Proliferative Response of Human Diploid Fibroblasts to Intermittent Light Exposure RAM PARSHAD AND KATHERINE K. SANFORD Department of Pathology, College of Medicine, Howard Uniuersity, Washington, D.C. 20059 and I Laboratory of Biochemistry, National Cancer Institute, Bethesda, Maryland 20014
ABSTRACT Three- to four-hour exposure to fluorescent light, one to three times weekly, reproducibly enhanced the proliferation rate of human diploid fibroblasts. This enhancement was observed in WI-38 and a line from whole embryo mince a t late population doubling level (PDL) as well as in a line from adult skin a t early PDL. Single or multiple exposures of short duration stimulated proliferation, whereas exposures of long duration were cytotoxic. This proliferative response is reversible, and is mediated through the culture medium, Dulbecco Vogt's supplemented with 10% fetal bovine serum. Apparently light produces some mitogenic substance(s1 in the culture medium that accumulates in the cells and is toxic or growth-stimulatory depending on its concentration per cell. Another possibility is that light produces in the medium both cytotoxic and growt h-stimulatory substances. Visible light has been shown to have lethal or growth-inhibitory effects on mammalian cells in vitro (Warburg, '68; Litwin, '72; Wang, '76; Pereira e t al., '76) and to cause chromosomal aberrations in mass culture (Parshad e t al., '77). These responses appear to be oxygen-dependent, and the lethal or growth-inhibitory effects reportedly result from photosensitization of riboflavin to produce in the medium toxic oxidation products of certain amino acids including tryptophan and tyrosine. Such toxic substances could conceivably be produced in culture medium during the routine handling of medium and/or cells in fluorescent-lighted rooms. Further, these substances might influence the proliferative response and life-span of human diploid fibroblasts in culture. We, therefore, explored the proliferative response of three lines of human fibroblasts to repeated exposures to fluorescent light when grown in mass culture. MATERIALS AND METHODS
Source of cells and culture procedures Line HU-274 was initiated from abdominal skin of a 50-year-old black male who died of hypertension and renal failure. The cell line was initiated and maintained continuously without exposure to light of wave lengths below 500 nm in t h a t cells and medium were routinely handled in rooms illuminated with J. CELL. PHYSIOL.. 92: 481-486.
gold or red fluorescent lights. The microscope lamp for culture examination was fitted with a red filter. WI-38 originally obtained from the American Type Culture Collection (12301 Parklawn Drive, Rockville, Maryland 20852) was frozen after 31 doublings, thawed, and carried under the above light conditions for a n additional 13 doublings (4 weeks). Line HU-278, obtained in primary culture from Microbiological Associates (Bethesda, Maryland) originated from a whole human embryo (lot no. 109068). The line was carried for 38 population doublings before cryopreservation. Cells were thawed and carried under the above light conditions for 11.5 additional doublings. Cells were grown in a modification of Dulbecco-Vogt's medium (Richter e t al., '72) supplemented with 10% fetal bovine serum (Flow Laboratories, Rockville, Maryland). The medium was prepared and filtered in gold light, and stored in the dark. Culture medium was renewed three times a week, except in experiments where the inoculum size was small (lo5cells per T-25); in these cultures, medium was renewed once a week. Cells at subculture were dispersed by rinsing with versene (1:5000) (Microbiological Associates, Bethesda, Maryland) and treating with trypsin: Received Dec. 13, '76.Accepted March 4, '77. 1 National Institutes of Health, US. Public Health Service, U S . Department of Health, Education, and Welfare.
