- Email: [email protected]
Confluent BALBc 3T3 cells monolayer provides selective and efficient growth of neoplastic cells
139
Cancer Letters, 52 (1990) 139- 143 Elsevier Scientific Publishers Ireland Ltd.
Confluent BALB/c 3T3 cells monolayer and efficient growth of neoplastic cells
provides
selective
Y. Kidoa, T. Mitsudomi”, H. Kuwano”, K. Yanoa, H. Matsuokab and K. Sugimachi” “Deportment bDepartment
of Surgery
II,
Faculty
of Medicine,
ofSurgery, MedicalInstitute
Kyushu
ojBioregulation,
University,
3-1-I
Maidashi
Higashi-ku,
Kyushu Uniuersity, 4546 Tsurumibaru,
Fukuoka
812
and
Beppu 874 (Japan)
(Received 19 February 1990) (Accepted
16 April 1990)
Summary
pared to other substrata and is also helpful in detecting incompletely transformed cells.
In an attempt to further characterize nonirradiated contact-inhibited confluent monolayer of BALB/c 3T3 cells (= ContactSensitiue Plates; CSP) as substrata for in oitro drug sensitivity testing, we compared the efficiency of colony formation with panels of cell lines on CSP with that on plastic dishes or in agar. Tumorlgenicity in athymic nude mice was also examined. We found that: (1) HeLa cells, 2 esophageal cancer lines, rat 3Y1 fibroblasts transformed by either adenovirus type 12, mouse polyoma virus, Rous avian sarcoma uirus, or plasmid DNA carrying v-Ha-ras oncogene all formed colonies on CSP and in agar and at the same time was tumorigenic. The efficiency of colony formation on CSP proued always to be higher than that in agar. (2) None of the 4 “normal” fibroplastic cell lines formed colonies on CSP or in agar and were tumorigenic. (3) Simian virus 40 and adenovirus ElA gene transformed rat 3Yl fibroblasts formed colonies on CSP but not in agar, and were not tumorigenic. Therefore, CSP was found to provide selective and eficient growth of neoplastic cells when com-
Correspondence
to: Y.
Kido.
Keywords: Contact inhibition; colony forma-
tion; neoplastic cell tumorigenicity; drug sensitivity testing. Introduction
BALB/c 3T3 A31 cells, originally isolated by Kakunaga [5] cease to proliferate when they reach confluence by the mechanism known as ‘ contact inhibition ‘ of cell division [ 1,4], This monolayer of cells referred to as (Contact-Sensitive Plates; CSP) remains in a resting state, even when not irradiated or when other cells are seeded on it [ll]. Conversely, while CSP do not allow normal cells to grow, they do allow tumor cells to effectively proliferate. Although contaminating stromal cells are sometimes a problem in determination the sensitivity to chemotherapeutic agents in clinical oncology, we were able to circumvent this problem by the use of CSP as a substrata of culture [lo, 111. In order to further characterize CSP in this study, we compared the cloning efficiency on CSP with that on plastic dishes or that in agar
0304-3835/90/$03.50 @ 1990 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
140
using various cell lines and while also examining the tumorigenicity of these cell lines in nude mice. Materials and methods Cell lines The history of the cell lines used in this study is summarized in Table 1. BALB/c 3T3 cells [5], NIH Swiss 3T3 cells [15], C3H lOTl/Z cells [13] and rat 3Yl cells [6] were normal ‘fibroblastic cell lines. KSE-1 [9] and KSE-2 cells, which were esophageal squamous cell carcinoma lines derived from two Japanese patients, and HeLa cells were of human neoplastic origin. As a model of the heterogeneity of cancer, 7 sublines of 3Yl cells transformed by various agents were also included in this study. SV-3Y l-C66 Py-3Y l-S2, Ad-3Y 1-219 cells were 3Yl cells transformed either by Simian virus 40 [18], mouse polyoma virus [18], or adenovirus type 12 (Ad 12) [18]. ElA-SYl-1 were 3Yl cells transformed by the ElA fragment of Ad 12 gene [14]. SR-3Yl-1 and HR-3Y l-l cells were 3Y 1 cells trans-
formed by Schmidt Ruppin strain Rous sarcoma virus [18] and v-Ha-ras provirus DNA [ 141, respectively. NG-3Y l-T12L were 3Y 1 cells transformed by chemical carcinogen N-methyl-N’-nitro-N-nitrosoguanidine. These derivatives of 3Yl cells were generous gifts from Dr. G. Kimura (Dept. of Virology Med. Inst. of Bioregulation Kyushu Univ.). The cells were routinely cultured in Dulbecco’s modified Eagle medium (DMEM) (Nissui, Osaka, Japan) supplemented with 10% fetal bovine serum (FBS) (Hazleton, KS, U.S.A.) with 20 pg/ml Gentamycin (Essex, Osaka, Japan) in a humidified atomosphere of 5% CO, and 95% air.
