Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Herpetic infection has been successfully treated with nucleoside analogs such as acyclovir (ACV) (Meyers et al., 1982; Dunkle et al., 1991; Whitley et al., 1991). However, resistant viruses appear in immunosuppressed patients, such as organ transplant recipients and patients with acquired immunodeficiency syndrome (Pass et al., 1979; Sibrack et al., 1982; Norris et al., 1988; Erlich et al., 1989; Oliver et al., 1989; Birch et al., 1990; Nugier et al., 1992; Reusser et al., 1996). Thus, the development of new anti-herpes simplex virus (HSV) agents with different modes of action is required.

We have been studying the antiviral activity of medicinal herb extracts for their possible use in the management of viral infection in humans. In our previous study, we selected 12 herbal extracts that show oral therapeutic anti-HSV type 1 (HSV-1) activity in a cutaneous infection model in mice (Simmons and Nash, 1984; Kumano et al., 1987; Kurokawa et al., 1993). Four of the 12 extracts augmented the oral therapeutic efficacy of ACV in mice (Kurokawa et al., 1995a). They showed potent antiviral activity against infection with ACV-phosphonoacetic acid (PAA)-resistant HSV-1 and wild-type HSV type 2 strains as well as wild-type HSV-1 in vitro and in vivo (Kurokawa et al., 1995a,b). These four herbal extracts also exhibited prophylactic efficacy against recurrent HSV-1 disease in mice (Kurokawa et al., 1997). We have recently purified and identified an anti-HSV compound, moronic acid, from one of the four herbal extracts that alleviated spontaneous and ultraviolet-induced recurrent HSV-2 genital disease in guinea pigs (Nakano et al., 1998; Kurokawa et al., 1999). This compound exhibited therapeutic anti-HSV-1 activity in a murine HSV infection model (Kurokawa et al., 1999). Furthermore, eugeniin was purified as an anti-HSV compound from the different species of plants, Geum japonicum Thunb. (G. japonicum) and Syzygium aromaticum (L.) MERR. et PERRY (S. aromaticum), of the four herbal extracts (Kurokawa et al., 1998). We showed that eugeniin inhibits HSV-1 DNA synthesis by interfering noncompetitively with HSV-1 DNA polymerase activity (Kurokawa et al., 1998). Thus, eugeniin may be a novel anti-HSV-1 compound showing an inhibitory action for viral DNA polymerase activity. However, its biological activity against HSV-1 infection has not been characterized in vivo.

In this study, we characterized the biological activity of eugeniin on a cutaneous HSV-1 infection model in mice and its precise interaction with HSV-1 DNA polymerase in vitro. In the murine model, we evaluated the bioavailability of eugeniin by comparing its therapeutic efficacy between oral and intraperitoneal administrations. Because eugeniin has been shown to inhibit HSV-1 DNA polymerase activity noncompetitively (Kurokawa et al., 1998), the inhibitory activity of eugeniin was further characterized by comparison with those of ACV and PAA.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Virus and Cells. Wild-type 7401H HSV-1 strain (Kurokawa et al., 1993) was used. The virus stock was prepared from infected Vero cells as reported previously (Kurokawa et al., 1993, 1998). Vero cells were grown and maintained in Eagle's minimum essential medium supplemented with 5 and 2% heat-inactivated calf serum, respectively. For the preparation of HSV-1 DNA polymerase, human embryonic lung (HEL) cells were grown and maintained in Eagle's minimum essential medium supplemented with 10 and 2% heat-inactivated fetal bovine serum, respectively.

Compounds. Eugeniin was purified from the herbal extract of G. japonicum or S. aromaticum using different chromatographic fractionations guided by anti-HSV activity, as described previously (Kurokawa et al., 1998). The purified eugeniin was dissolved in dimethyl sulfoxide (DMSO) at 20 mg/ml and used for the plaque reduction assay as described below. For administration to mice, eugeniin was dissolved in 0.4 ml of DMSO and then pyrogen-free distilled water or phosphate-buffered saline (PBS) was added to make a final volume of 40 ml for oral or intraperitoneal administration, respectively. DMSO solution (1%) was used as a control for the animal experiments.

