Mutatwn Research, 260 (1991) 25-38 ©1991Elsewer Science Pubhshers B.V 0165-1218/91/$03 50 ADONIS 016512189100076S
25
MUTGEN 01637
Mutagenicity of coolwhite fluorescent light for Salmonella Zlata Hartman, Philip E. Hartman and Wesley L. McDermott Department of Bwlogy, The Johns Hopkins Umverslty, Balttmore, MD 21218 (U S A )
(Recewed 4 June 1990) (Revision received 24 August 1990) (Accepted 5 September 1990)
Keywords Ames strains; Hlstldme reversion, Coolwhlte, Softwtute, Plasrmd pKM101, uvrB, Classroom experiment
Summary The most common fluorescent lamps in use today in homes and businesses in the United States, 'coolwhite' fluorescent lamps, emit light that is mutagenlc for Salmonella. Strains that carry both a uvrB mutation and plasmid pKM101 are extremely susceptible to tbas light-induced mutation. Both base substitution and frameshift mutations can be induced w~thout substantial lethal effects on the bacteria. Induced mutations accumulate essentially as a linear function of the time bacteria are exposed to illumination. Of Salmonella histldine-requiring strains with known nucleotide target sequences ( H a r t m a n et al., 1986; Cebula and Koch, 1989, 1990), strains either carrying one of the base substitution mutations, hlsG428 and htsG46, or one of the frameshifts, htsC3076 and hlsD6610, are most haghly mutagenized whereas frameshlft strains with htsD6580 and htsD3052 exhibit lower rates of mutagenesis. Mutagenlcity does not appear to require the presence of oxygen. A filter blocking wavelengths below 370 nm eliminates mutagenesis. Polystyrene, cellulose acetate and, especially, mylar and glass filters reduce mutagenesls, indicating that at least some of the mutagenlc effects can be attributed to leakage of radiations below 290 nm (far-ultraviolet light) from 'coolwhite' lamps. The more recently introduced fluorescent 'softwhlte' lamps are roughly 10-fold less mutagenic at approximately equal hght intensity. Incandescent light bulbs are much less mutagenic than are these fluorescent lamps. Our mutational data correlate closely with previous results in eukaryotic cells (Jacobson and Krell, 1982). A uvrB recA Salmonella double mutant is hypersensitive to the lethal effects of coolwhate fluorescent light, even when illuminated through the lids of glass Petri dishes. Thus, a p p r o p n a t e Salmonella strains would appear to be simple and useful screens for both the mutagenic and the lethal activities of fluorescent lamps. These systems are amenable to classroom laboratory use as relatively safe and effective means of demonstrating environmental mutagenesis.
Correspondence" P.E Hartman, Department of Biology, The Johns Hopkins Umverslty, 34th and Charles Streets, Baltimore, MD 21218 (U S A) Abbrevtattons CAT, catalase, SOD, superoxxdedlsmutase, TA,
fl-(2-thlazolyl)-DL-alamne,TRA, DL-1,2,4-trmzole-3-alamne.
The pioneering study of Speck and Rosenkranz (1975) and other studies that have followed demonstrate mutagenicity for Salmonella of emissions from fluorescent lamps designed for specialized uses in phototherapy or to stimulate plant growth,
26 TABLE 1 I N D U C T I O N OF PROTOTROPHIC REVERTANTS BY FLUORESCENT L I G H T IN VARIOUS SALMONELLA H I S T I D I N E AUXOTROPHS Illumanatlon was with coolwhate hght and through hds of glass Petn dishes unless otherwise indicated a Strain
1° 2. 3 4. 5 6 7 8 9. 10
11. 12. 13. 14
TA1975 TA1950 TA1535 TA92 TA2410
TA100 SB5428 TA2659 TA102
15 16 17 18 19. 20
TA2898
21 22. 23 24.
TA104
25 26 27 28. 29 30. 31. 32 33 34.
SB8052 TA1538 SB4312 SB4313 TA94 TA98
35 36 37. 38 39. 40. 41 42.
SB8076 TA1537 TA2425 TA2637
43 44
TA2637
SB4315 TA2678
