Non-enzymatic DNA strand breaks induced in mammalian cells by fluorescent light

11

Biochimica et Biophysica Acta, 520 (1978) 11--20 © Elsevier/North-Holland Biomedical Press

BBA 99228

NON-ENZYMATIC DNA STRAND BREAKS INDUCED IN MAMMALIAN CELLS BY FLUORESCENT LIGHT

MATTHEWS O. BRADLEY *, LEONARD C. ERICKSON and KURT W. KOHN

Laboratory of Molecular Pharmacology, Developmental The rapeu tics Program, Division of Cancer Treatment, National Institutes of Health, National Cancer Institute, Bethesda, Md. 20014 (U.S.A.) (Received August 19th, 1977) (Revised manuscript received February 13th, 1977)

Summary Fluorescent light (5.4 J . m -2. s-') induces 0.041 single-strand breaks per 108 daltons per h in the DNA of cultured Chinese hamster cells ( 4 . 8 . 1 0 -6 breaks per 108 daltons per J • m-2). The breaks are induced at I°C and hence are not likely to be the result of endonuclease incision. When the cells are incubated at 37°C, the breaks are rejoined within 2 h. At least two lesions are responsible for the observed effects. One lesion has the ability to break DNA subsequently treated with alkali but is neither toxic nor mutagenic. This lesion is produced by light of wavelength greater than approx. 350 nm. The other lesion(s) produce mutagenicity and/or toxicity, but do not necessarily produce strand breaks. These lesion(s) are produced by light of wavelength less than 350 nm.

Introduction The radiation emitted from fluorescent lamps (fluorescent light) is both toxic and mutagenic to V-79 Chinese hamster lung cells [1], The effects are mainly due to an interaction of light with the cells themselves rather than to photochemical reactions with components of the medium [ 1]. Little is known about the nature of DNA damage in mammalian cells produced by wavelengths above 295 nm, presumably the biologically active region of the fluorescent light spectrum [1]. It is known that 313 nm light will produce thymine dimers in Escherichia coli DNA [2] and that 365 nm light will produce single~trand breaks [3], dimers [4], and non
w h o m correspondence should be addressed. Abbreviation: HEPES, N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid.

12 products in procaryotic DNA [5]. Since fluorescent light consists of a broad mixture of wavelengths, complicated synergisms might occur between different portions of the spectrum. For example, Peak et al. [6] showed that either preirradiation or concomitant irradiation at 334 nm produces a synergistic inactivation of transforming DNA with 365 nm light. In this work we have examined the relationship between the mutagenicity and toxicity of fluorescent light in mammalian cells and the production of DNA damage. We show that fluorescent light produces single-strand breaks, presumably w i t h o u t enzymatic action, and that these breaks are rapidly rejoined. There are at least two types of lesions, one of which is neither mutagenic nor toxic and the other(s) of which may be. We further show that incandescent light is neither toxic nor mutagenic b u t that it produces DNA singlestrand breaks. Materials and Methods