Animals. Female BALB/c mice (6 week old, 17-19 g) were purchased from Sankyo Labo Service Co., Ltd., Tokyo, Japan. The mice were housed five per cage in a temperature-controlled room, with food and pyrogen-free water ad libitum, under a 12-h light/dark diurnal cycle (light at 7:00 AM). The temperature in the room was kept at 23 ± 2°C. The mice were acclimated for at least 3 to 4 days before starting an experimental procedure. The animal experimentation guidelines of Toyama Medical and Pharmaceutical University were followed in the animal studies.

Mouse HSV-1 Infection. BALB/c mice were cutaneously infected with wild type HSV-1 [1 × 10⁶ or 10⁴ plaque-forming units (PFU)/mouse] after scarification of the shaved right midflank with 27-gauge needles as described previously (Kurokawa et al., 1993, 1995a). The development of skin lesions and death were observed three times daily with the severity of the lesions being scored as described previously (Kurokawa et al., 1993, 1995a): 0, no lesion; 2, vesicles in local region; 4, erosion and/or ulceration in local region; 6, mild zosteriform lesion; 8, moderate zosteriform lesion; 10, severe zosteriform lesion; 12, death. The infected mice were fed and observed for at least 20 days to determine their mortality.

Determination of Virus Yields in Skin and Brain. Virus yields in the skin and brain were determined in infected mice. Mice were cutaneously infected with wild-type HSV-1 and eugeniin was orally administered at doses of 0.3 to 50 mg/kg following the same schedule as described above. The brain and skin [whole lesions that include the area (2 × 2 cm) encompassing the inoculation site of infected mice] were removed under ether anesthesia on days 3 and 6 after infection and homogenized in 2 ml of PBS as described previously (Kurokawa et al., 1995a). The homogenate was centrifuged at 3000 rpm for 15 min and the virus yield in the supernatant was determined by plaque assay on Vero cells (Kurokawa et al., 1993).

Toxicity Assay in Mice. Eugeniin was examined for its toxicity in uninfected mice. Five mice in each group were orally administered with eugeniin (50 mg/kg) for 7 days following the same schedule used in infected mice. Similarly, eugeniin (3, 10, and 20 mg/kg) was intraperitoneally administered. In either case, the uninfected mice were weighed on days 7, 14, and 20 after initial administration on day 0 and the mortality of mice was calculated on day 20.

Plaque Reduction Assay for Combination. The combined effects of eugeniin with ACV or PAA were examined for anti-HSV-1 activity in the plaque reduction assay. Duplicate cultures of Vero cells in 60-mm plastic dishes were infected with 100 PFU/0.2 ml of wild-type HSV-1 for 1 h. The cells were overlaid with 5 ml of nutrient methylcellulose (0.8%) medium containing various concentrations of eugeniin and/or ACV or PAA, and then cultured at 37°C for 2 to 3 days. The infected cells were fixed with 5% formalin solution and stained with 0.03% methylene blue solution; the number of plaques was counted under a dissecting microscope (Kurokawa et al., 1993). The effective concentrations for 50% plaque reduction (EC₅₀) were determined from a curve relating the plaque number to the concentration of drugs.

Cytotoxicity Assay of Drugs in Cell Culture. Cytotoxicity was examined by the growth inhibition of Vero cells as descried previously (Kurokawa et al., 1993). Briefly, Vero cells were seeded at a concentration of 2.5 × 10⁴ cells/well in 24-well plates and grown at 37°C for 2 days. The culture medium was replaced with fresh medium containing eugeniin and/or ACV or PAA at various concentrations. Cells were further grown for 2 days. The cells were treated with trypsin and the number of viable cells was determined by the trypan blue exclusion test. Cytotoxic concentrations for 50% growth reduction (CC₅₀) were determined from a curve relating percentage of cell viability to the concentration of drugs.