48 1
482
RAM PARSHAD AND KATHERINE K. SANFORD
SKIN LINE HU274
FETAL LUNG LINE WI-38
FETAL LINE HU 278
-
UNEXPOSED LIGHT-EXPOSED
P
d'
1
_
1
_
2
3
L
I
4
1
2
3
4
TIME IN WEEKS
Fig. 1 Effect of intermittent light exposure on proliferation of human cells in mass culture. Cumulative increase in cell number is plotted against time in weeks. At each weekly subculture, the inoculum was 2 X lo5 cellsiT-25 flask. The skin line was exposed after 7, WI-38 after 44, and fetal line after 49 population doublings. 4 r
9
UNEXPOSED LIGHT-EXPOSED LIGHT-4 Hrs LIGHT -24 Hrs
mined intervals, cells were dispersed as above and counted with a hemacytometer. Each determination is the average of four cultures. Light exposure Experimental cultures or flasks containing medium without cells were exposed at 37°C to two cool-white fluorescent light tubes (GE F 15TWCW) at 27 cm from the floor of the flask with an exposure intensity of 6 to 9 W/m2. EXPERIMENTAL
Effect of intermittent light exposure on proliferation rate Three hours of light exposure three times a week (24 hours after subculturing and twice 0.5 0 2 4 6 thereafter following fluid renewals a t 2 or 3 TIME I N D A Y S day intervals) increased the proliferation rate Fig. 2 Effect of !single light exposure of 4 or 24 hours on of all cell lines examined (fig. 1).This effect cytotoxicity and proliferation rate of celis of skin line HUwas observed by quantifying a t weekly inter274 never previously exposed to light of wavelengths below vals the increase in cell numbers from a con500 nm. stant inoculum size. Light did not stimulate EDTA mixture (Gibco, Grand Island, New proliferation until after the first week of exposures and initially appeared to inhibit proYork). liferation of cell line HU-274 which had never Quantitation of cell proliferation been exposed in vitro to light of wave lengths To determine the rate of cell proliferation, below 500nm. Subsequently, both lines of T-25 (Falcon) flasks were inoculated with lo5 fetal cells a t high population doubling levels or 2 x l o 5cells in 5 ml medium. At predeter- approaching phase 111, and adult cells after
483
PROLIFERATIVE RESPONSE OF HUMAN CELLS TO LIGHT
UNEXPOSED
-
0-4LIGHT-EXPOSED
+ 4 EXPOSED UNEXPOSED AT TIME
I I
I
C.
I
I I
0-0
c-..-c1
-
UNEXPOSED 4 Hrs
, I
I
I
48 Hrs
1
~
2
0
3
TIME IN WEEKS
Fig. 3 Relationship between duration of light exposures and proliferation of skin line HU-274. Cumulative increase in cell number is plotted for cultures exposed to light once a week, 24 hours after medium renewal and suhculturing. Inoculum size a t each subculture was l o 5 cells/ T25 flask. LIGHT EXPOSURE
0
0
0 hrslwk
4 hrslwk
8 hrsiwk
24 hrsiwk
48 hrsiwk
TIME IN WEEKS UUNEXPOSED
c--* &,4
-
LIGHT-EXPOSED EXPOSED UNEXPOSED
Fig. 4 Effect of transferring skin line HU274 from light exposure to no exposure. Cells in figure 3 after a single exposure during the first week were given a second exposure (0-- -0)or no exposure ( 0 - . .); unexposed controls (0-0). Inoculum size was lo5 cells per flask. - a
-
only seven population doublings showed this enhanced proliferative response t o light. To explore further the initial inhibitory effect of light on skin line HU-274, a quantitative growth experiment was carried out in
2 TIME IN WEEKS 1
3
Fig. 5 Proliferation rate of skin cells after transfer from intermittent light exposure to unexposed condition. Skin cells in figure 1 after four weeks of exposure three times a week were continued under the same condition of exposure, (0-- -0)or were transferred to unexposed 1 condition (0 - . - . - 0 ) ;unexposed control (0-0 inoculum size was 2 x los cells per flask.
which cells were given a single light exposure of 4 or 24 hours on the day following inoculation. Two hours after exposure and a t intervals thereafter for a week, cell counts were made of replicate cultures. These results indicate an initial cytotoxic effect especially after the longer exposure period, and little or no effect on subsequent proliferation rate (fig. 2).
Relationship of exposure time to growth stimulation To determine the effect of duration of light exposure on cell proliferation, replicate cultures of the skin line, HU 274, were given a single exposure of 0 , 4 , 8 , 2 4or 48 hours once a week. At weekly intervals cells were counted and subcultured with a n inoculum of l o 5cells per flask. Cultures were exposed 24 hours after inoculation. Under these conditions a stimulatory effect on cell proliferation was observed by the second week in culture with light exposures of four hours. With increased exposure time, a corresponding decrease in proliferation rate occurred (fig. 3).