Preparation of Contact-Sensitive Plates (CSP) CSPs were prepared as described [lo]. Briefly, 1 x lo5 of BALB/c 3T3 cells were inoculated onto 60-mm tissue culture dishes (Corning, Tokyo, Japan). The medium was changed at confluence and the plates were incubated for an additional 2 days with daily changes of media to ensure the complete resting state of the cells. The monolayer cultures thus obtained were referred to as CSP.
Table 1.
Cell lines used and their characters.
Category
Cell line
Reference
Tissue of origin
Transforming agent
Established lines from normal tissue
BALB/c 3T3 Swiss 3T3 C3H lOT1/2 3Yl Hela KSE- 1 KSE-2 Ad 3Yl ElA 3Yl
5 15 13 6
18 14
Mouse embryo fibroblasts Mouse embryo fibroblasts Mouse embryo fibroblasts Fischer rat fibroblasts Human cervical carcinoma Human esophageal carcinoma Human esophageal carcinoma Fischer rat fibroblasts Fischer rat fibroblasts
Py 3Yl sv 3Yl SR 3Yl HR 3Yl
18 18 18 14
Fischer Fischer Fischer Fischer
None None None None Spontaneous Spontaneous Spontaneous Adeno virus type 12 Plasmid carrying ElA region of Ad 12 in vitro Mouse polyoma virus Simian virus 40 Rous sarcoma virus Harvey murine sarcoma virus N-methyl-N’nitroN-nitrosoguanidine
Established lines of neoplastic origin Lines transformed
NG 3Yl
9
rat fibroblasts rat fibroblasts rat fibroblasts rat fibroblasts
Fischer rat fibroblasts
141
Cloning efficiency on plastic dishes or CSPs The randomly proliferating cells to be tested were dispersed by 0.25% Trypsin (Nakarai Kagaku, Kyoto, Japan) and 1 x lo2 or 2 X lo2 cells were plated onto dishes or CSPs with 5 ml of fresh media in triplicate. After 7-10 days of incubation without a medium change (plastic dishes) or with medium change every 3 days (CSP) , the cultures were fixed with methanol and stained by Giemsa. Then the number of cdlonies consisting of more than 50 cells were determined. Cloning efficiency in agar This was performed as described by MacPherson and Montagnier [8]. Briefly, 1 X lo2 -2 x lo3 cells suspended in 4 ml of complete medium containing 0.3% Difco agar (Sank0 Junyaku, Tokyo, Japan) were plated in 60mm dishes over a layer of 0.5% agar medium. Colonies containing at least 50 cells were scored after 2-3 weeks. Tumorigenicity
in nude mice
The tumorigenic potential examined by injecting 2 x lo6 -5 nude mice at 4-6 weeks scapular region. All mice were weekly for at least 4 months.
of cells was cells S.C. into 3 of age in the then observed
Results The cloning efficiency of various cell lines on plastic dishes, CSPs and in agar (Fig. 1, Fig. 2) None of the “normal” fibroblastic cell lines (C3H lOT1/2, NIH Swiss 3T3, BALB/c 3T3, 3Yl) formed visible colonies either on CSP or in agar. On the other hand, cell lines of human neoplastic origin (HeLa, KSE-1, KSE-2) formed colonies both on CSP and in agar. The efficiency of colony formation on CSP was as high as that on plastic dishes and 5 to 10 times higher than that in agar. 3Y 1 cells transformed by polyomavirus (Py-3Y 1)) adenovirus (Ad3Y 1), Rous sarcoma virus (SR-3Y 1), Harveyras gene (HR-3Y 1) and nitrosoguanidine (NG3Y 1) formed colonies both on CSP and in agar and their cloning efficiency on CSP was higher
Normal
cell
line
% 1OOr
5
5
m
: Agar
.-0 .-
m
:
CSP
0
:
plastic
5
50-
E .5 0
1
I
0
Tumorigenicity
0 0
lOTl/Z O/3
Neoplastic
oofi
00
S-3T3 o/3
cell
Ei-3T3 o/3
00
3Yl o/3
line
50 .-? 5 0 0 Tumorigenicity
KSE-1 5/5
KSE-2 5/5
HeLa 5/5
Fig. 1. Cloning efficiency of “normal” (upper) and neoplastic (lower) cell lines on plastic dish, contact-sensitive plate (CSP) and in agar. None of the “normal cell lines formed colonies on CSP nor in agar. Cell lines of human neoplastic origin formed colonies both on CSP and in agar. The efficiency of colony formation on CSP was as high as that on plastic dish and 5- 10 times higher than that in agar. The tumorigenicity of each cell is indicated under the figures. lOT1/2; C3H lOT1/2, S3T3; Swiss 3T3, B-3T3; BALB/c 3T3, Bars; S.D.
