Spontaneous Maturation and Differentiation of B16 Melanoma Cells in Culture John W. Kreider
3, 4
and Mary E. Schmoyer
1, 2
5, 6
Tumors are composed of functionally and morphologically heterogeneous populations of neoplastic cells. These cell populations are in different stages of replication, maturation, and death. The growth rate of a given tumor is directly proportional to the degree by which the rate of new cell production exceeds the rate of cell death. In some tumors, cell death is primarily the result of local ischemia, immune response, or mitotic accidents. However, tumors that retain phenotypic properties of the tissue of origin often continue relatively orderly patterns of differentiation and maturation. This results in the production of significant numbers of mature, "end-stage" cells with the reduced or absent capacity for further proliferation. The rate of neoplastic cell maturation can be an important factor in the regulation of tumor growth (1) because the mature tumor cells a) are produced at the expense of the proliferating population, b) may have irreversibly lost proliferative capacity, and c) often possess a limi ted life-span. Microscopic sections of tumors often demonstrate a peripheral region of actively replicating cells with inner layers of cells in different maturational stages. The innermost portions of the tumor contain fully mature cells, whereas the core is sometimes composed of necrotic tissue. This maturational "stratification" is especially conspicuous in the keratinous cysts of well-differentiated squamous cell carcinomas (2) and melanomas (3). The maturational stratification of neoplastic cells is highly analogous to that seen in normal tissues, e.g., epidermis (4) and colon epithelium (5). It seems reasonable to conclude that similar mechanisms are operative in the maturation of normal and neoplastic cells. However, the possibility cannot be excluded that neoplastic cell maturation may somehow result from an inhibition of cell proliferation, e.g., moderate ischemia and resultant anoxia and/or nutrient deprivation. The putative inverse relationship between cell proliferation and differentiation is not absolute and has little operational value (6). Cell differentiation and maturation are primarily the consequences of local environmental factors. Since it is
difficult to dissociate these factors in vivo, cell cultures are useful tools in the development of our understanding of the mechanisms involved. Malignant melanoma cells are especially appropriate for the study of neoplastic cell maturation and differentiation. These cells are readily cultured in vitro, are easily don ed, and produce highly malignant tumors when inoculated into suitable hosts. Melanin, the special product of melanocytes, is a conspicuous marker of cytodifferentiation and is often visible with the unaided eye. The biosynthetic pathway is well characterized and is principally the result of the interaction of a single enzyme (tyrosinase, E.C. 1.10.3.1) with the principal substrate L-tyrosine (7). Our purpose was to characterize the changes that occur in malignant melanocytes as they differentiated and matured in monolayer cultures. As a function of the age of the culture, we studied the interrelationships of population density, melanin production, cell size and melanin content, morphology, replicative activity, and the reversibility of melanoma cell maturation under the influence of transfer to a reduced population density in vitro or after transplantation to a syngeneic host. MATERIALS AND METHODS
Cells, media, and reagents.-BI6 melanoma is a spontaneous transplantable tumor syngeneic to C57BLJ6J mice and available from The Jackson Laboratory, Bar Harbor, Maine. We used a strain of Bl6 melanoma (BI6C3) which has been successively subdoned in our laboratory on three separate occasions. Each subdone was selected for maximal stability of the timing and intensity of melanogenesis. The cells were routinely cultured in supplemented medium in 16-ounce prescription or roller bottles as described in (8). In all experiments, the final concentration of L-tyrosine was raised to 2.0 mM (8). This level of L-tyrosine does not inhibit cell replication and is necessary for maximal melanogenic activity. Clump-free cell suspensions were prepared from monolayers for subculture and for experiments with the aid of Versene (0.25%) in phosphate-buffered saline (PBS). Cells were enumerated with a Coulter counter Model B (Coulter Electronics, Hialeah, Fla.) or, alternatively, try pan blue-excluding cells were evaluated in a hemacytometer. Incubation conditions in experiments.