Preparation of HSV DNA Polymerase and Assay of Its Activity. HSV-1 DNA polymerase was prepared to examine the enzymological inhibition by eugeniin and PAA. This polymerase was partially purified from wild HSV-1-infected HEL cells by high-salt extraction, phosphocellulose and DNA cellulose (Pharmacia Biotech) column chromatography as described previously (Kurokawa et al., 1998). The specific activity of HSV-1 DNA polymerase was 2360 units/mg of protein. One unit of activity was defined as an amount that catalyzes the incorporation of 1 nmol of [³H]dTMP into an acid-insoluble fraction in an hour at 37°C under the conditions specific for enzyme with 6.84 µM [³H]dTTP (2.99 TBq/mmol; Moravek Biochemicals, Inc., Brea, CA) and 50 µM dATP, dCTP, and dGTP. Reaction mixture (25 µl) for an assay of partially purified HSV-1 DNA polymerase activity contained 50 mM Tris-HCl (pH 8.0); 8 mM MgCl₂; 0.5 mM dithiothreitol; 100 mM ammonium sulfate; 50 µM each of dATP, dGTP, and dCTP; 0.1 to 3.0 µM [³H]dTTP; 25 µg/ml activated calf thymus DNA (Sigma); various concentrations of eugeniin or PAA; and enzyme. Reaction velocities were measured at various concentrations of [³H]dTTP and an apparent K_m value was graphically evaluated from Lineweaver-Burk plots. Apparent K_i values were graphically determined by measuring reaction velocities in the presence and absence of fixed concentrations of eugeniin or PAA (Dixon and Webb, 1979). All linear lines in each graph were drawn by using computer software (CA-Cricket Graph III 1.5.2.) and each line was statistically evaluated by the determination of correlation coefficient using a computer software (StatView 4.5). Finally, K_i and K_m values were calculated from the equation obtained by the computer analyses. For the combination assay of eugeniin and PAA against HSV-DNA polymerase activity, the reaction mixtures containing both eugeniin and PAA at various concentrations were incubated at 37°C for 10 min and then the acid-insoluble radioactivity was measured as described previously (Kurokawa et al., 1998). The percentage of activities of HSV DNA polymerase for the various concentrations of PAA and eugeniin in the presence of the fixed concentrations of eugeniin and PAA, respectively, were calculated from the activity in their absence as 100%.