RAM FARSHAD AND KATHERINE K. SANFORD
C. UNEXPOSED
0.-a LIGHT--EXPOSE0
o-.-o MEDIUM PRE-EXPOSED
-Unexposed ---oLight-exposed
I
,’ II
P
-.--+Medium
Pre-exposed
/.p
/
4, ,tf
1
2
3
TIME IN WEEKS
1
0
1 2 TIME IN WEEKS
3
Fig. 6 Effect of culture medium preexposed to light on proliferation rate of skin line HU274. Cells never exposed to light were transferred to culture medium that had been preexposed for four hours to light (0- 0). Control cultures were either exposed to light for four hours weekly (0--0), or never exposed to light ( 0 - 0 ) . Culture medium was renewed once a week.
Reversibility of the proliferative response
To ascertain whether the response t o light was permanent or reversible, some of the cells in the experiment of figure 3 after the first week of light exposure were maintained unexposed for the following week. The proliferative or inhibitory effect of the previous light exposure partially reverted during the second week in the absence of light (fig. 4). Similarly, some of the skin cells t h a t had been exposed, three times weekly for four weeks (fig. 1) were continued for another three weeks without further exposure to light. By the end of t,he second week the proliferation rate had partially reverted to that of the control cells never exposed t o light (fig. 5). Role of culture medium in the proliferative response to light
To determine whether the proliferative response to light was a direct effect on the cells or mediated through the culture medium, skin
Fig. 7 Effect of culture medium preexposed to light on proliferation rate of skin line HU274. Unexposed cells and cells intermittently exposed to light of figure 5 were continued under the same experimental conditions. In addition, light-exposed cells were transferred to preilluminated medium. Medium was renewed three times a week.
cells never exposed to light were carried in medium that had been exposed to light for four hours immediately before use; cells were subcultured a t weekly intervals. By the end of the second week, these cells showed a proliferative response approaching that of cells exposed to light four hours weekly. By this time, both cell populations had a higher proliferation rate than unexposed controls (fig. 6). Similarly, cells intermittently exposed to light in the experiment of figure 5 were continued with intermittent light exposure or were transferred to culture medium preexposed to light for three hours; culture medium was renewed three times a week. Cells in the preexposed culture medium showed a proliferation rate approaching that of the light-exposed cultures and significantly higher than that of the unexposed controls (fig. 7). These results indicated that the enhancement of proliferation by visible light could occur through the culture medium. DISCUSSION
Several investigators have reported cyto-
PROLIFERATIVE RESPONSE OF HUMAN CELLS TO LIGHT
toxic effects of fluorescent light on cells in culture. These deleterious effects resulted from the production of photooxidative products in the culture medium. However, in all studies, the cells were plated a t low inoculum sizes (Wang, '75, '76; Pereira e t al., '761, or the medium was exposed for long periods (Litwin, '72; Wang, '76) or the intensity of light far exceeded that used in the present study (Litwin, '72; Pereira et al., '76). Litwin ('721, on the other hand, observed a n increased lifespan but no enhanced proliferation rate in human embryonic fibroblasts when exposed to fluorescent light for two or four hours daily; when exposed for 6 hours daily the growth rate decreased as compared with that of unexposed cells. Similarly, in the present study, exposure of cells to fluorescent light for eight hours or more was cytotoxic andfor growth-inhibitory. Also, cells never exposed to light of wavelength below 500 nm showed a decreased proliferation rate during the first week of exposure (9). In contrast to these deleterious effects of fluorescent light, a stimulatory influence on proliferation of human diploid fibroblasts was seen under the present experimental conditions. These conditions include a relatively high inoculum size, low light intensity and short exposure period(s1. This stimulation has been observed repeatedly in adult skin fibroblasts a t low population doubling level as well as in WI-38 and in a line from minced whole fetus a t high population doubling level. Thus, enhanced proliferation rate is a reproducible response of human diploid cells to intermittent exposure to fluorescent light. Although the precise effective wave lengths have not, as yet, been identified, they lie between 290 and 500 nm since medium a n d cells were exposed through plastic flasks (Parshad e t al., '77) and control cultures were routinely handled in gold or red room light. These results further indicate a dose-response relationship in which a short exposure to light stimulates proliferation, whereas longer exposures are cytotoxic or growth-inhibitory. Since visible light has been shown to induce mutations both in bacteria (Webb. et al., '72) and mammalian cells in culture (Bradley e t al., '77), we explored the possibility that the enhanced proliferation might be a permanent heritable change. We found that the enhanced
