181
Mutation Research, 235 (1990) 181-186 DNA Repair
Elsevier MUTDNA 06379
Wavelength dependence for UV-induced pyrimidine dimer formation in DNA of human peripheral blood lymphocytes S t e v e n E. F r e e m a n
a n d S h a r o n L. R y a n
Lovelace Medical Foundation, 2425 Ridgecrest Drive, S.E., Albuquerque, N M 87108 (U.S.A.)
(Received 20 July 1989) (Revision received 15 November 1989) (Accepted 15 November 1989)
Keywords: Pyrimidinedimer formation in DNA; Wavelength dependence; Lymphocytes,peripheral blood, human; Action spectrum,
UV induction Summary An action spectrum for the induction of pyrimidine dimers in h u m a n peripheral lymphocytes was determined between 254 and 405 nm. The presence of pyrimidine dimers was determined as UV-induced lesions that were sensitive to the dimer-specific endonuclease from M i c r o c o c c u s luteus in conjunction with agarose gel electrophoresis. The rate of induction of pyrimidine dimers was maximal at 254 nm. These values can be compared with action spectra for UV-induced in vitro responses of lymphocytes.
Ultraviolet radiation (UV) from sunlight (290400 nm) can induce erythema, melanization and skin cancer in man. Recent increases in the incidence of sunlight induced skin cancer and the prospect that humans will be exposed to increased amounts of UV because of possible depletion of ozone in the stratosphere make it important that we understand fully the interactions of UV with biological molecules and the results impacted on cellular functions.
This research was supported by an intramural grant from the Lovelace Medical Foundation. The authors wish to acknowledge the helpful suggestions and the critical reading of this manuscript by Dr. R.D. key. Correspondence: Dr. Steven E. Freeman, Lovelace Medical Foundation, 2425 Ridgecrest Drive, S.E., Albuquerque, NM 87108, Phone (505) 262-7155, Fax (505) 262-7616 (U.S.A.).
The exposure of human peripheral blood lymphocytes to U V in vitro reduces viability (Morison et al., 1979), causes the loss of the ability to stimulate allogeneic lymphocytes in the mixed lymphocyte culture reaction (Lindahl-Kiessling and Safwenberg, 1971) and reduces the ability to proliferate in response to the mitogen phytohemagglutinin (Morison et al., 1980). In addition UV exposure induces unscheduled D N A synthesis in lymphocytes which is thought to represent repair of D N A damage. Although the study of the effects of U V exposure on lymphocyte function has yielded valuable information, the specific UV photoproducts that lead to these effects have not been determined. Determination of the UV-action spectrum for a particular photoproduct may help identify the chromophore involved in a light induced biologic event. One of the most widely characterized of the UV-induced D N A photoproducts is the cyclobu-
0921-8777/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
182 tyl pyrimidine dimer. The action spectrum for pyrimidine dimer induction has been determined for Escherichia coli (Peak et al., 1984), Chinese Hamster V79 cells in culture (Rothman and Setlow, 1979) and skin of hairless mice in situ (Ley et al., 1983). The action spectrum for UV-induced pyrimidine dimers in human skin in situ for wavelengths in the UVB (280-320 nm) and UVA (320-400 nm) regions has also been determined (Freeman et al., 1989). Pyrimidine dimers can be detected by use of a dimer-specific endonuclease from M. luteus (Carrier and Setlow, 1970) that makes a single-strand break adjacent to each dimer (Ahmed and Setlow, 1977; Setlow et al., 1975). An alkaline agarose gel (McDonell et al., 1977) method can be used to measure the number of resulting single-strand breaks in non-radioactive DNA (Sutherland and Shih, 1983; Freeman et al., 1986b). The alkaline agarose gel method has been developed to determine dimer formation in human skin (Sutherland et al., 1980; Gange et al., 1985; Freeman et al., 1986a). We have used the agarose gel method to determine the action spectrum for UV-induced pyrimidine dimer formation in human peripheral blood lymphocytes exposed in vitro to UV of wavelengths between 254 and 405 nm. Materials and methods
