Defensins and viral infection: dispelling common misconceptions.

Abstract

Human α- and β-defensins are cationic, amphipathic effector peptides of the innate immune system with broad antimicrobial activity [1]. α-defensins are produced by neutrophils (human neutrophil peptides [HNP] 1–4), as well as by epithelial cells in the gut and genitourinary tract (human defensins [HD] 5 and 6). β-defensins (human β-defensins [HBD] 1–4) are constitutively expressed by epithelial cells of skin and mucosal surfaces. Originally discovered due to their antibacterial activity, defensins are also active against both enveloped and non-enveloped viruses. Mechanisms important for bacterial inhibition have been historically assumed to be responsible for defensin antiviral activity; however, this assumption has not held up to experimentation. Our purpose is to address some persistent myths regarding the activity of human defensins against viruses, as well as to identify areas requiring further investigation (Figure 1). Figure 1 Common misconceptions of defensin antiviral activity. Misconception: Antiviral Activity Is Only Due to Defensin Positive Charge The initial antibacterial mechanism was proposed to be largely dependent on simple charge–charge attraction to the bacterial membrane [2]. Accordingly, defensins are most potently antibacterial in hypotonic media [1], [3]. Antiviral activity, in contrast, is generally preserved at physiological salt concentrations in normal cell culture media, arguing against a dominant charge–charge component to their interaction with viruses [4]–[6]. There is additional strong evidence that charge alone does not dictate antiviral activity: linearized α-defensins that lack a disulfide-stabilized 3-D structure are nonfunctional against all viruses tested [6]. Additionally, obligate monomer forms of α-defensins are highly attenuated for binding to and/or inactivation of viral pathogens. [7], [8]. If antiviral activity were simply due to charge–charge interactions, these defensin mutants would still be active, as their charge is conserved. A second line of investigation using direct mutational analysis showed that arginine to lysine substitutions at specific residues attenuate the antiviral activity of HD5 [7]. These results, coupled with the marked preference for arginines over lysines in α-defensins, imply that other aspects of arginine residues are more important than simple charge [6], [9]. Furthermore, β-defensins are, on average, more charged than α-defensins yet largely exhibit less antiviral activity, especially against non-enveloped viruses. Thus, while charge is important for defensin function, it is not the only determining factor in antiviral activity.