SB4317 SB4318
Genotype b
htsG46 htsG46 htsG46 hlsG46 htsG46 htsG46 hlsG46 hlsG46 htsG46 htsG46
rfa uvrB uvrB rfa (pKM101) uvrB (pKM101) uvrB (pKM101) uvrB (pKM101) uvrB (pKM101) uvrB (pKM101) rfa
Plating medium
Average number colomes per plate
(% nutrient broth)
Spontaneous
90-man dlununatlon
1 25 1.25 1.25 1.25 1 25 1.25 0.25 0.25 0 25 1.25
10 7 31 29 97 464 116 155 155 251
3 7 29 24 89 1215 1032 1869 1159 1112 22 3 28
htsG428 ara-9 htsG428 ara-9 hzsG428 uvrB1057 rfa-1028 ara-9 htsG8476 galE503 rfa-1027 (pKM101) (pAQ1 htsG428) hlsG428 recA1 srl-2 T n l O ara-9 h~sG428 recA1 srl-2" T n l O ara-9 htsG428 f e c a l srl-2 "TnlO ara-9 (pKM101) htsG428 recA1 srl-2"'TnlO ara-9 (pKM101) htsG428 uvrB ara-9 (pKM101) htsG428 uorB ara-9 (pKM101)
1 25 0 25 1.25
15 2 17
0 25 1.25 0.25 1.25 0 25 1 25 0.25
74 14 6 14 10 628 87
87 10 6 9 10 - 1986 d 790
hlsG428 htsG428 htsG428 hlsG428
0 25 0.25 0 25 0.25
96 116 102 78
1092 -- 2998 d 215 186
uvrB1057 uvrB1057 uorB1057 uvrB1057
rfa-1028 rfa-1028 rfa-1028 rfa-1028
ara-9 ara-9 ara-9 ara-9
(pKM101) (pKM101) (pKM101) (pKM101)
htsD3052 htsD3052 htsD3052 htsD3052 htsD3052 htsD3052 htsD3052 hlsD3052 hlsD3052 htsD3052
ara-9 uvrB rfa ara-9 uvrB gsh-6 ara-9 uvrB gsh-7 ara-9 ara-9 (pKM101) uorB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB gsh-6 ara-9 (pKM101) uvrB gsh- 7 ara-9 (pKM101)
1.25 1 25 0 25 0.25 1 25 1.25 0 25 0.25 0 25 0.25
27 50 4 2 35 53 13 7 13 19
25 38 4 3 33 154 132 13 62 51
htsC3076 htsC3076 htsC3076 hlsC3076 htsC3076 htsC3076 hlsC3076 htsC3076
ara-9 uvrB rfa ara-9 ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101)
1 25 1 25 1.25 1 25 1 25 1.25 1 25 1 25
12 12 21 65 64 64 67 109
7 13 34 595 987 1258 1110 117
htsC3076 urvB rfa ara-9 (pKM101) hzsC3076 urvB rfa ara-9 (pKM101)
1.25 1 25
109 167
1365 2953 d
Average number reduced revertants/man/lO 8 cells c
0 0 0 0 0 5.3 6.5 17.3 No hd 10 1 Plashc hd 65 0 03 0 0 07 0 07 0 0 0 0 -87 8.0 95 N o lid 1 5 Soft white hght 0 9 Incandescent hght
- 20 9
0 0 0 0 0 07 08 0 04 Soft whtte hght 08 04 0 0 0 05 6.0 93 11 4 97 0
+ SOD +CAT +SOD+CAT + KV370 filter
14 8 Plastic hd N o hd
- 38 0
27 TABLE 1 (continued) Strata
Genotype
Plating medmm (% nutrient broth)
Averagenumber colomesper plate Sponta- 90-mmdluneous manatlon
45. TA2637 hlsC3076 urvB rfa ara-9 (pKM101) 46 hzsC3076 urvB rfa ara-9 (pKM101) 47 hlsC3076 urvB rfa ara-9 (pKM101) 48 hlsC3076 uroB rfa ara-9 (pKM101) 49. htsC3076 uroB rfa ara-9 (pKM101) 50. htsC3076 urvB rfa ara-9 (pKM101) 51. hlsC3076 urvB rfa ara-9 (pKM101) 52 hlsC3076 urvB rfa ara-9 (pKM101)
0 25 0.25 0.25 0.25 0 25 0.25 0 25 0.25
24 36 46 46 30 20 30 31
53. 54. 55 56 57
TA88 TA89 TA90 TAll0 TA97
0.25 0 25 0 25 0 25 0.25
26 42 36 40 54
29 35 27 72 305
58. 59.
TR2705 hlsO1242 hlsD6580 TA2635 htsO1242 hlsD6580 uvrB rfa
0.25 0 25
7 7
3 2
60 61.
TA95 TA96
0.25 0.25
14 55
12 233
htsO1242htsD6610 hssO1242hzsD6610 uvrB gal htsO1242htsD6610 uvrB rfa htsO1242 htsD6610 (pKM101) htsO1242hlsD6610 uvrB rfa (pKM101)
hlsO1242h~sD6580 (pKM101) htsO1242htsD6580 uvrB rfa (pKM101)
732 69 1158 - 2846 d 247 1156 30 40
Average number reduced revertants/man/108 cells c
6.2 0.39 Incandescent hght 11 8 Plastic lid -- 39 0 NO hd 1.9 + Cellulose acetate 7 9 Cellulose acetate alone 0 + Mylar 0 06 Mylar alone 0 0 0 0.15 50 0 0 0 13
a Speoal condlUons are Indicated at the end of each appropriate hne and explained m the text. b Genotypes are described m Materials and methods, Bacterial stratus c Increase m average number of colomes per plate after lllurmnatlon (spontaneous control subuacted) and adjusted for number of viable cells plated. d Crowded plates, therefore colony count only approxtmate.