Statistical Analyses. Student's t test was used to evaluate the significance of the differences in mean survival times and mean times at which skin lesions were initially scored as 2 or 6 after infection. Significance of differences in virus yields in organs and the mean weights of the mice groups were also evaluated by the Student's t test. The repeated measures ANOVA with Dunn's procedure as a multiple comparison procedure was used to analyze the interaction between eugeniin and water in mean skin lesions for 3 to 14 days after infection. Statistical differences in the mortality were evaluated using Fisher's exact test. A p value of less than 0.05 was statistically defined as significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Therapeutic Efficacy of Eugeniin in Mice Infected with HSV-1. Therapeutic efficacy of eugeniin was examined through the two different routes of administration in a cutaneous HSV-1 infection model in mice and its oral bioavailability was evaluated with respect to the therapeutic efficacy. When mice were infected with wild-type HSV-1 at 10⁶ PFU/mouse in this model, all of the infected mice ultimately died. Therefore, we used this lethal infection to evaluate the distinct activity of eugeniin against HSV-1 infection. We previously showed that oral administration of the hot-water extract (250 mg/kg) of G. japonicum, which corresponded to the dose for human use in mice, exhibited significant therapeutic efficacy in this murine model (Kurokawa et al., 1995a,b). Because the eugeniin content of the hot-water extract was 0.04%, 0.1 mg of eugeniin would be contained in 250 mg of the hot-water extract. Thus, we used 0.3 mg/kg eugeniin as one of the doses for human use of the hot-water extract in mice. As shown in Table 1, the oral administration of eugeniin at 0.3 mg/kg significantly delayed the development of skin lesions compared with the control (p < 0.01 by the Student's t test and the repeated measures ANOVA). When the dose of eugeniin was increased to 50 mg/kg, eugeniin not only delayed the development of skin lesions but also significantly prolonged mean survival times (p < 0.01 or 0.05 by the Student's t test and/or the repeated measures ANOVA). Furthermore, the intraperitoneal administration of eugeniin at 0.3 mg/kg was significantly effective in retarding the development of skin lesions as observed in the oral administration (p < 0.05 and 0.01 by the Student's t test and the repeated measures ANOVA, respectively). The intraperitoneal administration of 3 mg/kg as well as the oral administration of 50 mg/kg was significant in retarding the development of skin lesions and mean survival times (p < 0.05 by the Student's t test and the repeated measures ANOVA). In either case of the administrations, the higher doses (3 or 50 mg/kg) showed significant therapeutic efficacy on two different aspects for the symptom of infected mice, which exhibited trends suggesting a dose dependence with respect to therapeutic efficacy. Eugeniin exhibited therapeutic efficacy in the lethal HSV-1 infection model and the effects were similar in the two different administrations.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Therapeutic efficacy of eugeniin on development of skin lesions caused by HSV-1 infection in mice. Mice were cutaneously infected with wild-type HSV-1 at 104 PFU/mouse. Eugeniin at 3 mg/kg () and 6 mg/kg () and PBS () as a control were intraperitoneally dosed three times daily for 7 days after infection. ACV at 5 mg/kg () was orally dosed three times daily for 7 days after infection. The skin lesions of infected mice were observed as described in the text. Parentheses indicate mortality in each group on days 20 after infection. Bars indicate the S.E. of each mean score. *p < 0.005 versus control at 104 PFU/mouse by the repeated measures ANOVA; **p < 0.0005 versus control at 104 PFU/mouse by the repeated measures ANOVA; ***p < 0.0001 versus control at 104 PFU/mouse by the repeated measures ANOVA; p < 0.05 versus controls by Fisher's exact test; p < 0.001 versus controls by Fisher's exact test.

Toxicity of Eugeniin in Mice. The oral administration of eugeniin was examined for its toxicity in mice (Table 2). At a 50 mg/kg oral dose of eugeniin, there was no significant difference in the mean weights between treated and untreated mice and lethal toxicity was not observed for at least 20 days after the initial administration. Thus, the oral administration of eugeniin was not significantly toxic.

Effect of Eugeniin on Virus Yields in the Skin and Brain of HSV-1-Infected Mice. We examined the anti-HSV-1 activity of eugeniin in the skin and brain of wild-type HSV-1-infected mice. Table 3 shows the virus yields in the skin and brain removed from infected mice on days 3 and 6 after infection. Since the values of virus yield in the skin and brain varied somewhat in repeating experiments, we evaluated the effect of eugeniin on virus yields in two independent experiments using different doses. In the skin, 10 mg/kg eugeniin was significantly effective in reducing virus yield compared with the controls (p < 0.01 by the Student's t test). The oral administration of eugeniin at 0.3 to 50 mg/kg showed a tendency to reduce virus yields in the skin compared with the controls, although the reduction of virus yields was not statistically significant at some doses. In the brain, 50 mg/kg eugeniin significantly reduced virus yield compared with the controls (p < 0.01 by the Student's t test). Eugeniin at 1, 3, 10, and 50 mg/kg reduced virus yield to 95.5, 81.6, 78.9, and 59.1%, respectively, of the controls on day 6 after infection. Thus, the oral administration of eugeniin showed a tendency to reduce virus yields in both of the skin and brain compared with the controls. As shown in Table 3, the ratio of percentage of virus yields in skin and brain was calculated to compare the anti-HSV-1 activity of eugeniin with each organ. The percentage of virus yields in the brain of mice treated with eugeniin (10 and 50 mg/kg) were 1.15- to >1.61-fold less than those in the skin on day 6 after infection compared with those in eugeniin-untreated mice (controls). Eugeniin exhibited a stronger anti-HSV-1 activity in the brain than in the skin of infected mice.

