In, J Rudiurron OIW,lkl,Qa RIOI Phw Vol. Pnnted ,n the USA All rights reserved.
20. PP. 733-73X Copyright
0360.3016191 $3.00 + .Xl (c’ 1991 Pergamon Press plc
0 Original Contribution VARIATION IN RADIATION SENSITIVITY DURING THE CELL OF TWO HUMAN SQUAMOUS CELL CARCINOMAS CORAL
A. QUIET,
M.D.,’ RALPH
R. WEICHSELBAUM,
M.D.’ AND DAVID
CYCLE
J. GRDINA,
PH.D.‘,*
‘Department of Radiation and Cellular Oncology, Michael Reese Medical Center and the University of Chicago Center for Radiation Therapy, University of Chicago; and *Department of Biological and Medical Sciences, Argonne National Laboratory, Argonne, IL Changes in the radiation age response are described in two cell lines derived from human squamous cell carcinomas. A radioresistant tumor cell line, JSQ3, has a De of 240 cGy and is polyploid with a DNA content of 2.68. A relatively radiosensitive tumor cell line, SCC-61, has a D0 of 126 cGy and has a DNA index of 1.16. Tumor cells were separated and synchronized by centrifugal elutriation; flow cytometry was used to determine cell-cycle parameters and relative synchrony. The radioresistant cell line, JSQ3B, was found to have twice the number of cells in S-phase than the more sensitive cell line (28% and 13% for JSQ3B and SCC-61B, respectively). Both cell lines, despite differences in intrinsic radiosensitivity, were most resistant during S-phase (Des of 258 and 157 cGy for JSQ3B and SCC-61B, respectively) and were maximally sensitive during G1 (Des of 193 and 95 cGy for JSQ3B and SCCdlB, respectively). Clinical implications of our findings are discussed. Radiation age response, Human squamous cell carcinoma, Predictive assays.
INTRODUCTION
but with different inherent radiation sensitivities differ in their age response to radiation.
Using a variety of mammalian cell lines, the variation in X-ray sensitivity as a function of cell age has been well characterized. In general, many of these cell lines exhibit relatively increased resistance to radiation during S-phase as compared to other phases of the cell (4, 16, 17). Several exceptions to this have been reported by investigators using L-cell and murine fibrosarcoma cell lines. Both of these cell systems exhibited relative radiation sensitivity during S-phase and resistance during G, (8, 24). These findings demonstrate that the relative radiation age response can vary as a function of the cell systems studied. The age response has not been well characterized in human tumors. A cell line derived from a human uterine cervix carcinoma has been reported to exhibit marked radioresistance throughout G, , whereas cells in the G2+M phases were sensitive to radiation (13). Recently a cell line isolated from a human adenocarcinoma of the colon was shown to be resistant during mid-S phase and demonstrated sensitivity during the G, and G,+M phases (22). It was of interest to us, therefore, to investigate whether cell lines derived from human tumors of similar histologies
METHODS
AND
would
MATERIALS
Cell lines used were JSQ-3 and SCC-6 1. These two human squamous cell carcinoma lines were derived from patients with head and neck cancer and have been previously described in detail (20, 2 1). Cells were routinely cultured in Delbecco’s modified Eagles medium with 20% fetal calf serum. Cells were passaged every 3 to 4 days to maintain exponential growth. Centrifugal elutriation is a non-toxic method of synchronizing cells, provided that cell size directly correlates with the DNA content. As has been seen in other tumors, subpopulations of cells can exist that vary in size such that the purity of synchronization is limited (9, 12). To obtain cell lines that could be synchronized using centrifugal elutriation, cells were elutriated and G, cells of a homogenous size were collected and expanded in mass culture. The cells were re-elutriated and evaluated by flow cytometry. The results of flow cytometry showed that these
Reprint requests to: Coral A. Quiet, M.D., University of Chicage Medical Center, Dept. Radiation and Cellular Oncology, 5841 S. Maryland Ave., Box 442, Chicago, IL 60637.
Accepted
133
for publication
12 October
1990.
734
1. J. Radiation
Oncology 0 Biology 0 Physics
cells were amendable to synchronization using centrifugal elutriation as cell size correlated with DNA content. These lines were designated JSQ-3B and SK-6lB after their respective parent lines.
For elutriation, cells were harvested in exponential growth using 0.05% Trypsin and 0.02% Disodium-ethylenediaminetetraacetic acid (EDTA) at 37°C. Following trypsinization, cells were pelleted and resuspended in lo15 ml of modified Eagles medium without serum and passed through a 21 gauge needle to ensure a single cell suspension. Cells were then separated by centrifugal elutriation using a centrifugal rotor. The system and all associated tubing was sterilized with 20% ethanol and maintained at 4°C. With a rotor speed of 2100 rpm. 3 X 10’ cells were introduced into the elutriation chamber using a flow rate of 5 ml/min. The flow rate was held constant as the rotor speed was decreased in decrements of 100 rpm from 2000 rpm to 900 rpm. Fifty mls were collected at each rotor speed and 12 fractions were collected in total. Fractions were stored at 4°C. The number of cells collected in each fraction was counted using both a hemocytometer and an automated cell counter.
April I YY I ( Volume
10, Number
regression analysis of the curve.