Cell culture and labeling. The male Chinese hamster lung cells, V-79-4, were cultured as monolayers in a modified Eagles minimal essential medium containing 5% fetal bovine serum (CM) as previously described [1]. L1210 cells, used as internal controls for the elution assay, were grown in spinner culture with RPMI 1630 medium containing 20% heat-inactivated fetal bovine serum as previously described [7]. The DNA of V-79 or L1210 cells was labeled by a 20 h incubation of exponentially growing cells in medium containing [2-14C]thymi dine (58 Ci/mol, 0.02 pCi/ml, New England Nuclear, Boston, Mass.) or [Me3H]thymidine (20 Ci/mmol, 0.1 ~Ci/ml, New England Nuclear, Boston, Mass.), containing enough unlabeled thymidine to give a final concentration of 10 -6 M in the culture medium. Toxicity and mutagenicity. The determinations of toxicity as colony survival after illumination and mutagenicity as frequency of resistance to 6-thioguanine were made as previously described [1]. Illumination. Fluorescent illumination with Sylvania Cool White F15T8-CW lamps was as previously described [1] except that the cells were held at either I or 37°C during illumination. Coming 150 cm: plastic flasks containing 250 ml of distilled water (2.5 cm) were placed between the lamps and the cells in order to prevent heating of the medium. The same flasks containing p-aminobenzoic acid {12.5% (w/v), 2.5 cm path length) dissolved in 95% ethanol were used as filters between the fluorescent light and the cells. The cells were 8 cm from the fluorescent bulbs. The visible light intensity impinging on the cells was measured with an Eppley Thermopile (The Eppley Laboratory, Inc., Newport, R.I.) under the conditions used in each of the experiments. Visible light intensity was determined by subtracting the thermopile reading obtained with a K o d a k 70 filter (opaque to wavelengths below 660 nm) from the reading obtained w i t h o u t the filter. In this way the visible light intensity was determined w i t h o u t including the infrared intensity. Ultraviolet irradiation was from a General Electric 254 nm germicidal lamp calibrated with a short wave ultraviolet intensity meter (J-225 UV meter, UV Products, Inc., Calif.) to deliver 1 W/m 2. Prior to irradiation the cells were chilled to I ° C on an ice-water bath and irradiated in 2 ml of CM. Cells were

13 either harvested immediately for elution or kept at I ° C for 3 h until harvest. Incandescent illumination of V-79 cells was performed as described above for fluorescent light except that a 60 W bulb was used in place of the fluorescent tubes. The amount of light at the level of the cells was measured as above with the Eppley Thermopile. Alkaline elution. The alkaline elution assay has been extensively described and characterized in previous publications [7,9--12]; therefore, we will only describe aspects that are pertinent to this work. For illumination at 1°C or 37°C the cultures were first temperature equilibrated in Falcon 100-ram plastic petri plates containing 10 ml of CM supplemented with 50/~M HEPES buffer, pH 7.2. After illumination the plates were rinsed once with 10 ml of balanced saline solution (1°C) containing 0.02% EDTA; 2 ml more of this solution was added and the cells were gently scraped off the plates with a rubber policeman. The clumps were aspirated to a single cell suspension in 8 ml of CM; the suspensions were counted with a Coulter Counter Model B and kept on ice until placed on the filter for lysis. Single-strand break frequencies were calculated from the relationship Break frequency = 4.05 • 10 -1° BT - - B0 dalton-~ where B T and Bo are relative elution rates of treated and control 14C-labeled cells, respectively, and B~ and B~ are relative elution rates of 150 rad-irradiated and unirradiated L 1 2 1 0 cells [14]. The break frequency produced by 150 rad in L1210 cells is assumed to be 2.7 • 10 -12 dalton -I • tad -1 × 150 rad = 4.05 • 10 -1° dalton -I [10]. B ~ - - B ~ = 1 . 0 0 0 - - 0 . 1 8 5 = 0.185. Relative elution rate (B) was calculated from the formula: B = log R0.8 -- log R0.s log(0.8) -- log(0.5) where R0.8 and Ro.s are the fraction of the [14C]DNA that remained on the filter when 0.8 or 0.5 o f the reference [SH]DNA remained on the filter. A relative elution rate of 1.0 is obtained after 150 R X-ray. X-irradiation was delivered to cells at 0°C, by two vertically opposed Phillips RT-250 X-ray Tubes operating at 200 keV, 15 mA, with 0.25 mm Cu and 0.55 mm A1 Filters. Fluorometric assay for DNA after alkaline elution. The elution assay fractionates the total DNA from 106 cells into a number of vials containing approx. 3.2 ml of tetrapropylammonium hydroxide at pH 12.1. To chemically assay this DNA it must be quantitatively recovered from the tetrapropyl solution, washed of any fluorescent contaminants, dried, and reacted with 3,5-diaminobenzoic acid dihydrochloride in a small volume. The method we developed was based on that of Kissane and Robins [13]. After the elution was completed, 2 vols (6.5 ml) 95% ethanol containing 0.3 M sodium acetate and 0.055 M glacial acetic acid were added to all vials which were then placed at --20°C for 2 h. An acidic pH was necessary to achieve quantitative recovery of the DNA. The precipitated DNA from 2--3 fractions was collected on a Gelman alpha-6-metricel 25 mm 0.45 ~M pore size filter held in a 30 position Mfllipore No. 3025 sampling manifold (Millipore Corp., Bedford, Mass.). The vials and the filters were rinsed twice with 4 ml

