ABSTRACT
We here examined whether exposure of mice to UV-B affected their susceptibility to the murine malaria parasite Plasmodium chabaudi. When BALB/c mice with depilated skin were irradiated with UV-B and subsequently infected with the parasite, 80 to 100% of the UV-B-irradiated mice died within 12 days of infection with a sublethal dose. In addition, UV-B irradiation of C57BL/10 (B-10) mice, which are otherwise naturally resistant to the parasites, rendered them susceptible, and 100% of irradiated B-10 mice died within 11 days postinfection. The level of plasma gamma interferon (IFN-γ) in unirradiated B-10 mice at 5 days after infection increased to 566 pg/ml, whereas the UV-B exposure of mice impaired the production of IFN-γ, which showed a maximum level of 65 pg/ml in response to the parasite infection. The maximum level of plasma interleukin-10 in UV-B-irradiated mice in response to the parasite infection was ∼1,100 pg/ml, which was approximately fourfold higher than the maximum level in unirradiated control mice. When UV-B-irradiated B-10 mice were administered murine recombinant IFN-γ after infection, the mice regained parasite resistance. These results demonstrated that the UV-B exposure of mice enhances the susceptibility to the malaria parasites and suggested that the enhanced susceptibility following UV-B exposure was mediated by impairment of IFN-γ production in response to the parasite infection.
Environmental concerns regarding the global depletion of stratospheric ozone and the seasonal hole in the ozone layer over the Antarctic have been growing. A significant (4 to 7%) increase in surface erythemal UV radiation has occurred during the latest 2 decades in the mid-latitude areas of both hemispheres of the Earth (6), and there is apprehension that this increase, especially the increase in UV-B rays (290- to 320-nm wavelength) as opposed to UV-A rays (320- to 400-nm wavelength), will harmfully influence human health. A body of evidence that has been accumulating over the past 20 years indicates that the exposure of humans and experimental animals to UV-B irradiation can modify various immune responses, raising the possibility that UV-B irradiation affects susceptibility to infectious diseases. In fact, it has been well recognized that latent herpes simplex virus is reactivated by natural exposure to sunlight (15). Moreover, in many infections, UV-B irradiation appears to affect the clinical progression of the disease, with a decreased clearance of microorganisms, a decreased survival time, and a recrudescence of lesions from a latent state (3), although little is known about the mechanisms of these unfavorable effects on infectious diseases.
The United Nations Environmental Programme has cautioned that malaria is a disease for which there is an increasing risk of infection due to ozone depletion and the resulting UV-B increase on Earth (14). However, the reference source was a personal communication, and no epidemiological or experimental reports demonstrating that UV-B irradiation enhances malarial infection have so far been published. In this study, therefore, we determined whether UV-B irradiation of animals enhances susceptibility to malaria parasites in mouse model systems.
Female BALB/c and C57BL/10 (B-10) mice (SLC Co., Hamamatsu, Japan), 7 to 10 weeks old, were used. In all experiments, the animals were handled as specified in the institute's Guidelines for Animal Experiments. UV-B irradiation was performed as described in our previous report (16). Briefly, the dorsal hair was clipped, and the remaining hair was removed with a depilatory cream. The next day, the exposed skin was irradiated with a UV-B lamp equipped with two bulbs (FL20SE, Toshiba Leitec, Tokyo, Japan). The bulbs have a broad emission spectrum (285 to 350 nm, with peak emission at 314 nm). The highest output is primarily in the UV-B range (290 to 320 nm), corresponding to 64% of the total emission, while the remaining 36% is in the UV-A zone. UV-B was administered in a single dose of 200 mJ/cm2 (40-cm distance for 2 h) with the aid of a UV radiometer (UV 103; Macum Photometrics Ltd., Livingston, Scotland). This level of irradiation induced apparent erythema with edema in both strains of mice on the day following irradiation, and the inflammation became encrusted within a week.
Mouse malaria parasite Plasmodium chabaudi subsp.chabaudi AS (P. chabaudi) was maintained by syringe passage every week in B-10 mice (5). Parasitized red blood cells (PRBC) were collected by heart puncture, washed, and suspended in Dulbecco's phosphate-buffered saline. Mice were infected intraperitoneally (i.p.) with 105 PRBC/ml. The same infection procedure was performed in each experiment, unless otherwise specified. For determination of parasitemia of mice, the PRBC among more than 1,000 red blood cells were counted after Diff-Quik (International Reagents Co., Kobe, Japan) staining of tail blood smears.
