The Journal of Dermatology Vol. 17: 143-148, 1990
Human Squamous Cell Carcinoma from Skin: Establishment and Characterization of a New Cell Line (HSC-5) Yutaka Hozumi, Shigeo Kondo, Takako Shimoura and Kazuo Aso Abstract
A new cell line, designated as HSC-5 and derived from human skin squamous cell carcinoma (SCC), has been established in vitro and maintained proliferative in continuous tissue culture for over two years. The cells grow in a monolayer in vitro and have anaplastic epithelioid features. The doubling time was about 35hr at the 30th passage. Chromosome analysis showed hypotetraploidy with a modal number of 76. A trial of transplantation of the cultured cells into nude mice was not successful. Analysis of cytokeratins from HSC-5 by two-dimensional gel electrophoresis revealed polypeptides No.5, 8, 13, 18 and 19. The cell line is available to other investigators.
Key words:
human skin squamous cell carcinoma; cell line; cytokeratin; HSC-5
Introduction
Most of the SCC cell lines existing at present are derived from cervix, mouth, larynx, esophagus, urinary bladder or lung (1-7). Establishments of cell lines from human skin SCC have been rare (8-11). Since human skin carcinomas are easily observed, diagnosed, and treated with a relatively good prognosis compared with other internal tumors, they may be good material for studying human tumors immunologically, biologically, and therapeutically. We have succeeded in establishing a new cell line, named HSC-5 and derived from human skin, and we report its characteristics in this paper. Materials and Methods Materials Dulbecco's modified essential medium (DMEM), Received October 2, 1989; accepted for publication November 10, 1989. Department of Dermatology, Yamagata University School of Medicine, Iida-nishi, Yamagata-city, Yamagata 990-23,]apan. Reprint requests to: Yutaka Hozumi, Department of Dermatology, Yamagata University School of Medicine, Iida-nishi, Yamagata-city, Yamagata 990-23,]apan.
Iscove's modification of Dulbecco's medium (IMDM), Ham's F12, Hank's MEM, 199, PRMI 1640, and fetal calf serum (FCS) were purchased from Dainihon Chemical Company, Osaka, japan. GIT medium was purchased from Nippon Chemical Company, Tokyo. Plastic culture dishes are Falcon (Becton Dickinson). 3T3 cells derived from Swiss Albino were purchased from Dainihon Chemical Company, Osaka, japan. Nude mice (BALB/c) were purchased from japan Clear, Tokyo. EDTA and trypsin (1:250) were obtained from Wako Chemical Company, Osaka, Japan. QY-l (infiltrating media) and Quetol 812 (embedding media) were obtained from Nisshin EM Company, Tokyo. HSC-l cell line: This cell line was established in our laboratory from human skin SCC (9). Methods Primary culture: An ulcerated tumor on the upper part of the right auricle of a 75-year-old japanese man (Fig. I-A) was surgically excised on 2 February, 1987, and minced into about 1 mm" portions, which were placed in 6 em plastic dishes and cultured in DMEM supplemented with 20% FCS, penicillin (100 units/ml), and streptomycin (50 ,ug/ml), in a gas phase of 95% air and 5% CO 2 , By about 5 days, cells with an epithelial pattern appeared from the tumor material. Fibroblasts surrounded the outgrowth completely by about 12 days. The culture was then
144
Hozumi et al
exposed to 0.02% EDTA in phosphate buffered saline (PBS) at pH 7.4 for 15 sec and washed by blowing PBS through a pipette to remove the fibroblasts. After removing more than 90% of the fibroblasts from the culture, 3T3 cells treated with mitomycin C (MMC) (16) were added at the rate of3 x 105 cells/6 cm dish as a feeder layer for epithelioid cell outgrowth. The same procedure was repeated twice until the first subculture was performed. Subculture: The first subculture was performed on the 29th day of the primary culture, when the sheet of epithelioid cells occupied about