In vitro virus inactivation effects of aqueous arginine solution were evaluated by incubating the virus preparations with various solvents as a function of pH or temperature. After incubation, the surviving infectious viruses were determined by infecting appropriate host cells with viruses and measuring the plaque formation of the infected cells. With enveloped viruses, arginine has consistently shown a synergistic effect with solution pH and temperature on virus inactivation. Typical inactivation data with an enveloped virus, HSV-2 (herpes simplex virus type-2), are shown in Figure 1 [30]. Similar trend has been observed for other enveloped viruses, including HSV-1 (HSV type-1) and influenza virus [10,11,12,13], and at various incubation temperatures. The results in Figure 1 were obtained following incubation of the virus with the indicated solvents as a function of pH at 20 ºC for 5 min. Virus inactivation was enhanced at a higher temperature and longer incubation time. Acid titration was conducted using 0.1 M citrate and the pH was adjusted by NaOH. As shown in Figure 1, virus inactivation was marginal at pH 4.2 (open triangle), below which a sharp decrease in virus yield was observed. Nearly 5 log₁₀ reduction was observed at pH 3.8. Neither the increased citrate concentration at 0.7 M (open square) nor the addition of 0.6 M NaCl (closed triangle) resulted in a shift on the pH titration curve. Although marginal, higher concentration of citrate and NaCl appeared to suppress pH-induced virus inactivation above pH 4.2.

The above results indicate that neither 0.6 M NaCl nor excess citrate (i.e., 0.6 M) have a synergistic effect with pH (i.e., 0.1 M citrate) on virus inactivation. Conversely, arginine demonstrated synergy with pH in inactivating viruses. Figure 1 shows pH titration of virus inactivation in 0.7 M arginine (arginine HCl solution titrated with HCl). It is evident from the figure that the pH dependence was shifted to a higher pH, by about 0.3–0.4 pH unit. This suggests that to achieve a 5 log₁₀ reduction, the solution pH can be maintained at ~4.0–4.1 in the presence of 0.7M arginine, while the same reduction in infectivity can only be achieved at a pH of ~3.8–3.9 in its absence. Although this difference in effective pH range appears to be marginal, the difference could have a significant impact on patient safety if these acidic solutions are utilized in circumstances that require application for human body surface, or perhaps accidental contact. Similar synergistic effects of arginine with a variety of environmental conditions (i.e., pH and temperature) have been observed with several other enveloped viruses [32,33].

Arginine at neutral pH also demonstrates synergy with temperature in inactivating viruses [12,13]. Figure 2 shows an example of HSV-1 inactivation with temperature (This sentence was moved here from the top of Figure 2). In phosphate-buffered saline (PBS), HSV-1 was stable up to 40 ºC, above which the infectivity of the virus decreased sharply. There were few surviving viruses at 50 ºC. The temperature dependence of virus inactivation was gradually shifted to lower temperatures with increasing arginine concentration at neutral pH. In the presence of 1.2 M arginine, there was already 50% virus inactivation observed at 30 ºC. The temperatures at which 50 (second column) and 90% virus inactivation (fourth column) occurred are summarized in Table 1.

Both temperatures decreased with increasing arginine concentration, leading to 16–17 ºC lower temperatures in 1.2 M arginine. Table 1 contains the inactivation data with NaCl. Aqueous NaCl solutions at the indicated concentrations (0.5–1.2 M) also decreased the inactivation temperature, but to a lesser extent. For example, 1.2 M NaCl reduced the temperatures of 50 and 90% inactivation only by 7 (third column) and 4 ºC (fifth column). The results with NaCl suggest that increasing ionic strength is responsible for inactivating the virus. At least a part of virus inactivation by arginine may be due to its ionic property, although virus inactivation by NaCl and arginine may be due to an entirely different mechanism. We will analyze the possible mechanism of arginine effects from the types of damages that viruses suffer and how arginine, or ionic strength, may interrelate to these damages.