-Except where otherwise specified, lOX 106 trypan blue-excluding cells in 5 ml medium were planted into 60-mm petri dishes (#3002, Falcon Plastics, Oxnard, Cali f.). The cultures were incubated at 36° C in a humidified 95% air-5% CO 2 Received February 13, 1975; accepted May 21, 1975. Supported by Public Health Service research grant 2R01CAII097 from the National Cancer Institute and funds from the Jake Gittlen Golf Tournament. 3 Recipient of Career Development Award 7·K4-CA38809 from the National Cancer Institute. 4 Departments of Pathology and Microbiology, College of Medi· cine, The Milton S. Hershey Medical Center, The Pennsylvania State University, Hershey, Pa. 17033. 5 Department of Pathology, The Milton S. Hershey Medical Center. 6 We acknowledge the technical assistance of Nancy Lengle. 1
2
JOURNAL OF THE NATIONAL CANCER INSTITUTE, VOL. 55, NO.3, SEPTEMBER 1975
641
Downloaded from http://jnci.oxfordjournals.org/ at Dalhousie University on July 9, 2014
SUMMARY-SI6 melanoma tumors and cultures are composed of cells with different melanin contents and replicative activIties. The hypothesis was tested in vitro that these various cells constituted a poputation in the process of differentiation and maturation. Early cultures were predominantly composed of small, amelanotic cells with high replicative activity. Older cultures contained mostly larger and heavily melanotic cells with little or no replicative activity. Replicative activity, as measured by the uptake of tritiated thymidine, was inversely proportional to cell size and melanin content. Colony-forming ability was impaired if the original cells were melanotic. TumorigeniCity was unaffected except in very old (9-day) cultures. Our results support the concept th'8t malignant melanocytes undergo a sequence of developmental changes which eventuates in the production of mature cells characterized by enlargement, elevated melanin content, and reduced replicative and colony-forming capacity.-J Natl Cancer Inst 55: 641-647, 1975.
642
KREIDER AND SCHMOYER
Melanin assay.-Melanin was quantitated by a fluorimetric technique described in (10). The amount of melanin present was calculated in arbitrary fluorescence units. Melanin assayed in cultures was the total amount of melanin produced by that culture during the entire period since planting. The amount of melanin reported was the amount recovered from the cell sheet and culture medium, as well as that in the medium removed from that particular culture at the periodic medium replacements. Platting efficiency and tumorigenicity assays.-Plating efficiency is a measure of the replicative potential of a measured aliquot of a cell suspension. In contrast to the assessment of a colony-forming ability (as previously described), the plating efficiency assay does not permit any prior determination of individual cell morphology. Plating efficiency assays in BI6 melanoma are a more sensitive assessment of cell replicative potential than tumorigenicity because of the greater numbers of cells required to form tumors (l03-1Q4 dye-excluding cells). Cells were planted at an initial concentration of l.0 X 106 trypan blue-excluding cells per dish. At intervals of I, 3, 5, 7, and 9 days, the monolayers were washed once with PBS and the cells suspended with 0.25% Versene in PBS. The cell suspension was washed, counted, adjusted to appropriate concentrations, and assayed for plating efficiency and tumorigenicity. A total of 200 cells were planted into 60-mm dishes with 5 ml preequilibrated medium. After 7-15 days' growth, the dishes were washed, fixed, and stained with Giemsa, and the numbers of colonies enumerated. Plating efficiency was calculated as the number of resulting colonies divided by the number of cells inoculated into the dish (X 100). Trypan blue-excluding cells (l.Ox 106 ) per 25,...1 of serum-free medium obtained from the original cell suspensions were inoculated into the left hind footpads of 4- to 5-week-old C57BL/6J male mice. The animals were examined weekly and scored for incidence of tumors at the inoculation site. RESULTS Cell Population Density and Melanin Production