485
proliferative response to one 4-hour exposure partially disappeared by the end of the first week of growth in the absence of light. However, cells exposed intermittently for four weeks required more than one week to revert to the lower proliferation rate. These results suggest a cumulative effect of repeated exposures. Thus, the proliferative response to light of
ABSTRACT Three- to four-hour exposure to fluorescent light, one to three times weekly, reproducibly enhanced the proliferation rate of human diploid fibroblasts. This enhancement was observed in WI-38 and a line from whole embryo mince a t late population doubling level (PDL) as well as in a line from adult skin a t early PDL. Single or multiple exposures of short duration stimulated proliferation, whereas exposures of long duration were cytotoxic. This proliferative response is reversible, and is mediated through the culture medium, Dulbecco Vogt's supplemented with 10% fetal bovine serum. Apparently light produces some mitogenic substance(s1 in the culture medium that accumulates in the cells and is toxic or growth-stimulatory depending on its concentration per cell. Another possibility is that light produces in the medium both cytotoxic and growt h-stimulatory substances. Visible light has been shown to have lethal or growth-inhibitory effects on mammalian cells in vitro (Warburg, '68; Litwin, '72; Wang, '76; Pereira e t al., '76) and to cause chromosomal aberrations in mass culture (Parshad e t al., '77). These responses appear to be oxygen-dependent, and the lethal or growth-inhibitory effects reportedly result from photosensitization of riboflavin to produce in the medium toxic oxidation products of certain amino acids including tryptophan and tyrosine. Such toxic substances could conceivably be produced in culture medium during the routine handling of medium and/or cells in fluorescent-lighted rooms. Further, these substances might influence the proliferative response and life-span of human diploid fibroblasts in culture. We, therefore, explored the proliferative response of three lines of human fibroblasts to repeated exposures to fluorescent light when grown in mass culture. MATERIALS AND METHODS
Source of cells and culture procedures Line HU-274 was initiated from abdominal skin of a 50-year-old black male who died of hypertension and renal failure. The cell line was initiated and maintained continuously without exposure to light of wave lengths below 500 nm in t h a t cells and medium were routinely handled in rooms illuminated with J. CELL. PHYSIOL.. 92: 481-486.
gold or red fluorescent lights. The microscope lamp for culture examination was fitted with a red filter. WI-38 originally obtained from the American Type Culture Collection (12301 Parklawn Drive, Rockville, Maryland 20852) was frozen after 31 doublings, thawed, and carried under the above light conditions for a n additional 13 doublings (4 weeks). Line HU-278, obtained in primary culture from Microbiological Associates (Bethesda, Maryland) originated from a whole human embryo (lot no. 109068). The line was carried for 38 population doublings before cryopreservation. Cells were thawed and carried under the above light conditions for 11.5 additional doublings. Cells were grown in a modification of Dulbecco-Vogt's medium (Richter e t al., '72) supplemented with 10% fetal bovine serum (Flow Laboratories, Rockville, Maryland). The medium was prepared and filtered in gold light, and stored in the dark. Culture medium was renewed three times a week, except in experiments where the inoculum size was small (lo5cells per T-25); in these cultures, medium was renewed once a week. Cells at subculture were dispersed by rinsing with versene (1:5000) (Microbiological Associates, Bethesda, Maryland) and treating with trypsin: Received Dec. 13, '76.Accepted March 4, '77. 1 National Institutes of Health, US. Public Health Service, U S . Department of Health, Education, and Welfare.