two-thirds of the entire surface the culture dish. Treatment with 0.02% EDTA was performed as described in 'primary culture' to remove most of the contaminating fibroblasts. Next, the culture was exposed to 0.02% EDTA-0.05% trypsin solution (1:1) for about 3 min; then cells were isolated. The isolated cells were subcultured together with MMC-treated 3T3 cells in 6 cm plastic dishes at a ratio of 1 x 105-3 X 105 cells to 3 x 105 3T3 cells. When colonies of epithelioid cells grew, pushing away 3T3 cells at the periphery and occupying nearly the entire surface of the culture dish, the same procedure was repeated for the next subculture. After the 7th subculture, 3T3 cells were omitted from the culture and only tumor cells were cultured. Tumor cells grew very slowly in the absence of 3T3 feeders. In order to stimulate growth, various media such as Ham's F12, IMDM, HMEM, 199, RPMI 1640 and GIT were tried. We found that IMDM + 20% FCS was the best Thus, from the 18th subculture the medium was changed from DMEM to IMDM. Cells were subcultured approximately every week. Subsequently, the concentration of FCS in the media was changed from 20% to 10% from the 50th passage. Electron micrography: Tumor specimens excised from the patient were fixed first with 2.5% glutaraldehyde in 2mM cacodylate buffer and 2mM osmium tetraoxide at 4°C for 2h. The specimens were then dehydrated in an ethanol series followed by QY-l and embedded in Quetol 812. For the culture cells, dishes with cells at 7 and 52 passages were added to 2.5% glutaraldehyde in 2mM cacodylate buffer at pH 7.4 for 30 min and with 2mM osmium tetraoxide at 4°C for 60 min. Then the dishes were dehydrated by passage through a graded ethanol series and embedded in Quetol 812. Ultrathin sections were stained with uranyl acetate and lead citrate and
examined under an electron microscope. Cytokeratin extraction and electrophoresis: Cultured cells were washed 3 times with PBS, followed by extraction of cytokeratins from the Triton X-I00 insoluble fraction accordingfo the methods described by Achtstaetter et al. (17). Concentration of extracted cytokeratins was estimated by absorbance at 278 nm. Cytokeratins solubilized in lysis buffer (17) were applied to SDS-PAGE and two-dimensional gel electrophoresis, using non-equilibrium pH gradient gel for the first dimension, by the methods of Laemmli (18) and O'Farrell et al. (19), respectively. Chromosome analysis: Cells were cultured in a medium containing 1 ,ug/ml colcemid for 2hr, dispersed with 0.02% EDTA-0.05% trypsin solution (1:1) in PBS, and treated with 75mM sodium citrate. They were then fixed in a mixture of methanol: acetic acid (3:1 vol/vol). A small drop of the fixed cell suspension was air dried on a cold glass slide and stained with Giemsa. More than 100 cells in metaphase were chosen under a microscope, photographed, and subjected to chromosome analysis. Mycoplasma assay: To test the contamination of mycoplasma in the cultured cells, we used the method of DNA fluorochrome staining. Cells were cultured on a coverslip. After 2-3 days, the cultured cells on the coverslip were fixed in a 1:1 mixture of methanol and acetate and stained for 30 min with Hoechst 33258 (0.05 ,ug/ml). The specimens were observed by fluorescence microscopy at 400x under oil immersion. Growth kinetics: Cell kinetics were obtained by the use of standard techniques. Briefly, 1 x 105 cells were seeded into 8 em" tissue culture dishes and the adherent cells were counted at appropriate intervals in triplicate dishes; then the doubling time of the cell population was estimated in the logarithmic growth phase. Transplantation experiment on nude mice: 12 million cells of the 45th and 54th subcultures suspended in 0.5 ml PBS were inoculated through a syringe subcutaneously into the back of two nude mice. The experiments were repeated twice.