than that in agar the same as with human neoplastic cell lines. However, Simian virus 40 transformed 3Y 1 (SV-3Yl) and 3Y 1 cells transformed by ElA gene of adenovirus 12 (ElA-3Yl) cells showed a different phenotype. These two cell lines formed colonies on CSP at a lower efficiency than that on plastic dishes and they did not form colonies in agar.
142
3Yl
transformant
% r 100
q
Tumorigenicity in nude mice Cell lines which did not form colonies in agar (C3H lOTl/Z., NIH Swiss 3T3, BALB/c 3T3, 3Y1, SV-3Y1, ElA-3Yl) were not tumorigenic in athymic nude mice, while the other cell lines all formed tumors in nude mice. All these results are summarized in Table 2.
: Agar
E .!? 2 “,
50
.-t
Discussion
2 0 0 Tumorigenicity
SR 3/3
Ad 3/3
PY 3/3
o/3
o/3
NG 2/3
HR 3/3
%I
I
0 0 3Yl Tumorigenicity O/3
Fig. 2. Cloning efficiency of 3Yl and its transformants on plastic dishes, CSP and in agar. SR-, Ad-, Py-, NG-, HRSY 1 cells formed colonies both on CSP and in agar. SV- and ElA-3Yl cells formed colonies on CSP, while not in agar. These two cell lines were also found not to be tumorigenic in nude mice.
Table 2.
Relationships between growth properties vitro and tumorigenicity in nude mouse. Cell lines
Colony formation
Normal cell lines SV3Y1, ElA, 3Yl Ad3Yl,Py3Yl, HR3Y1, SR 3Y1, NG 3Y1, Neoplastic cell lines
+ , Formed tumor.
colonies
and/or
Tumorigenicity
Plates
CSP
+
---
+
+
-
-
+
+
+
+
tumor;
in
Agar
- , no colonies
or
In this study we investigated the cloning efficiency on various substrata along with tumorigenicity in nude mice using various cell lines, to characterize CSPs as an effective substrata for chemosensitivity testing. Tumorigenicity was specifically correlated by anchorage independent growth in agar as has been previously reported [Z]. CSP did not allow “normal” cells to grow, which provides a great advantage for determining chemosensitivity in vitro, because tumor specimens_usually contain varying amounts of stromal cells. This inhibition of CSP on growth of normal cells may be mediated by a cell-to-cell interaction, such as gap junctional communication. Yamasaki et al. [17] reported that there is a complete lack of communication between transformed and normal cells as observed by fluorescent dye injection. In addition, when there is communication between the two different cell types, there is no expression of the transformed phenotype. Conversely, CSPs did not act as inhibititors of colony formation for neoplastic cells. Klein et al. [7] successfully characterized human bronchial carcinoma cells on X-ray irradiated C3H lOTl/Z feeder layer cells. Their feeder layer culture system proved to be selective for the growth of malignant cells and all non malignant epithelial and connective tissue cells were observed to die within two days. However, we found that the irradiated confluent monolayers (cell mat) were not able to inhibit the proliferation of all normal cells, as described by Chang [3]. Therefore we used contact-inhibited confluent non-irradiated monolayers of BALB/c 3T3 cells as “CSP”. The fact that all colony formers in agar also formed colonies on CSP with a much higher
143
efficiency suggests that CSP may serve as a feeder layer for neoplastic cells which are seeded on them. This is major advantage of CSP over agar culture which has limits in clinical application [ 12,161. SV-3Y 1 cells and E l A-3Y 1 were unique in that they formed colonies on CSP but not in agar.These cells are apparently abnormal with respect to reduced serum requirements high saturation density and expression of viral antigens [ 14,181. These cells are considered to be incomplete transformants and it is noteworthy that the CSP were able to detect such cells. Carcinogenesis has been generally thought to occur in a step by step fashion. In this context, CSP may also become a useful tool in screening and selecting these low malignant cells which may be developing into fully malignant cells for studying cancer.