Effects of Eugeniin with ACV or PAA on Anti-HSV-1 Activity in Vitro. Eugeniin was previously shown to exhibit antiviral activity against ACV- and PAA-resistant HSV-1 as well as the wild type in vitro (Kurokawa et al., 1998). Thus, it was suggested to have a different mode of anti-HSV-1 action from those of ACV and PAA. Furthermore, to assess the mode of its anti-HSV-1 action, combinations of eugeniin with ACV or PAA were examined for their anti-HSV-1 activities by the plaque reduction assay. As shown in Figs. 2 and 3, data were analyzed by isobolograms based on EC₅₀ values. FIC indexes for combination of eugeniin with ACV were 0.50 to 0.89 at the EC₅₀ levels. Eugeniin enhanced the anti-HSV-1 activity of ACV synergistically (Fig. 2). On the other hand, FIC indexes for the combination of eugeniin with PAA were 1.16 to 1.31 (Fig. 3). The combination with PAA was not synergistic, but rather antagonistic.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Isobologram showing the anti-HSV-1 action of eugeniin with ACV. The FIC value of 1.0 corresponds to the EC50 (µg/ml) of each drug. The straight lines between FIC values of 1.0 for two drugs represent the dose for combinations that would be needed to produce an EC50 value if the interaction of the two drugs was additive. The actual doses for combinations that produced an EC50 value are shown by closed circles.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Isobologram showing the anti-HSV-1 action of eugeniin with PAA. The FIC value of 1.0 corresponds to the EC50 (µg/ml) of each drug. The straight lines between FIC values of 1.0 for two drugs represent the dose for combinations that would be needed to produce an EC50 value if the interaction of the two drugs was additive. The actual doses for combinations that produced an EC50 value are shown by closed circles.

Interaction of Eugeniin and PAA on HSV-1 DNA Polymerase Activity. Eugeniin in combination with PAA was examined for its inhibitory activity of HSV DNA polymerase to evaluate the interaction between eugeniin and PAA on HSV DNA polymerase activity. HSV-1 DNA polymerase was partially purified from HSV-1-infected HEL cells and the apparent K_m value was 1.32 µM with respect to dTTP. Using this partially purified HSV-1 DNA polymerase, apparent K_i values for eugeniin and PAA were determined as 0.52 and 1.59 µM, respectively. Eugeniin and PAA alone were confirmed to inhibit the activity of HSV DNA polymerase noncompetitively with respect to dTTP as reported previously (Leinbach et al., 1976; Reno et al., 1978; Kurokawa et al., 1998). In combination of eugeniin with PAA, data were presented as percentage of activities of HSV DNA polymerase without K_i and K_m analyses (Fig. 4, A and B) to show clearly no enhancement of inhibitory effect of eugeniin and PAA alone. The percentages of activities of HSV DNA polymerase were plotted versus the various concentrations of PAA and eugeniin in the presence of the fixed concentrations of eugeniin and PAA, respectively. Eugeniin and PAA did not strengthen the inhibition of HSV-1 DNA polymerase activity in their combination but rather weakened the inhibition. Thus eugeniin and PAA in combination were confirmed not to potentiate each other's inhibitory activity and each of them might interfere with the interaction with HSV DNA polymerase and the other.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Combined effect of eugeniin with PAA against HSV DNA polymerase activity. Eugeniin and PAA in combination were examined for the inhibitory activity of partially purified HSV DNA polymerase as described in text. The percentage of activities of HSV DNA polymerase were plotted versus the various concentrations of PAA (A) and eugeniin (B) in the presence of the fixed concentrations of eugeniin and PAA, respectively.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Eugeniin has been purified from the hot-water extracts of G. japonicum and S. aromaticum and exhibited anti-HSV-1 activity in vitro (Kurokawa et al., 1998). In this study, we demonstrated that the purified compound exhibited therapeutic anti-HSV-1 activity in a cutaneous HSV infection model in mice. We previously showed that the extracts of G. japonicum and S. aromaticum exhibited oral therapeutic efficacy in a cutaneous infection model and a recurrent HSV infection model in mice (Kurokawa et al., 1993, 1995a,b, 1997). The extracts of medicinal herbs are popular in China and Japan and have been safely and cautiously managed for the treatment of chronic diseases based on the information on adverse reactions accumulated historically in traditional therapy (Jiangxu New Medical College, 1978; Kurokawa et al., 1993). For their possible use in the management of viral infection in humans, we have been evaluating the antiviral activity of extracts of medicinal herbs. Thus, it was verified that the two herbal extracts containing eugeniin are effective against HSV-1 infection in vivo.