Results Original tumor: Figure 1 shows the original tumor and its histological findings. The cells
A Cell Line from Human Skin Cancer
145
Fig. 1. 'Original tumor. (A) Picture of an ulcerated mushroom-like tumor on the upper part of the right auricle. (B) Tumor cells are polygonal in shape. Hom pearls and individual cell keratinization are frequently seen. (H & E, x320). (C) The ultrastructure of the excised tumor (x15,OOO). Bundles of tonofilaments (F), desmosomes (D), and nucleus (N) are seen.
possessed a slightly eosinophilic cytoplasm and few intercellular bridges. They were polygonal or elongate in shape; the majority were atypical with pyknotic nuclei. Horn pearls and individual cell keratinization occupied about 50 to 70% of whole cells (Fig. I-B). The electron micrograph (Fig. l-C) showed many bundles of tonofilaments and desmosomes in the cytoplasm. Histologically, this tumor was relatively differentiated. Tissue culture: Tumor cells grew steadily up to the 7th subculture in the presence of 3T3 feeder; thereafter they grew very slowly in the absence of 3T3 feeder. Therefore the culture medium was changed from DMEM to IMDM. Cells were epithelioid in shape and varied in size (Fig. 2-A). Electron micrographically, the cells had relatively rounded nuclei, tonofilaments, microvilli and desmosome-like structures (Fig. 2-B, C and D). As shown in Fig. 2-B, No evidence of keratinization, e.g. horny cells or keratohyalin granules, was obtained by electron micrography of the cultured cells. Neither viral particles nor inclusion bodies were observed, electron micrographically, in the tumor mass or in individual cultured cells. Cytokeratins extracted from HSC-l and HSC-
5: Cytokeratins extracted from HSC-5 were analysed by SDS-PAGE and two-dimensional gel electrophoresis, and the resolved polypeptides were numbered according to the cytokeratin catalog described by Moll et a1. (20). Cytokeratins from HSC-5 (86th) contained polypeptides No.5, 8, 13, 18 and 19 and traces of No.6, 14 and 17. Cytokeratins No. 1/2 and lOllI, which are expressed in keratinizing epidermis, were not detected. Cytokeratins from HSC-I (78th) contained polypeptides No. 5, 6, 14, 17 and 18 (Fig. 3-A, B and C). Chromosome analysis: The distribution of the chromosome number ranged from 21 to 119 in the cells of the 14th passage. Figure 4 shows the distribution number between 67 and 80, with a peak at 76. No distinct marker chromosomes were observed. Mycoplasma assay: Absence of mycoplasmal contamination was verified by DNA fluorochrome staining as described in Materials and Methods. Growth kinetics: The growth kinetics of HSC-5 were studied by the technique describedin Materials and Methods. At the 30th passage, the doubling time was about 35 hr (Fig. 5). Transplantation experiments: Culture cells of the 45th passage were transplanted into two
146
Hozumi et a1
Fig. 2. Cultured cells (HSC-5). (A) Cells of the 42th passage at 5 days after inoculation (phase contrast, x320). They are polygonal or elongated in shape. (phase contrast, x320). (B) (C) (D) The ultrastructure of the cultured cell (HSC-5) of the 7th (B), (C), and 52th (D) passages: (C) and (D) are in a parallel to the layer, while (B) is in a plane perpendicular; (C) shows desmosomes (D), tonofilaments (F), mitochondria (M), elongated microvilli (Mv), and nuclei (N) (xI5,OOO); (B) Piling up or multilayering of cells. (x4,OOO); (D) shows desmosome (D), tonofilament (F), ribosomes (R), and microtubles (MT). (xI5,OOO)
BALBIc nude mice. Another day, the same experiment (culture cells of the 54th passage) was repeated. None of the transplantation experiments on four nude mice was successful.
Storage of the cell line: Extra cells of HSC-5 were frozen-stored at -80°C in IMDM containing 20% FCS, 10% dimethyl sulfoxide, penicillin, and streptomycin. The survival rate of
A Cell Line from Human Skin Cancer
147
92.5k
Uk 45k 40k
25k Fig. 3. Cytokeratins pattern. (A) SDS-PAGE. Lane: standard proteins; phosphorylase A (92,500), bovine serum albumin (68,000), ovalbumin (45,000), aldorase from rabbit muscle (40,000) and chymotrypsinogen A (25,000). Lane 2: HSC-l. Lane 3: HSC-5. (B), (C) 2-dim. electrophoresis, B: HSC-5, C: HSC-l. b, p, a: standard protein; b: bovine serum albumin, p: 3 phosphoglyceric phosphokinase from yeast, a: a-actin from rabbit muscle. Number represents Moll's cytokeratin number. (%) 20 18
16
...•.