14

66% ethanol at --20°C. The filters were removed to clean scintillation vials, dried at 37°c, and reacted at 60°C in an air oven for 45 rain with 0.2 ml of 2 M 3,5-diaminobenzoic acid dihydrochloride that had been previously passed by centrifugation four times through activated charcoal (Sigma No. C4386 washed with HC1) to remove impurities and lower blank values. Once the reactants had cooled to room temperature, 1.8 ml 1 M HC1 was added to the vials. The resulting fluorescence was measured in an Aminco-Bowman spectrofluorometer (American Instrument Co., Silver Spring, Md.). Standards were made from calf t h y m u s DNA (Sigma Chemical Co., St Louis, Mo.). dissolved in 0.3 M NH4OH, sonicated, and stored at 4 ° C. The recovery of DNA from the filters was approx. 91%. Alkaline sedimentation. The cells were prepared for linear alkaline sucrose gradient sedimentation by the same procedure as for alkaline elution. The cells were lysed, digested with proteinase-K, layered, and centrifuged as described previously [7,10] except that the 4% sucrose solution contained 0.2 M NaOH instead of 0.1 M NaOH.

Results

DNA strand breakage by fluorescent light Fig. 1 shows the alkaline elution patterns of DNA from the cells after 0, 1, 2 and 3 h of illumination with 5.4 J • m -2 • s -1 of fluorescent light in complete medium at I°C. The rate of elution after 3 h of illumination (5.9 • 104 J / m 2) corresponded to the effect of 450 rad of X-ray (compare Figs. 1 and 2) which

:::1 0.8

0.7

~"e-. e

0.6 90,5

O.2

I

5

L

I

10 15 Hours of Elution

Fig. I. Kinetics of alkaline elution after fluorescent light. T h e cells were irradiated in m e d i u m at 1 ° C for 0, control at I ° C with no irradiation; e-----.•, I h; • . . . . . . •, 2 h ; • •, 3 h (2.0 • 10 4 , 4.0 • 10 4 and 5.9 • 10 4 J/m 2 , respectively), The fraction of DNA remaining on t h e filter is p l o t t e d s e m i - l o g a r i t h m i c a l l y as a f u n c t i o n o f t i m e o f e l u t i o n .

the following times: ©

15 TABLE I THE

EFFECTS

STRAND

BREAK

OF

VARIOUS

PRODUCTION

LIGHT

SOURCES

ON

SURVIVAL,

MUTAGENICITY,

AND

DNA

IN V-79 CELLS

I r r a d i a t i o n s w e r e d o n e a t I ° C o n ice a s d e s c r i b e d i n M a t e r i a l s a n d M e t h o d s . T h e c a l c u l a t i o n s o f b r e a k s / d a l t o n were m a d e as d e s c r i b e d in M a t e r i a l s a n d M e t h o d s a n d t h e c a l c u l a t i o n s of b r e a k s • cell- I " m 2 • J - I w e r e m a d e o n t h e a s s u m p t i o n t h a t a V - 7 9 c e l l c o n t a i n s 1 0 p g o f D N A . F o r s u r v i v a l s t u d i e s the c e l l s w e r e allowed to attach for 4 h before illumination began. Source

Dose (J/m 2 )