Eugeniin was orally and intraperitoneally applied to infected mice to evaluate its oral bioavailability with respect to therapeutic efficacy. In either case of its oral or intraperitoneal administration, eugeniin at 0.3 mg/kg showed similar therapeutic efficacy in retarding the development of herpetic skin lesions (Table 1). When the oral administration of eugeniin at 0.3 to 50 mg/kg was applied to HSV-1-infected mice, the mean survival time of infected mice was significantly extended at a dose of 50 mg/kg (Table 1). In the case of intraperitoneal administration, eugeniin at 3 and 6 mg/kg used was significantly effective in prolonging the mean survival times, even in reducing mortality (Fig. 1; Table 1). Eugeniin showed similar therapeutic efficacy in the oral and intraperitoneal administrations at the different doses more than 0.3 mg/kg. Alternatively, eugeniin at 50 mg/kg was not toxic when administered orally, although its intraperitoneal doses at more than 10 mg/kg were toxic in mice (Table 2). The modes of absorption and clearance of eugeniin at the different higher doses may be different in the two routes of administration. However, the oral doses of eugeniin examined exhibited therapeutic anti-HSV-1 efficacy similar to the intraperitoneal doses without toxicity (Table 1). This suggested that oral bioavailability of eugeniin with respect to therapeutic efficacy was similar to its intraperitoneal bioavailability and might be high compared with oral bioavailability assumed from the difference between oral and intraperitoneal administration in general. Such oral bioavailability of eugeniin in detail should be confirmed by the pharmacokinetic and metabolical studies.

We used lethal and mild HSV infection models in mice in this study. Using the lethal infection model, we first evaluated the biological activity of eugeniin against HSV infection in vivo and the activity was confirmed by comparison with that of ACV as a positive control in the mild infection model. In the mild infection model, all infected mice did not always die and the development of herpetic symptoms was delayed. In this mild model, the intraperitoneal dose of eugeniin at 6 mg/kg exhibited significant therapeutic efficacy in mice and reduced the mortality of infected mice (Fig. 1). There was no significant difference between the efficacies of intraperitoneal dose of eugeniin at 6 mg/kg and the oral dose of ACV at 5 mg/kg in mice (Fig. 1). Thus, eugeniin at 6-mg/kg intraperitoneal dose may be as biologically active as ACV at a 5 mg/kg oral dose in the mild model.

The oral dose of eugeniin at 50 mg/kg was not toxic in mice (Table 2). Its intraperitoneal dose at 10 mg/kg reduced the mean weight of mice but did not show lethal toxicity (Table 2). At a 20 mg/kg intraperitoneal dose, all mice were dead (Table 2). However, when eugeniin at 6 mg/kg was intraperitoneally administered to HSV-1-infected mice, no significant weight loss was observed (data not shown). Both the oral dose of eugeniin at 50 mg/kg and its intraperitoneal dose at 6 mg/kg were significantly effective in alleviating herpetic symptoms (Fig. 1; Table 1). Thus, eugeniin at the oral dose at 50 mg/kg and the intraperitoneal dose at 6 mg/kg exhibited therapeutic efficacy in mice without toxicity.