etc. B r u n n e r a n d Brehm (1985) d e m o n s t r a t e d an e n h a n c e m e n t of mutagenesis b y plasmid p K M 1 0 1 m Salmonella strain htsG46 exposed to glassfiltered blue fluorescent light. Here we c o n c e n t r a t e o n e x a m i n a t i o n of fluorescent lamps to winch there is a m u c h b r o a d e r exposure. Coolwhite a n d softwhite fluorescent l a m p s are the models that are very widely used in homes, businesses, a n i m a l quarters, tissue culture hoods a n d elsewhere in the e n v i r o n m e n t . G r o w i n g cultures of Eschertchta coli a n d S a l m o n e l l a t y p h l m u r l u m have been shown to be m u t a g e m z e d b y c o n t i n u o u s exposure to unshielded coolwhite fluorescent light; an excess of exogenous t r y p t o p h a n was present i n some of these experiments ( W e b b a n d Malina, 1967, 1970) while an excess of histidine was present in the others ( M o r t e l m a n s a n d Stocker, 1976). Chinese h a m s t e r V-79 l u n g a n d ovary cells have b e e n shown to be m u t a g e n i z e d b y exposure to coolwhite fluorescent light through the lids of plastic Petri dishes (Bradley a n d Sharkey, 1977; Hsie et al.,
1977). I n a thorough study of the m u t a g e n i c t t y a n d toxicity of sunlight a n d five artificial hght sources, J a c o b s o n a n d coworkers f o u n d that light from coolwhite fluorescent l a m p s was roughly 4.5-fold more m u t a g e n i c for L5178Y m o u s e l y m p h o m a cells (at a c o m p a r a b l e total irradiance) t h a n was sunlight (Jacobson et al., 1978; J a c o b s o n a n d Krell, 1979; Krell a n d Jacobson, 1980; Lugo et al., 1980). These a n d other effects of fluorescent l a m p a n d related r a d i a t i o n s o n m a m m a l i a n cells i n vitro have been s u m m a r i z e d ( J a c o b s o n a n d Krell, 1982; Coohill et al., 1987). I n studies of the genetic significance of superoxide g e n e r a t i o n d u r i n g p h o t o m e d i a t e d redox cycling of a n t h r a q u i n o n e s ( H a r t m a n a n d G o l d s t e l n , 1989), we n o t e d that coolwhlte fluorescent light exposure of s t a t i o n a r y phase S a l m o n e l l a susp e n d e d in p h o s p h a t e buffer led to m u t a g e n e s i s in the absence of exogenous photosensltizers. M u t a genicity was readily detected even though ill u m i n a t i o n was carried out t h r o u g h the lids of
28 glass Petri dishes. We decided to investigate this phenomenon further because: (a) The mutagenicity that we observed was of surprisingly high magnitude, and the relative responses of various Salmonella strains had not been studied previously. (b) Coolwhate fluorescent lamp emissions are largely above about 340 nm, that is they are sometimes inferred to be essentially devoid of farand mid-ultraviolet light (Wurtman, 1975; General Electric Co., 1987) and mutagenesis was observed m glass dishes that we would have expected to screen out considerable far-ultraviolet light. (c) Wavelengths in the mid- (290-320 nm) and nearultraviolet (320-400 nm) are implicated in human photocarcinogenesls and other photobiological effects on mammals (Setlow, 1974; Amblard et al., 1987; Cascinelli and Marcheslnl, 1989; Council on Scientific Affairs, 1989; Koh et al., 1990). (d) It is these higher ultraviolet light wavelengths that are predominantly the ones that are increasing in solar radiations reaching the surface of the poles of the earth due to ozone depletion (Peak and Peak, 1989; Koh et al., 1990). (e) Coolwhite fluorescent lamps are ubiquitously present in households and business establishments throughout the United States. Fluorescent lights are suspect m causing eye lesions (cataracts and retinopathy) in rats and mice occupying the top rows of cages in animal quarters (Haseman et al., 1989). The incidence of melanoma is increasing in the United States at a rate of 3.4% per year (American Cancer Society, 1989). Several epidemiologacal studies, but not others, have implicated fluorescent lighting in the causation of melanomas (discussed in Maxwell and Elwood, 1983; Elwood, 1986; Swerdlow et al., 1988). Since different types of fluorescent tubes vary considerably in their emissions of shorter wavelength light (Maxwell and Elwood, 1983, 1986), a negative outcome in one epidemiological study need not be in conflict with a positive correlation in another study unless the same type of lamps and shielding were common to both studies. Since our experiments were initiated, two other laboratories have noted the mutagenicity of coolwhlte fluorescent lamps for non-growing suspensions of E. cob and Salmonella (Stankowski et al., 1989; De Flora et al., 1990a, b).
Materials and methods
Bactertal
strams.
Salmonella
typhtmurium
strain LT-2 histidine-requirlng auxotrophs are listed in Table 1 (Ames et al., 1975; Maron and Ames, 1983; Hartman et al., 1986). Nucleotide sequences participating in reversion to prototrophy of these auxotrophs are given in Hartman et al. (1986) and, for hisC3076, in Cebula and Koch (1989, 1990). Additional genetic markers are: (a) ara-9=inability to ferment arabinose (Hartman, 1987), (b) hlsO1242 = small deletion in the histidine operon promoter sequence that eliminates attenuation and thus leads to elevated transcription of the histidlne operon (Roth et al., 1966; Johnston and Roth, 1981), (c) uvrB= deletions of the uvrB DNA repair gene and adjacent base sequences including gene(s) for biotin synthesis and, in some cases, also for chlorate resistance (Maron and Ames, 1983), (d) rfa = defect in cell wall biosynthesis (Maron and Ames, 1983), (e) plasmid pKM101 (McCann et al., 1975; Maron and Ames, 1983; Walker, 1984), (f) plasmid pAQ1 (Maron and Ames, 1983), (g) srl2: :Tn 10 -- transposon insertion inactivating a gene involved with sorbltol utilization, and (h) gsh = defective in the first step of giutattuone biosynthesis (Dahl et al., 1989). Plasmid pKM101 was transferred into strains TA2898, SB4312 and SB4313 to create strains SB4315, SB4317 and SB4318, respectively, by mating with strain TA2000 (McCann et al., 1975). A multiply mutant strain containing uvrB and recA (recombination deficient) and the pKM101 plasmld (strain SB4320) was constructed by cotransduction into strain TA2410 of srl-202::TnlO and recA1. Integration-defective, high-transducing P22 phage grown on strain TT521 were used to infect TA2410 as recipient. Colonies resistant to 10 /~g/ml tetracycline (Sigma) in nutrient agar were screened for phage sensitivity, for hypersensitivity to 254 nm ultraviolet light ( u v r B - recA-), and amplcillln resistance (content of pKM101). The presence of markers affecting mutagen susceptibility and D N A repair was confirmed in tests as outlined by Ames et al. (1975) and Maron and Ames (1983).