6
7
8
9
10
11
12
Acknowledgement The authors wish to thank Prof. G. Kimura, Department of Virology, Medical Institute of Bioregulation, Kyushu University for providing the 3Yl and the transformed cell lines; B.T. Quinn, Kyushu University for valuable comments on the manuscript. References Abercrombie, M. (1979) Contact nancy. Nature, 281,259-262.
13
14
15 inhibition
and
malig-
Banet, J.C., Crawfold, B.D., Mixter, L.O., Schechman, L.M., Ts’o, P.O.P. and Pofiack, R. (1979) Correlation in vitro growth properties and tumorigenicity of Syrian hamster cell lines. Cancer Res., 39, 1504-1510.
16
Chang, C., Trosko, J.E., El-Fouly, H.H., GibsonD’Ambrosio, R.E. and D’Ambrosio, S.M. (1987) Contact insensivity of a subpopulation of normal human fetal epithelial cells and of human carcinoma cell lines. Cancer Res., 47, 1634-1645. Holly, R.W. and Kiernan, J.W. (1968) “Contact inhibition”, of cell division in 3T3 cells. Proc. Natl. Acad. Sci. USA, 60,300-304.
17
Kakunaga, T. (1973) A quantitative system for assay of malignant transformation by chemical carcinogens using a clone derived from BALB/3T3. Int. J. Cancer, 12,463473.
Kimura, G., Itagaki, A. and Summers, J. (1975) Rat cell line 3Y 1 and its virogenic polyoma-and SV40transformed derivatives. fnt. J. Cancer, 26,435-442. Klein, J.C., Zurcher, C. and van Bekkum, D.W. (1987) Differential behavior of human bronchial carcinoma cells in culture. Cancer Res., 47,3251-3258. MacPherson, 1. and Montagnier, L. (1964) Agar suspension culture for the selective assay of cells transformed by polyoma virus. Virology, 23,291-294. Matsuoka, H., Sugimachi, K., Ueo, H.. Kuwano, H., Nakano, S. and Nakayama, M. (1987) Sex hormone response of newly established squamous cell line derived from clinical esophageal carcinoma. Cancer 4130-4140. Matsuoka, H., Sugimachi, K. and Yano. Chemosensitivity testing for anticancer drugs
Res.,
47,
K. (1987) using con-
fluent monolayers. Med. Sci. Res., 15, 837-838. Matsuoka. H., Sugimachi, K., Yano, K., Kido, Y ., Shirabe. K. and Mitsudomi, T. (1988) Influence of oestradiol upon contact-sensitive plates of confluent BALB/c 3T3 cell monolayers. Med. Sci. Res., 16, 1017-1018. Pavelic, Z.P., Slocum, H.K., Rustum, Y .M., Creaven, P.J., Novak, N.J., Karakousis, C., Takita, H. and Mittelman, A. (1980) Growth of cell colonies in soft agar from biopsies of different human solid tumors. Cancer Res., 40, 4151-4158. Reznikoff, C.A., Brankow, D.W. and Heidelberger, C. (1973) Establishment and characterization of a clonal line of C3H mouse embryo cell sensitive to post confluence inhibition of division. Cancer Res., 33, 3321-3328. Shimura, H., Ohtsu, M., Matsuzaki, A., Mitsudomi. T., Onodera, K. and Kimura. G. (1988) Selective cytotoxicity of phospholipids and diacyl glycerols to rat 3Y 1 fibroblasts transformed by adenovirus type 12 or its ElA gene. Cancer Res., 48,578-583. Todaro, G.J. and Green, H. (1963) Quantitative studies of the growth of mouse embryo cells in culture and their development into established lines. J. Cell Biol., 17, 299313. Von Hoff, D.D., Weisenthal, L.M., Ihde, D.C., Mathews, M.J., Layard, M. and Makuch, R. (1981) Growth of lung cancer colonies from bronchoscopy washings. Cancer, 48, 400-403. Yamasaki. H., Hollstein, M., Mesnil, M.. Martel. N. and Aguelon, A.M. (1987) Selective lack of inter cellular communication between transformed and nontransformed cells as a common property of chemical and oncogene transformation of BALB/c 3T3 cells. Cancer Res., 47.
18
5658-5664. Zaitsu. H., Tanaka, H., Mitsudomi, T., Matsuzaki, A., Ohtsu. M. and Kimura, G. (1988) Differences in proliferation properties among sublines of rat 3Yl fibroblasts transformed by various agents in vitro. Biomed. 197.