Eugeniin reduced virus yields in the skin and brain of infected mice in a dose-dependent manner. The reduction was significant in the skin and brain at the doses of 10 and 50 mg/kg, respectively, for oral administration (Table 3). As shown in Table 1, the oral doses of eugeniin at 0.3 to 10 mg/kg significantly retarded the development of herpetic skin lesions and the oral dose at 50 mg/kg was significantly effective in both retarding the development of skin lesions and prolonging the mean survival times. The lethal HSV infection causes encephalitis and death in infected mice (Kurokawa et al., 1993, 1995a). Thus, the reduction of virus yields in the skin and brain was consistent with the amelioration of herpetic symptoms in mice. Eugeniin exhibited therapeutic anti-HSV-1 activity in both the skin and brain of infected mice. Furthermore, eugeniin showed a stronger anti-HSV-1 activity in the brain than in the skin of HSV-1-infected mice at the higher doses (Table 3). Previously, we showed that the hot-water extract of G. japonicum, in which eugeniin is involved, had a stronger anti-HSV-1 activity in the brain than in the skin. Thus, both the extract and eugeniin itself exhibited stronger anti-HSV-1 activity in the brain than in the skin. Eugeniin may represent therapeutic efficacy of the extract and be beneficial in preventing central nervous system complications followed by HSV infection.

Eugeniin was previously shown to exhibit anti-HSV activity against ACV-PAA-resistant HSV-1 strain as well as the wild type in vitro (Kurokawa et al., 1998). In vivo, the therapeutic activity of eugeniin was confirmed in mice infected with ACV-PAA-resistant HSV-1 as well as the wild type in our preliminary study (data not shown). The mode of its anti-HSV action was indicated to be different from those of ACV and PAA (Kurokawa et al., 1998). Previously, we showed that eugeniin noncompetitively inhibits the activity of HSV-1 DNA polymerase (Kurokawa et al., 1998). Thus, the anti-HSV-1 action of eugeniin was further analyzed in comparison with those of ACV and PAA. In an isobologram analyzing the combined effect of eugeniin with ACV in Vero cells (Fig. 2), eugeniin enhanced the anti-HSV-1 activity of ACV synergistically. Thus, the mode of anti-HSV-1 action of eugeniin was strongly confirmed to be different from that of ACV. In combination of eugeniin with PAA on Vero cells, on the other hand, eugeniin acted rather antagonistically with PAA in Vero cells (Fig. 3). The antagonistic activity of eugeniin with PAA was confirmed by their combinations on HSV-1 DNA polymerase activity (Fig. 4). Each of eugeniin and PAA did not strengthen the other's inhibitory activity but rather weakened it. PAA, a pyrophosphate analog, interacts noncompetitively with HSV DNA polymerase to directly inhibit the polymerase by interacting with the pyrophosphate binding site, blocking the binding of the pyrophosphate moiety that is cleaved from deoxyribonucleoside triphosphates during DNA synthesis (Leinbach et al., 1976; Datta and Hood, 1981). Eugeniin may interact with HSV-1 DNA polymerase in a manner interfering directly and indirectly with the binding of PAA to the polymerase and alter the stereo-structure of the HSV DNA polymerase to inhibit the activity. Therefore, eugeniin was suggested to interact with the HSV-1 DNA polymerase in the vicinity of PAA binding site and was indicated to possess the mode of different anti-HSV-1 action from those of ACV and PAA.

Eugeniin showed therapeutic anti-HSV-1 activity in infected mice. It was characterized to exhibit a novel anti-HSV action, including different anti-HSV activities in the brain and skin of infected mice. Thus, eugeniin was verified to be a candidate of new anti-HSV compounds. Eugeniin is a kind of ellagitannin. Some ellagitannins have been reported to exhibit anti-human immunodeficiency virus activity as inhibitors of reverse transcriptase (Kashiwada et al., 1992; Lee et al., 1992; Xie et al., 1995). There may be a possibility that some of the ellagitannins are more effective in inhibiting HSV-1 DNA polymerase activity. A further study of derivatives of eugeniin may allow us to study much more effective anti-HSV agents from natural products.