Cell population density was defined as the number of cells that remained firmly attached to the vessel substrate after three washes with PBS but which were suspended with 0.25% Versene. Within 2 days, cultures initiated with 1.0 X 10 6 cells doubled in population density (textfig. I). Also at 2 days, the amount of melanin recoverable from cells and medium decreased by about 50%. Within about I week of culture, the cell population temporarily stabilized at 3.0x 106 cells/culture at confluency, but later increased to 4.0 X 106 cells, possibly due to the stimulus of medium replacement at 10 days. This population density was not maintained, and equilibrium was achieved at 3.0X106 cells. Replacement of medium at 15 days did not affect the cell population. Melanin accumulated at a relatively steady rate in the postconfluency period. Relationship of Cell Morphology to Replicative Activity
Melanoma cells in cultures initiated with 1.0 X 10 6 cells underwent a consistent pattern of morphologic changes that were associated with replicative activity. When first planted, the predominant cell type in B16 melanoma cultures was comparatively small and amelanotic. It had a scanty cytoplasm and a high frequency of nuclear 3H-TDR uptake and often formed mitotic figures. These cells, which we termed a cells (figs. 1, 2), diminished in
Downloaded from http://jnci.oxfordjournals.org/ at Dalhousie University on July 9, 2014
atmosphere. Medium was replaced at approximately 4- to 5-day intervals. Morphologic observations.-Cover-slip cultures were washed three times with PBS and fixed in a 3: I (vol/vol) ethanol:acetic acid solution; the DNA was stained with the Feulgen-auramine 0 method (9). The preparations were examined and photographed with a Zeiss fluorescence microscope equipped with an HBO-200 lamp, dark-field condenser, exciter filter BG-12, and barrier filter 47. Under these conditions, the nucleus was goldenyellow and the melanin granules were a brilliant cobalt blue. Autoradiography.-Living cells were centrifuged directly onto glass slides with the Cytocentrifuge (Shandon Southern Instruments, Inc., Sewickley, Pa.). The slides were washed three times in PBS, fixed for I hour in a 3: I ethanol: acetic acid solution, and extracted overnight in multiple changes of 70% ethanol. Slides were then air dried, later dipped in liquid photographic emulsion (NTB; Eastman Kodak, Rochester, N.Y.), exposed for I month, and then developed in Dektol (Eastman Kodak). Measurement of cell size and estimation of melanin content.-Petri dishes (60-mm) were planted with 1.0 X 106 try pan blue-excluding cells in 5 ml medium. On days 1 or 6, cultures were pulsed with tritiated thymidine (3H-TDR) (1.0 ,...Ci/ml) for I hour. The monolayer was then washed three times with PBS, and the cells were suspended with 0.25% Versene. The suspensions were adjusted to l.Ox 106 cells/ml and centrifuged onto glass slides with the Cytocentrifuge. The glass slides were processed for autoradiography. Unstained cells were examined under phase optics. The nuclear and cytoplasmic diameters were measured with a linear ocular scale calibrated with a stage micrometer and the surface areas calcula ted (= 7fr2). Nuclei wi th greater than 5 grains were considered labeled with 3H-TDR. The cell was subjectively classified with respect to its melanin content on an arbitrary scale of 0 to 4 (O=no melanin; 1 = < 12 melanin granules/cell; 2= 12100 melanin granules/cell; 3 = heavily melanotic cytoplasm but perinuclear space clear of melanin; and 4 = heavily melanotic cytoplasm with obscuration of nucleus). Colony-forming ability of individual cells in different maturational stages.-Measurement of the colony-forming ability of individual, identified cells permits an assessment of the relationship of the original cell's morphology to its potential ability to form colonies. Petri dishes (60-mm) were planted with 1.25 X 10 5 try pan blue-excluding cells and incubated overnight, and the medium was replaced on days 1 and 5 of the experiment. At days 5-9, dilute suspensions of cells from the petri dishes in medium (20 cells/25 ,...1) were inoculated into the wells of Microtest II culture plates (#3040; Falcon Plastics). After we allowed 3 hours for the cells to attach to the bottom of the wells, the medium was replaced; in each well a single cell, cleary separated from its neighbors, was marked on the underside of the dish by encirclement with a Trident Cell Finder (Aloe Products, St. Louis, Mo.) in an inverted phase-contrast microscope. Single cells were examined and classified on the basis of melanin content with the subjective scale previously described. Cell size was obtained with the ocular micrometer. Cells were then allowed to grow for 7 additional days. The number of cells developing from the original identified cell were counted.