Res., 9, 181-
Cancer Letters, 52 (1990) 139- 143 Elsevier Scientific Publishers Ireland Ltd.
Confluent BALB/c 3T3 cells monolayer and efficient growth of neoplastic cells
provides
selective
Y. Kidoa, T. Mitsudomi”, H. Kuwano”, K. Yanoa, H. Matsuokab and K. Sugimachi” “Deportment bDepartment
of Surgery
II,
Faculty
of Medicine,
ofSurgery, MedicalInstitute
Kyushu
ojBioregulation,
University,
3-1-I
Maidashi
Higashi-ku,
Kyushu Uniuersity, 4546 Tsurumibaru,
Fukuoka
812
and
Beppu 874 (Japan)
(Received 19 February 1990) (Accepted
16 April 1990)
Summary
pared to other substrata and is also helpful in detecting incompletely transformed cells.
In an attempt to further characterize nonirradiated contact-inhibited confluent monolayer of BALB/c 3T3 cells (= ContactSensitiue Plates; CSP) as substrata for in oitro drug sensitivity testing, we compared the efficiency of colony formation with panels of cell lines on CSP with that on plastic dishes or in agar. Tumorlgenicity in athymic nude mice was also examined. We found that: (1) HeLa cells, 2 esophageal cancer lines, rat 3Y1 fibroblasts transformed by either adenovirus type 12, mouse polyoma virus, Rous avian sarcoma uirus, or plasmid DNA carrying v-Ha-ras oncogene all formed colonies on CSP and in agar and at the same time was tumorigenic. The efficiency of colony formation on CSP proued always to be higher than that in agar. (2) None of the 4 “normal” fibroplastic cell lines formed colonies on CSP or in agar and were tumorigenic. (3) Simian virus 40 and adenovirus ElA gene transformed rat 3Yl fibroblasts formed colonies on CSP but not in agar, and were not tumorigenic. Therefore, CSP was found to provide selective and eficient growth of neoplastic cells when com-
Correspondence
to: Y.
Kido.
Keywords: Contact inhibition; colony forma-
tion; neoplastic cell tumorigenicity; drug sensitivity testing. Introduction
BALB/c 3T3 A31 cells, originally isolated by Kakunaga [5] cease to proliferate when they reach confluence by the mechanism known as ‘ contact inhibition ‘ of cell division [ 1,4], This monolayer of cells referred to as (Contact-Sensitive Plates; CSP) remains in a resting state, even when not irradiated or when other cells are seeded on it [ll]. Conversely, while CSP do not allow normal cells to grow, they do allow tumor cells to effectively proliferate. Although contaminating stromal cells are sometimes a problem in determination the sensitivity to chemotherapeutic agents in clinical oncology, we were able to circumvent this problem by the use of CSP as a substrata of culture [lo, 111. In order to further characterize CSP in this study, we compared the cloning efficiency on CSP with that on plastic dishes or that in agar
0304-3835/90/$03.50 @ 1990 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
140
using various cell lines and while also examining the tumorigenicity of these cell lines in nude mice. Materials and methods Cell lines The history of the cell lines used in this study is summarized in Table 1. BALB/c 3T3 cells [5], NIH Swiss 3T3 cells [15], C3H lOTl/Z cells [13] and rat 3Yl cells [6] were normal ‘fibroblastic cell lines. KSE-1 [9] and KSE-2 cells, which were esophageal squamous cell carcinoma lines derived from two Japanese patients, and HeLa cells were of human neoplastic origin. As a model of the heterogeneity of cancer, 7 sublines of 3Yl cells transformed by various agents were also included in this study. SV-3Y l-C66 Py-3Y l-S2, Ad-3Y 1-219 cells were 3Yl cells transformed either by Simian virus 40 [18], mouse polyoma virus [18], or adenovirus type 12 (Ad 12) [18]. ElA-SYl-1 were 3Yl cells transformed by the ElA fragment of Ad 12 gene [14]. SR-3Yl-1 and HR-3Y l-l cells were 3Y 1 cells trans-
formed by Schmidt Ruppin strain Rous sarcoma virus [18] and v-Ha-ras provirus DNA [ 141, respectively. NG-3Y l-T12L were 3Y 1 cells transformed by chemical carcinogen N-methyl-N’-nitro-N-nitrosoguanidine. These derivatives of 3Yl cells were generous gifts from Dr. G. Kimura (Dept. of Virology Med. Inst. of Bioregulation Kyushu Univ.). The cells were routinely cultured in Dulbecco’s modified Eagle medium (DMEM) (Nissui, Osaka, Japan) supplemented with 10% fetal bovine serum (FBS) (Hazleton, KS, U.S.A.) with 20 pg/ml Gentamycin (Essex, Osaka, Japan) in a humidified atomosphere of 5% CO, and 95% air.