Human Squamous Cell Carcinoma from Skin: Establishment and Characterization of a New Cell Line (HSC-5) Yutaka Hozumi, Shigeo Kondo, Takako Shimoura and Kazuo Aso Abstract
A new cell line, designated as HSC-5 and derived from human skin squamous cell carcinoma (SCC), has been established in vitro and maintained proliferative in continuous tissue culture for over two years. The cells grow in a monolayer in vitro and have anaplastic epithelioid features. The doubling time was about 35hr at the 30th passage. Chromosome analysis showed hypotetraploidy with a modal number of 76. A trial of transplantation of the cultured cells into nude mice was not successful. Analysis of cytokeratins from HSC-5 by two-dimensional gel electrophoresis revealed polypeptides No.5, 8, 13, 18 and 19. The cell line is available to other investigators.
Key words:
human skin squamous cell carcinoma; cell line; cytokeratin; HSC-5
Introduction
Most of the SCC cell lines existing at present are derived from cervix, mouth, larynx, esophagus, urinary bladder or lung (1-7). Establishments of cell lines from human skin SCC have been rare (8-11). Since human skin carcinomas are easily observed, diagnosed, and treated with a relatively good prognosis compared with other internal tumors, they may be good material for studying human tumors immunologically, biologically, and therapeutically. We have succeeded in establishing a new cell line, named HSC-5 and derived from human skin, and we report its characteristics in this paper. Materials and Methods Materials Dulbecco's modified essential medium (DMEM), Received October 2, 1989; accepted for publication November 10, 1989. Department of Dermatology, Yamagata University School of Medicine, Iida-nishi, Yamagata-city, Yamagata 990-23,]apan. Reprint requests to: Yutaka Hozumi, Department of Dermatology, Yamagata University School of Medicine, Iida-nishi, Yamagata-city, Yamagata 990-23,]apan.
Iscove's modification of Dulbecco's medium (IMDM), Ham's F12, Hank's MEM, 199, PRMI 1640, and fetal calf serum (FCS) were purchased from Dainihon Chemical Company, Osaka, japan. GIT medium was purchased from Nippon Chemical Company, Tokyo. Plastic culture dishes are Falcon (Becton Dickinson). 3T3 cells derived from Swiss Albino were purchased from Dainihon Chemical Company, Osaka, japan. Nude mice (BALB/c) were purchased from japan Clear, Tokyo. EDTA and trypsin (1:250) were obtained from Wako Chemical Company, Osaka, Japan. QY-l (infiltrating media) and Quetol 812 (embedding media) were obtained from Nisshin EM Company, Tokyo. HSC-l cell line: This cell line was established in our laboratory from human skin SCC (9). Methods Primary culture: An ulcerated tumor on the upper part of the right auricle of a 75-year-old japanese man (Fig. I-A) was surgically excised on 2 February, 1987, and minced into about 1 mm" portions, which were placed in 6 em plastic dishes and cultured in DMEM supplemented with 20% FCS, penicillin (100 units/ml), and streptomycin (50 ,ug/ml), in a gas phase of 95% air and 5% CO 2 , By about 5 days, cells with an epithelial pattern appeared from the tumor material. Fibroblasts surrounded the outgrowth completely by about 12 days. The culture was then
144
Hozumi et al
exposed to 0.02% EDTA in phosphate buffered saline (PBS) at pH 7.4 for 15 sec and washed by blowing PBS through a pipette to remove the fibroblasts. After removing more than 90% of the fibroblasts from the culture, 3T3 cells treated with mitomycin C (MMC) (16) were added at the rate of3 x 105 cells/6 cm dish as a feeder layer for epithelioid cell outgrowth. The same procedure was repeated twice until the first subculture was performed. Subculture: The first subculture was performed on the 29th day of the primary culture, when the sheet of epithelioid cells occupied about two-thirds of the entire