Lymphocyte isolation. Blood samples (20-40 ml) were collected from healthy consenting volunteers into heparinized tubes. The lymphocytes were separated from polymorphonuclear leukocytes and erythrocytes by layering 10 ml of whole blood onto Histopaque gradients (Sigma Chemical Co.) and centrifuging at 450 g for 30 min at room temperature. Lymphocytes and other mononuclear cells were aspirated from the gradient-plasma interfaces and washed twice in phosphate-buffered saline (PBS) (0.15 M NaC1, 0.01 M sodium phosphate buffer, pH 7.3). The final cell pellet was resuspended in PBS at a concentration of 1.0 x 106/ml and the optical density of the cells determined at the desired wavelengths. Irradiation conditions. The cells were then exposed while being stirred in a quartz cuvette (1-cm light-path length) to various doses of UV from
either a low pressure mercury germicidal lamp (254 nm), a 2500 W Xe arc lamp (280, 290, 300, 310 or 320 nm) or a 2500-W Hanovia high-pressure Hg lamp (334, 365 or 405 nm). The Xe and Hg lamps were used in conjunction with a quarter-meter, diffraction-grating monochromator (Kratos Analytical Instruments, Inc.). The UV exciting the monochromator was filtered with Schott glass filters (WG280, 1 mm; WG295, 2 mm; WG305, 3 mm; WG320, 2 mm; WG335, 2 mm; GG375, 3 mm; and GG400, 3 mm, respectively) to reduce any scattered light of shorter wavelengths. Since the germicidal lamp emits principally 254-nm radiation, no filter was necessary. Dose rates were determined with an Optronic Model 742 spectroradiometer (Optronics Laboratories, Inc., Orlando, FL). The fluence rates were 0.43, 1, 5, 6, 20, 38, 260, 300 and 240 W / m 2, respectively. The final UV exposure was corrected by the method of Morowitz (1950) using the optical density of the cell suspension. DNA preparation and agarose gel electrophoresis. After irradiation the cells in 1-ml aliquots were collected by centrifugation and resuspended in 0.5 ml of a Tris buffer (0.05 M Tris pH 8.0, 0.01 M EDTA, 0.04 M NaC1) and lysed by the addition of 25/~1 of a 10% (w/v) solution of sodium dodecyl sulfate. The D N A was then isolated as described by Freeman and Ryan (1988) and the concentrations determined with a spectrofluorometer (Hoefer Scientific Instruments, Sacramento, CA). Aliquots of D N A (40-60 ng) were treated with enough pyrimidine dimer-specific M. luteus UVendonuclease determined to give stoichiometric cleavage at dimer sites (Carrier and Setlow, 1970). The DNA was then denatured with an alkaline buffer (0.13% ( w / v ) bromocresol green, 25% (v/v) glycerol, 0.5 N N a O H ) and added to wells of a 0.4% alkaline agarose gel in 2 mM EDTA, 30 mM N a O H as electrophoresis solution. The gel containing the treated DNA, along with DNA controls not treated with UV-endonuclease and DNA molecular length standards (bacteriophage DNA from T4, T7 and X D N A digested with the restriction enzyme HindIII) were subjected to static field electrophoresis in a BioRad Minigel (BioRad,
183
Richmond, CA) apparatus for 2 h at 40 V. The gel was neutralized (0.1 M Tris, pH 8.0, 30 rain), stained with ethidium bromide (1 /~g/ml) in distilled water, destained in distilled water for 1 h and photographed using Polaroid Type 55 posit i v e / negative film. The D N A distribution along each lane was determined by scanning the photographic negative with a Hoefer GS 300 Scanning Densitometer (Hoefer Scientific Instruments). The voltage output of the densitometer is digitized with an 8-bit analog-to-digital ( A / D ) converter and stored using the GS-350 Data System (Hoefer Scientific Instruments) on disc using an IBM X T / A T . The digitized gels scans were then transferred to a Micro Vax II (Digital Equipment Corporation) microcomputer and the number average molecular lengths of the UV endonuclease-treated and untreated samples were determined from analysis of the DNA profiles (Freeman et al., 1986b; Freeman and Thompson, 1988; Freeman and Ryan, 1988). The frequency of UV-endonuclease sensitive sites per 103 bases ( D i m e r s / l O 3 bases) was determined using the equation: Gel scans.
Dimers ~ - ~ bases
1
1
Z n ( + endo)
Zn( - endo)
(1)
where Ln(+endo) is the number average molecular length of the DNA sample treated with UV endonuclease, and Ln(_endo) is the number average molecular length of the corresponding untreated sample. Results D o s e - r e s p o n s e curves. Determination of an action spectrum for pyrimidine dimer formation in human lymphocytes irradiated in vitro requires measurement of dose-response curves for damage production in lymphocytes from different volunteers using a wide wavelength range. We determined the extent of DNA damage induced in lymphocytes exposed to UV radiation using wavelengths selected with a monochr0mator or from a germicidal lamp. D N A was extracted from the lymphocytes and treated with the dimer-specific
36
I
~2.7!