4
of all points
on the exponential
part
RESULTS Following the establishment of cell lines JSQ-3B and SK-61B, flow cytometry was used to analyze the DNA distribution as shown in Figure 1. The relative quantity of DNA per cell, or DNA index, was calculated using transformed human lymphocytes as the standard for diploid; the DNA index for the cell line KC-6 1B is 1.16 and for JSQ-3B it is 2.68 times diploid. Under exponential growth conditions. the distribution of cells in the cell cycle also varied as shown in Table 1. In cell line JSQ-3B, 28% of the population is in the S-phase. while SK-6 1B has only 13% of cells in the S-phase. The percentage of cells in the G?+M phase is similar. Radiobiologic parameters of JSQ-3B and KC-6 1B in exponential growth are similar to their respective parent populations as shown in Table 2. Following elutriation, the recovery of cells was typically greater than 90%. Flow cytometry was used to measure the relative DNA content of the cells in each elutriator fraction. Presented for comparison in Figure 2 are data
The DNA content of individual cells in suspension was measured by flow cytometry (FCM) after they were fixed in 70% ethanol and stained with DAPI (4-6-diamidino2-phenylindole). The distribution of DNA fluorescence and intensity were analyzed using a computer program and presented as DNA histograms. From these histograms the proportion of cells in G,, S, and G2+M phase was calculated using the method of Gohda (5, 6). The DNA index, defined as the proportion of DNA content in the tumor G, cells to the DNA content of lymphocytes in G, , was determined by flow cytometry for both cell lines.
SCC6 1.B
JSQ-3B
X-ray survival curves were determined for an unseparated control and for each cell population enriched in either G, , S, or G,+M phase cells. Following separation, cells were immediately plated and irradiated with a 250 kV x-ray generator operating at 26 mA yielding a dose rate of 107 cGy/min. Immediately after irradiation, the cells were returned to the incubator. After 10 to 14 days, the cells were fixed, stained with crystal violet, and counted. Colonies of at least 50 cells were considered survivors. Data points were determined from the average of 2 to 3 experiments. The single hit multitarget model of survival curve analysis was used. In this model, survival curve parameters measured are the Do, the inverse of the slope, and n, the extrapolation of the slope to the ordinate. These parameters were determined by a least squares
c
human lymphocytes
95
212
Fig. I Representative flow cytometry, profiles of exponentially growing tumor lines SCC-6 1B, of human lymphocytes.
JSQ-3B,
and a diploid
control
Radiation age response 0 C. A.
Table I. Cell cycle distribution
of exponentially
S
67.6 52.6
13.2 28
SCC-6 I B JSQ-3B
cells in G2-M 19.2 19.4
describing the percentages of cells from each fraction in G, , S, and G2+M phases. For both cell lines, an enriched G, population was collected in fractions 2 and 3. With increasing fraction number, the collected population contained increasing proportions of S- and G2+M phase cells. Flow cytometry was performed on fractions of interest after each elutriation to verify that the separation and results were found to be consistent between experiments. The three fractions analyzed from each cell line were chosen to minimize volume overlap and maximize the purity of the subpopulations analyzed. While fraction 1 has the most enriched population of G, cells. 95%) pure, the presence of small debris from the separation as well as the few cells collected in this fraction make it a sub-optimal fraction to analyze thus fractions 2 & 3 were pooled. Similarly, the last two fractions, while most enriched in Gz/ M cells, are not optimal for analysis. Fractions 2 and 3 were pooled and used as an enriched G, population. Similarly, the last two fractions (fraction I I and 12) are not optimal for analysis due to the presence of cell clumps and debris. Fraction 10 was used as the enriched G,/M population and fraction 7 as the enriched S-phase population. To ensure that centrifugal elutriation would not alter clonigenicity or radiosensitivity, some cells were retained as an unseparated control (USC), while the rest were elu-
Table 2. Summary of clonigenic survival curves
JSQ-3B F#3 (G
I enriched) F#7 (S enriched) F# IO (G2-M enriched) Unseperated (control) JSQ-3 (parent cell line)
193 * 0.5 313 k 6.3 258 * 14.2 240 k 6.7 253
Do pure (cGy)*
n
Do (cGy)
2.2 * 1.4 * 2.2 * 2.3 *
0.2 0.1 0.2 0.2
127 464 210 236
2.1 Do pure n
Do (cGy) KC-6 I B F#3 (G 1 enriched) F#7 (S enriched) F#IO (G2-M enriched) Unseperated (control) SCC-6 1 (parent cell line) * Do determined mathematically cells as described in text.
95 157 126 126 117
t 3.6 + 10.8 + 2.5 + 4.5
6.3 1.9 2.5 2.5 2.0
135
ef u/
growing cells
Percent GI (%)
QUIET
k * k *
for a pure population
(cGy)*
0.7 0.3 0.4 0.4
84 215 102 120
of synchronized
0.0
40
2.0
FRACTION
6.0
_c -f
JSQ-JB
-0.0
20
4.0
FRACTION
Fig. 2. Percentage
cell-cycle number.