Germicidal Fluorescent

7.5 2.0. 104 5.9. 104

P-Aminobenzoic acid filtered fluorescent Incandescent Control T

1.6 • 104 3.2 • 104 9.8 • 104 0

Percent * survival

3 5 -+ 72 ± 35 + 104 99 95 100

2.6 5 *** 3.2'**

+ 7 ± 13 ± 7 + 11

Mutant * * frequency

3.6 . 10 -4 7.2. 10 -5.** 2.4.10 --4*** 9.4 7,8 3.9 4.5 ± 2.6

• • • •

10 -6 10 -6 10 -6 10 -6

Breaks/ dalton

Breaks • c e l l - 1. j-1. m 2

2.7 . 10 -11 0.3 • 10 -9 1 . 2 . 1 0 -9

3 . 6 . 101 1 . 5 • 1 0 -1 2 . 0 . 1 0 -1

0 . 2 • 10 -9 0.4 • 10 -9 0.3 ' 10 -9 0

1.3 " 1 0 -1 1.3 • 1 0 -1 3.1 • 10 -2 0

* T h e m e a n a n d s t a n d a r d d e v i a t i o n o f six d e t e r m i n a t i o n s in t w o separate e x p e r i m e n t s . ** T h e m e a n o f t w o s e p a r a t e e x p e r i m e n t s e x c e p t f o r t h e c o n t r o l w h i c h a r e t h e m e a n a n d s t a n d a r d d e v i ation of four determinations and four experiments. *** Data from Bradley and Sharkey [1]. t T h e c o n t r o l s are d e f i n e d b y t h i s a n a l y s i s a s c o n t a i n i n g n o b r e a k s i n t h e i r D N A .

corresponds to a singie-strand break frequency of 1.2 • 10 -9 breaks per dalton of DNA (Table I). The same effect is produced whether the cells are irradiated on plastic or on glass. Because the elution kinetics are almost linear and are very similar to those produced by X-ray, we presume that the majority of the DNA lesions are n o t alkali sensitive but rather are formed in vivo by an interaction between light and the cells themselves. This reasoning is based on the observation that methyl nitrosourea (MNU, NSC No. 23909), an agent known to produce alkali-sensitive lesions, causes the rate of elution to increase with time as one would expect if the lesions are being progressively broken at pH 12.1 [12,14]. Such lesions produce a shouldered elution curve as opposed to the linear or slightly convex shape caused by X-ray or fluorescent light [12,14]. If the cultures are illuminated at 37°C for 3 h, and harvested on ice immediately after illumination, then the expected number of breaks is decreased by a b o u t t w o thirds. This observation suggests that the breaks are being rapidly rejoined, so that a steady-state break frequency develops that is dependent upon the relative rates of break production and repair. Since fluorescent light produces DNA strand breaks at I ° C (Fig. 1 ) a temperature at which enzymatic reactions are slow, breaks are n o t likely to be due to enzymatic action. Furthermore, the number of breaks do n o t increase after irradiation if the cells are held at I ° C for 2 h. These observations are both consistent with the idea that the breaks are produced by a non-enzymatic process. Radiation of wavelength 254 nm is known to produce thymine dimers [15] in DNA with breaks appearing within cells by a repair-endonuclease incision of DNA in the region of the thymine dimer [16]. Since fluorescent light produces apparently non-enzymatic breaks, we tested whether 254 nm light from the germicidal lamp at I ° C would also produce non-enzymatic breaks. We used a dose of germicidal light (7.5 J / m 2) that kills the same number of cells as 3 h of

16

0.9

-'~.

0.8 if \~',,.

\ .

"'18t

~ 0.4 O

g 0.3

0.2

t

5

I

i

10 15 Hours of Elution

'

5

:

