor growth d id not a lways remain progressive [25].
. The doubl ing t imes of the cell l ines in the present s tudy are also in contrast to those of earl ier reports that cells removed from tumors from Sincla i r swine at approx imate ly 3 months of age and older exhib i ted a b iphas ic growth pat tern in which there was an ini t ia l popu la t ion doubl ing t ime of between 120 and 160 hours, wh ich decreased to 80-110 hours from approx imate ly passage 7 unt i l growth ceased at passages 18-20 [12]. Tumors removed from Sin- clair swine aged 2 months or younger repor ted ly exhibi ted a monophas ic doubl ing t ime of be tween 120 and 180 hours unti l d ivis ion ceased [12]. UISO-SSCM-5052 and UISO- SSCM-8093 (lesions removed at 6 weeks of age and 15 weeks gestation, respect ively) had a s lower ini t ial doubl ing time, but exhibi ted an increase in growth rate wi th in- creased passage number. The converse was true for l ines UISO-SSCM-433 (26 weeks) and 438 (8 weeks). In addi- t ional s tudies where SSCM has repor ted ly been adapted to long-term cul ture (19-46 passages), cell l ines had a short doubl ing t ime ( - 3 6 hours), but were d ip lo id , exhibi t ing a uniform modal number of 38 [13]. Premelanosomes and melanosomes were absent and dope oxidase was only pres- ent in a frequency of 1 in 200 to 1 in 500 cells /13], sug- gesting that a significant number of cells d id not remain phenotypica l ly differentiated.
All UISOoSSCM cell l ines were pheno typ ica l ly hetero- geneous, exhibi t ing the dendri t ic , cuboidal , and spindle- shaped cells observed in previous s tudies of SSCM [12] and in a significant number of human me lanoma cell l ines [10, 15, 26]. Premelanosomes and melanosomes were pres- ent in all cells after extensive passage. DOPA oxidase histo- chemist ry revealed dopa -ox idase -pos i t i ve cells present in all SSCM lines, at least through passage 25, wi th vi r tual ly all cells exhibi t ing some activity. All four SSCM cell l ines in this series differed karyotypica l ly from one another. The chromosomal number varied signif icant ly between cell l ines and each represented its own unique set of transloca-
Swine Melanoma Cytogenetics 91
t ions and abnormal karyotypes, inc luding mul t ip le homo- logs, t ranslocat ions, r inged, and iso-chromosomes. None of the current cell l ines has ceased growth in monolayer cul- ture after a m i n i m u m of 40 passages, which supports karyo- typic and morphologic evidence for a t ransformed cell.
Al though all four cell l ines showed similar modal num- bers for those early and later passages screened, abnormal karyotypes were present in all four cell l ines and evolut ion- ary, karyotypic changes were associated with increased passage number. For example, UISO-SSCM-438 harvested at p 11 showed very few chromosomal alterations, whereas in later passage (p 29), significant chromosome anomalies had arisen. In contrast to an earl ier report [13], present results show no normal mitosis, wi th all SSCM lines exhib- it ing structural and numer ica l chromosomal anomalies. Chromosomes 2, 6, 13, and 14 were most frequently affected (60-88%), wi th aberrat ions inc luding homolog addi t ions and delet ions, t ranslocat ions, and i sochromosome produc- tion. Chromosome 2 presented the highest frequency of involvement (88%), wi th both the p and q arms, as well as the centromeric region, being affected. Chromosomes 6 and 13 had the next highest number of aberrat ions (65-76%), wi th the terminal end of chromosome 6p and both terminal ends of 13q being affected. Chromosome 14q (62%) termi- nal ends were also involved in mul t ip le translocations. Chromosomal addi t ions, delet ions, and translocat ions were also detected in chromosomes 4, 5, 7, 15, and 18. However, the percentage of involvement is significantly lower (13-26%, data not shown). At present, whi le we can not be sure whether involvement of the latter chromosomes is more an evolut ionary than pr imary change involving ind iv idua l tumors, their appearance later in passage would suggest the former. Al though SSCM cell l ines did not ap- pear to show a consis tent pat tern of karyotypic change with increasing passage, chromosomes 13 and 14 cont inued to show the greatest number of al terations. In addit ion, each cell l ine presented unique markers even though many of the same chromosomes were involved. We were not able to ident i fy specific chromosomal al terat ions associated with the changes in cell prol i ferat ion in vitro, however.