The results described above were obtained in the test tubes. It would be a challenging idea to test the effects of arginine in vivo. The time course of inactivation demonstrated that inactivation occurred shortly after the exposure of the virus to arginine solution, suggesting that even a brief exposure of the virus to arginine solution is sufficient. Such a fast action of virus inactivation suggests the potential usefulness of arginine as a non-irritative disinfectant. Conventional disinfectants generally cannot be used on the mucosal membranes or on the injured sites of the body due to their severe cell and tissue toxicities. We anticipated that arginine, being a natural product of lower toxicity, would have insignificant adverse effects on these surfaces upon topical applications and hopefully may exhibit some efficacy. To our surprise, viruses were significantly inactivated by acidic arginine solution on the body surface in two model systems, i.e., HSV-1 infection on epithelial keratinocytes and HSV-2 genital infection [31].

In an epithelial keratinocyte model, HSV-1 was inoculated in the cornea of rabbit eyes and treated with an acidic arginine solution. Those eyes treated with saline (negative control: eyes without treatment) showed severe lesion. In contrast, the eyes treated with a commercially available acycloguanosine ointment demonstrated potent efficacy with no lesion. The application of an acidic arginine solution onto the infected eyes resulted in lesion on the cornea, however to a lesser extent than those treated with saline alone. Citrate of the same acidity did not exhibit such a clear suppression. Although the concentration and pH of the acidic arginine solution were not at physiological conditions, it should be noted that no apparent injuries to the cornea of the treated rabbit’s eyes were observed [31]. The decreased efficacy and toxicity may be in part due to the dilution of acidic arginine solution, upon administration, by the massive liquid secretions. Nevertheless, these results indicate that arginine can suppress the occurrence of infection and the development of keratinocytic symptom by HSV-1 infection in rabbit’s cornea and may warrant further investigation on a more effective formulation and administration route.

We then examined the effects of arginine treatment on the establishment of genital infection with HSV-2 [31]. HSV-2 is also sensitive to the virucidal effects of arginine, significantly more so than HSV-1 [30]. BALB/c mice are sensitive to HSV-2 infection and the virus first multiplies at the site of infection in the vagina, then invades into the central nervous system through the sacral ganglia neuronal cells, and finally causing lethal encephalitis at the brain stem. Mice, if untreated, start dying at 8 days post infection. The infected mice treated with 1.5 M arginine solution (pH 3.5) demonstrated a greatly enhanced survival rate [31]. These results clearly show that the treatment with acidic arginine effectively suppressed the establishment and development of HSV-2 infection in the genital organs of mice, suggesting its potential development as a human therapeutic against genital HSV-2 infection.

Low pH alone, or in combination with elevated temperature, has been implicated as the major factor influencing the structure of viral proteins and virus infectivity. Adenovirus 2 (Ad2) underwent a structural transition at elevated temperatures, accompanied by the loss of penton bases and their fibres at the icosahederal vertices, which are required for the maintenance of viral infectivity [34]. While this structural change caused Ad2 aggregation, the structure change, not the aggregation, appeared to be the direct cause of virus inactivation (Figure 3). This temperature-induced structure change was greater at increasing pH.

Similarly, Ad5 (Adenovirus 5) demonstrated structural change of capsid with increasing temperature, which was enhanced further at higher pH [35]. Using a fluorescence probe, the temperature dependence of fluorescence showed that increasing temperature resulted in initial loosening of the capsid structure and then more complete capsid disassembly, both leading to virus inactivation. Although these capsid structure changes led to virus aggregation, it appears that the structure changes themselves were responsible for inactivation. Since both cases are concerned with non-enveloped viruses, the effects of temperature are primarily on viral proteins.