643
SPONTANEOUS DIFFERENTIATION OF B16 MELANOMA
DAY I
•
20.0
DAY 6
18.0
MELANIN PRODUCTION
3, 4
and Mary E. Schmoyer
1, 2
5, 6
Tumors are composed of functionally and morphologically heterogeneous populations of neoplastic cells. These cell populations are in different stages of replication, maturation, and death. The growth rate of a given tumor is directly proportional to the degree by which the rate of new cell production exceeds the rate of cell death. In some tumors, cell death is primarily the result of local ischemia, immune response, or mitotic accidents. However, tumors that retain phenotypic properties of the tissue of origin often continue relatively orderly patterns of differentiation and maturation. This results in the production of significant numbers of mature, "end-stage" cells with the reduced or absent capacity for further proliferation. The rate of neoplastic cell maturation can be an important factor in the regulation of tumor growth (1) because the mature tumor cells a) are produced at the expense of the proliferating population, b) may have irreversibly lost proliferative capacity, and c) often possess a limi ted life-span. Microscopic sections of tumors often demonstrate a peripheral region of actively replicating cells with inner layers of cells in different maturational stages. The innermost portions of the tumor contain fully mature cells, whereas the core is sometimes composed of necrotic tissue. This maturational "stratification" is especially conspicuous in the keratinous cysts of well-differentiated squamous cell carcinomas (2) and melanomas (3). The maturational stratification of neoplastic cells is highly analogous to that seen in normal tissues, e.g., epidermis (4) and colon epithelium (5). It seems reasonable to conclude that similar mechanisms are operative in the maturation of normal and neoplastic cells. However, the possibility cannot be excluded that neoplastic cell maturation may somehow result from an inhibition of cell proliferation, e.g., moderate ischemia and resultant anoxia and/or nutrient deprivation. The putative inverse relationship between cell proliferation and differentiation is not absolute and has little operational value (6). Cell differentiation and maturation are primarily the consequences of local environmental factors. Since it is
difficult to dissociate these factors in vivo, cell cultures are useful tools in the development of our understanding of the mechanisms involved. Malignant melanoma cells are especially appropriate for the study of neoplastic cell maturation and differentiation. These cells are readily cultured in vitro, are easily don ed, and produce highly malignant tumors when inoculated into suitable hosts. Melanin, the special product of melanocytes, is a conspicuous marker of cytodifferentiation and is often visible with the unaided eye. The biosynthetic pathway is well characterized and is principally the result of the interaction of a single enzyme (tyrosinase, E.C. 1.10.3.1) with the principal substrate L-tyrosine (7). Our purpose was to characterize the changes that occur in malignant melanocytes as they differentiated and matured in monolayer cultures. As a function of the age of the culture, we studied the interrelationships of population density, melanin production, cell size and melanin content, morphology, replicative activity, and the reversibility of melanoma cell maturation under the influence of transfer to a reduced population density in vitro or after transplantation to a syngeneic host. MATERIALS AND METHODS
Cells, media, and reagents.-BI6 melanoma is a spontaneous transplantable tumor syngeneic to C57BLJ6J mice and available from The Jackson Laboratory, Bar Harbor, Maine. We used a strain of Bl6 melanoma (BI6C3) which has been successively subdoned in our laboratory on three separate occasions. Each subdone was selected for maximal stability of the timing and intensity of melanogenesis. The cells were routinely cultured in supplemented medium in 16-ounce prescription or roller bottles as described in (8). In all experiments, the final concentration of L-tyrosine was raised to 2.0 mM (8). This level of L-tyrosine does not inhibit cell replication and is necessary for maximal melanogenic activity. Clump-free cell suspensions were prepared from monolayers for subculture and for experiments with the aid of Versene (0.25%) in phosphate-buffered saline (PBS). Cells were enumerated with a Coulter counter Model B (Coulter Electronics, Hialeah, Fla.) or, alternatively, try pan blue-excluding cells were evaluated in a hemacytometer. Incubation conditions in experiments.