phases
8.0
100
NUMBER
6.0
GI-PHASE S-PHASE G2tWPHASE
80
10.0
NUMRER
of tumor cells distributed among the various plotted as a function of elutriation fraction
triated. After the standard elutriator procedure, the elutriated cells were repooled. Radiation survival curves were constructed for both the USC and the repooled elutriated cells. Survival curves were not significantly different (data not shown). Radiosensitivity throughout the cell cycle was characterized for both cell lines and is shown in Figure 3 and summarized in Table 2. Both cell lines, despite their marked differences in radiosensitivity, have increased resistance to radiation killing when cells were in S-phase, (Do of 3 I3 and 157 cGy for JSQ-3B and SCC-61 B, respectively) and relative radiosensitivity when they were in G, (Do of 193 and 95 cGy for JSQ-3B and SCC-61B, respectively) and G,+M phases (Do of 258 and 97 cGy for JSQ-3B and SCC-6 1B, respectively). The ratio of the Do of the enriched population to the Do of the asynchronous population varied similarly for each cell line throughout the cell cycle. Centrifugal elutriation is useful in separating cells enriched at various phases of the cell cycle. However, it is difficult to obtain a 100% enrichment for cells in each cell phase without resorting to additional synchronizing agents, such as phase specific chemicals, and thus altering
736
1. J. Radiation Oncology
0 Biology 0 Physics
CENTIGREY
April 199 I. Volume 20. Number 4
CENTIGREY
Fig. 3. Clonogenic survival curves for JSQ-3B and SCC-6 1B tumor lines irradiated in exponential growth and in enriched populations of G I-, S-, and G2SM phase cells. Data points are from the average of 2 to 3 experiments. Standard errors are indicated when they are greater than the symbol. Lines through the points represent the least squares fit of points of the exponential part of the curve
the clonogenicity and/or the radiosensitivity of the cells. To estimate the radiosensitivity of a pure cell-cycle phase, it is possible to weight the Do of each enriched population by the proportion of cells in G, , S, and G2+M phase cells. Three equations are obtained with three variables that can be solved simultaneously. In this manner, the Do of pure S-phase cells was determined to be even more markedly radioresistant (Do of 464 and 215 cGy for JSQ-3B and SCC-6 1B, respectively). G, cells in contrast are quite radiosensitive (Do of 127 and 84 cGy for JSQ-3B and SCC-6lB, respectively). As shown in Table 2, when the theoretical Do for G, , S, and G2+M are weighted by the proportion of cells in each phase in exponentially growing populations, the calculated Do for an asynchronous population closely resembles the experimental value. The calculated Do for JSQ-3B is 236 cGy, which is within the standard error of the experimental value of 240 cGy.
DISCUSSION In this report we have characterized the radiation age response using two cell lines derived from human squamous cell carcinomas. The age response of both cell lines was quite similar despite marked differences in their intrinsic radiosensitivities. Both cell lines were more resistant to X-ray damage during S-phase and maximally sensitive during G, and G2+M phases. These findings suggest that the age response is independent of their intrinsic radiosensitivity. The mechanism which accounts for cell cycle variation in radiation killing is still unknown. The use of centrifugal elutriation as a separation technique results in a relatively pure G1 subpopulation but impure subpopulation of S- + Gz+M phase cells. The limited purity achievable in the S- and G2+M phase population appears to be limited by the inherent size heter-
Radiation age response 0 C. A. QUIET et al.
ogeneity of cells within these phases of the cell cycle and is more pronounced in human tumor cell lines than in murine tumor lines (8, 12). Other techniques for separating S-phase cells such as serum deprivation or hydroxyurea are inadequate due to metabolic perturbator and radiosensitization. Determination of radiosensitivity at various phases of the cell cycle is not exact and as shown here only represent enriched populations. In an attempt to account for differences in the intrinsic radiosensitivity between these two cell lines, cell cycle distribution and DNA content were analyzed. The radioresistant cell line has 28% of the asynchronous population in S-phase whereas the sensitive cell line has 13% of the asynchronous population in S-phase. Therefore, the in vitro radiosensitivity of a cell line may be influenced by the proportion of cells in radioresistant and sensitive parts of the cell cycle as well as differences in the intrinsic radiosensitivity throughout the cell cycle. The tumor cell lines studied herein showed marked differences in their DNA content. The resistant tumor cell line is polyploid with a DNA index of 2.68 whereas the sensitive tumor cell line, SCC-6 1B, has a DNA index of 1.16. While the association between ploidy and radiosensitivity is unclear, there is evidence that tumor lines with increased DNA content are less radiocurable than diploid tumors (3. 7, 15, 1X). Several investigators have correlated a poor prognosis in tumors that are aneuploid and polyploid (1, 11, 15, 18). It has been observed that
737
in vivo polyploid carcinomas of the oral cavity are less radioresponsive than aneuploid tumors (3). Percentage of cells in S-phase and ploidy are two parameters which should be considered in the development of predictive assays for tumor response. Recently, several clinical studies have demonstrated that S-phase specific agents given concominantly with radiation enhance the efficacy of radiation (10, 14, 19). 5FU and hydroxyurea have been found to act as radiosensitizers in several clinical studies possibly be blocking cells from entering S-phase or by a direct cytotoxic effect on S-phase cells. Similarly, hyperthermia offers an increased cytotoxic effect when used concominantly with radiation as S-phase cells have been shown to be selectively heat sensitive (2, 23). The mechanism by which heat and chemotherapy interact with radiation is not known, but based on the data shown here, they may be active against the dominant radioresistant population. The work presented here demonstrates that despite the differences in inherent radiosensitivity, these two tumor lines with similar histologies exhibit similar and marked changes in radiosensitivity throughout the cell cycle with relative radioresistance during S-phase and sensitivity during G, phase. As more in vivo data on tumor cell cycle distribution and radiosensitivity become available, it may be possible to combine chemotherapy and/or hyperthermia with radiation to kill cell-cycle resistant tumor subpopulations and improve therapeutic ratio.