Norwalk virus like particles (VLP), a non-enveloped virus, demonstrates a unique dependence on pH and temperature stresses. When the virus preparation was brought to pH 8.0, intact VLP virus disappeared, most likely due to changes in the local conformational changes of the viral proteins [36]. When this preparation was brought back to pH 7.0, fully assembled infectious virus was recovered: i.e., pH-induced structure change is reversible. However, response to high temperature stress is different. Both the local structural changes of the proteins and the capsid structure upon heating were irreversible. It appears that this irreversible heat-induced change is due to aggregation of viral proteins.

Another stress that causes virus inactivation is dehydration, which can occur both in vitro and in vivo. Measles virus is unstable in the dried state at elevated temperatures [40]. Although spray drying procedure causes little changes in its viral infectivity, virus was rapidly inactivated during storage at 37 ºC [41]. Since drying itself had no effect on virus infectivity, there must be a stress or several stresses in the storage condition responsible for time-dependent virus inactivation. Arginine at 0.23 M in pre-drying concentration, insufficient by itself to cause virus inactivation, resulted in the protection of the virus during storage in the dried state [41]. In this case, arginine did not demonstrate synergy with stress (i.e., dehydration) to cause virus inactivation.

Aggregation of virus is a common observation [42], although its relation to inactivation may depend on how the virus samples are prepared (see Figure 4). Virus aggregation can now be more readily and accurately characterized using modern technologies, including AUC (analytical ultracentrifugation), DLS (dynamic light scattering) and FFF (field flow fractionation). For example, sedimentation velocity analysis (using AUC) showed that adenovirus aggregates into highly heterogeneous sizes with the monomer population accounting for only ~50% (Alliance Protein Laboratories homepage, www.ap-lab.com).

Aggregation of infectious poliovirus has been extensively studied [37]. A monoclonal antibody, MoAb 35-1f4, was shown to cause polymerization of poliovirus, presumably through its effects on surface proteins, i.e., bivalent binding and/or consequent conformational changes. This polymerization resulted in loss of infectivity. As the size of oligomers increased, infectivity was reduced (e.g. the dimeric virus has 50% reduced infectivity), suggesting that the loss of infectivity was due to steric hindrance of virus accessibility to the host cells and not due to irreversible damages that the virus has suffered from the bound antibody. This was confirmed by the release of infectious virus from the polymerized structure through papain treatment. Polymerization of poliovirus was at least in part due to bivalent antibody binding, as excess antibody resulted in monomeric virion particles: this is expected from a large excess of antibody per binding site on viral particle [43]. Aggregation of infectious viruses is problematic for inactivation. Aggregated viruses are often non-neutralizable by antiserum and cause persistent and delayed infectivity [44]. In this case, viruses may have to be disaggregated for efficient virus inactivation, as will be discussed later.

Structural damages on viruses are often accompanied by virus aggregation. In such cases, virus inactivation and aggregation are associated with each other and hence it is difficult to distinguish between damage-induced inactivation and aggregation-induced inactivation, as illustrated in the following examples. Pinto et al. have used DLS to study the aggregation behavior of MS2 bacteriophage, which ranged in size up to 500 nm, and have correlated aggregation to its reduced efficacy [45]. DLS was also employed to correlate the aggregation of influenza virus particles to loss in antigen activity [46]. Furthermore, a 2-log decrease in infectivity of baculovirus was reported to occur concurrently with virus aggregation, upon incubation above 45 ºC [47]. The size of the virus particle was shown to increase from 150 to 250 nm by DLS measurements. Both low pH and high temperature have been shown to be the major factors responsible for virus inactivation associated with aggregation. Measles virus (MV) was shown to aggregate by DLS measurements immediately upon lowering the solution pH from 6 to 4 [32]. With increasing temperature, both the aggregate size and measurement variability of MV increased further. The cause of aggregation is still unknown, however membrane destabilization due to the instability of the membrane proteins, and thus the lipid-protein interaction, may help to explain the observation. Thermally-induced changes in the secondary structure of MV viral proteins was detected using circular dichroism (CD) at 222nm as a function of temperature, and the temperature at which this transition occurred was decreased by more than 30 ºC upon lowering the pH from 6 to 4. Namely, low pH may cause the conformational changes of MV viral proteins at physiological temperatures, which in turn affect membrane structure of MV and lead to virus aggregation.