2 . 4 0 x 10 -21
< iS-
._.._°---'-'-'""
Ill D
0.9- .......... .-~....... ........ ~.......... ;, ...... .22 0.08
9
12
UV-EXPOSURE(photons/era 2 xlffl~ Fig. 1. Production of U V endonuclease sensitive sites (pyrimidine dimers) in DNA of h u m a n lymphocytes to UV from a monochromator. D N A was extracted from irradiated lymphocytes, treated with U V endonuclease, and subjected to alkaline agarose gel electrophoresis. The gels were photographed and the negatives scanned. The n u m b e r of UV endonuclease sensitive sites per 1000 bases [(Dirners/lO 3) bases] determined for each dose at 320 n m (O) and 334 n m (11). The lines and the slopes of the lines were determined by the method of least squares. The error estimates are standard errors of the mean.
enzyme M. luteus UV-endonuclease which makes a single-strand nick adjacent to prymidine dimers. The D N A was then subjected to alkaline agarose gel electrophoresis and the number of nicks were then determined by computer assisted analysis of photographs of the agarose gels (Sutherland et al., 1984; Freeman et al., 1986b; Freeman and Thompson, 1988). Dose-response curves at 320 nm and 334 nm for lymphocytes from one of the volunteers are shown in Fig. 1. Each point represents the averaged number of pyrimidine dimers expressed as [ D i m e r s / l O 3] bases as determined from analysis of at least 3 independent gels with D N A obtained from the lymphocytes. The error bars represent standard error of the mean. We estimate the number of [ D i m e r s / l O 3] bases produced per UV exposure ( [ D i m e r s / l O 3] bases per Quantum) by using the method of least squares to generate lines for the dose-response. The slopes of the lines yield those estimates (Fig. 1). Determination of pyrimidine dimer action spectrum. To determine which wavelengths were most
184
effective in producing pyrimidine dimers we generated an action spectrum by determining the slopes of the least squares generated lines as a function of wavelength for each subject and averaging the slopes at each wavelength (Fig. 2). The effectiveness of light in inducing dimers in lymphocytes varies over 7 orders of magnitude and the maximum of this action spectrum is 254
Iz
< -1(5 ~.-2 ~
uJ
>.-4I.,
Mutation Research, 235 (1990) 181-186 DNA Repair
Elsevier MUTDNA 06379
Wavelength dependence for UV-induced pyrimidine dimer formation in DNA of human peripheral blood lymphocytes S t e v e n E. F r e e m a n
a n d S h a r o n L. R y a n
Lovelace Medical Foundation, 2425 Ridgecrest Drive, S.E., Albuquerque, N M 87108 (U.S.A.)
(Received 20 July 1989) (Revision received 15 November 1989) (Accepted 15 November 1989)
Keywords: Pyrimidinedimer formation in DNA; Wavelength dependence; Lymphocytes,peripheral blood, human; Action spectrum,
UV induction Summary An action spectrum for the induction of pyrimidine dimers in h u m a n peripheral lymphocytes was determined between 254 and 405 nm. The presence of pyrimidine dimers was determined as UV-induced lesions that were sensitive to the dimer-specific endonuclease from M i c r o c o c c u s luteus in conjunction with agarose gel electrophoresis. The rate of induction of pyrimidine dimers was maximal at 254 nm. These values can be compared with action spectra for UV-induced in vitro responses of lymphocytes.
Ultraviolet radiation (UV) from sunlight (290400 nm) can induce erythema, melanization and skin cancer in man. Recent increases in the incidence of sunlight induced skin cancer and the prospect that humans will be exposed to increased amounts of UV because of possible depletion of ozone in the stratosphere make it important that we understand fully the interactions of UV with biological molecules and the results impacted on cellular functions.
This research was supported by an intramural grant from the Lovelace Medical Foundation. The authors wish to acknowledge the helpful suggestions and the critical reading of this manuscript by Dr. R.D. key. Correspondence: Dr. Steven E. Freeman, Lovelace Medical Foundation, 2425 Ridgecrest Drive, S.E., Albuquerque, NM 87108, Phone (505) 262-7155, Fax (505) 262-7616 (U.S.A.).