REFERENCES 1. Atkin, N. B.; Baker, M. C. Prognostic significance of modal DNA value and other factors in malignant tumors based on 1,465 cases. Br. J. Cancer 40:2 lo-22 1: 1979. 2. Dewey. W.; Hopwood, L. E.: Saparero, S. A.: Gerweck, L. E. Cellular responses to combinations hyperthermia and radiation. Radiology 123:463-474, 1977. 3. Franzen. G.; Klintenberg, C.: Olofsson, J.; Risberg, B. DNA measurement-an objective predictor of response to irradiation? A review of 24 squamous cell carcinomas of the oral cavity. Br. J. Cancer 53:643-65 1; 1986. 4. Freyer, J. P.; Wilder, M. E.: Raju, M. R. Rapid assay for cell age response to radiation by electronic volume flow cell sorting. Int. J. Radiat. Biol. 52:91-106; 1987. 5. Gohde, W.; Schumann, J.; Buchner, T.: Otto, F.; Barlogie, B. Pulse cytophotometry: application in tumor and cell biology and clinical oncology. In: Melamed, M. R., Mullaney. P. F., Mendelsohn, M., eds. Flow cytometry and sorting. New York: John Wiley; 1979:599. 6. Gohda, M.; Schumann, J.: Zante. J. The use of DAPI in pulse cytophotometry. In: Lutz, B., ed. Pulse cytometry. Proceedings of the 34th International Symposium. Ghent, Belgium: European Press: 1975:229. 7. Helm, L. E. Cellular DNA amounts of squamous cell carcinoma ofthe head and neck region in relation to prognosis. Layngoscope 92: 1064- 1069; 1982. 8. Hunter, N.; Peters, L. J.: Grdina, D. J.; White, R. A. Radiation sensitivity of murine fibrosarcoma cells separated by centrifugal elutriation. Radiat. Res. 80:389-397: 1979.
9. Keng, P. C.; Wheeler, K. T.; Siemann, D. W.; Lord, E. Direct synchronization of cells from solid tumors by centrifugal elutriation. Explt. Cells Res. 134: 15-22; 198 1. 10. Lo, T. C.; Wiley, A. L.; Ansfield, F. J.; Brandenburg, J. H.; Davis, H. L., Jr.; Gollin, F. F.; Johnson, R. 0.; Ramirez. G.: Vermund. H. Combined radiation therapy and 5-fluorouracil for advanced squamous cell carcinoma of the oral cavity and oropharynx: a randomized study. Am. J. Radiat. 126:229-235; 1976. Il. Look, A. T.; Douglass, E. C.; Meyer, W. H. Clinical importance of near diploid tumor stem lines in patients with osteosarcoma of an extremity. N. Eng. J. Med. 3 18: 15671572: 1988. 12. Meistrich, M. L.; Grdina, D. J.; Meyn, R. E.; Barlogie, B. Separation of cells from mouse solid tumors by centrifugal elutriation. Cancer Res. 371429 l-4296; 1977. 13. Pettersen, E. 0.: Christensen, T.; Bakke, 0.: Oftebro, R. A change in the oxygen effect throughout the cell-cycle of human cells of the line NHIK 3025 cultivated in vitro. Int. J. Radiat. Biol. 31:171-184; 1977. 14. Piver, M. S.; Barlow, J. J.; Vongtama, V.: Blumenson, L. Hydroxyurea: a radiation potentiator in carcinoma of the uterine cervix. Am. J. Obstet. Gynecol. 147:803-808; 1983. 15. Sickle-Santanello, B. J.; Farrar, W. B.; Dodson, J. L.; O’Toole, R. V.; Keyhani-Rofagha, S. Flow cytometric analysis of DNA content as a prognostic indicator in squamous cell carcinoma of the tongue. Am. J. Surg. 152:393-395: 1986.
738
I. J.
Radiation
Oncology 0 Biology 0 Physics
16. Sinclair, W.; Morton, R. X-ray sensitivity during the ceil generation cycle of cultured Chmese hamster cells. Radiat. Res. 291450-474; 1966.
April 199 I. Volume
21.
17. Terasima. R.: Tolmach, L. J. Variations in several responses of HeLa cells to x-irradiation during the division cycle. Biophys. J. 3: I l-33: 1963. 18. Tytor. M.: Franzen. G.; Olofsson, J.: Brunk, U.: Nordenskjold, B. DNA content, malignancy grading and prognosis in Tl and T2 oral cavity carcinomas. Br. J. Cancer 56:647652: 1987.
22.
19. Vokes. E. E.; Panje. W. R.: Schilsky, R. L.: Mick. R.: Awan, A. M.: Moran, W. J.: Goldman. M. D.; Tyhor. A. G.: Wcichselbaum, R. R. Hydroxyurea, Flourouracil, and concomitant radiotherapy in poor-prognosis head and neck cancer: A phase I-II study. J. Clin. Oncol. 7:761-768: 1989.
23.
R. R.; Beckett. M. A.: Schwartz, J. L.: Drit20. Weichselbaum, schilo. A. Radioresistant tumor cells are present in head
24.
20. Number
4
and neck carcinomas that recur after radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 15:575-579: 1988. Weichselbaum. R. R.; Dahlberg, W.; Little. J. B.: Ervin, T. J.: Miller. D.; Hellman. S.; Rheinwald. J. G. Cellular xray repair parameters of early passage squamous cell carcinoma lines derived from patients with known responses to radiotherapy. Br. J. Cancer 49:595-60 I: 1984. West, C. M. L.: Keng. P. C.: Sutherland, R. M. Growth phase related variation in the radiation sensitivity of human colon adenocarcinoma cells. Int. J. Radiat. Oncol. Biol. Phys. 14:1213-1219: 1988. Westra. A.; Dewey. W. C. Variation in sensitivity to heat shock during the cell-cycle of Chinese hamster cells in vitro. Int. J. Radiat. Biol. Phys. 19:467-477: I97 I. Whitmore. G. F.: Gulyas. S.: Botond. J. Radiation sensitivity throughout the cell cycle and its relationship to recovery. In: Cellular radiation biology. Baltimore, MD: Williams and Wilkins: 1965:423-44 I.
