HEp-2 and Vero cells were grown in Eagle’s minimum essential medium (MEM) containing 5% fetal bovine serum (FBS). HHV-2, strain 186, was used throughout the experiments. HHV-2 was propagated in Vero cells cultured in MEM supplemented with 0.5% FBS. The virus was stored at −80°C until use. The amount of virus was measured by a plaque assay on Vero cells as previously described (12,13).

L-arginine hydrochloride (simply described as arginine) was obtained from Ajinomoto Co., Inc. Aqueous solutions containing arginine or NaCl were prepared in 20 mM acetic acid. The pH of the arginine solutions was adjusted to the indicated values with HCl; 20 mM acetic acid is insufficient to titrate arginine solution to the desired pH. The pH of the NaCl in 20 mM acetic acid and aqueous citrate solutions was adjusted to the indicated pH with NaOH. The pH meter was routinely calibrated using pH calibration standards.

All the starting materials were stored on ice prior to the virus inactivation experiments. A 190 μl aliquot of the solutions to be evaluated was placed in 2.0-ml plastic tube on ice and received 10 μl of virus preparation [approximately 10⁸ plaque-forming units (PFU)/ml)]. This was immediately followed by vigorous mixing and the sample mixture was incubated at the indicated temperatures. After the incubation, aliquots of these virus samples were 100-fold diluted with Dulbecco’s phosphate-buffered saline (PBS) without Ca²⁺ and Mg²⁺ containing 0.5% FBS to terminate the virucidal action of the solutions. This dilution step helped stabilize the infectivity of the surviving viruses after the inactivation treatment. The viruses were further diluted with ice-cold PBS containing 0.5% FBS and the number of infectious viruses in the treated preparation was measured by a plaque assay on Vero cells. All the experiments were conducted in duplicate or triplicate.

The effect of arginine on the multiplication of HHV-2 was examined as follows. The monolayered HEp-2 cells in 35-mm dishes were infected with the virus at an indicated multiplicity of infection (MOI). The infected cells were further incubated at 37°C for the indicated period in the serum-free MEM containing 0.1% bovine serum albumin (BSA) and in the indicated concentrations of arginine. At the end of the incubation, the amounts of progeny virus in the infected cultures were determined as previously described (14). Briefly, after two or three cycles of freezing and thawing the infected cells along with the culture media, the number of infectious viruses in the lysate was measured by a plaque assay.

The cytopathic effects (CPEs) were determined by microscopic observation of the cells; approximate amounts of rounded cells on monolayers were estimated under a phase-contrast microscope. The extent of cell death in the cultures was determined as follows. The monolayerd cells were trypsinized to obtain a single cell suspension. After the addition of MEM containing 5% FBS to the suspension to neutralize trypsin and stabilize the cells, the numbers of living and dead cells were measured by a dye-exclusion method with trypan blue (15).

Seven-week-old female BALB/c mice were infected with HHV-2 as follows. The mice, whose vaginal walls were scratched, were infected with 6×10⁵ PFU HHV-2 in 10 μl by vaginal instillation. For the treatment with test solutions, the mice were treated with 20 μl of the indicated solutions by vaginal instillation at 2, 12 and 24 h post-infection (h p.i.) and then once every day for 20 days. In case of acycloguanosine, it was administrated daily intraperitoneally at a dose of 0.05 mg/200 μl/head. The number of mice that survived and the symptoms of these infected mice were examined daily. All animal experiments were performed in compliance with the principles of laboratory animal care (NIH publication No. 86, revised 1985) and with the guidelines of Teikyo University School of Medicine Animal Experiments Committee.

Previously we observed that arginine and arginine derivatives are effective in inactivating a variety of enveloped viruses, although a number of them, such as NDV, are not sensitive to arginine (6). When HHV-2 was incubated with mildly acidic arginine solution, this virus was also sensitive to arginine (Fig. 1). Incubation of the virus with 1.0 M arginine at pH 4.3 at 30°C for 10 min reduced the infectivity to less than one-thousandth the level of the PBS control, while 0.1 M citrate at the same pH revealed marginal inactivation (0.25 relative to the PBS control) and 0.7 and 1.0 M sodium chloride solutions at the same pH were not effective. These results clearly indicate that mildly acidic arginine effectively inactivates HHV-2 and that neither the acidic pH nor the high salt concentration was effective. This inactivation by arginine was concentration-dependent, since 0.7 M arginine solution reduced the infectivity to only one-hundredth the level of the PBS control (Fig. 1). A comparison with other viruses indicates that HHV-2 is the most sensitive virus examined thus far (data not shown).