The exposure of human peripheral blood lymphocytes to U V in vitro reduces viability (Morison et al., 1979), causes the loss of the ability to stimulate allogeneic lymphocytes in the mixed lymphocyte culture reaction (Lindahl-Kiessling and Safwenberg, 1971) and reduces the ability to proliferate in response to the mitogen phytohemagglutinin (Morison et al., 1980). In addition UV exposure induces unscheduled D N A synthesis in lymphocytes which is thought to represent repair of D N A damage. Although the study of the effects of U V exposure on lymphocyte function has yielded valuable information, the specific UV photoproducts that lead to these effects have not been determined. Determination of the UV-action spectrum for a particular photoproduct may help identify the chromophore involved in a light induced biologic event. One of the most widely characterized of the UV-induced D N A photoproducts is the cyclobu-
0921-8777/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
182 tyl pyrimidine dimer. The action spectrum for pyrimidine dimer induction has been determined for Escherichia coli (Peak et al., 1984), Chinese Hamster V79 cells in culture (Rothman and Setlow, 1979) and skin of hairless mice in situ (Ley et al., 1983). The action spectrum for UV-induced pyrimidine dimers in human skin in situ for wavelengths in the UVB (280-320 nm) and UVA (320-400 nm) regions has also been determined (Freeman et al., 1989). Pyrimidine dimers can be detected by use of a dimer-specific endonuclease from M. luteus (Carrier and Setlow, 1970) that makes a single-strand break adjacent to each dimer (Ahmed and Setlow, 1977; Setlow et al., 1975). An alkaline agarose gel (McDonell et al., 1977) method can be used to measure the number of resulting single-strand breaks in non-radioactive DNA (Sutherland and Shih, 1983; Freeman et al., 1986b). The alkaline agarose gel method has been developed to determine dimer formation in human skin (Sutherland et al., 1980; Gange et al., 1985; Freeman et al., 1986a). We have used the agarose gel method to determine the action spectrum for UV-induced pyrimidine dimer formation in human peripheral blood lymphocytes exposed in vitro to UV of wavelengths between 254 and 405 nm. Materials and methods
Lymphocyte isolation. Blood samples (20-40 ml) were collected from healthy consenting volunteers into heparinized tubes. The lymphocytes were separated from polymorphonuclear leukocytes and erythrocytes by layering 10 ml of whole blood onto Histopaque gradients (Sigma Chemical Co.) and centrifuging at 450 g for 30 min at room temperature. Lymphocytes and other mononuclear cells were aspirated from the gradient-plasma interfaces and washed twice in phosphate-buffered saline (PBS) (0.15 M NaC1, 0.01 M sodium phosphate buffer, pH 7.3). The final cell pellet was resuspended in PBS at a concentration of 1.0 x 106/ml and the optical density of the cells determined at the desired wavelengths. Irradiation conditions. The cells were then exposed while being stirred in a quartz cuvette (1-cm light-path length) to various doses of UV from
either a low pressure mercury germicidal lamp (254 nm), a 2500 W Xe arc lamp (280, 290, 300, 310 or 320 nm) or a 2500-W Hanovia high-pressure Hg lamp (334, 365 or 405 nm). The Xe and Hg lamps were used in conjunction with a quarter-meter, diffraction-grating monochromator (Kratos Analytical Instruments, Inc.). The UV exciting the monochromator was filtered with Schott glass filters (WG280, 1 mm; WG295, 2 mm; WG305, 3 mm; WG320, 2 mm; WG335, 2 mm; GG375, 3 mm; and GG400, 3 mm, respectively) to reduce any scattered light of shorter wavelengths. Since the germicidal lamp emits principally 254-nm radiation, no filter was necessary. Dose rates were determined with an Optronic Model 742 spectroradiometer (Optronics Laboratories, Inc., Orlando, FL). The fluence rates were 0.43, 1, 5, 6, 20, 38, 260, 300 and 240 W / m 2, respectively. The final UV exposure was corrected by the method of Morowitz (1950) using the optical density of the cell suspension. DNA preparation and agarose gel electrophoresis. After irradiation the cells in 1-ml aliquots were collected by centrifugation and resuspended in 0.5 ml of a Tris buffer (0.05 M Tris pH 8.0, 0.01 M EDTA, 0.04 M NaC1) and lysed by the addition of 25/~1 of a 10% (w/v) solution of sodium dodecyl sulfate. The D N A was then isolated as described by Freeman and Ryan (1988) and the concentrations determined with a spectrofluorometer (Hoefer Scientific Instruments, Sacramento, CA). Aliquots of D N A (40-60 ng) were treated with enough pyrimidine dimer-specific M. luteus UVendonuclease determined to give stoichiometric cleavage at dimer sites (Carrier and Setlow, 1970). The DNA was then denatured with an alkaline buffer (0.13% ( w / v ) bromocresol green, 25% (v/v) glycerol, 0.5 N N a O H ) and added to wells of a 0.4% alkaline agarose gel in 2 mM EDTA, 30 mM N a O H as electrophoresis solution. The gel containing the treated DNA, along with DNA controls not treated with UV-endonuclease and DNA molecular length standards (bacteriophage DNA from T4, T7 and X D N A digested with the restriction enzyme HindIII) were subjected to static field electrophoresis in a BioRad Minigel (BioRad,