29 Illummatlon condttions. Bacteria were grown overnight with aeration in Difco-nutrient broth (Difco Laboratories, Detroit, MI) at 37°C. The stationary phase cultures were sedlmented by centrifugation and resuspended at a concentration of about 1 0 9 cells/ml in 0.1 M phosphate buffer, pH 7.4, at ambient temperature (24 ° C). Aliquots of 4 ml of culture suspensions were pipetted into 60 × 15 mm Petrl dishes. In most experiments these were Kamball Kimax or Pyrex glass dishes; where specifically noted, Falcon No. 3002 (Becton Dickinson Labware, Lincoln, N J) polystyrene plastic dishes were used. When added, bovine erythrocyte superoxade dismutase (EC 1.15.1.1) and bovine liver catalase (EC 1.11.1.6) (Sigma Chemical Co., St. Louis, MO) were present in the cell suspensions at 315 units and 1719 units per ml, respectively. Bacteria were illuminated with the suspension 10 cm below a standard white-enameled hght fixture with no reflectors and containing two General Electric 15W coolwhite No. F15T8CW fluorescent lamps (General Electric Co., Cleveland, OH). Petri dishes were placed on a black tabletop. In a few experiments, where noted, two General Electric 15W softwhite No. F15T12SW lamps replaced the coolwhite lamps and, in order to maintain constant fight intensity, the distance from the lamp to the bacterial suspension was adjusted to 7.6 cm. Also specifically noted in the text was utilization in a few experiments of a single 75-W General Electric coolwhite incandescent light bulb 13 cm from the bacterial suspension, also giving equal hght intensity in the blue-blue green portion of the spectrum. Measured intensities at the surfaces of the bacterial suspensions were 480 footcandles (5165 lumen/m2). This is about 10fold higher than office illuminance (Maxwell and Elwood, 1983, 1986). Equalization of light intensities utilized measurements with a Gossen tri-lux footcandle model 3.66-365 light meter with a 20 × adaptor (P. Gossen and Co., Erlangen, Germany). Fresh lamps were used. Long-term aged lamps had outputs of measurable light about 20% lower than did fresh lamps, but results were parallel where tested. Total radiation from the lamps was measured
with an Eppley calibrated linear thermopile (Eppley Laboratories, Newport, RI) and using a Kiethley model 195A digital voltmeter as readout device (Kiethley Instruments Inc., Cleveland, OH). Total irradiance at 10 cm from coolwhite fluorescent lamps was 284/.tW/cm 2 (2.8 J m - 2 sec-l), at 7.6 cm from softwhlte fluorescent lamps was 314 /~W/cm 2 (3.1 J m -2 sec -1) and 13 cm from a 75-W coolwhite incandescent bulb was 22107 /~W/cm 2 (221 J m 2 sec-1). Total radiation fluence from the coolwhite fluorescent lamps was decreased about 20% by glass Petri dish lids, 23% by the KV370 filter, and 32% by plastic Petri dish lids. A light nonpermeable blank was subtracted from each of these measurements. Bacterial survival was measured before and after appropriate intervals of illumination. After serial dilution of each bacterial suspension, 0.1-ml aliquots were spread on the surfaces of duplicate or triplicate Difco nutrient agar plates followed by incubation at 37°C for 24 h before scoring. Reversion assays utilized spreading of duplicate or triplicate 0.1-ml aliquots of undiluted bacterial suspensions on the surfaces of minimal E medium (Vogel and Bonner, 1956) supplemented with 1.25% ( v / v ) or 0.25% (v/v) hquid Difco nutrient broth, 0.2% glucose, and 8 /~M biotin. Nutrient broth supplementation has been found to enhance recovery of induced mutants (Demerec and Cahn, 1953), including those induced with near-ultraviolet light (Turner and Webb, 1981; Leonardo et al., 1984). Revertant colonies were scored after 48-h incubation at 37°C. All operations were carried out in subdued room lighting and incubations were in the dark. The bacterial suspensions were illuminated m closed Petri dishes except where specifically noted. Thus, Petri dish lids served as light reflectors and filters in most experiments. Transmittance of light as a function of wavelength was measured for the lids versus air as a standard in a Varlan dual-beam spectrophotometer (series 634; Varian Instrument Group, Columbia, MD). Transmittance of Schott sharp cut-on filters KV370 and KV389 (Schott Glaswerke, Mainz, Germany) and cellulose acetate and mylar filters were measured analogously. In experiments where these filters were used, they
30 [ 800
were individually taped to the tops of glass Petri dish lids except where specifically noted as used singly.
Charactertzatwn of hlsG428 revertants. In one experiment, all revertant colonies of strain TA104 on a portion of a Petri dish were picked and streaked out on E mmimal medium (Vogel and Bonner, 1956) supplemented with glucose and biotin. Isolated single colonies were inoculated into nutrient broth. After overnight incubation, the unaerated broth cultures were dduted 1 : 1 0 in physiological saline soluuon (0.85% NaC1) and then examined for resistance to each of two hlstldine analogues m tests designed after Levin et al. (1984) as modified by H a r t m a n et al. (1984). Minimal-glucose-biotin plates (Falcon No. 3025 = 150 × 25 m m with 20-mm grids) were first streaked along grid markings at opposite edges with 0.05 ml of a solution of 25 m g / m l DL-1,2,4-triazole-3alanine (Sigma Chemical Co., St. Louis, MO) or streaked twice with 0.05 ml of 20 m g / m l fl-(2thiazolyl)-DL-alanine (Sigma). After these streaks dried, culture suspensions were streaked starting at the center of the plates on parallel grids vertical to the analogue streaks. Plates were scored for inhibition patterns after overnight incubation at 37°C and after further incubation. Results
Calculatton of mutation rates. Table 1 presents results that m the vast majority of cases reflect averages from 2 - 6 indwidual experiments. The average number of spontaneous revertant colonies per plate (2-3 plates per assay) is compared with the average number of revertant colonies obtained with the same bacterial suspension after 90 rain illumination (2-3 plates per assay) for 35 different histidine-requiring derivatives of the parent Salmonella typhtmurlum strain LT-2. An illulmnatlon time of 90 mln was chosen for m a n y experiments because tins gave a significant number of reduced revertants over spontaneous background without saturating the assay plates with revertant colonies. Induction of mutations proceeded approximately linearly with time up to 90 min in a number of experiments where this was measured.