The rate of the virus inactivation was much faster with arginine. HHV-2 revealed no noticeable inactivation in the PBS during 20 min of incubation on ice (Fig. 2). In both 1.0 M sodium chloride and 0.1 M citrate solutions at pH 4.3, the virus was inactivated only slightly with time at the same temperature. Conversely, the virus inactivation occurred rapidly in 1.0 M arginine solution at pH 4.3; it decreased in monotone with time, reaching one-thousandth the level of the PBS control at 10 min of the incubation on ice and approximately 10⁻⁵ at 20 min of the incubation.

We previously demonstrated that the ability of arginine and citrate to inactivate HHV-1 diminishes rapidly when the pH of the solution was increased (2). The effect of pH on the inactivation of HHV-2 by 0.1 M citrate or 0.7 M arginine is displayed in Fig. 3. To compensate for the difference in ionic strength between these two solutions, 0.6 M sodium chloride was added to the series of 0.1 M citrate. At or below pH 4.0, all three solutions equally and efficiently inactivated HHV-2. Above pH 4.0, 0.1 M citrate solution by itself or supplemented with 0.6 M sodium chloride lost the ability to inactivate the virus to a much greater extent than 0.7 M arginine; 0.7 M arginine reduced virus infectivity below the detection level even at pH 4.2. A higher concentration of citrate at 0.7 M was least effective, possibly due to the salting-out effect of citrate that protected the virus from the inactivation, as discussed previously (5). Differences in the degree of virus inactivation among different solutions decreased at higher pH, e.g., above 4.6.

As described above, virucidal activity of arginine was much weaker at a higher pH. However, even at a neutral pH, a prolonged incubation time resulted in significant inactivation. The time course of the virus inactivation on ice by arginine at pH 7.0 is displayed in Fig. 4. Consistent with the results shown in Fig. 1, the ability depended an arginine concentration. HHV-2 was inactivated quickly with arginine solution at 1.2 M, reaching one-thousandth within 20 min of the incubation while the virus infectivity decreased slightly, although significantly (only 50–60% of the PBS control), in 0.75 M arginine solution. At 1.0 M, arginine at pH 7.0 inactivated the virus more effectively compared to that at 0.75 M, but less effectively compared to that at 1.0 M arginine at pH 4.3 (Fig. 2), reaching one hundredth the level of the PBS control at 20 min of the incubation. These results clearly reveal that, even at a neutral pH, arginine inactivates HHV-2, provided that higher arginine concentration and a longer incubation time were employed than those at an acidic pH.

From the results described above, it is evident that arginine directly inactivates HHV-2 in the test tube. Next, we examined the effects of arginine on the virus multiplication in infected cells by incubating HHV-2-infected HEp-2 cells with arginine at various concentrations. The progeny virus yield as a function of arginine concentration (note that the arginine concentration is mM, not M as in the virus inactivation experiments) is displayed in Fig. 5. Arginine reduced the virus yield dose-dependently in the infected cells, reaching one-hundredth the yield of the control untreated culture at 16 mM. As a control of ionic strength of arginine, the effects of sodium chloride were examined. As shown in Fig. 5, sodium chloride also concentration-dependently inhibited the multiplication of HHV-2, but to a much smaller extent than the level achieved by arginine. Although HHV-2 is closely related to HHV-1, HHV-2 is more sensitive to arginine than is HHV-1; only 1-log reduction of the relative virus yield was observed for HHV-1 in the presence of arginine at 25 mM (10). Although the arginine concentration is not within a physiological level, it is considered to be within a tolerable range for the infected cells, as described in the following section (i.e., cytotoxic effects of arginine).

It should be noted that the antiviral activity of arginine described above has no relation to the observed direct virucidal activity of arginine, as observed at higher concentrations required to inactivate the virus (Figs. 1–4). Thus, these antiviral effects of arginine are not mediated by virucidal activity, but rather by other factors.

Arginine induced notable morphological alterations in HEp-2 cells which included both rounding and shrinkage of the cells, but without cell detachment from the dishes. CPEs were enhanced with increasing concentrations of arginine and, at the concentration of 20 mM, arginine induced the CPEs in a small, but significant population of uninfected HEp-2 cells. However, we previously demonstrated, that upon incubation of HEp-2 cells with arginine for 26 h, the cell death was induced only by a small (2–5%) fraction of the total cell population independently of arginine concentrations (0–40 mM), a magnitude similar to the cell death observed in the sodium chloride-treated cells at the same concentrations (10). These results suggest that the observed decrease in the virus yield by arginine is unlikely due to the degeneration (cell death) of the infected cells by arginine. This suggestion is further confirmed by the observations that the infection with HHV-2 completely abolished arginine-induced CPEs: namely, arginine does not induce CPEs in HEp-2 cells infected with HHV-2, possibly due to the expression of anti-apoptotic genes of HHV-2 in the infected cells (16).