In another example of low pH-mediated aggregation, respiratory syncytial virus (RSV) underwent extensive aggregation under acidic conditions; aggregation of RSV at pH 3–5 during dialysis was confirmed by elevated optical density values [33]. Further evidence of membrane destabilization leading to aggregation was obtained from the intrinsic fluorescence data; the temperature at which the changes in the tertiary structure of viral proteins was detected correlated with the temperature at which increases in turbidity occurred, which suggest that the exposure of the apolar tryptophan (Trp) residues led to aggregation of RSV particles. CD analysis revealed that the temperature (T_d) at which the RSV membrane proteins lost secondary structural content varied depending on the solution pH; significant loss was observed below 40 ºC at pH 3–5 and in between 40–60 ºC for solution pH between 6 and 8. FTIR (Fourier transform infrared spectroscopy) analysis indicated that the cause of this shift in T_d was primarily due to the formation of intermolecular β-sheets, which may be due to aggregation. Thus, there is an abundance of data implicating the structural instability of viral proteins to be the main cause for virus aggregation, which consequently leads to its inactivation. It is still unclear, however, why the viral membrane is unstable under acidic condition and not under alkaline condition.

Freezing causes loss of infectivity of enveloped viruses [48], suggesting that ice crystal formation damages membrane lipids, as depicted in Figure 5 [49,50]. Consistent with this notion, DMSO (dimethyl sulfoxide), which is commonly used to preserve cultured cells, protected several viruses form freezing-induced loss of infectivity [51].

Defensins, which comprise a family of broad-spectrum antimicrobial peptides, are among the key mediators of the innate immune system. One of the defensin family member, human neutrophil peptide (HNP-1), was shown to be effective in inactivating several viruses, including HSV types 1 and 2, cytomegalovirus, vesicular stomatitis virus, and influenza virus A/WSN [38]. Similarly to arginine, an enhancement in HNP-1’s virucidal efficacy was observed upon modifying either the solution pH or the temperature; increasing the incubation temperature from 10 to 40 ºC resulted in >3Log₁₀ reduction in infectivity of HSV-1 in the presence of 25 μg/mL HNP-1 at pH 7.4, while increasing the pH from 6 to 8 in the presence of HNP-1, reduced the infectivity by >1 Log₁₀ at 37 ºC. Defensins are not effective against all viruses, however. HNP-1 was not effective against echovirus and reovirus, which are both non-enveloped viruses. Similarly, arginine was shown to be ineffective against poliovirus, a non-enveloped virus [10]. The difference in susceptibility observed between enveloped and non-enveloped viruses suggests that the viral envelope may be the target of both defensins and arginine, as described later. The correlation between reduction in infectivity and the interaction of these classes of compounds to the viral envelope suggests that the interaction leads to a breach in the integrity of the viral envelope. Evidence to support this statement can be found in a study that was conducted to determine the effect of human defensin HNP-2 on the integrity of unilamellar vesicles [39]. A fluorescence quenching study, utilizing ANTS-DPX (8-aminonaphthalene-1,3,6 trisulfonic acid - p-xylene-bis-pyridinium bromide) system encapsulated within a POPG (palmitoyloleoylphosphatidylglycerol) vesicle, was conducted to observe the effect of increasing HNP-2 concentration on the integrity of the liposome. The addition of 1.3 μM HNP-2 to POPG vesicles resulted in enhanced fluorescence, which further increased with incubation time. Decreasing the HNP-2 concentration from 1.3 to 0.3 μM resulted in a decrease in both the fluorescence signal and the release kinetics, suggesting that the interaction is concentration dependent. The similarity observed between arginine and defensins may not be unexpected, as mature neutrophil defensins are reported to comprise of high arginine content [39].