1400 L=J
1200 FA2410= hlsG46
I000
BOO
600 Z 4OO
TA
25
MUTGEN 01637
Mutagenicity of coolwhite fluorescent light for Salmonella Zlata Hartman, Philip E. Hartman and Wesley L. McDermott Department of Bwlogy, The Johns Hopkins Umverslty, Balttmore, MD 21218 (U S A )
(Recewed 4 June 1990) (Revision received 24 August 1990) (Accepted 5 September 1990)
Keywords Ames strains; Hlstldme reversion, Coolwhlte, Softwtute, Plasrmd pKM101, uvrB, Classroom experiment
Summary The most common fluorescent lamps in use today in homes and businesses in the United States, 'coolwhite' fluorescent lamps, emit light that is mutagenlc for Salmonella. Strains that carry both a uvrB mutation and plasmid pKM101 are extremely susceptible to tbas light-induced mutation. Both base substitution and frameshift mutations can be induced w~thout substantial lethal effects on the bacteria. Induced mutations accumulate essentially as a linear function of the time bacteria are exposed to illumination. Of Salmonella histldine-requiring strains with known nucleotide target sequences ( H a r t m a n et al., 1986; Cebula and Koch, 1989, 1990), strains either carrying one of the base substitution mutations, hlsG428 and htsG46, or one of the frameshifts, htsC3076 and hlsD6610, are most haghly mutagenized whereas frameshlft strains with htsD6580 and htsD3052 exhibit lower rates of mutagenesis. Mutagenlcity does not appear to require the presence of oxygen. A filter blocking wavelengths below 370 nm eliminates mutagenesis. Polystyrene, cellulose acetate and, especially, mylar and glass filters reduce mutagenesls, indicating that at least some of the mutagenlc effects can be attributed to leakage of radiations below 290 nm (far-ultraviolet light) from 'coolwhite' lamps. The more recently introduced fluorescent 'softwhlte' lamps are roughly 10-fold less mutagenic at approximately equal hght intensity. Incandescent light bulbs are much less mutagenic than are these fluorescent lamps. Our mutational data correlate closely with previous results in eukaryotic cells (Jacobson and Krell, 1982). A uvrB recA Salmonella double mutant is hypersensitive to the lethal effects of coolwhate fluorescent light, even when illuminated through the lids of glass Petri dishes. Thus, a p p r o p n a t e Salmonella strains would appear to be simple and useful screens for both the mutagenic and the lethal activities of fluorescent lamps. These systems are amenable to classroom laboratory use as relatively safe and effective means of demonstrating environmental mutagenesis.
Correspondence" P.E Hartman, Department of Biology, The Johns Hopkins Umverslty, 34th and Charles Streets, Baltimore, MD 21218 (U S A) Abbrevtattons CAT, catalase, SOD, superoxxdedlsmutase, TA,
fl-(2-thlazolyl)-DL-alamne,TRA, DL-1,2,4-trmzole-3-alamne.
The pioneering study of Speck and Rosenkranz (1975) and other studies that have followed demonstrate mutagenicity for Salmonella of emissions from fluorescent lamps designed for specialized uses in phototherapy or to stimulate plant growth,
26 TABLE 1 I N D U C T I O N OF PROTOTROPHIC REVERTANTS BY FLUORESCENT L I G H T IN VARIOUS SALMONELLA H I S T I D I N E AUXOTROPHS Illumanatlon was with coolwhate hght and through hds of glass Petn dishes unless otherwise indicated a Strain
1° 2. 3 4. 5 6 7 8 9. 10
11. 12. 13. 14
TA1975 TA1950 TA1535 TA92 TA2410
TA100 SB5428 TA2659 TA102
15 16 17 18 19. 20
TA2898
21 22. 23 24.
TA104
25 26 27 28. 29 30. 31. 32 33 34.
SB8052 TA1538 SB4312 SB4313 TA94 TA98
35 36 37. 38 39. 40. 41 42.
SB8076 TA1537 TA2425 TA2637
43 44
TA2637
SB4315 TA2678
SB4317 SB4318
Genotype b
htsG46 htsG46 htsG46 hlsG46 htsG46 htsG46 hlsG46 hlsG46 htsG46 htsG46
rfa uvrB uvrB rfa (pKM101) uvrB (pKM101) uvrB (pKM101) uvrB (pKM101) uvrB (pKM101) uvrB (pKM101) rfa
Plating medium
Average number colomes per plate
(% nutrient broth)
Spontaneous
90-man dlununatlon
1 25 1.25 1.25 1.25 1 25 1.25 0.25 0.25 0 25 1.25
10 7 31 29 97 464 116 155 155 251
3 7 29 24 89 1215 1032 1869 1159 1112 22 3 28
htsG428 ara-9 htsG428 ara-9 hzsG428 uvrB1057 rfa-1028 ara-9 htsG8476 galE503 rfa-1027 (pKM101) (pAQ1 htsG428) hlsG428 recA1 srl-2 T n l O ara-9 h~sG428 recA1 srl-2" T n l O ara-9 htsG428 f e c a l srl-2 "TnlO ara-9 (pKM101) htsG428 recA1 srl-2"'TnlO ara-9 (pKM101) htsG428 uvrB ara-9 (pKM101) htsG428 uorB ara-9 (pKM101)
1 25 0 25 1.25
15 2 17
0 25 1.25 0.25 1.25 0 25 1 25 0.25
74 14 6 14 10 628 87
87 10 6 9 10 - 1986 d 790
hlsG428 htsG428 htsG428 hlsG428
0 25 0.25 0 25 0.25
96 116 102 78
1092 -- 2998 d 215 186
uvrB1057 uvrB1057 uorB1057 uvrB1057
rfa-1028 rfa-1028 rfa-1028 rfa-1028
ara-9 ara-9 ara-9 ara-9
(pKM101) (pKM101) (pKM101) (pKM101)