20. PP. 733-73X Copyright
0360.3016191 $3.00 + .Xl (c’ 1991 Pergamon Press plc
0 Original Contribution VARIATION IN RADIATION SENSITIVITY DURING THE CELL OF TWO HUMAN SQUAMOUS CELL CARCINOMAS CORAL
A. QUIET,
M.D.,’ RALPH
R. WEICHSELBAUM,
M.D.’ AND DAVID
CYCLE
J. GRDINA,
PH.D.‘,*
‘Department of Radiation and Cellular Oncology, Michael Reese Medical Center and the University of Chicago Center for Radiation Therapy, University of Chicago; and *Department of Biological and Medical Sciences, Argonne National Laboratory, Argonne, IL Changes in the radiation age response are described in two cell lines derived from human squamous cell carcinomas. A radioresistant tumor cell line, JSQ3, has a De of 240 cGy and is polyploid with a DNA content of 2.68. A relatively radiosensitive tumor cell line, SCC-61, has a D0 of 126 cGy and has a DNA index of 1.16. Tumor cells were separated and synchronized by centrifugal elutriation; flow cytometry was used to determine cell-cycle parameters and relative synchrony. The radioresistant cell line, JSQ3B, was found to have twice the number of cells in S-phase than the more sensitive cell line (28% and 13% for JSQ3B and SCC-61B, respectively). Both cell lines, despite differences in intrinsic radiosensitivity, were most resistant during S-phase (Des of 258 and 157 cGy for JSQ3B and SCC-61B, respectively) and were maximally sensitive during G1 (Des of 193 and 95 cGy for JSQ3B and SCCdlB, respectively). Clinical implications of our findings are discussed. Radiation age response, Human squamous cell carcinoma, Predictive assays.
INTRODUCTION
but with different inherent radiation sensitivities differ in their age response to radiation.
Using a variety of mammalian cell lines, the variation in X-ray sensitivity as a function of cell age has been well characterized. In general, many of these cell lines exhibit relatively increased resistance to radiation during S-phase as compared to other phases of the cell (4, 16, 17). Several exceptions to this have been reported by investigators using L-cell and murine fibrosarcoma cell lines. Both of these cell systems exhibited relative radiation sensitivity during S-phase and resistance during G, (8, 24). These findings demonstrate that the relative radiation age response can vary as a function of the cell systems studied. The age response has not been well characterized in human tumors. A cell line derived from a human uterine cervix carcinoma has been reported to exhibit marked radioresistance throughout G, , whereas cells in the G2+M phases were sensitive to radiation (13). Recently a cell line isolated from a human adenocarcinoma of the colon was shown to be resistant during mid-S phase and demonstrated sensitivity during the G, and G,+M phases (22). It was of interest to us, therefore, to investigate whether cell lines derived from human tumors of similar histologies
METHODS
AND
would
MATERIALS
Cell lines used were JSQ-3 and SCC-6 1. These two human squamous cell carcinoma lines were derived from patients with head and neck cancer and have been previously described in detail (20, 2 1). Cells were routinely cultured in Delbecco’s modified Eagles medium with 20% fetal calf serum. Cells were passaged every 3 to 4 days to maintain exponential growth. Centrifugal elutriation is a non-toxic method of synchronizing cells, provided that cell size directly correlates with the DNA content. As has been seen in other tumors, subpopulations of cells can exist that vary in size such that the purity of synchronization is limited (9, 12). To obtain cell lines that could be synchronized using centrifugal elutriation, cells were elutriated and G, cells of a homogenous size were collected and expanded in mass culture. The cells were re-elutriated and evaluated by flow cytometry. The results of flow cytometry showed that these
Reprint requests to: Coral A. Quiet, M.D., University of Chicage Medical Center, Dept. Radiation and Cellular Oncology, 5841 S. Maryland Ave., Box 442, Chicago, IL 60637.
Accepted
133
for publication
12 October
1990.
734
1. J. Radiation
Oncology 0 Biology 0 Physics
cells were amendable to synchronization using centrifugal elutriation as cell size correlated with DNA content. These lines were designated JSQ-3B and SK-6lB after their respective parent lines.
For elutriation, cells were harvested in exponential growth using 0.05% Trypsin and 0.02% Disodium-ethylenediaminetetraacetic acid (EDTA) at 37°C. Following trypsinization, cells were pelleted and resuspended in lo15 ml of modified Eagles medium without serum and passed through a 21 gauge needle to ensure a single cell suspension. Cells were then separated by centrifugal elutriation using a centrifugal rotor. The system and all associated tubing was sterilized with 20% ethanol and maintained at 4°C. With a rotor speed of 2100 rpm. 3 X 10’ cells were introduced into the elutriation chamber using a flow rate of 5 ml/min. The flow rate was held constant as the rotor speed was decreased in decrements of 100 rpm from 2000 rpm to 900 rpm. Fifty mls were collected at each rotor speed and 12 fractions were collected in total. Fractions were stored at 4°C. The number of cells collected in each fraction was counted using both a hemocytometer and an automated cell counter.
April I YY I ( Volume
10, Number
regression analysis of the curve.