Serum samples from each group of mice were collected by heart puncture. Gamma interferon (IFN-γ) was assayed with a commercially available enzyme-linked immunosorbent assay (ELISA) kit for murine IFN-γ (R & D Systems Inc., Minneapolis, Minn.). The IFN-γ titer was calculated from the ratio of the serum sample to the reference value of the standard sample in the kit. A similar experiment was also performed for interleukin-10 (IL-10), and the levels of IL-10 were assayed using a murine IL-10 ELISA kit (R & D systems Inc.). For the IFN-γ tests, a 400-ng dose of murine recombinant IFN-γ (Pepro Tech EC Ltd., London, England) was administered i.p. every day from days 3 to 7 postinfection. Control mice were injected with the same amount of RPMI 1640 medium including 5% normal mouse serum, which was used in the IFN-γ preparation.
Results are presented as means ± standard errors of the means. The statistical significance of the serum IFN-γ levels was calculated using Student's t test. P < 0.05 was considered significant.
First, to explore whether UV-B irradiation of experimental mice affected malarial infection, we compared postinfection survival curves for UV-B-irradiated and unirradiated control mice using the BALB/c strain, which is susceptible to the murine malaria parasite P. chabaudi (11). When unirradiated mice were injected i.p. with 104 PRBC, a sublethal dose, four of five infected mice survived for at least 12 days after infection. In contrast, when BALB/c mice were preexposed to UV-B at 200 mJ/cm2 and then infected with 104 PRBC, all five mice died within 10 days of infection (Fig. 1a). The exposure of mice to UV-B without parasite infection caused no lethal effects (data not shown). When mice were infected with a higher parasite dose (105 PRBC), all irradiated and unirradiated mice died within 10 days postinfection but the irradiated mice appeared to die earlier than the unirradiated mice (Fig. 1b). These observations showed that the preexposure of BALB/c mice to UV-B rendered the mice more susceptible to the malaria parasite.
Survival profiles of P. chabaudi-infected BALB/c and B-10 mice. Groups of five BALB/c (a and b) and B-10 (c and d) mice were exposed to UV-B (200 mJ/cm2) and the next day were infected i.p. with 104 (a and c) or 105PRBC (b and d) Symbols: ○, unirradiated mice; ●, UV-B-irradiated mice.
To determine whether the enhancement of parasite susceptibility after UV-B exposure occurred in other mouse strains, we also tested the B-10 mouse strain, which is naturally resistant to the murine malaria (11). When unirradiated B-10 mice were i.p. injected with 104 or 105 PRBC, four of the mice infected with 104 PRBC and all mice infected with 105 PRBC survived (Fig. 1c and d). Preirradiation of B-10 mice with UV-B clearly enhanced the parasite susceptibility; all five mice infected with 105 PRBC died within 11 days postinfection (Fig. 1d), and four of the five mice infected with 104 PRBC also died (Fig. 1c).
To see the parasite growth in infected mice, parasitemia was assayed by staining tail blood smears. Until the middle stage (∼6 days postinfection), there was no difference in parasitemia between irradiated and unirradiated mice of both BALB/c and B-10 strains (Fig.2). Parasitemia reached the maximum level during the late stage (7 to 10 days), when the UV-B-irradiated mice of both strains showed slightly higher parasitemia than the unirradiated controls, but the differences were not statistically significant (Fig.2). However, all of the irradiated mice died 1 to 2 days after the peak of parasitemia, whereas all of the unirradiated mice survived, with a decrease in parasitemia to near the basal level 3 days after the peak (Fig. 1 and 2). These observations suggested that, in UV-B-irradiated mice, an immune mechanism to eliminate the parasite at the late infection stage was impaired.
Parasitemia in infected mice. BALB/c (a) and B-10 (b) mice (five of each) preirradiated by UV-B (200 mJ/cm2) (●) or unirradiated (○) were infected with sublethal doses of the parasite, 104 and 105 PRBC, respectively. After various periods, the parasitemia of each mouse was determined by staining tail blood smears. The mean parasitemia values from five mice are shown. Vertical lines indicate standard errors.
IFN-γ is suggested to be an important factor in determining the susceptibility of mice to murine malaria, because the average level of plasma IFN-γ in BALB/c mice (susceptible to P. chabaudi) is significantly lower than the level in the B-10 strain (naturally resistant to the parasites) (12, 13). We therefore examined whether the preexposure of mice to UV-B affected the plasma IFN-γ level in response to the parasite infection. For this, B-10 mice, either irradiated with UV-B or unirradiated, were infected with 105 PRBC and assayed for plasma IFN-γ levels between 3 and 6 days after infection. The plasma IFN-γ level in the unirradiated control mice began to increase on the 4th day postinfection, reached a maximum level (566 ± 59 pg/ml; n = 3) on the 5th day, and then decreased on the 6th day (Fig. 3). Interestingly, the plasma IFN-γ level in irradiated mice was lower than that in the unirradiated control between days 4 and 6 postinfection, and the maximum level (65 ± 26 pg/ml; n = 3) was only ∼12% of the maximum level of the unirradiated control. Statistical differences between irradiated and unirradiated mice on the 5th and 6th days postinfection were significant (P < 0.02 and P < 0.05, respectively). We did not examine the effects of UV-B exposure on IFN-γ production in BALB/c mice, since the plasma IFN-γ production in response to the parasite infection was too low to be accurately determined, even in unirradiated BALB/c mice (13).