htsD3052 htsD3052 htsD3052 htsD3052 htsD3052 htsD3052 htsD3052 hlsD3052 hlsD3052 htsD3052
ara-9 uvrB rfa ara-9 uvrB gsh-6 ara-9 uvrB gsh-7 ara-9 ara-9 (pKM101) uorB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB gsh-6 ara-9 (pKM101) uvrB gsh- 7 ara-9 (pKM101)
1.25 1 25 0 25 0.25 1 25 1.25 0 25 0.25 0 25 0.25
27 50 4 2 35 53 13 7 13 19
25 38 4 3 33 154 132 13 62 51
htsC3076 htsC3076 htsC3076 hlsC3076 htsC3076 htsC3076 hlsC3076 htsC3076
ara-9 uvrB rfa ara-9 ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101) uvrB rfa ara-9 (pKM101)
1 25 1 25 1.25 1 25 1 25 1.25 1 25 1 25
12 12 21 65 64 64 67 109
7 13 34 595 987 1258 1110 117
htsC3076 urvB rfa ara-9 (pKM101) hzsC3076 urvB rfa ara-9 (pKM101)
1.25 1 25
109 167
1365 2953 d
Average number reduced revertants/man/lO 8 cells c
0 0 0 0 0 5.3 6.5 17.3 No hd 10 1 Plashc hd 65 0 03 0 0 07 0 07 0 0 0 0 -87 8.0 95 N o lid 1 5 Soft white hght 0 9 Incandescent hght
- 20 9
0 0 0 0 0 07 08 0 04 Soft whtte hght 08 04 0 0 0 05 6.0 93 11 4 97 0
+ SOD +CAT +SOD+CAT + KV370 filter
14 8 Plastic hd N o hd
- 38 0
27 TABLE 1 (continued) Strata
Genotype
Plating medmm (% nutrient broth)
Averagenumber colomesper plate Sponta- 90-mmdluneous manatlon
45. TA2637 hlsC3076 urvB rfa ara-9 (pKM101) 46 hzsC3076 urvB rfa ara-9 (pKM101) 47 hlsC3076 urvB rfa ara-9 (pKM101) 48 hlsC3076 uroB rfa ara-9 (pKM101) 49. htsC3076 uroB rfa ara-9 (pKM101) 50. htsC3076 urvB rfa ara-9 (pKM101) 51. hlsC3076 urvB rfa ara-9 (pKM101) 52 hlsC3076 urvB rfa ara-9 (pKM101)
0 25 0.25 0.25 0.25 0 25 0.25 0 25 0.25
24 36 46 46 30 20 30 31
53. 54. 55 56 57
TA88 TA89 TA90 TAll0 TA97
0.25 0 25 0 25 0 25 0.25
26 42 36 40 54
29 35 27 72 305
58. 59.
TR2705 hlsO1242 hlsD6580 TA2635 htsO1242 hlsD6580 uvrB rfa
0.25 0 25
7 7
3 2
60 61.
TA95 TA96
0.25 0.25
14 55
12 233
htsO1242htsD6610 hssO1242hzsD6610 uvrB gal htsO1242htsD6610 uvrB rfa htsO1242 htsD6610 (pKM101) htsO1242hlsD6610 uvrB rfa (pKM101)
hlsO1242h~sD6580 (pKM101) htsO1242htsD6580 uvrB rfa (pKM101)
732 69 1158 - 2846 d 247 1156 30 40
Average number reduced revertants/man/108 cells c
6.2 0.39 Incandescent hght 11 8 Plastic lid -- 39 0 NO hd 1.9 + Cellulose acetate 7 9 Cellulose acetate alone 0 + Mylar 0 06 Mylar alone 0 0 0 0.15 50 0 0 0 13
a Speoal condlUons are Indicated at the end of each appropriate hne and explained m the text. b Genotypes are described m Materials and methods, Bacterial stratus c Increase m average number of colomes per plate after lllurmnatlon (spontaneous control subuacted) and adjusted for number of viable cells plated. d Crowded plates, therefore colony count only approxtmate.
etc. B r u n n e r a n d Brehm (1985) d e m o n s t r a t e d an e n h a n c e m e n t of mutagenesis b y plasmid p K M 1 0 1 m Salmonella strain htsG46 exposed to glassfiltered blue fluorescent light. Here we c o n c e n t r a t e o n e x a m i n a t i o n of fluorescent lamps to winch there is a m u c h b r o a d e r exposure. Coolwhite a n d softwhite fluorescent l a m p s are the models that are very widely used in homes, businesses, a n i m a l quarters, tissue culture hoods a n d elsewhere in the e n v i r o n m e n t . G r o w i n g cultures of Eschertchta coli a n d S a l m o n e l l a t y p h l m u r l u m have been shown to be m u t a g e m z e d b y c o n t i n u o u s exposure to unshielded coolwhite fluorescent light; an excess of exogenous t r y p t o p h a n was present i n some of these experiments ( W e b b a n d Malina, 1967, 1970) while an excess of histidine was present in the others ( M o r t e l m a n s a n d Stocker, 1976). Chinese h a m s t e r V-79 l u n g a n d ovary cells have b e e n shown to be m u t a g e n i z e d b y exposure to coolwhite fluorescent light through the lids of plastic Petri dishes (Bradley a n d Sharkey, 1977; Hsie et al.,
1977). I n a thorough study of the m u t a g e n i c t t y a n d toxicity of sunlight a n d five artificial hght sources, J a c o b s o n a n d coworkers f o u n d that light from coolwhite fluorescent l a m p s was roughly 4.5-fold more m u t a g e n i c for L5178Y m o u s e l y m p h o m a cells (at a c o m p a r a b l e total irradiance) t h a n was sunlight (Jacobson et al., 1978; J a c o b s o n a n d Krell, 1979; Krell a n d Jacobson, 1980; Lugo et al., 1980). These a n d other effects of fluorescent l a m p a n d related r a d i a t i o n s o n m a m m a l i a n cells i n vitro have been s u m m a r i z e d ( J a c o b s o n a n d Krell, 1982; Coohill et al., 1987). I n studies of the genetic significance of superoxide g e n e r a t i o n d u r i n g p h o t o m e d i a t e d redox cycling of a n t h r a q u i n o n e s ( H a r t m a n a n d G o l d s t e l n , 1989), we n o t e d that coolwhlte fluorescent light exposure of s t a t i o n a r y phase S a l m o n e l l a susp e n d e d in p h o s p h a t e buffer led to m u t a g e n e s i s in the absence of exogenous photosensltizers. M u t a genicity was readily detected even though ill u m i n a t i o n was carried out t h r o u g h the lids of