48 1
482
RAM PARSHAD AND KATHERINE K. SANFORD
SKIN LINE HU274
FETAL LUNG LINE WI-38
FETAL LINE HU 278
-
UNEXPOSED LIGHT-EXPOSED
P
d'
1
_
1
_
2
3
L
I
4
1
2
3
4
TIME IN WEEKS
Fig. 1 Effect of intermittent light exposure on proliferation of human cells in mass culture. Cumulative increase in cell number is plotted against time in weeks. At each weekly subculture, the inoculum was 2 X lo5 cellsiT-25 flask. The skin line was exposed after 7, WI-38 after 44, and fetal line after 49 population doublings. 4 r
9
UNEXPOSED LIGHT-EXPOSED LIGHT-4 Hrs LIGHT -24 Hrs
mined intervals, cells were dispersed as above and counted with a hemacytometer. Each determination is the average of four cultures. Light exposure Experimental cultures or flasks containing medium without cells were exposed at 37°C to two cool-white fluorescent light tubes (GE F 15TWCW) at 27 cm from the floor of the flask with an exposure intensity of 6 to 9 W/m2. EXPERIMENTAL
Effect of intermittent light exposure on proliferation rate Three hours of light exposure three times a week (24 hours after subculturing and twice 0.5 0 2 4 6 thereafter following fluid renewals a t 2 or 3 TIME I N D A Y S day intervals) increased the proliferation rate Fig. 2 Effect of !single light exposure of 4 or 24 hours on of all cell lines examined (fig. 1).This effect cytotoxicity and proliferation rate of celis of skin line HUwas observed by quantifying a t weekly inter274 never previously exposed to light of wavelengths below vals the increase in cell numbers from a con500 nm. stant inoculum size. Light did not stimulate EDTA mixture (Gibco, Grand Island, New proliferation until after the first week of exposures and initially appeared to inhibit proYork). liferation of cell line HU-274 which had never Quantitation of cell proliferation been exposed in vitro to light of wave lengths To determine the rate of cell proliferation, below 500nm. Subsequently, both lines of T-25 (Falcon) flasks were inoculated with lo5 fetal cells a t high population doubling levels or 2 x l o 5cells in 5 ml medium. At predeter- approaching phase 111, and adult cells after
483
PROLIFERATIVE RESPONSE OF HUMAN CELLS TO LIGHT
UNEXPOSED
-
0-4LIGHT-EXPOSED
+ 4 EXPOSED UNEXPOSED AT TIME
I I
I
C.
I
I I
0-0
c-..-c1
-
UNEXPOSED 4 Hrs
, I
I
I
48 Hrs
1
~
2
0
3
TIME IN WEEKS
Fig. 3 Relationship between duration of light exposures and proliferation of skin line HU-274. Cumulative increase in cell number is plotted for cultures exposed to light once a week, 24 hours after medium renewal and suhculturing. Inoculum size a t each subculture was l o 5 cells/ T25 flask. LIGHT EXPOSURE
0
0
0 hrslwk
4 hrslwk
8 hrsiwk
24 hrsiwk
48 hrsiwk
TIME IN WEEKS UUNEXPOSED
c--* &,4
-
LIGHT-EXPOSED EXPOSED UNEXPOSED
Fig. 4 Effect of transferring skin line HU274 from light exposure to no exposure. Cells in figure 3 after a single exposure during the first week were given a second exposure (0-- -0)or no exposure ( 0 - . .); unexposed controls (0-0). Inoculum size was lo5 cells per flask. - a
-
only seven population doublings showed this enhanced proliferative response t o light. To explore further the initial inhibitory effect of light on skin line HU-274, a quantitative growth experiment was carried out in
2 TIME IN WEEKS 1
3
Fig. 5 Proliferation rate of skin cells after transfer from intermittent light exposure to unexposed condition. Skin cells in figure 1 after four weeks of exposure three times a week were continued under the same condition of exposure, (0-- -0)or were transferred to unexposed 1 condition (0 - . - . - 0 ) ;unexposed control (0-0 inoculum size was 2 x los cells per flask.
which cells were given a single light exposure of 4 or 24 hours on the day following inoculation. Two hours after exposure and a t intervals thereafter for a week, cell counts were made of replicate cultures. These results indicate an initial cytotoxic effect especially after the longer exposure period, and little or no effect on subsequent proliferation rate (fig. 2).
Relationship of exposure time to growth stimulation To determine the effect of duration of light exposure on cell proliferation, replicate cultures of the skin line, HU 274, were given a single exposure of 0 , 4 , 8 , 2 4or 48 hours once a week. At weekly intervals cells were counted and subcultured with a n inoculum of l o 5cells per flask. Cultures were exposed 24 hours after inoculation. Under these conditions a stimulatory effect on cell proliferation was observed by the second week in culture with light exposures of four hours. With increased exposure time, a corresponding decrease in proliferation rate occurred (fig. 3).
RAM FARSHAD AND KATHERINE K. SANFORD
C. UNEXPOSED
0.-a LIGHT--EXPOSE0
o-.-o MEDIUM PRE-EXPOSED
-Unexposed ---oLight-exposed
I
,’ II
P
-.--+Medium
Pre-exposed
/.p
/
4, ,tf
1
2
3
TIME IN WEEKS
1
0
1 2 TIME IN WEEKS
3
Fig. 6 Effect of culture medium preexposed to light on proliferation rate of skin line HU274. Cells never exposed to light were transferred to culture medium that had been preexposed for four hours to light (0- 0). Control cultures were either exposed to light for four hours weekly (0--0), or never exposed to light ( 0 - 0 ) . Culture medium was renewed once a week.