Two major classes of cel lular genes, oncogenes, and those regulat ing stabil i ty are suggested to be affected by chromosomal rearrangement. Stabi l i ty genes include those involved with recombinat ion, repl icat ion, d ivis ion control, and repair [27]. Cells wi th rearrangement of a stabil i ty gene(s) are l ikely to show evolving karyotypes with mult i - ple chromosomal al terat ions as seen in our cell lines, whereas those rearrangements involved with oncogenes are more l ikely to show stable karyotypes wi th few addi t ional rearrangements [27, 28]. Recurring sites of chromosome change have been suggested to p inpoin t locations of cellu- lar oncogenes in human melanoma [17-20, 29-31]. This may also be true for comparable genes in swine. If chromo- somes exhibi t recurr ing sites of structural and numeric change which p inpoin t areas of regulatory sequences in human tumorigenesis , p inpoin t ing specific regions of chro- mosomal al terat ions in SSCM may help to define the abnor- mali t ies d iscovered dur ing the growth and maintenance of SSCM cell l ines. Areas in which one would look for oncogenes or s tabi l iz ing genes present in swine chromo-
somes may be located at the te rminal ends of chromosomes 6, 13, and 14, at chromosomes 6 (p15 and q35), 13 ( q l l and q49), 14 ( q l l and q29), and chromosome 2 at (p17, p11, and q29).
Interest ingly, chromosomes 13, 14, and 15 are also in- volved in the abnormal i t ies observed in WBCs of affected animals. The presence of anomal ies in those chromosomes which are also most affected in SSCM cells suggests that inherent abnormal i t ies exist in the genome of these ani- mals, which may be associated wi th the presence of mela- noma. Further, cytogenetic s tudies of affected and unaf- fected Sinclair swine should help to define this associat ion. Final ly, unl ike UISO-SSCM-438 and 8093, UISO-SSCM- 433 and UISO-SSCM-5052 exhib i ted heterogeneous geno- types wi th each line present ing h y p o d i p l o i d and hypotet ra- p lo id karyotypes. Whi le it is current ly not poss ible to ver- ify, the hypod ip lo id subl ines may be sequent ia l ly evolving into the more complex hypote t rap lo id lines. Extensive sub- cloning of each cell l ine may provide a better under s t and ing of the evolut ionary cytogenetic processes and un ique char- acteristics present in each of these cell l ines.
This research was supported in part by NIH-NCI-PO1-CA31827, RO1-CA-33764, and the Texas Agricultural Experiment Station. The authors gratefully acknowledge the generosity of Dr. L. Chris- man (deceased) in reviewing the karyotypes and D. Gardner for manuscript preparation.
R E F E R E N C E S
1. Millikan LE, Boylon JL, Hook JJ Jr, Manning PJ (1974): Mela- noma in sinclair swine, a new model. J Invest Dermatol 62:20-30.
2. Hook JJ Jr, Aultman MD, Millikan RW, Oxenhandler RW, Adelstein EH (1977): Development and regression of cutane- ous exophytic melanoma in sinclair (S-1] miniature swine. Yale J Biol Med. 50:561.
3. Hook JJ Jr, Aultmann MD, Adelstein EH, Oxehandler RW, Millikan LE, Middleton CC (1979); Influence of selective breeding on the incidence of melanomas in the sinclair minia- ture swine. Int J Cancer 24:668-672.
4. Middleton CC, Hood RR Jr (1981): The miniature pig: a mela- noma model. Lab Anim 10:48-53.
5. Berkelhammer J, Ensign BM, Hook RR Jr, Hecker CJ, Smith GD, Oxenhandler RW (1982): Growth and spontaneous regression of swine melanoma: relationship of in vitro leucocyte reactiv- ity. J Nat Cancer lust 68:461-468.
6. Hook JJ Jr, Berkelhammer J, Oxenhandler RW (1982): Animal model for human disease: sinclair swine melanoma. Am J Pathol 108:130-133.
7. Oxenhandler RW, Adelstein EH, Gaigh JP, Hook RR, Clark WH (1979): Malignant melanoma in sinclair swine. Am J Pathol 96:707-714.