28 glass Petri dishes. We decided to investigate this phenomenon further because: (a) The mutagenicity that we observed was of surprisingly high magnitude, and the relative responses of various Salmonella strains had not been studied previously. (b) Coolwhate fluorescent lamp emissions are largely above about 340 nm, that is they are sometimes inferred to be essentially devoid of farand mid-ultraviolet light (Wurtman, 1975; General Electric Co., 1987) and mutagenesis was observed m glass dishes that we would have expected to screen out considerable far-ultraviolet light. (c) Wavelengths in the mid- (290-320 nm) and nearultraviolet (320-400 nm) are implicated in human photocarcinogenesls and other photobiological effects on mammals (Setlow, 1974; Amblard et al., 1987; Cascinelli and Marcheslnl, 1989; Council on Scientific Affairs, 1989; Koh et al., 1990). (d) It is these higher ultraviolet light wavelengths that are predominantly the ones that are increasing in solar radiations reaching the surface of the poles of the earth due to ozone depletion (Peak and Peak, 1989; Koh et al., 1990). (e) Coolwhite fluorescent lamps are ubiquitously present in households and business establishments throughout the United States. Fluorescent lights are suspect m causing eye lesions (cataracts and retinopathy) in rats and mice occupying the top rows of cages in animal quarters (Haseman et al., 1989). The incidence of melanoma is increasing in the United States at a rate of 3.4% per year (American Cancer Society, 1989). Several epidemiologacal studies, but not others, have implicated fluorescent lighting in the causation of melanomas (discussed in Maxwell and Elwood, 1983; Elwood, 1986; Swerdlow et al., 1988). Since different types of fluorescent tubes vary considerably in their emissions of shorter wavelength light (Maxwell and Elwood, 1983, 1986), a negative outcome in one epidemiological study need not be in conflict with a positive correlation in another study unless the same type of lamps and shielding were common to both studies. Since our experiments were initiated, two other laboratories have noted the mutagenicity of coolwhlte fluorescent lamps for non-growing suspensions of E. cob and Salmonella (Stankowski et al., 1989; De Flora et al., 1990a, b).
Materials and methods
Bactertal
strams.
Salmonella
typhtmurium
strain LT-2 histidine-requirlng auxotrophs are listed in Table 1 (Ames et al., 1975; Maron and Ames, 1983; Hartman et al., 1986). Nucleotide sequences participating in reversion to prototrophy of these auxotrophs are given in Hartman et al. (1986) and, for hisC3076, in Cebula and Koch (1989, 1990). Additional genetic markers are: (a) ara-9=inability to ferment arabinose (Hartman, 1987), (b) hlsO1242 = small deletion in the histidine operon promoter sequence that eliminates attenuation and thus leads to elevated transcription of the histidlne operon (Roth et al., 1966; Johnston and Roth, 1981), (c) uvrB= deletions of the uvrB DNA repair gene and adjacent base sequences including gene(s) for biotin synthesis and, in some cases, also for chlorate resistance (Maron and Ames, 1983), (d) rfa = defect in cell wall biosynthesis (Maron and Ames, 1983), (e) plasmid pKM101 (McCann et al., 1975; Maron and Ames, 1983; Walker, 1984), (f) plasmid pAQ1 (Maron and Ames, 1983), (g) srl2: :Tn 10 -- transposon insertion inactivating a gene involved with sorbltol utilization, and (h) gsh = defective in the first step of giutattuone biosynthesis (Dahl et al., 1989). Plasmid pKM101 was transferred into strains TA2898, SB4312 and SB4313 to create strains SB4315, SB4317 and SB4318, respectively, by mating with strain TA2000 (McCann et al., 1975). A multiply mutant strain containing uvrB and recA (recombination deficient) and the pKM101 plasmld (strain SB4320) was constructed by cotransduction into strain TA2410 of srl-202::TnlO and recA1. Integration-defective, high-transducing P22 phage grown on strain TT521 were used to infect TA2410 as recipient. Colonies resistant to 10 /~g/ml tetracycline (Sigma) in nutrient agar were screened for phage sensitivity, for hypersensitivity to 254 nm ultraviolet light ( u v r B - recA-), and amplcillln resistance (content of pKM101). The presence of markers affecting mutagen susceptibility and D N A repair was confirmed in tests as outlined by Ames et al. (1975) and Maron and Ames (1983).