[

15 Fraction Number

I

It

25

Fig. 2. E f f e c t s of v a r i o u s i r r a d i a t i o n c o n d i t i o n s . T h e cells w e r e p r e c h i l i e d t o 1 ° C b e f o r e i r r a d i a t i o n a n d w e r e h e l d a t t h a t t e m p e r a t u r e d u r i n g i r r a d i a t i o n a n d u n t i l lysis. T h e c o n t r o l s w e r e h e l d a t l ° C f o r 3.5 h . In order from the b o t t o m : • e , 3 h f l u o r e s c e n t light at I ° C (5.9 • 104 j / m 2 ) ; • . . . . . .z~ 3 0 0 t a d X-ray at I ° C w i t h o u t r e p a i r ; • - - - - - ~ , f l u o r e s c e n t light f i l t e r e d b y 2.5 c m o f 1 2 . 5 % ( w / v in 95% e t h a n o l ) p - a m i n o b e n z o i c acid f o r 3 h ( 3 . 2 , 1 0 4 J / m 2 ) ; • . . . . . - o 3 h i n c a n d e s c e n t light ( 9 . 8 - 1 0 4 J / m 2 ) ; ..... -~, 7.5 J / m 2 g e r m i c i d a l l a m p ; • "-, 3 h f l u o r e s c e n t light b l o c k e d w i t h o n e t h i c k n e s s o f b l a c k c o n s t r u c t i o n p a p e r ; o . . . . . . o, c o n t r o l ; • A 3 h f l u o r e s c e n t light b l o c k e d b y a l u m i n u m foil at I ° C . Fig. 3. Alkaline s u c r o s e gradients of D N A f r o m cells i r r a d i a t e d w i t h f l u o r e s c e n t light, o o, c o n t r o l ; e--.--.•, l h f l u o r e s c e n t light; $ - - - - - - o , 2 h f l u o r e s c e n t light (2.0 . 1 0 4 a n d 4 , 0 . 1 0 4 J / m 2, r e s p e c tively ).

fluorescent light (35% survival; 5.9 • 104 J/m2). Fig. 2 and Table I show that cells irradiated with 7.5 J/m 2 of germicidal light and held until lysis at I°C contain only 2% of the number of single-strand breaks as do fluorescent light at doses giving the same inactivation (35% survival). No time-dependent enzymatic action occurs since the breaks produced immediately after irradiation do n o t increase during 3 h of incubation at 1 ° C, suggesting that germicidal light produces DNA strand breaks w i t h o u t enzymatic action. The efficiency of germicidal light induced breaks is 36 b r e a k s - c e l l -1. m 2- j-1 or approx. 100 times higher than fluorescent ligher per J / m 2 of incident energy. The t w o light sources can thus be differentiated on the bases of efficiency and that at equal toxicities fluorescent light produces 45 times more breaks than does germicidal light (Table I). Because fluorescent light produces so many apparent single-strand DNA breaks we wondered whether the breaks could also be detected with the less sensitive alkaline sucrose gradient method. Fig. 3 shows that 1 and 2 h of fluorescent light produces dose-dependent decreases in the apparent DNA molecular weight. In human cells, similar shifts in sedimentation profile were seen after 500 rad of X-ray [7]. Therefore, t w o independent methods of analysis show that fluorescent light produces similar numbers and types of single-strand breaks in mammalian cells.

17 1.0 1.0' 0.9

0.9 0.8

\

~

\\ ,\ \\b

i

..~

•\

"'•.. oN •\

\

i.7. c 0.6 o

~.~

0.

"o

' O,•

"oN

~,

•~ 0.5 r~

._c 0.5

e~ zm 0.4 "6

~ 0.4

._~

._~

u. 0.3

ft. 0.3

0.2

l

°\ °

0.7 u. 0.6

l

-II..~

0.8 0.7

1

F o.. O

5

L J 15 I0 HoursofElufion

20

0.2

I

5

1

I

10 15 Hours of Elution

Fig. 4 . K i n e t i c s o f a l k a l i n e e l u t i o n a f t e r f l u o r e s c e n t light: u n l a b e l e d D N A . T h e cells w e r e i r r a d i a t e d as in Fig. 1 for 3 h (5.9 • 104 J/m2). The DNA was not radioactively labeled and was assayed by the fluorometric m e t h o d o f d e s c r i b e d in M a t e r i a l s a n d M e t h o d s . c o control at I°C with no irradiation; o ..... -~, 3 0 0 t a d o f X - r a y ; • -', 3 h o f f l u o r e s c e n t l i g h t a t 1 ° C . Fig. 5. K i n e t i c s o f s i n g l e ~ t r a n d r e j o i n i n g a f t e r f l u o r e s c e n t l i g h t . C u l t t t r e s w e r e i r r a d i a t e d f o r 3 h a t I ° C ( 5 . 9 • 1 0 4 J / m 2) a n d t h e n r e t u r n e d t o 3 7 ° C f o r v a r l o u s t i m e s o f r e p a i r . In o r d e r f r o m t h e b o t t o m t o the top: • "-, 3 h f l u o r e s c e n t l i g h t , I ° C , n o r e p a i r ; • - - - o , 1 5 m i n r e p a i r ; 0 - - - 4 , 3 0 m i n r e p a i r ; • - - - - o , 8 0 m i n r e p a i r ; 0 - - -e, 1 3 0 m i n r e p a i r .