183
Richmond, CA) apparatus for 2 h at 40 V. The gel was neutralized (0.1 M Tris, pH 8.0, 30 rain), stained with ethidium bromide (1 /~g/ml) in distilled water, destained in distilled water for 1 h and photographed using Polaroid Type 55 posit i v e / negative film. The D N A distribution along each lane was determined by scanning the photographic negative with a Hoefer GS 300 Scanning Densitometer (Hoefer Scientific Instruments). The voltage output of the densitometer is digitized with an 8-bit analog-to-digital ( A / D ) converter and stored using the GS-350 Data System (Hoefer Scientific Instruments) on disc using an IBM X T / A T . The digitized gels scans were then transferred to a Micro Vax II (Digital Equipment Corporation) microcomputer and the number average molecular lengths of the UV endonuclease-treated and untreated samples were determined from analysis of the DNA profiles (Freeman et al., 1986b; Freeman and Thompson, 1988; Freeman and Ryan, 1988). The frequency of UV-endonuclease sensitive sites per 103 bases ( D i m e r s / l O 3 bases) was determined using the equation: Gel scans.
Dimers ~ - ~ bases
1
1
Z n ( + endo)
Zn( - endo)
(1)
where Ln(+endo) is the number average molecular length of the DNA sample treated with UV endonuclease, and Ln(_endo) is the number average molecular length of the corresponding untreated sample. Results D o s e - r e s p o n s e curves. Determination of an action spectrum for pyrimidine dimer formation in human lymphocytes irradiated in vitro requires measurement of dose-response curves for damage production in lymphocytes from different volunteers using a wide wavelength range. We determined the extent of DNA damage induced in lymphocytes exposed to UV radiation using wavelengths selected with a monochr0mator or from a germicidal lamp. D N A was extracted from the lymphocytes and treated with the dimer-specific
36
I
~2.7!
2 . 4 0 x 10 -21
< iS-
._.._°---'-'-'""
Ill D
0.9- .......... .-~....... ........ ~.......... ;, ...... .22 0.08
9
12
UV-EXPOSURE(photons/era 2 xlffl~ Fig. 1. Production of U V endonuclease sensitive sites (pyrimidine dimers) in DNA of h u m a n lymphocytes to UV from a monochromator. D N A was extracted from irradiated lymphocytes, treated with U V endonuclease, and subjected to alkaline agarose gel electrophoresis. The gels were photographed and the negatives scanned. The n u m b e r of UV endonuclease sensitive sites per 1000 bases [(Dirners/lO 3) bases] determined for each dose at 320 n m (O) and 334 n m (11). The lines and the slopes of the lines were determined by the method of least squares. The error estimates are standard errors of the mean.
enzyme M. luteus UV-endonuclease which makes a single-strand nick adjacent to prymidine dimers. The D N A was then subjected to alkaline agarose gel electrophoresis and the number of nicks were then determined by computer assisted analysis of photographs of the agarose gels (Sutherland et al., 1984; Freeman et al., 1986b; Freeman and Thompson, 1988). Dose-response curves at 320 nm and 334 nm for lymphocytes from one of the volunteers are shown in Fig. 1. Each point represents the averaged number of pyrimidine dimers expressed as [ D i m e r s / l O 3] bases as determined from analysis of at least 3 independent gels with D N A obtained from the lymphocytes. The error bars represent standard error of the mean. We estimate the number of [ D i m e r s / l O 3] bases produced per UV exposure ( [ D i m e r s / l O 3] bases per Quantum) by using the method of least squares to generate lines for the dose-response. The slopes of the lines yield those estimates (Fig. 1). Determination of pyrimidine dimer action spectrum. To determine which wavelengths were most
184
effective in producing pyrimidine dimers we generated an action spectrum by determining the slopes of the least squares generated lines as a function of wavelength for each subject and averaging the slopes at each wavelength (Fig. 2). The effectiveness of light in inducing dimers in lymphocytes varies over 7 orders of magnitude and the maximum of this action spectrum is 254
Iz
< -1(5 ~.-2 ~
uJ
>.-4I.,