29 Illummatlon condttions. Bacteria were grown overnight with aeration in Difco-nutrient broth (Difco Laboratories, Detroit, MI) at 37°C. The stationary phase cultures were sedlmented by centrifugation and resuspended at a concentration of about 1 0 9 cells/ml in 0.1 M phosphate buffer, pH 7.4, at ambient temperature (24 ° C). Aliquots of 4 ml of culture suspensions were pipetted into 60 × 15 mm Petrl dishes. In most experiments these were Kamball Kimax or Pyrex glass dishes; where specifically noted, Falcon No. 3002 (Becton Dickinson Labware, Lincoln, N J) polystyrene plastic dishes were used. When added, bovine erythrocyte superoxade dismutase (EC 1.15.1.1) and bovine liver catalase (EC 1.11.1.6) (Sigma Chemical Co., St. Louis, MO) were present in the cell suspensions at 315 units and 1719 units per ml, respectively. Bacteria were illuminated with the suspension 10 cm below a standard white-enameled hght fixture with no reflectors and containing two General Electric 15W coolwhite No. F15T8CW fluorescent lamps (General Electric Co., Cleveland, OH). Petri dishes were placed on a black tabletop. In a few experiments, where noted, two General Electric 15W softwhite No. F15T12SW lamps replaced the coolwhite lamps and, in order to maintain constant fight intensity, the distance from the lamp to the bacterial suspension was adjusted to 7.6 cm. Also specifically noted in the text was utilization in a few experiments of a single 75-W General Electric coolwhite incandescent light bulb 13 cm from the bacterial suspension, also giving equal hght intensity in the blue-blue green portion of the spectrum. Measured intensities at the surfaces of the bacterial suspensions were 480 footcandles (5165 lumen/m2). This is about 10fold higher than office illuminance (Maxwell and Elwood, 1983, 1986). Equalization of light intensities utilized measurements with a Gossen tri-lux footcandle model 3.66-365 light meter with a 20 × adaptor (P. Gossen and Co., Erlangen, Germany). Fresh lamps were used. Long-term aged lamps had outputs of measurable light about 20% lower than did fresh lamps, but results were parallel where tested. Total radiation from the lamps was measured
with an Eppley calibrated linear thermopile (Eppley Laboratories, Newport, RI) and using a Kiethley model 195A digital voltmeter as readout device (Kiethley Instruments Inc., Cleveland, OH). Total irradiance at 10 cm from coolwhite fluorescent lamps was 284/.tW/cm 2 (2.8 J m - 2 sec-l), at 7.6 cm from softwhlte fluorescent lamps was 314 /~W/cm 2 (3.1 J m -2 sec -1) and 13 cm from a 75-W coolwhite incandescent bulb was 22107 /~W/cm 2 (221 J m 2 sec-1). Total radiation fluence from the coolwhite fluorescent lamps was decreased about 20% by glass Petri dish lids, 23% by the KV370 filter, and 32% by plastic Petri dish lids. A light nonpermeable blank was subtracted from each of these measurements. Bacterial survival was measured before and after appropriate intervals of illumination. After serial dilution of each bacterial suspension, 0.1-ml aliquots were spread on the surfaces of duplicate or triplicate Difco nutrient agar plates followed by incubation at 37°C for 24 h before scoring. Reversion assays utilized spreading of duplicate or triplicate 0.1-ml aliquots of undiluted bacterial suspensions on the surfaces of minimal E medium (Vogel and Bonner, 1956) supplemented with 1.25% ( v / v ) or 0.25% (v/v) hquid Difco nutrient broth, 0.2% glucose, and 8 /~M biotin. Nutrient broth supplementation has been found to enhance recovery of induced mutants (Demerec and Cahn, 1953), including those induced with near-ultraviolet light (Turner and Webb, 1981; Leonardo et al., 1984). Revertant colonies were scored after 48-h incubation at 37°C. All operations were carried out in subdued room lighting and incubations were in the dark. The bacterial suspensions were illuminated m closed Petri dishes except where specifically noted. Thus, Petri dish lids served as light reflectors and filters in most experiments. Transmittance of light as a function of wavelength was measured for the lids versus air as a standard in a Varlan dual-beam spectrophotometer (series 634; Varian Instrument Group, Columbia, MD). Transmittance of Schott sharp cut-on filters KV370 and KV389 (Schott Glaswerke, Mainz, Germany) and cellulose acetate and mylar filters were measured analogously. In experiments where these filters were used, they
30 [ 800
were individually taped to the tops of glass Petri dish lids except where specifically noted as used singly.
Charactertzatwn of hlsG428 revertants. In one experiment, all revertant colonies of strain TA104 on a portion of a Petri dish were picked and streaked out on E mmimal medium (Vogel and Bonner, 1956) supplemented with glucose and biotin. Isolated single colonies were inoculated into nutrient broth. After overnight incubation, the unaerated broth cultures were dduted 1 : 1 0 in physiological saline soluuon (0.85% NaC1) and then examined for resistance to each of two hlstldine analogues m tests designed after Levin et al. (1984) as modified by H a r t m a n et al. (1984). Minimal-glucose-biotin plates (Falcon No. 3025 = 150 × 25 m m with 20-mm grids) were first streaked along grid markings at opposite edges with 0.05 ml of a solution of 25 m g / m l DL-1,2,4-triazole-3alanine (Sigma Chemical Co., St. Louis, MO) or streaked twice with 0.05 ml of 20 m g / m l fl-(2thiazolyl)-DL-alanine (Sigma). After these streaks dried, culture suspensions were streaked starting at the center of the plates on parallel grids vertical to the analogue streaks. Plates were scored for inhibition patterns after overnight incubation at 37°C and after further incubation. Results
Calculatton of mutation rates. Table 1 presents results that m the vast majority of cases reflect averages from 2 - 6 indwidual experiments. The average number of spontaneous revertant colonies per plate (2-3 plates per assay) is compared with the average number of revertant colonies obtained with the same bacterial suspension after 90 rain illumination (2-3 plates per assay) for 35 different histidine-requiring derivatives of the parent Salmonella typhtmurlum strain LT-2. An illulmnatlon time of 90 mln was chosen for m a n y experiments because tins gave a significant number of reduced revertants over spontaneous background without saturating the assay plates with revertant colonies. Induction of mutations proceeded approximately linearly with time up to 90 min in a number of experiments where this was measured.
1400 L=J
1200 FA2410= hlsG46
I000
BOO
600 Z 4OO
TA