-Except where otherwise specified, lOX 106 trypan blue-excluding cells in 5 ml medium were planted into 60-mm petri dishes (#3002, Falcon Plastics, Oxnard, Cali f.). The cultures were incubated at 36° C in a humidified 95% air-5% CO 2 Received February 13, 1975; accepted May 21, 1975. Supported by Public Health Service research grant 2R01CAII097 from the National Cancer Institute and funds from the Jake Gittlen Golf Tournament. 3 Recipient of Career Development Award 7·K4-CA38809 from the National Cancer Institute. 4 Departments of Pathology and Microbiology, College of Medi· cine, The Milton S. Hershey Medical Center, The Pennsylvania State University, Hershey, Pa. 17033. 5 Department of Pathology, The Milton S. Hershey Medical Center. 6 We acknowledge the technical assistance of Nancy Lengle. 1
2
JOURNAL OF THE NATIONAL CANCER INSTITUTE, VOL. 55, NO.3, SEPTEMBER 1975
641
Downloaded from http://jnci.oxfordjournals.org/ at Dalhousie University on July 9, 2014
SUMMARY-SI6 melanoma tumors and cultures are composed of cells with different melanin contents and replicative activIties. The hypothesis was tested in vitro that these various cells constituted a poputation in the process of differentiation and maturation. Early cultures were predominantly composed of small, amelanotic cells with high replicative activity. Older cultures contained mostly larger and heavily melanotic cells with little or no replicative activity. Replicative activity, as measured by the uptake of tritiated thymidine, was inversely proportional to cell size and melanin content. Colony-forming ability was impaired if the original cells were melanotic. TumorigeniCity was unaffected except in very old (9-day) cultures. Our results support the concept th'8t malignant melanocytes undergo a sequence of developmental changes which eventuates in the production of mature cells characterized by enlargement, elevated melanin content, and reduced replicative and colony-forming capacity.-J Natl Cancer Inst 55: 641-647, 1975.
642
KREIDER AND SCHMOYER
Melanin assay.-Melanin was quantitated by a fluorimetric technique described in (10). The amount of melanin present was calculated in arbitrary fluorescence units. Melanin assayed in cultures was the total amount of melanin produced by that culture during the entire period since planting. The amount of melanin reported was the amount recovered from the cell sheet and culture medium, as well as that in the medium removed from that particular culture at the periodic medium replacements. Platting efficiency and tumorigenicity assays.-Plating efficiency is a measure of the replicative potential of a measured aliquot of a cell suspension. In contrast to the assessment of a colony-forming ability (as previously described), the plating efficiency assay does not permit any prior determination of individual cell morphology. Plating efficiency assays in BI6 melanoma are a more sensitive assessment of cell replicative potential than tumorigenicity because of the greater numbers of cells required to form tumors (l03-1Q4 dye-excluding cells). Cells were planted at an initial concentration of l.0 X 106 trypan blue-excluding cells per dish. At intervals of I, 3, 5, 7, and 9 days, the monolayers were washed once with PBS and the cells suspended with 0.25% Versene in PBS. The cell suspension was washed, counted, adjusted to appropriate concentrations, and assayed for plating efficiency and tumorigenicity. A total of 200 cells were planted into 60-mm dishes with 5 ml preequilibrated medium. After 7-15 days' growth, the dishes were washed, fixed, and stained with Giemsa, and the numbers of colonies enumerated. Plating efficiency was calculated as the number of resulting colonies divided by the number of cells inoculated into the dish (X 100). Trypan blue-excluding cells (l.Ox 106 ) per 25,...1 of serum-free medium obtained from the original cell suspensions were inoculated into the left hind footpads of 4- to 5-week-old C57BL/6J male mice. The animals were examined weekly and scored for incidence of tumors at the inoculation site. RESULTS Cell Population Density and Melanin Production