4
of all points
on the exponential
part
RESULTS Following the establishment of cell lines JSQ-3B and SK-61B, flow cytometry was used to analyze the DNA distribution as shown in Figure 1. The relative quantity of DNA per cell, or DNA index, was calculated using transformed human lymphocytes as the standard for diploid; the DNA index for the cell line KC-6 1B is 1.16 and for JSQ-3B it is 2.68 times diploid. Under exponential growth conditions. the distribution of cells in the cell cycle also varied as shown in Table 1. In cell line JSQ-3B, 28% of the population is in the S-phase. while SK-6 1B has only 13% of cells in the S-phase. The percentage of cells in the G?+M phase is similar. Radiobiologic parameters of JSQ-3B and KC-6 1B in exponential growth are similar to their respective parent populations as shown in Table 2. Following elutriation, the recovery of cells was typically greater than 90%. Flow cytometry was used to measure the relative DNA content of the cells in each elutriator fraction. Presented for comparison in Figure 2 are data
The DNA content of individual cells in suspension was measured by flow cytometry (FCM) after they were fixed in 70% ethanol and stained with DAPI (4-6-diamidino2-phenylindole). The distribution of DNA fluorescence and intensity were analyzed using a computer program and presented as DNA histograms. From these histograms the proportion of cells in G,, S, and G2+M phase was calculated using the method of Gohda (5, 6). The DNA index, defined as the proportion of DNA content in the tumor G, cells to the DNA content of lymphocytes in G, , was determined by flow cytometry for both cell lines.
SCC6 1.B
JSQ-3B
X-ray survival curves were determined for an unseparated control and for each cell population enriched in either G, , S, or G,+M phase cells. Following separation, cells were immediately plated and irradiated with a 250 kV x-ray generator operating at 26 mA yielding a dose rate of 107 cGy/min. Immediately after irradiation, the cells were returned to the incubator. After 10 to 14 days, the cells were fixed, stained with crystal violet, and counted. Colonies of at least 50 cells were considered survivors. Data points were determined from the average of 2 to 3 experiments. The single hit multitarget model of survival curve analysis was used. In this model, survival curve parameters measured are the Do, the inverse of the slope, and n, the extrapolation of the slope to the ordinate. These parameters were determined by a least squares
c
human lymphocytes
95
212
Fig. I Representative flow cytometry, profiles of exponentially growing tumor lines SCC-6 1B, of human lymphocytes.
JSQ-3B,
and a diploid
control
Radiation age response 0 C. A.
Table I. Cell cycle distribution
of exponentially
S
67.6 52.6
13.2 28
SCC-6 I B JSQ-3B
cells in G2-M 19.2 19.4
describing the percentages of cells from each fraction in G, , S, and G2+M phases. For both cell lines, an enriched G, population was collected in fractions 2 and 3. With increasing fraction number, the collected population contained increasing proportions of S- and G2+M phase cells. Flow cytometry was performed on fractions of interest after each elutriation to verify that the separation and results were found to be consistent between experiments. The three fractions analyzed from each cell line were chosen to minimize volume overlap and maximize the purity of the subpopulations analyzed. While fraction 1 has the most enriched population of G, cells. 95%) pure, the presence of small debris from the separation as well as the few cells collected in this fraction make it a sub-optimal fraction to analyze thus fractions 2 & 3 were pooled. Similarly, the last two fractions, while most enriched in Gz/ M cells, are not optimal for analysis. Fractions 2 and 3 were pooled and used as an enriched G, population. Similarly, the last two fractions (fraction I I and 12) are not optimal for analysis due to the presence of cell clumps and debris. Fraction 10 was used as the enriched G,/M population and fraction 7 as the enriched S-phase population. To ensure that centrifugal elutriation would not alter clonigenicity or radiosensitivity, some cells were retained as an unseparated control (USC), while the rest were elu-
Table 2. Summary of clonigenic survival curves
JSQ-3B F#3 (G
I enriched) F#7 (S enriched) F# IO (G2-M enriched) Unseperated (control) JSQ-3 (parent cell line)
193 * 0.5 313 k 6.3 258 * 14.2 240 k 6.7 253
Do pure (cGy)*
n
Do (cGy)
2.2 * 1.4 * 2.2 * 2.3 *
0.2 0.1 0.2 0.2
127 464 210 236
2.1 Do pure n
Do (cGy) KC-6 I B F#3 (G 1 enriched) F#7 (S enriched) F#IO (G2-M enriched) Unseperated (control) SCC-6 1 (parent cell line) * Do determined mathematically cells as described in text.
95 157 126 126 117
t 3.6 + 10.8 + 2.5 + 4.5
6.3 1.9 2.5 2.5 2.0
135
ef u/
growing cells
Percent GI (%)
QUIET
k * k *
for a pure population
(cGy)*
0.7 0.3 0.4 0.4
84 215 102 120
of synchronized
0.0
40
2.0
FRACTION
6.0
_c -f
JSQ-JB
-0.0
20
4.0
FRACTION
Fig. 2. Percentage
cell-cycle number.