A previous characterization of the kinetics of the viral multiplication process in HHV-infected Vero cells (17) revealed that viral DNA replication occurs exclusively between 3 and 6 h p.i and a large amount of DNA is accumulated in the infected cells when the replication of virus DNA is completed. The formation of nucleocapsids as well as the envelopment of the nucleocapsids begins at 5 h p.i. simultaneously with the formation of infectious progeny virus and continues until approximately 12 h p.i. In HEp-2 cells, the onset as well as the completion of the formation of the progeny virus is slightly delayed and a small but significant degree of the progeny virus formation continues even after 12 h p.i., compared with those in Vero cells (Koyama, unpublished data).

To examine the arginine-sensitive step in the multiplication process of HHV-2 in HEp-2 cells, the ‘time of addition’ experiment was performed. Arginine was added to the infected culture at various times after the onset of HHV-2 multiplication, and the virus yield at 26 h p.i. was compared to the amount of virus formed in the infected cells at the time of addition of arginine (Fig. 6). The formation of the progeny virus was almost immediately suppressed by the addition when the infected cells received arginine at 0, 2 or 4 h p.i., as noted in small additional increases of the virus yield. In contrast, when arginine was added at or after 6 h p.i., the suppression in the progeny virus formation was evident but not complete; the formation of the progeny virus continued for a short time after arginine addition. For example, the amount of the infectious virus at 8 h p.i. was 1.42×10⁶. When arginine was added at this time point and cell culture was continued for 26 h, the final virus yield was 4.4×10⁶, meaning that the progeny virus formation continued after the addition to a significant extent. However, this virus yield (4.4×10⁶) was less than the virus yield without arginine addition (35×10⁶), although it was more than 3-fold higher than the amount of the infectious virus formed at 8 h p.i. (1.42×10⁶). A similar result was obtained from the previous ‘time of addition experiment’ with HHV-1 (10) and an additional characterization of the one-step growth of HHV-1 revealed that the formation of the progeny virus in the HHV-1-infected HEp-2 cells continued normally for 3–4 h after the addition of arginine and then stopped (10). Both the current and the previous results involving HHV-1 agree with the conclusion that the formation of the progeny virus remains sensitive to arginine in the late stage (6, 8, 10 and 12 h p.i.) of the infection, even after the completion of the viral DNA replication and the formation of nucleocapsids. However, it should be noted that our results do not exclude the possible arginine-sensitive step in the early stages (2 and 4 h p.i.) of the HHV infection.

The in vitro studies as described above clearly indicate the antiviral and virucidal activities of arginine against HHV-2 infection. Although both of these activities require higher arginine concentration than the physiological level, we carried out a pilot study to examine the potential application of arginine as a therapeutic or preventive medicine in vivo. Mouse genital herpes infection was used as a model disease for the treatment with arginine, since arginine at high concentrations is not tolerable for an in vivo systemic application but can be used to treat superficial virus infection on the body surface.

BALB/c mice are very sensitive to HHV-2 following infection through the vaginal route; the virus first multiplies at the site of infection in the vagina, followed by invasion into the central nervous system through sacral ganglia neuronal cells, and finally causes lethal encephalitis at the brain stem. To examine arginine as a possible treatment for genital herpes infection, we examined the effects of acidic arginine on the establishment of HHV-2 infection in a mouse model. To ensure the virus infection, the vaginal walls of the mice were scratched before inoculation with HHV-2. Two hours after the inoculation, the mice received 20 μl arginine solution by vaginal instillation according to the schedule described in Materials and method. Considering a potential dilution of the arginine solution by vaginal secretion, arginine at slightly higher concentrations and a lower pH was used in this in vivo study compared to those used in the in vitro studies described above.

When the infected mice were treated with PBS (control without arginine treatment), the mortality of mice was noted at 8 days p.i. and the number of surviving mice was 1 out of 5 at 20 days p.i. (Fig. 7). However, when the infected mice were treated with arginine solution (pH 3.5), only one mouse died at 8 days p.i., with four mice surviving until 20 days p.i. In contrast to the infected mice treated with arginine, those treated with citrate buffer at the same pH began to die at 8 days p.i., and all infected mice died by 10 days p.i.. Although acycloguanosine is known to be an effective antiviral drug for human HSV infections, this medicine is toxic to BALB/c mice and treatment with this compound may not rescue mice from the infection. The number of surviving mice was almost similar to those treated with PBS, although the mice treated with acycloguanosine began to die at 10 days p.i. and the mice treated with PBS began to die at 8 days p.i. It should be noted that no apparent tissue damage was observed in the genital organs following arginine treatment. These results clearly indicate that the arginine treatment may effectively rescue mice from the infection of HHV-2 through the vaginal route and suggest a potential use of arginine as a preventive medicine against genital HHV-2 infection.