Reversibility of the proliferative response
To ascertain whether the response t o light was permanent or reversible, some of the cells in the experiment of figure 3 after the first week of light exposure were maintained unexposed for the following week. The proliferative or inhibitory effect of the previous light exposure partially reverted during the second week in the absence of light (fig. 4). Similarly, some of the skin cells t h a t had been exposed, three times weekly for four weeks (fig. 1) were continued for another three weeks without further exposure to light. By the end of t,he second week the proliferation rate had partially reverted to that of the control cells never exposed t o light (fig. 5). Role of culture medium in the proliferative response to light
To determine whether the proliferative response to light was a direct effect on the cells or mediated through the culture medium, skin
Fig. 7 Effect of culture medium preexposed to light on proliferation rate of skin line HU274. Unexposed cells and cells intermittently exposed to light of figure 5 were continued under the same experimental conditions. In addition, light-exposed cells were transferred to preilluminated medium. Medium was renewed three times a week.
cells never exposed to light were carried in medium that had been exposed to light for four hours immediately before use; cells were subcultured a t weekly intervals. By the end of the second week, these cells showed a proliferative response approaching that of cells exposed to light four hours weekly. By this time, both cell populations had a higher proliferation rate than unexposed controls (fig. 6). Similarly, cells intermittently exposed to light in the experiment of figure 5 were continued with intermittent light exposure or were transferred to culture medium preexposed to light for three hours; culture medium was renewed three times a week. Cells in the preexposed culture medium showed a proliferation rate approaching that of the light-exposed cultures and significantly higher than that of the unexposed controls (fig. 7). These results indicated that the enhancement of proliferation by visible light could occur through the culture medium. DISCUSSION
Several investigators have reported cyto-
PROLIFERATIVE RESPONSE OF HUMAN CELLS TO LIGHT
toxic effects of fluorescent light on cells in culture. These deleterious effects resulted from the production of photooxidative products in the culture medium. However, in all studies, the cells were plated a t low inoculum sizes (Wang, '75, '76; Pereira e t al., '761, or the medium was exposed for long periods (Litwin, '72; Wang, '76) or the intensity of light far exceeded that used in the present study (Litwin, '72; Pereira et al., '76). Litwin ('721, on the other hand, observed a n increased lifespan but no enhanced proliferation rate in human embryonic fibroblasts when exposed to fluorescent light for two or four hours daily; when exposed for 6 hours daily the growth rate decreased as compared with that of unexposed cells. Similarly, in the present study, exposure of cells to fluorescent light for eight hours or more was cytotoxic andfor growth-inhibitory. Also, cells never exposed to light of wavelength below 500 nm showed a decreased proliferation rate during the first week of exposure (9). In contrast to these deleterious effects of fluorescent light, a stimulatory influence on proliferation of human diploid fibroblasts was seen under the present experimental conditions. These conditions include a relatively high inoculum size, low light intensity and short exposure period(s1. This stimulation has been observed repeatedly in adult skin fibroblasts a t low population doubling level as well as in WI-38 and in a line from minced whole fetus a t high population doubling level. Thus, enhanced proliferation rate is a reproducible response of human diploid cells to intermittent exposure to fluorescent light. Although the precise effective wave lengths have not, as yet, been identified, they lie between 290 and 500 nm since medium a n d cells were exposed through plastic flasks (Parshad e t al., '77) and control cultures were routinely handled in gold or red room light. These results further indicate a dose-response relationship in which a short exposure to light stimulates proliferation, whereas longer exposures are cytotoxic or growth-inhibitory. Since visible light has been shown to induce mutations both in bacteria (Webb. et al., '72) and mammalian cells in culture (Bradley e t al., '77), we explored the possibility that the enhanced proliferation might be a permanent heritable change. We found that the enhanced
485
proliferative response to one 4-hour exposure partially disappeared by the end of the first week of growth in the absence of light. However, cells exposed intermittently for four weeks required more than one week to revert to the lower proliferation rate. These results suggest a cumulative effect of repeated exposures. Thus, the proliferative response to light of