Preparation of Contact-Sensitive Plates (CSP) CSPs were prepared as described [lo]. Briefly, 1 x lo5 of BALB/c 3T3 cells were inoculated onto 60-mm tissue culture dishes (Corning, Tokyo, Japan). The medium was changed at confluence and the plates were incubated for an additional 2 days with daily changes of media to ensure the complete resting state of the cells. The monolayer cultures thus obtained were referred to as CSP.
Table 1.
Cell lines used and their characters.
Category
Cell line
Reference
Tissue of origin
Transforming agent
Established lines from normal tissue
BALB/c 3T3 Swiss 3T3 C3H lOT1/2 3Yl Hela KSE- 1 KSE-2 Ad 3Yl ElA 3Yl
5 15 13 6
18 14
Mouse embryo fibroblasts Mouse embryo fibroblasts Mouse embryo fibroblasts Fischer rat fibroblasts Human cervical carcinoma Human esophageal carcinoma Human esophageal carcinoma Fischer rat fibroblasts Fischer rat fibroblasts
Py 3Yl sv 3Yl SR 3Yl HR 3Yl
18 18 18 14
Fischer Fischer Fischer Fischer
None None None None Spontaneous Spontaneous Spontaneous Adeno virus type 12 Plasmid carrying ElA region of Ad 12 in vitro Mouse polyoma virus Simian virus 40 Rous sarcoma virus Harvey murine sarcoma virus N-methyl-N’nitroN-nitrosoguanidine
Established lines of neoplastic origin Lines transformed
NG 3Yl
9
rat fibroblasts rat fibroblasts rat fibroblasts rat fibroblasts
Fischer rat fibroblasts
141
Cloning efficiency on plastic dishes or CSPs The randomly proliferating cells to be tested were dispersed by 0.25% Trypsin (Nakarai Kagaku, Kyoto, Japan) and 1 x lo2 or 2 X lo2 cells were plated onto dishes or CSPs with 5 ml of fresh media in triplicate. After 7-10 days of incubation without a medium change (plastic dishes) or with medium change every 3 days (CSP) , the cultures were fixed with methanol and stained by Giemsa. Then the number of cdlonies consisting of more than 50 cells were determined. Cloning efficiency in agar This was performed as described by MacPherson and Montagnier [8]. Briefly, 1 X lo2 -2 x lo3 cells suspended in 4 ml of complete medium containing 0.3% Difco agar (Sank0 Junyaku, Tokyo, Japan) were plated in 60mm dishes over a layer of 0.5% agar medium. Colonies containing at least 50 cells were scored after 2-3 weeks. Tumorigenicity
in nude mice
The tumorigenic potential examined by injecting 2 x lo6 -5 nude mice at 4-6 weeks scapular region. All mice were weekly for at least 4 months.
of cells was cells S.C. into 3 of age in the then observed
Results The cloning efficiency of various cell lines on plastic dishes, CSPs and in agar (Fig. 1, Fig. 2) None of the “normal” fibroblastic cell lines (C3H lOT1/2, NIH Swiss 3T3, BALB/c 3T3, 3Yl) formed visible colonies either on CSP or in agar. On the other hand, cell lines of human neoplastic origin (HeLa, KSE-1, KSE-2) formed colonies both on CSP and in agar. The efficiency of colony formation on CSP was as high as that on plastic dishes and 5 to 10 times higher than that in agar. 3Y 1 cells transformed by polyomavirus (Py-3Y 1)) adenovirus (Ad3Y 1), Rous sarcoma virus (SR-3Y 1), Harveyras gene (HR-3Y 1) and nitrosoguanidine (NG3Y 1) formed colonies both on CSP and in agar and their cloning efficiency on CSP was higher
Normal
cell
line
% 1OOr
5
5
m
: Agar
.-0 .-
m
:
CSP
0
:
plastic
5
50-
E .5 0
1
I
0
Tumorigenicity
0 0
lOTl/Z O/3
Neoplastic
oofi
00
S-3T3 o/3
cell
Ei-3T3 o/3
00
3Yl o/3
line
50 .-? 5 0 0 Tumorigenicity
KSE-1 5/5
KSE-2 5/5
HeLa 5/5
Fig. 1. Cloning efficiency of “normal” (upper) and neoplastic (lower) cell lines on plastic dish, contact-sensitive plate (CSP) and in agar. None of the “normal cell lines formed colonies on CSP nor in agar. Cell lines of human neoplastic origin formed colonies both on CSP and in agar. The efficiency of colony formation on CSP was as high as that on plastic dish and 5- 10 times higher than that in agar. The tumorigenicity of each cell is indicated under the figures. lOT1/2; C3H lOT1/2, S3T3; Swiss 3T3, B-3T3; BALB/c 3T3, Bars; S.D.