['4C]Thymidine was used to label DNA in all of the foregoing experiments with fluorescent light. Because certain manufacturing procedures for [14C]thymidine may result in 5-bromodeoxyuridine contamination [17], we tested whether this might account for the observed single-strand breaks produced by fluorescent light. A fluorometric m e t h o d to assay eluted, unlabeled DNA (see Materials and Methods) shows that the elution kinetics after fluorescent light are the same when the DNA is unlabeled as when it is radioactively labeled (Fig. 4). Therefore, the observed strand breaks are not due to contamination of the isotopic thymidine.

Rejoining of DNA strand breaks produced by fluorescent light If the cells are returned to 37°C after 3 h of fluorescent illumination at 1°C, then the rate of strand rejoining is very rapid as shown in Fig. 5. Rejoining is virtually completed by 130 min. Similar rates of repair were seen after equivalent numbers of breaks were introduced by both X-ray and fluorescent light (unpublished data) suggesting that both types of damage are repaired by the same cellular system. Because fluorescent light produces DNA damage that appears very similar to that produced by X-ray, we wondered whether the effect of fluorescent light would be blocked by X-ray transparent but light opaque materials. Fig. 2 shows

18 that both black construction paper and aluminum foil block measurable fluorescent light damage to DNA. We presume, therefore, that fluorescent lights are not producing levels of energy in the X-ray region of the electromagnetic spectrum that are detectable by this m e t h o d o l o g y (a lower limit of approx. 20 rad).

Incandescent light Since fluorescent light causes mutations, toxicity, and DNA strand breaks in hamster cells, it is possible that incandescent light would have similar effects. However, as shown in Table I light from a 60 W incandescent bulb is neither toxic nor mutagenic to V-79 cells held at 1°C in medium during a 3 h illumination. By comparison, fluorescent light under similar conditions is both toxic and mutagenic [1]. As seen in Fig. 2, incandescent light for 3 h at I ° C {9.8. 104 J / m 2) produces a frequency of DNA strand breaks equivalent to approx. 1 h of illumination with fluorescent light ( 2 . 0 . 1 0 4 J/m2). This observation shows that incandescent light can produce a lesion in DNA that is either not toxic and mutagenic at all, or only very slightly so when compared to fluorescent light.

Effective spectral regions As shown above, incandescent light produces neither mutagenicity nor toxicity b u t some strand breaks. This observation implies that the wavelengths producing toxicity and mutagenicity are less than 350 nm, the lowest wavelength effectively emitted by incandescent bulbs. We attempted to test this hypothesis by using a second flask, placed as a filter between the cells and the fluorescent light source, containing 12.5% p-aminobenzoic acid (2.5 cm thick). This procedure reduces the percent transmittance to less than 1% at wavelengths below 345 nm. If cultures are illuminated for 3 h at 1 ° C with fluorescent light filtered through p-aminobenzoic acid, then the incident light is neither toxic nor mutagenic as shown in Table I. However, under the same conditions DNA strand breaks are produced at a frequency equivalent to approx. 1.5 h of unfiltered fluorescent light (Table I, Figs. 1 and 2). This result confirms the conclusion drawn from the incandescent light studies that wavelengths greater than 350 nm can produce a lesion in DNA, expressed as DNA strand breaks, that is neither toxic nor mutagenic. Discussion