Time courses of plasma IFN-γ and IL-10 production byP. chabaudi-infected B-10 mice with or without UV-B irradiation. For the IFN-γ assay, 12 B-10 mice preirradiated by UV-B (200 mJ/cm2) or unirradiated were infected with 105 PRBC. At each time point indicated, three mice from each group were bled, and the levels of IFN-γ in the sera were determined by ELISA. For the IL-10 assay, 33 B-10 mice, irradiated or unirradiated, were infected, bled, and assayed as described above. The mean values ± standard errors from one set of experiments are shown. Two additional sets of experiments were performed, and similar results were obtained. Symbols: ○, IFN-γ of unirradiated mice; ●, IFN-γ of UV-B-irradiated mice; □, IL-10 of unirradiated mice; ■, IL-10 of UV-B-irradiated mice.
To our knowledge, the present paper is the first to show that plasma IFN-γ levels decrease following UV irradiation, although a recent report by Reeve et al. suggested that IFN-γ production was enhanced by UV-A irradiation in mice (8). UV-B rays act as a stress and induce releases of various cytokines including IL-10 (1, 7, 9, 10), which is suggested to be involved in UV-B-induced immunosuppression through the dysfunction of T-helper lymphocytes or monocytes (2). We thus determined whether UV-B irradiation in our experimental system affected the level of circulating IL-10. For this, UV-B-preirradiated or unirradiated mice infected with 105 PRBC of P. chabaudi were bled and their plasma IL-10 levels were assayed. In the early stage of infection (days 1 to 4 postinfection), plasma IL-10 was undetectable. The level of plasma IL-10 increased during the middle (5th and 6th days postinfection) and late stages (after 7 days postinfection) (Fig. 3). The IL-10 level for UV-B-irradiated mice rose slower than that for unirradiated mice around the middle stage, and, at the late stage, the plasma IL-10 level for UV-irradiated mice increased rapidly, reaching an approximately fourfold-higher level than that for unirradiated mice on day 8 postinfection. Unfortunately, statistical significance could not be calculated because only one mouse in the irradiated group survived in this experiment. These observations indicate that UV-B irradiation also affected IL-10 production in response to parasite infection, although it is currently difficult to precisely determine whether the enhancement of the IL-10 level in infected mice by UV-B irradiation is prior to the suppression of IFN-γ production.
We next addressed the question of whether the impairment of IFN-γ production in UVB-irradiated B-10 mice was responsible for the acquisition by the mice of susceptibility to the parasite. For this purpose, mice preexposed to UV-B were infected with 105PRBC, the infected mice were i.p. administered murine recombinant IFN-γ (400 ng per mouse) or control vehicle (RPMI 1640 containing 5% normal mouse serum) every day for 3 to 7 days postinfection, and the survival ratios were determined. Four of five UV-B-irradiated mice administered IFN-γ survived, while four of five mice administered control vehicle died within 12 days of infection. The latter result was consistent with our findings shown in Fig. 1d. These results demonstrated that the parasite susceptibility of UV-B-irradiated B-10 mice is suppressed by postinfection administration of IFN-γ. IFN-γ activates macrophages (MØ), and the primed MØ produce nitric oxide (NO). NO radicals kill various infectious agents (4). A similar scenario might operate in our murine model system with P. chabaudi, although it remains unknown whether MØ activation to produce NO is suppressed by UV-B irradiation.
ACKNOWLEDGMENTS
We gratefully acknowledge Yukio Takizawa for his useful advice and valuable information for advancing this study. We are also grateful to Kentaro Hanada (Department of Biochemistry and Cell Biology of our institute) for his critical reading of this manuscript and helpful comments.
This study was supported by a grant from the Global Environment Program of the Environment Agency of Japan.
Notes
Editor: S. H. E. Kaufmann
FOOTNOTES
- Received 30 August 1999.
- Returned for modification 29 September 1999.
- Accepted 3 January 2000.
- Copyright © 2000 American Society for Microbiology