than that in agar the same as with human neoplastic cell lines. However, Simian virus 40 transformed 3Y 1 (SV-3Yl) and 3Y 1 cells transformed by ElA gene of adenovirus 12 (ElA-3Yl) cells showed a different phenotype. These two cell lines formed colonies on CSP at a lower efficiency than that on plastic dishes and they did not form colonies in agar.
142
3Yl
transformant
% r 100
q
Tumorigenicity in nude mice Cell lines which did not form colonies in agar (C3H lOTl/Z., NIH Swiss 3T3, BALB/c 3T3, 3Y1, SV-3Y1, ElA-3Yl) were not tumorigenic in athymic nude mice, while the other cell lines all formed tumors in nude mice. All these results are summarized in Table 2.
: Agar
E .!? 2 “,
50
.-t
Discussion
2 0 0 Tumorigenicity
SR 3/3
Ad 3/3
PY 3/3
o/3
o/3
NG 2/3
HR 3/3
%I
I
0 0 3Yl Tumorigenicity O/3
Fig. 2. Cloning efficiency of 3Yl and its transformants on plastic dishes, CSP and in agar. SR-, Ad-, Py-, NG-, HRSY 1 cells formed colonies both on CSP and in agar. SV- and ElA-3Yl cells formed colonies on CSP, while not in agar. These two cell lines were also found not to be tumorigenic in nude mice.
Table 2.
Relationships between growth properties vitro and tumorigenicity in nude mouse. Cell lines
Colony formation
Normal cell lines SV3Y1, ElA, 3Yl Ad3Yl,Py3Yl, HR3Y1, SR 3Y1, NG 3Y1, Neoplastic cell lines
+ , Formed tumor.
colonies
and/or
Tumorigenicity
Plates
CSP
+
---
+
+
-
-
+
+
+
+
tumor;
in
Agar
- , no colonies
or
In this study we investigated the cloning efficiency on various substrata along with tumorigenicity in nude mice using various cell lines, to characterize CSPs as an effective substrata for chemosensitivity testing. Tumorigenicity was specifically correlated by anchorage independent growth in agar as has been previously reported [Z]. CSP did not allow “normal” cells to grow, which provides a great advantage for determining chemosensitivity in vitro, because tumor specimens_usually contain varying amounts of stromal cells. This inhibition of CSP on growth of normal cells may be mediated by a cell-to-cell interaction, such as gap junctional communication. Yamasaki et al. [17] reported that there is a complete lack of communication between transformed and normal cells as observed by fluorescent dye injection. In addition, when there is communication between the two different cell types, there is no expression of the transformed phenotype. Conversely, CSPs did not act as inhibititors of colony formation for neoplastic cells. Klein et al. [7] successfully characterized human bronchial carcinoma cells on X-ray irradiated C3H lOTl/Z feeder layer cells. Their feeder layer culture system proved to be selective for the growth of malignant cells and all non malignant epithelial and connective tissue cells were observed to die within two days. However, we found that the irradiated confluent monolayers (cell mat) were not able to inhibit the proliferation of all normal cells, as described by Chang [3]. Therefore we used contact-inhibited confluent non-irradiated monolayers of BALB/c 3T3 cells as “CSP”. The fact that all colony formers in agar also formed colonies on CSP with a much higher
143
efficiency suggests that CSP may serve as a feeder layer for neoplastic cells which are seeded on them. This is major advantage of CSP over agar culture which has limits in clinical application [ 12,161. SV-3Y 1 cells and E l A-3Y 1 were unique in that they formed colonies on CSP but not in agar.These cells are apparently abnormal with respect to reduced serum requirements high saturation density and expression of viral antigens [ 14,181. These cells are considered to be incomplete transformants and it is noteworthy that the CSP were able to detect such cells. Carcinogenesis has been generally thought to occur in a step by step fashion. In this context, CSP may also become a useful tool in screening and selecting these low malignant cells which may be developing into fully malignant cells for studying cancer.