surface the culture dish. Treatment with 0.02% EDTA was performed as described in 'primary culture' to remove most of the contaminating fibroblasts. Next, the culture was exposed to 0.02% EDTA-0.05% trypsin solution (1:1) for about 3 min; then cells were isolated. The isolated cells were subcultured together with MMC-treated 3T3 cells in 6 cm plastic dishes at a ratio of 1 x 105-3 X 105 cells to 3 x 105 3T3 cells. When colonies of epithelioid cells grew, pushing away 3T3 cells at the periphery and occupying nearly the entire surface of the culture dish, the same procedure was repeated for the next subculture. After the 7th subculture, 3T3 cells were omitted from the culture and only tumor cells were cultured. Tumor cells grew very slowly in the absence of 3T3 feeders. In order to stimulate growth, various media such as Ham's F12, IMDM, HMEM, 199, RPMI 1640 and GIT were tried. We found that IMDM + 20% FCS was the best Thus, from the 18th subculture the medium was changed from DMEM to IMDM. Cells were subcultured approximately every week. Subsequently, the concentration of FCS in the media was changed from 20% to 10% from the 50th passage. Electron micrography: Tumor specimens excised from the patient were fixed first with 2.5% glutaraldehyde in 2mM cacodylate buffer and 2mM osmium tetraoxide at 4°C for 2h. The specimens were then dehydrated in an ethanol series followed by QY-l and embedded in Quetol 812. For the culture cells, dishes with cells at 7 and 52 passages were added to 2.5% glutaraldehyde in 2mM cacodylate buffer at pH 7.4 for 30 min and with 2mM osmium tetraoxide at 4°C for 60 min. Then the dishes were dehydrated by passage through a graded ethanol series and embedded in Quetol 812. Ultrathin sections were stained with uranyl acetate and lead citrate and
examined under an electron microscope. Cytokeratin extraction and electrophoresis: Cultured cells were washed 3 times with PBS, followed by extraction of cytokeratins from the Triton X-I00 insoluble fraction accordingfo the methods described by Achtstaetter et al. (17). Concentration of extracted cytokeratins was estimated by absorbance at 278 nm. Cytokeratins solubilized in lysis buffer (17) were applied to SDS-PAGE and two-dimensional gel electrophoresis, using non-equilibrium pH gradient gel for the first dimension, by the methods of Laemmli (18) and O'Farrell et al. (19), respectively. Chromosome analysis: Cells were cultured in a medium containing 1 ,ug/ml colcemid for 2hr, dispersed with 0.02% EDTA-0.05% trypsin solution (1:1) in PBS, and treated with 75mM sodium citrate. They were then fixed in a mixture of methanol: acetic acid (3:1 vol/vol). A small drop of the fixed cell suspension was air dried on a cold glass slide and stained with Giemsa. More than 100 cells in metaphase were chosen under a microscope, photographed, and subjected to chromosome analysis. Mycoplasma assay: To test the contamination of mycoplasma in the cultured cells, we used the method of DNA fluorochrome staining. Cells were cultured on a coverslip. After 2-3 days, the cultured cells on the coverslip were fixed in a 1:1 mixture of methanol and acetate and stained for 30 min with Hoechst 33258 (0.05 ,ug/ml). The specimens were observed by fluorescence microscopy at 400x under oil immersion. Growth kinetics: Cell kinetics were obtained by the use of standard techniques. Briefly, 1 x 105 cells were seeded into 8 em" tissue culture dishes and the adherent cells were counted at appropriate intervals in triplicate dishes; then the doubling time of the cell population was estimated in the logarithmic growth phase. Transplantation experiment on nude mice: 12 million cells of the 45th and 54th subcultures suspended in 0.5 ml PBS were inoculated through a syringe subcutaneously into the back of two nude mice. The experiments were repeated twice.