8. Beattie CW, Tissot R, Amoss MS Jr (1988): Experimental mod- els in human melanoma research: a logical perspective. Semin Oncol 15:500-511.
9. Amoss MS Jr, Ronan SG, Beattie CW (1988): Growth of sinclair swine melanoma as a function of age, histopathological staging and gonadal status. Cancer Res 48:1708-1711.
10. DasGupta TK, Ronan SG, Beattie CW, Shilkaitis A, Amoss MS Jr (1989): Comparative histopathology of porcine and human cutaneous melanoma. J Ped Dermatol 6:289-299.
11. Berkelhammer J, Caines SM, Adelstein E, Oxenhandler RW,
9 2 A. G r e e n et al.
and Hook RR Jr (1978): Cultivation in vitro of sinclair swine melanoma cells. Proc Am Assoc Cancer Res 19:120.
12. Kovacs SH, Agris PF (1979): Long term in vitro cell culture of sinclair swine melanoma. In Vitro 15:329-341.
13. Berkelhammer J, Caines SM, Dexter DL, Adelstein E, Oxenhan- dler R, Hook RR Jr (1979): Adaption of sinclair swine mela- noma to long term growth in vitro. Cancer Res 39:4960-4964.
14. Gustavsson I (1988): Department of animal breeding and genet- ics, standard karyotype of the domestic pig. Heredity 109:151-157.
15. Quinn LA, Woods LK, Merrick SB, Arabasz NM, Moore GE (1977): Cytogenetic analysis of twelve human malignant mela- noma cell lines. J Natl Cancer Inst 59:301-307.
16. Chen TR, Shaw MW (1973): Stable chromosome changes in a human malignant melanoma. Cancer Res 33:2042-2047.
17. Becher R, Gibas Z, Karakousis C, Sandberg AA (1983): Nonran- dom changes in malignant melanoma. Cancer Res 43:5010-5016.
18. Trent JM, Rosenfield SB, Meyskens FL (1983): Chromosome 6q involvement in human malignant melanoma. Cancer Genet Cytogenet 8:177-180.
19. Limon J, Dal Cin P, Sait SNJ, Karakousis C, Sandberg AA (1988): Chromosome changes in metastatic human melanoma. Cancer Genet Cytogenet 30:201-211.
20. Sozzi G, Miozzo M, Calderone C, Fossati G, Pierotti MA, Casci- nelli N, Della Porta G (1990): Chromosome abnormalities and fragile sites in human melanoma. Cancer Genet Cytogenet 44:61-67.
21. Kern OH, Bertelson CA, Mann BD, Campbell MA, Morton
DL, Cochran AJ (1983): Clinical application of the clonogenic assay. Ann Clin Lab Sci 13:10-15.
22. Hamburger AH, Salmon SE (1977): Primary essay of human tumor stem cells. Science 197:461-463.
23. Sheenan D, Hraphak B (1980): The Theory of Practice of Histol- ogy. 2nd Edition.
24. Verma RS, Babu A (1989): Human Chromosomes: Manual of Basic Techniques. Pergamon Press, Inc. pp. 47-48.
25. Berkelhammer J, Hook RR Jr (1980): Growth of Sinclair swine melanoma in the hamster cheek pouch. Transplantation. 29:193-195.
26. Romsdahl MM, Hsu TC (1967): Establ ishment and biological properties of human melanoma cell lines grown in vitro. Surg Forum 18:78-79.
27. Cowan JM, Beckett MA, Tarbell N, Weichselbaum RR (1990): Symmetrical chromosome rearrangements in cell lines estab- lished from human radiat ion-induced sarcomas. Cancer Genet Cytogenet. 50:125-137.
28. Volpe JPG (1988): Genetic instability of cancer. Why a meta- static tumor is unstable and a benign tumor is stable. Cancer Genet Cytogenet 34:125-134.
29. Pathak S, Drwinga FIL, Hsu TC (1983): Involvement of chromo- some 6 in rearrangements in malignant melanoma cell lines. Cytogenet Cell Genet 36:573-579.
30. Balaban GB, Herlyn M, Clark WH, Nowell PC (1986): Karyo- typic evolution in human malignant melanoma. Cancer Genet Cytogenet 19:113-122.
31. Sandberg AA, Turc-Carel C, Gemmill R (1988): Chromosomes in solid tumors and beyond. Cancer Res 48:1049-1059.