phases
8.0
100
NUMBER
6.0
GI-PHASE S-PHASE G2tWPHASE
80
10.0
NUMRER
of tumor cells distributed among the various plotted as a function of elutriation fraction
triated. After the standard elutriator procedure, the elutriated cells were repooled. Radiation survival curves were constructed for both the USC and the repooled elutriated cells. Survival curves were not significantly different (data not shown). Radiosensitivity throughout the cell cycle was characterized for both cell lines and is shown in Figure 3 and summarized in Table 2. Both cell lines, despite their marked differences in radiosensitivity, have increased resistance to radiation killing when cells were in S-phase, (Do of 3 I3 and 157 cGy for JSQ-3B and SCC-61 B, respectively) and relative radiosensitivity when they were in G, (Do of 193 and 95 cGy for JSQ-3B and SCC-61B, respectively) and G,+M phases (Do of 258 and 97 cGy for JSQ-3B and SCC-6 1B, respectively). The ratio of the Do of the enriched population to the Do of the asynchronous population varied similarly for each cell line throughout the cell cycle. Centrifugal elutriation is useful in separating cells enriched at various phases of the cell cycle. However, it is difficult to obtain a 100% enrichment for cells in each cell phase without resorting to additional synchronizing agents, such as phase specific chemicals, and thus altering
736
1. J. Radiation Oncology
0 Biology 0 Physics
CENTIGREY
April 199 I. Volume 20. Number 4
CENTIGREY
Fig. 3. Clonogenic survival curves for JSQ-3B and SCC-6 1B tumor lines irradiated in exponential growth and in enriched populations of G I-, S-, and G2SM phase cells. Data points are from the average of 2 to 3 experiments. Standard errors are indicated when they are greater than the symbol. Lines through the points represent the least squares fit of points of the exponential part of the curve
the clonogenicity and/or the radiosensitivity of the cells. To estimate the radiosensitivity of a pure cell-cycle phase, it is possible to weight the Do of each enriched population by the proportion of cells in G, , S, and G2+M phase cells. Three equations are obtained with three variables that can be solved simultaneously. In this manner, the Do of pure S-phase cells was determined to be even more markedly radioresistant (Do of 464 and 215 cGy for JSQ-3B and SCC-6 1B, respectively). G, cells in contrast are quite radiosensitive (Do of 127 and 84 cGy for JSQ-3B and SCC-6lB, respectively). As shown in Table 2, when the theoretical Do for G, , S, and G2+M are weighted by the proportion of cells in each phase in exponentially growing populations, the calculated Do for an asynchronous population closely resembles the experimental value. The calculated Do for JSQ-3B is 236 cGy, which is within the standard error of the experimental value of 240 cGy.
DISCUSSION In this report we have characterized the radiation age response using two cell lines derived from human squamous cell carcinomas. The age response of both cell lines was quite similar despite marked differences in their intrinsic radiosensitivities. Both cell lines were more resistant to X-ray damage during S-phase and maximally sensitive during G, and G2+M phases. These findings suggest that the age response is independent of their intrinsic radiosensitivity. The mechanism which accounts for cell cycle variation in radiation killing is still unknown. The use of centrifugal elutriation as a separation technique results in a relatively pure G1 subpopulation but impure subpopulation of S- + Gz+M phase cells. The limited purity achievable in the S- and G2+M phase population appears to be limited by the inherent size heter-
Radiation age response 0 C. A. QUIET et al.
ogeneity of cells within these phases of the cell cycle and is more pronounced in human tumor cell lines than in murine tumor lines (8, 12). Other techniques for separating S-phase cells such as serum deprivation or hydroxyurea are inadequate due to metabolic perturbator and radiosensitization. Determination of radiosensitivity at various phases of the cell cycle is not exact and as shown here only represent enriched populations. In an attempt to account for differences in the intrinsic radiosensitivity between these two cell lines, cell cycle distribution and DNA content were analyzed. The radioresistant cell line has 28% of the asynchronous population in S-phase whereas the sensitive cell line has 13% of the asynchronous population in S-phase. Therefore, the in vitro radiosensitivity of a cell line may be influenced by the proportion of cells in radioresistant and sensitive parts of the cell cycle as well as differences in the intrinsic radiosensitivity throughout the cell cycle. The tumor cell lines studied herein showed marked differences in their DNA content. The resistant tumor cell line is polyploid with a DNA index of 2.68 whereas the sensitive tumor cell line, SCC-6 1B, has a DNA index of 1.16. While the association between ploidy and radiosensitivity is unclear, there is evidence that tumor lines with increased DNA content are less radiocurable than diploid tumors (3. 7, 15, 1X). Several investigators have correlated a poor prognosis in tumors that are aneuploid and polyploid (1, 11, 15, 18). It has been observed that
737
in vivo polyploid carcinomas of the oral cavity are less radioresponsive than aneuploid tumors (3). Percentage of cells in S-phase and ploidy are two parameters which should be considered in the development of predictive assays for tumor response. Recently, several clinical studies have demonstrated that S-phase specific agents given concominantly with radiation enhance the efficacy of radiation (10, 14, 19). 5FU and hydroxyurea have been found to act as radiosensitizers in several clinical studies possibly be blocking cells from entering S-phase or by a direct cytotoxic effect on S-phase cells. Similarly, hyperthermia offers an increased cytotoxic effect when used concominantly with radiation as S-phase cells have been shown to be selectively heat sensitive (2, 23). The mechanism by which heat and chemotherapy interact with radiation is not known, but based on the data shown here, they may be active against the dominant radioresistant population. The work presented here demonstrates that despite the differences in inherent radiosensitivity, these two tumor lines with similar histologies exhibit similar and marked changes in radiosensitivity throughout the cell cycle with relative radioresistance during S-phase and sensitivity during G, phase. As more in vivo data on tumor cell cycle distribution and radiosensitivity become available, it may be possible to combine chemotherapy and/or hyperthermia with radiation to kill cell-cycle resistant tumor subpopulations and improve therapeutic ratio.