In this report we have shown that fluorescent light produces single strand breaks in the D N A of Chinese hamster cells grown in tissue culture. The breaks are unrelated to pyrimidine dimers and are also apparently distinct from previously described chemically induced alkaline-sensitive lesions. Tyrell et al. [3] showed that 365 nm radiation causes single-strand breaks in extracted phage T4 DNA and in bacterial DNA in vivo. It is possible that these breaks are due to some tightly b o u n d photosensitizer(s). However, to our knowledge, this is the first report of such DNA damage in mammalian cells produced by fluorescent or near-ultraviolet light sources. The strand breaks produced by fluorescent light are presumably generated by non-enzymatic mechanisms. The evidence for this hypothesis is: (1) the

19 breaks are produced at 1 ° C, a temperature where most enzymatic reactions are very slow, (2) the number of breaks remains constant at I ° C for 2 h after illumination showing no delay between illumination and break production, and (3) breaks are produced by fluorescent light in DNA isolated from cells that have been lysed and deproteinized under conditions where no enzymes should survive (Bradley, M.O., unpublished). Given this evidence, it seems most likely to us that fluorescent light produces breaks by a non-enzymatic mechanism either directly in DNA itself or by the action of some unknown sensitizer. Other sorts of DNA damage have been shown to be produced by near ultraviolet light. For example, Cabrera-Ju~rez and Setlow [5] showed that irradiation with 334 and 365 nm of transforming DNA from Haemophilus influenzae produced a p h o t o p r o d u c t distinct from the cyclobutane thymine dimer. However, other evidence exists suggesting that pyrimidine dimers can also be induced by 365 nm irradiation of bacterial DNA [4]. H o w the strand breaks that we see in the elution and sedimentation assays are related to these previous findings remains to be determined. Both incandescent and p-aminobenzoic acid filtered fluorescent light, producing wavelengths greater than approx. 345 nm, cause single-strand breaks without concomitant mutagenicity or toxicity. As seen in Table I, similar break frequencies produce toxicity and mutagenicity with unfiltered b u t n o t with filtered fluorescent light. We interpret this result to mean that fluorescent light produces at least one DNA lesion resulting in a break that is neither toxic nor mutagenic or n o t detectably so under our conditions. The chemical properties of this lesion as well as its mechanism of strand rejoining may, by comparison, help us to understand the nature of more p o t e n t DNA lesions. A second implication is that other agents may exist which produce DNA strand breaks b u t are not mutagenic. The ability to produce DNA breaks has been reported to correlate well with mutagenicity/carcinogenicity [18], however, our results suggest that this correlation may n o t be invariant. Although the mutagenicity and toxicity of fluorescent light is removed when p-aminobenzoic acid is used as a filter, approximately half of the DNA singlestrand breaks still remain. We presume, therefore, that at least one other DNA lesion is responsible for the remaining strand breaks. Whether or not these breaks result in either toxicity or mutagenicity remains unanswered. When cells are irradiated with fluorescent light at I ° C and then returned to 37°C, the DNA strand breaks are repaired within 2 h. Although the rate of repair of V-79 cells after equivalent doses of X-ray has not been rigorously studied, it appears that strand break ligation after X-ray also occurs rapidly within 2--4 h [7]. It is, therefore, possible that X-ray and fluorescent lightinduced breaks are both repaired by the same system. Whether or n o t the mutagenic and toxic damage produced b y light of between 295 and 350 nm is repaired by the same system as is the non-mutagenic and non-toxic lesions produced by light above 350 nm is something for further study. It is possible that different enzyme systems repair the two types of damage by different mechanisms. Gantt et al. [19] have recently shown that fluorescent light causes DNA crosslinking in mouse e m b r y o cells. We have been able to confirm their observations with V-79 cells and have shown that a small number of crosslinks appear