Cell population density was defined as the number of cells that remained firmly attached to the vessel substrate after three washes with PBS but which were suspended with 0.25% Versene. Within 2 days, cultures initiated with 1.0 X 10 6 cells doubled in population density (textfig. I). Also at 2 days, the amount of melanin recoverable from cells and medium decreased by about 50%. Within about I week of culture, the cell population temporarily stabilized at 3.0x 106 cells/culture at confluency, but later increased to 4.0 X 106 cells, possibly due to the stimulus of medium replacement at 10 days. This population density was not maintained, and equilibrium was achieved at 3.0X106 cells. Replacement of medium at 15 days did not affect the cell population. Melanin accumulated at a relatively steady rate in the postconfluency period. Relationship of Cell Morphology to Replicative Activity
Melanoma cells in cultures initiated with 1.0 X 10 6 cells underwent a consistent pattern of morphologic changes that were associated with replicative activity. When first planted, the predominant cell type in B16 melanoma cultures was comparatively small and amelanotic. It had a scanty cytoplasm and a high frequency of nuclear 3H-TDR uptake and often formed mitotic figures. These cells, which we termed a cells (figs. 1, 2), diminished in
Downloaded from http://jnci.oxfordjournals.org/ at Dalhousie University on July 9, 2014
atmosphere. Medium was replaced at approximately 4- to 5-day intervals. Morphologic observations.-Cover-slip cultures were washed three times with PBS and fixed in a 3: I (vol/vol) ethanol:acetic acid solution; the DNA was stained with the Feulgen-auramine 0 method (9). The preparations were examined and photographed with a Zeiss fluorescence microscope equipped with an HBO-200 lamp, dark-field condenser, exciter filter BG-12, and barrier filter 47. Under these conditions, the nucleus was goldenyellow and the melanin granules were a brilliant cobalt blue. Autoradiography.-Living cells were centrifuged directly onto glass slides with the Cytocentrifuge (Shandon Southern Instruments, Inc., Sewickley, Pa.). The slides were washed three times in PBS, fixed for I hour in a 3: I ethanol: acetic acid solution, and extracted overnight in multiple changes of 70% ethanol. Slides were then air dried, later dipped in liquid photographic emulsion (NTB; Eastman Kodak, Rochester, N.Y.), exposed for I month, and then developed in Dektol (Eastman Kodak). Measurement of cell size and estimation of melanin content.-Petri dishes (60-mm) were planted with 1.0 X 106 try pan blue-excluding cells in 5 ml medium. On days 1 or 6, cultures were pulsed with tritiated thymidine (3H-TDR) (1.0 ,...Ci/ml) for I hour. The monolayer was then washed three times with PBS, and the cells were suspended with 0.25% Versene. The suspensions were adjusted to l.Ox 106 cells/ml and centrifuged onto glass slides with the Cytocentrifuge. The glass slides were processed for autoradiography. Unstained cells were examined under phase optics. The nuclear and cytoplasmic diameters were measured with a linear ocular scale calibrated with a stage micrometer and the surface areas calcula ted (= 7fr2). Nuclei wi th greater than 5 grains were considered labeled with 3H-TDR. The cell was subjectively classified with respect to its melanin content on an arbitrary scale of 0 to 4 (O=no melanin; 1 = < 12 melanin granules/cell; 2= 12100 melanin granules/cell; 3 = heavily melanotic cytoplasm but perinuclear space clear of melanin; and 4 = heavily melanotic cytoplasm with obscuration of nucleus). Colony-forming ability of individual cells in different maturational stages.-Measurement of the colony-forming ability of individual, identified cells permits an assessment of the relationship of the original cell's morphology to its potential ability to form colonies. Petri dishes (60-mm) were planted with 1.25 X 10 5 try pan blue-excluding cells and incubated overnight, and the medium was replaced on days 1 and 5 of the experiment. At days 5-9, dilute suspensions of cells from the petri dishes in medium (20 cells/25 ,...1) were inoculated into the wells of Microtest II culture plates (#3040; Falcon Plastics). After we allowed 3 hours for the cells to attach to the bottom of the wells, the medium was replaced; in each well a single cell, cleary separated from its neighbors, was marked on the underside of the dish by encirclement with a Trident Cell Finder (Aloe Products, St. Louis, Mo.) in an inverted phase-contrast microscope. Single cells were examined and classified on the basis of melanin content with the subjective scale previously described. Cell size was obtained with the ocular micrometer. Cells were then allowed to grow for 7 additional days. The number of cells developing from the original identified cell were counted.
643
SPONTANEOUS DIFFERENTIATION OF B16 MELANOMA
DAY I
•
20.0
DAY 6
18.0
MELANIN PRODUCTION