Results Original tumor: Figure 1 shows the original tumor and its histological findings. The cells
A Cell Line from Human Skin Cancer
145
Fig. 1. 'Original tumor. (A) Picture of an ulcerated mushroom-like tumor on the upper part of the right auricle. (B) Tumor cells are polygonal in shape. Hom pearls and individual cell keratinization are frequently seen. (H & E, x320). (C) The ultrastructure of the excised tumor (x15,OOO). Bundles of tonofilaments (F), desmosomes (D), and nucleus (N) are seen.
possessed a slightly eosinophilic cytoplasm and few intercellular bridges. They were polygonal or elongate in shape; the majority were atypical with pyknotic nuclei. Horn pearls and individual cell keratinization occupied about 50 to 70% of whole cells (Fig. I-B). The electron micrograph (Fig. l-C) showed many bundles of tonofilaments and desmosomes in the cytoplasm. Histologically, this tumor was relatively differentiated. Tissue culture: Tumor cells grew steadily up to the 7th subculture in the presence of 3T3 feeder; thereafter they grew very slowly in the absence of 3T3 feeder. Therefore the culture medium was changed from DMEM to IMDM. Cells were epithelioid in shape and varied in size (Fig. 2-A). Electron micrographically, the cells had relatively rounded nuclei, tonofilaments, microvilli and desmosome-like structures (Fig. 2-B, C and D). As shown in Fig. 2-B, No evidence of keratinization, e.g. horny cells or keratohyalin granules, was obtained by electron micrography of the cultured cells. Neither viral particles nor inclusion bodies were observed, electron micrographically, in the tumor mass or in individual cultured cells. Cytokeratins extracted from HSC-l and HSC-
5: Cytokeratins extracted from HSC-5 were analysed by SDS-PAGE and two-dimensional gel electrophoresis, and the resolved polypeptides were numbered according to the cytokeratin catalog described by Moll et a1. (20). Cytokeratins from HSC-5 (86th) contained polypeptides No.5, 8, 13, 18 and 19 and traces of No.6, 14 and 17. Cytokeratins No. 1/2 and lOllI, which are expressed in keratinizing epidermis, were not detected. Cytokeratins from HSC-I (78th) contained polypeptides No. 5, 6, 14, 17 and 18 (Fig. 3-A, B and C). Chromosome analysis: The distribution of the chromosome number ranged from 21 to 119 in the cells of the 14th passage. Figure 4 shows the distribution number between 67 and 80, with a peak at 76. No distinct marker chromosomes were observed. Mycoplasma assay: Absence of mycoplasmal contamination was verified by DNA fluorochrome staining as described in Materials and Methods. Growth kinetics: The growth kinetics of HSC-5 were studied by the technique describedin Materials and Methods. At the 30th passage, the doubling time was about 35 hr (Fig. 5). Transplantation experiments: Culture cells of the 45th passage were transplanted into two
146
Hozumi et a1
Fig. 2. Cultured cells (HSC-5). (A) Cells of the 42th passage at 5 days after inoculation (phase contrast, x320). They are polygonal or elongated in shape. (phase contrast, x320). (B) (C) (D) The ultrastructure of the cultured cell (HSC-5) of the 7th (B), (C), and 52th (D) passages: (C) and (D) are in a parallel to the layer, while (B) is in a plane perpendicular; (C) shows desmosomes (D), tonofilaments (F), mitochondria (M), elongated microvilli (Mv), and nuclei (N) (xI5,OOO); (B) Piling up or multilayering of cells. (x4,OOO); (D) shows desmosome (D), tonofilament (F), ribosomes (R), and microtubles (MT). (xI5,OOO)
BALBIc nude mice. Another day, the same experiment (culture cells of the 54th passage) was repeated. None of the transplantation experiments on four nude mice was successful.
Storage of the cell line: Extra cells of HSC-5 were frozen-stored at -80°C in IMDM containing 20% FCS, 10% dimethyl sulfoxide, penicillin, and streptomycin. The survival rate of
A Cell Line from Human Skin Cancer
147
92.5k
Uk 45k 40k
25k Fig. 3. Cytokeratins pattern. (A) SDS-PAGE. Lane: standard proteins; phosphorylase A (92,500), bovine serum albumin (68,000), ovalbumin (45,000), aldorase from rabbit muscle (40,000) and chymotrypsinogen A (25,000). Lane 2: HSC-l. Lane 3: HSC-5. (B), (C) 2-dim. electrophoresis, B: HSC-5, C: HSC-l. b, p, a: standard protein; b: bovine serum albumin, p: 3 phosphoglyceric phosphokinase from yeast, a: a-actin from rabbit muscle. Number represents Moll's cytokeratin number. (%) 20 18
16
...•.
