This paper provides a new information on the biological properties of HA, describing its antiviral in vitro activity against 5 viruses, with different structural and biological features.
HA is a non-sulfated glycosaminoglycan widely present in the extracellular matrix of soft tissues and in several biologic fluids . Experimental and clinical data demonstrate the involvement of HA in structure maintenance, moisturizing, tissue lubrication and wound healing. These properties, associated with an excellent safety profile, are exploited in medical practice as well as in aesthetic and cosmetic fields [2, 3]. Furthermore, initial evidence also ascribes antimicrobial properties to HA [11–13], adding further appeal to the HA-containing products because of the beneficial effects probably deriving from its antimicrobial properties. Here, we have demonstrated a wide spectrum antiviral activity of a high molecular weight HA. The most effective inhibition is observed against COXB5, MV and Influenza Virus WSN33 (A/H1N1) and it is interesting that the phenomenon is retained even at a concentration (1 mg/ml) lower than that one commonly used for clinical or cosmetic applications. The high SIs displayed by HA for these viruses document a strong antiviral activity not related to cytotoxicity. HSV-1 and PPV are also inhibited by HA, but only at the highest concentration (4 mg/ml): the SIs are consequently lower. No activity is ever observed against ADV-5, HHV-6, PRRSV. The HA failure to inhibit ADV is not unexpected since, according to Chaudhuri , HA enhances ADV replication in vivo and, though to a lesser extent, also in vitro. The different cell system we used may account for the lack of replication enhancement observed in our experiments.
Our findings provide evidence on the effects of HA on a variety of RNA and DNA viruses, with or without the lipidic envelope, characterized by very different replication strategies. Therefore, we can speculate that HA antiviral mechanism(s) probably involves general/non-specific host cell-virus interaction at membrane level, such as virus entry or release, rather than restricted, virus-specific events occurring inside the cell. This speculation is supported by the kinetic results showing that both COXB5 and HSV-1 are growth inhibited irrespective of the time of HA addition during the virus replication cycle. In line with this hypothesis, we may suggest that HA, known to have a heavy ionic charge, may alter the electrostatic interactions between virus and cell receptors and/or other cell membrane components, thus in turn affecting virus entry and/or exit. Literature data substantiate this hypothesis. Dengue virus uses heparan-sulphate (HS) as a cell receptor: different types of HA, interacting with the virus envelope glycoprotein, responsible for virus attachment to HS, are able to inhibit virus attachment and entry . HSV-1 recognizes HS as a receptor too and it has been reported that molecules, affecting the interaction between HSV-1 envelope glycoproteins and HS, are able to reduce the virus growth [15–17]. Several Enteroviruses (including some Coxsackieviruses) are reported to interact with HS for virus entry [18–20]. The results of our kinetic experiments seem to suggest that virus release is also inhibited by HA. Notoriously, this step involves host cell membrane. HSV-1 and COXB5 are released by different mechanisms: HSV-1 by trans-membrane trafficking of vesicles, while COXB5 by cell membrane lysis. Therefore we may assume that the mechanism, by which HA inhibits virus release, has to be non-specific. The results observed in the assay of cell protection from lysis, showing that cells exposed to HA are more resistant to cell lysis, suggest that HA stabilizes cell membranes. This modification could impair any membrane involving process, such as envelope fusion with cell membrane, vesicle fusion and membrane disruption. We have observed that in the antiviral experiments with MV, a syncytiogen virus, not only virus yield was reduced in the presence of HA, but syncytia size also appeared at light microscope observation smaller than those ones of the control cultures. This HA stabilizing activity on membranes indirectly implies that cell exposure to HA inhibits virus entry and/or exit. Finally, the lack of appreciable virucidal activity by HA, against all the viruses under study, rules out the possibility of a direct virus inactivation by HA and it also supports the idea that the steps, involving virus-cell membrane interaction, are preferentially, if not exclusively, affected by HA.
In conclusions, our study provides a wide spectrum demonstration of the antiviral activity by HA, opening new perspectives in prophylaxis and therapy of some viral diseases. The hypothesis of specifically counteracting cell-virus attachment and viral release, by local administration of HA, is very appealing especially for oral, genital and ano-rectal anatomical areas, where compound(s) can be administered as creams, gels or wash solutions. Many HA-based commercial products already available for topic use have HA concentrations much higher or equal to those we found active against different viruses: so it can be hypothesized that effective concentrations can be locally reached. The inhibitory activity observed against HSV-1 is particularly interesting since HA is the basic component of mouthwash solutions and ophthalmic drops, used as artificial tears. According to our findings, HA may be also considered/included as an antiviral agent in the treatment of HSV-1-associated stomatitis and kerato-conjunctivitis in this kind of preparations. Moreover, since Enteroviruses are often responsible for a childhood form of vesicular stomatitis as well as respiratory diseases, the present findings on Coxsackievirus inhibition may open new perspectives for oral administration of HA as natural atoxic medical treatment of newborns and babies. Similarly, the anti-influenza virus activity observed might be exploited in nasal sprays to locally reduce viral replication. Moreover, HA has been demonstrated to have pro-inflammatory activity. Low weight HA fragmentation products and, in the presence of IFNγ, even high molecular weight HA molecules can also activate innate immunity mechanisms through Toll-like Receptor 4 and CD44 [21, 22]. This pro-inflamamtory activity may contribute to counteract virus replication and spread in vivo.
In conclusion, our findings strongly support the use of this safe glycosaminoglycan in clinical practice as a potential antiviral compound, both for disease prevention and treatment. Further clinical trials on this topic are required to better understand the antiviral activity of this compound.