20 at 37°C 3 h after irradiation at I°C but that these crosslinks have disappeared by 21 h, presumably by some sort of repair process (unpublished observation). However, crosslinking may be relatively unimportant in the mutagenic effects of fluorescent light since cross-linking agents such as nitrogen mustard and phenylalanine mustard are much more toxic than mutagenic (Bradley, M.O., in preparation) whereas by comparison fluorescent light is much more mutagenic than toxic. As shown in Table I and Fig. 2, germicidal light produces single-strand breaks in the DNA of cultured V-79 cells at I°C. We presume these breaks are nonenzymatic since they are formed in cells held at l°C and since they do not increase at I°C between 0 and 3 h after irradiation. The efficiency of break production per unit energy flux per cell is 180 times higher for 254 nm light than it is for fluorescent light but the number of breaks produced per cell at the same survival is 45 times higher for fluorescent than germicidal light. Marmur et al. [20], and Moroson and Alexander [21] have previously shown that breaks are produced by very high doses of 254 nm light in isolated procaryotic DNA but to our knowledge this is the first report of non-enzymatic DNA strand breaks produced by germicidal light which intact mammalian cells at high survival levels. If the germicidal lamps emit near ultraviolet radiation (above 295 nm) then the same wavelength(s) may be responsible for strand break production by both light sources. Alternatively, 254 nm light may be a very efficient DNA strand breaker. Further work with monochromatic light sources should resolve these questions. Acknowledgements Thanks are given to Nancy A. Sharkey and Irene Clark for help with various parts of this work and to Madie Tyler for typing the manuscript. References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Bradley, M.O. and Sh~rkey, N.A. (1977) Nature 2 6 6 , 7 2 4 - - 7 2 6 Setlow, R.B. and Carrier, W.L. (1966) J. Mol. Biol. 1 7 , 2 3 7 - - 2 5 4 TyrreIl, R.M., Ley, R.O. and Webb, R.B. (1974) Photoch e m. Photobiol. 20, 395--398 TyrreU, R.M. (1973) P h o t o c h e m . Photobiol. 17, 69--73 Cabrera-Judrez, E. and Setlow, J.K. (1977) Biochim. Biophys. Acta 2 6 2 , 2 2 0 - - 2 2 6 Peak, M J . , Peak, J.G. and Webb, R.B. (1975) P h o t o c h e m . Photobiol. 2 1 , 1 2 9 - - 1 3 1 Bradley, M.O., Erickson, L.C. and Kohn, K.W. (1976) Murat. Res. 3 7 , 2 7 9 - - 2 9 2 Day, R.S. and Muel, B. (1974) P h o t o c h e m . Photoblol. 20, 95--102 Fornaee, A,J., K o h n , K.W. and Kann, H.E. (1976) Proc. Natl. Acad. Sci. U.S. 73, 39---43 K o h n , K.W., Erickson, L.C., Ewig, R.A.G. and F r i e d m a n , C.A. (1976) Biochemistry 15, 4629---4637 Fornace, A J . and Kohn, K.W. (1976) Biochim. Biophys. Acta 435, 95--103 Ewig, R.A.G. and Kohn, K.W. (1977) Cancer Res. 37, 2 1 1 4 - - 2 1 2 2 Kissane, J.M. and Robins, E. (1958) J. Biol. Chem. 2 3 3 , 1 8 4 - - 1 8 8 Erickson, L.C., Bradley, M.O. and Kohn, K.W. (1977) Cancer Res. 37, 3 7 4 4 - - 3 7 5 0 Beukers, R. and Berends, W. (1966) Biochim. Biophys. Acta 41,550---551 Setlow, R.B. and Carrier, W.L. (1964) Proc. Natl. Acad. Sci. U.S, 5 1 , 2 2 6 - - 2 3 1 Elkind, M.M. and Ley, R.D. (1976) Biochem. Biophys. Res. C ommun. 68,691---698 Swenbcrg, J.A., Petzold, G.L. and Harbaeh, P.R. (1976) B l o c h e m . Biophys. Res, C o m m u n . 7 2 , 7 3 2 - 738 19 Gan tt, R., Ewlg, R.A.G., Sanford, K.K. and Jones, G.M. (1977) Cancer Res. Abstr. 18, 204 20 Marmur, J., Anderson, W.F., Matthews, L., Berns, K., Gajewska, E., Lane, O. and Dory, P. (1961) J. CeU. Comp. Physiol. 58, Suppl. 1, 33--55 21 Moroson, H. and Alexander, P, (1961) Radiat. Res. 14, 29--49