6
7
8
9
10
11
12
Acknowledgement The authors wish to thank Prof. G. Kimura, Department of Virology, Medical Institute of Bioregulation, Kyushu University for providing the 3Yl and the transformed cell lines; B.T. Quinn, Kyushu University for valuable comments on the manuscript. References Abercrombie, M. (1979) Contact nancy. Nature, 281,259-262.
13
14
15 inhibition
and
malig-
Banet, J.C., Crawfold, B.D., Mixter, L.O., Schechman, L.M., Ts’o, P.O.P. and Pofiack, R. (1979) Correlation in vitro growth properties and tumorigenicity of Syrian hamster cell lines. Cancer Res., 39, 1504-1510.
16
Chang, C., Trosko, J.E., El-Fouly, H.H., GibsonD’Ambrosio, R.E. and D’Ambrosio, S.M. (1987) Contact insensivity of a subpopulation of normal human fetal epithelial cells and of human carcinoma cell lines. Cancer Res., 47, 1634-1645. Holly, R.W. and Kiernan, J.W. (1968) “Contact inhibition”, of cell division in 3T3 cells. Proc. Natl. Acad. Sci. USA, 60,300-304.
17
Kakunaga, T. (1973) A quantitative system for assay of malignant transformation by chemical carcinogens using a clone derived from BALB/3T3. Int. J. Cancer, 12,463473.
Kimura, G., Itagaki, A. and Summers, J. (1975) Rat cell line 3Y 1 and its virogenic polyoma-and SV40transformed derivatives. fnt. J. Cancer, 26,435-442. Klein, J.C., Zurcher, C. and van Bekkum, D.W. (1987) Differential behavior of human bronchial carcinoma cells in culture. Cancer Res., 47,3251-3258. MacPherson, 1. and Montagnier, L. (1964) Agar suspension culture for the selective assay of cells transformed by polyoma virus. Virology, 23,291-294. Matsuoka, H., Sugimachi, K., Ueo, H.. Kuwano, H., Nakano, S. and Nakayama, M. (1987) Sex hormone response of newly established squamous cell line derived from clinical esophageal carcinoma. Cancer 4130-4140. Matsuoka, H., Sugimachi, K. and Yano. Chemosensitivity testing for anticancer drugs
Res.,
47,
K. (1987) using con-
fluent monolayers. Med. Sci. Res., 15, 837-838. Matsuoka. H., Sugimachi, K., Yano, K., Kido, Y ., Shirabe. K. and Mitsudomi, T. (1988) Influence of oestradiol upon contact-sensitive plates of confluent BALB/c 3T3 cell monolayers. Med. Sci. Res., 16, 1017-1018. Pavelic, Z.P., Slocum, H.K., Rustum, Y .M., Creaven, P.J., Novak, N.J., Karakousis, C., Takita, H. and Mittelman, A. (1980) Growth of cell colonies in soft agar from biopsies of different human solid tumors. Cancer Res., 40, 4151-4158. Reznikoff, C.A., Brankow, D.W. and Heidelberger, C. (1973) Establishment and characterization of a clonal line of C3H mouse embryo cell sensitive to post confluence inhibition of division. Cancer Res., 33, 3321-3328. Shimura, H., Ohtsu, M., Matsuzaki, A., Mitsudomi. T., Onodera, K. and Kimura. G. (1988) Selective cytotoxicity of phospholipids and diacyl glycerols to rat 3Y 1 fibroblasts transformed by adenovirus type 12 or its ElA gene. Cancer Res., 48,578-583. Todaro, G.J. and Green, H. (1963) Quantitative studies of the growth of mouse embryo cells in culture and their development into established lines. J. Cell Biol., 17, 299313. Von Hoff, D.D., Weisenthal, L.M., Ihde, D.C., Mathews, M.J., Layard, M. and Makuch, R. (1981) Growth of lung cancer colonies from bronchoscopy washings. Cancer, 48, 400-403. Yamasaki. H., Hollstein, M., Mesnil, M.. Martel. N. and Aguelon, A.M. (1987) Selective lack of inter cellular communication between transformed and nontransformed cells as a common property of chemical and oncogene transformation of BALB/c 3T3 cells. Cancer Res., 47.
18
5658-5664. Zaitsu. H., Tanaka, H., Mitsudomi, T., Matsuzaki, A., Ohtsu. M. and Kimura, G. (1988) Differences in proliferation properties among sublines of rat 3Yl fibroblasts transformed by various agents in vitro. Biomed. 197.
Res., 9, 181-