REFERENCES 1. Atkin, N. B.; Baker, M. C. Prognostic significance of modal DNA value and other factors in malignant tumors based on 1,465 cases. Br. J. Cancer 40:2 lo-22 1: 1979. 2. Dewey. W.; Hopwood, L. E.: Saparero, S. A.: Gerweck, L. E. Cellular responses to combinations hyperthermia and radiation. Radiology 123:463-474, 1977. 3. Franzen. G.; Klintenberg, C.: Olofsson, J.; Risberg, B. DNA measurement-an objective predictor of response to irradiation? A review of 24 squamous cell carcinomas of the oral cavity. Br. J. Cancer 53:643-65 1; 1986. 4. Freyer, J. P.; Wilder, M. E.: Raju, M. R. Rapid assay for cell age response to radiation by electronic volume flow cell sorting. Int. J. Radiat. Biol. 52:91-106; 1987. 5. Gohde, W.; Schumann, J.; Buchner, T.: Otto, F.; Barlogie, B. Pulse cytophotometry: application in tumor and cell biology and clinical oncology. In: Melamed, M. R., Mullaney. P. F., Mendelsohn, M., eds. Flow cytometry and sorting. New York: John Wiley; 1979:599. 6. Gohda, M.; Schumann, J.: Zante. J. The use of DAPI in pulse cytophotometry. In: Lutz, B., ed. Pulse cytometry. Proceedings of the 34th International Symposium. Ghent, Belgium: European Press: 1975:229. 7. Helm, L. E. Cellular DNA amounts of squamous cell carcinoma ofthe head and neck region in relation to prognosis. Layngoscope 92: 1064- 1069; 1982. 8. Hunter, N.; Peters, L. J.: Grdina, D. J.; White, R. A. Radiation sensitivity of murine fibrosarcoma cells separated by centrifugal elutriation. Radiat. Res. 80:389-397: 1979.
9. Keng, P. C.; Wheeler, K. T.; Siemann, D. W.; Lord, E. Direct synchronization of cells from solid tumors by centrifugal elutriation. Explt. Cells Res. 134: 15-22; 198 1. 10. Lo, T. C.; Wiley, A. L.; Ansfield, F. J.; Brandenburg, J. H.; Davis, H. L., Jr.; Gollin, F. F.; Johnson, R. 0.; Ramirez. G.: Vermund. H. Combined radiation therapy and 5-fluorouracil for advanced squamous cell carcinoma of the oral cavity and oropharynx: a randomized study. Am. J. Radiat. 126:229-235; 1976. Il. Look, A. T.; Douglass, E. C.; Meyer, W. H. Clinical importance of near diploid tumor stem lines in patients with osteosarcoma of an extremity. N. Eng. J. Med. 3 18: 15671572: 1988. 12. Meistrich, M. L.; Grdina, D. J.; Meyn, R. E.; Barlogie, B. Separation of cells from mouse solid tumors by centrifugal elutriation. Cancer Res. 371429 l-4296; 1977. 13. Pettersen, E. 0.: Christensen, T.; Bakke, 0.: Oftebro, R. A change in the oxygen effect throughout the cell-cycle of human cells of the line NHIK 3025 cultivated in vitro. Int. J. Radiat. Biol. 31:171-184; 1977. 14. Piver, M. S.; Barlow, J. J.; Vongtama, V.: Blumenson, L. Hydroxyurea: a radiation potentiator in carcinoma of the uterine cervix. Am. J. Obstet. Gynecol. 147:803-808; 1983. 15. Sickle-Santanello, B. J.; Farrar, W. B.; Dodson, J. L.; O’Toole, R. V.; Keyhani-Rofagha, S. Flow cytometric analysis of DNA content as a prognostic indicator in squamous cell carcinoma of the tongue. Am. J. Surg. 152:393-395: 1986.
738
I. J.
Radiation
Oncology 0 Biology 0 Physics
16. Sinclair, W.; Morton, R. X-ray sensitivity during the ceil generation cycle of cultured Chmese hamster cells. Radiat. Res. 291450-474; 1966.
April 199 I. Volume
21.
17. Terasima. R.: Tolmach, L. J. Variations in several responses of HeLa cells to x-irradiation during the division cycle. Biophys. J. 3: I l-33: 1963. 18. Tytor. M.: Franzen. G.; Olofsson, J.: Brunk, U.: Nordenskjold, B. DNA content, malignancy grading and prognosis in Tl and T2 oral cavity carcinomas. Br. J. Cancer 56:647652: 1987.
22.
19. Vokes. E. E.; Panje. W. R.: Schilsky, R. L.: Mick. R.: Awan, A. M.: Moran, W. J.: Goldman. M. D.; Tyhor. A. G.: Wcichselbaum, R. R. Hydroxyurea, Flourouracil, and concomitant radiotherapy in poor-prognosis head and neck cancer: A phase I-II study. J. Clin. Oncol. 7:761-768: 1989.
23.
R. R.; Beckett. M. A.: Schwartz, J. L.: Drit20. Weichselbaum, schilo. A. Radioresistant tumor cells are present in head
24.
20. Number
4
and neck carcinomas that recur after radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 15:575-579: 1988. Weichselbaum. R. R.; Dahlberg, W.; Little. J. B.: Ervin, T. J.: Miller. D.; Hellman. S.; Rheinwald. J. G. Cellular xray repair parameters of early passage squamous cell carcinoma lines derived from patients with known responses to radiotherapy. Br. J. Cancer 49:595-60 I: 1984. West, C. M. L.: Keng. P. C.: Sutherland, R. M. Growth phase related variation in the radiation sensitivity of human colon adenocarcinoma cells. Int. J. Radiat. Oncol. Biol. Phys. 14:1213-1219: 1988. Westra. A.; Dewey. W. C. Variation in sensitivity to heat shock during the cell-cycle of Chinese hamster cells in vitro. Int. J. Radiat. Biol. Phys. 19:467-477: I97 I. Whitmore. G. F.: Gulyas. S.: Botond. J. Radiation sensitivity throughout the cell cycle and its relationship to recovery. In: Cellular radiation biology. Baltimore, MD: Williams and Wilkins: 1965:423-44 I.
