Peptide Inhibitors against Influenza Virus

Abstract

Influenza A virus is a particularly problematic virus because of its ability to cause high levels of morbidity on a global scale within a remarkably short period of time. It also has the potential to kill very large numbers of people as occurred in the Spanish influenza pandemic in 1918. Options for antiviral therapy are limited because of the paucity of available drugs and the rapid mutation rate of the virus leading to the emergence of drug-resistant strains. The current H1N1 pandemic and potential threats posed by other strains highlight the need to develop novel therapeutic and prophylactic strategies. Here, we summarize the current state and recent developments of peptide-based inhibitors of influenza A virus.

Peptide inhibitors

A wide range of organisms, such as protozoans, insects, invertebrates, plants, amphibians, birds, fish and mammals, including humans, produce peptides as their first line of defence against viral, fungal, bacterial and other parasitic infections [1]. These peptides usually range from 12 to 100 amino acids in length, are generally positively charged (+2 to +9) and are amphiphilic [2,3]. Hundreds of such peptides have been recently identified [4,5]. Peptides usually act on microbes by destabilizing the cell or by damaging the microbial membranes. However, they also target the immune system by modulating phagocytosis, increasing prostaglandin release, neutralizing the effect of lipopolysaccharides and by promoting accumulation of various immune cells at inflammatory sites [6–8]. In certain cases, they also promote angiogenesis [9] and influence wound repair [10].

Antiviral peptides are structurally diverse and broadly fall into four main categories, such as amphiphilic peptides with 2–4 β-strands, amphipathic α-helices, loop structures and extended structures [4]. Antiviral peptides often target components of the viral envelope and thus affect the absorption and entry processes [11–14]. Blocking the interaction of the virus with its receptor either by targeting the surface proteins of the virus or the host cell receptors prevents initial virus entry or subsequent cell-to-cell spread. Alternatively, they might target the replication machinery of the virus [15,16].

Peptide inhibitors of influenza virus

Peptides such as defensins have long been utilized to inhibit bacterial infections [17,18]. Recent studies demonstrated the usefulness of these peptides against viral infections [19,20].

Enveloped viruses, such as HIV, influenza A, severe acute respiratory syndrome-coronavirus, parainfluenza, respiratory syncytial and Ebola, infect cells by fusion with the host cell membrane using their envelope protein(s), which are commonly type-1 fusion proteins. Once the viral fusion protein attaches to the surface receptors of the host cell, it undergoes a programmed series of conformational changes that lead to membrane fusion [21–23]. Preventing these conformational changes at the attachment or entry step, or through blocking host cell proteins essential for the virus entry, can prevent virus fusion. In the wake of the first commercialized anti-HIV fusion inhibitor enfuvirtide (T-20) [24], a few peptide-based inhibitors have also been identified with great potential against influenza virus fusion.

Affinity chromatography used routinely in protein separation and purification can also be used for high-throughput screening. It has been used successfully to identify inhibitors and ligands against different targets from peptide libraries. A hexapeptide influenza A inhibitor (YRSKQA) was identified by this method using the first 11 N-terminal amino acid residues of HA2 (also known as a fusion peptide [FP]) as the ligand [25]. By using a positional scanning peptide library based on this lead compound, an improved peptide (GRGKHK) with a 3.3× higher affinity than the original lead compound was identified. The length of this improved peptide was further increased with scanning of extended peptide libraries using GRGKHK itself as the starting compound. The extended peptide library contained 18 different heptapeptide sequences with the first residue as any one of the 18 naturally occurring amino acids (excluding cysteine and tryptophan) and the other 6 amino acids fixed as GRGKHK. The library was screened using the same affinity chromatography method and the process was repeated until a peptide with a length equal to that of FP (1–11 residues) was identified. The anti-influenza activity of the resulting hendecapeptide was confirmed by its ability to prevent virus-induced cytopathic effect in Madin-Darby canine kidney cells and erythrocyte agglutination.

The θ-defensin, retrocyclin-2 (RC2), was shown to inhibit influenza virus infection by blocking membrane fusion mediated by the viral haemagglutinin (HA) protein [26]. A notable interesting point in this experiment is that RC2 was capable of inhibiting the viral fusion even after the HA protein attained a fusogenic conformation or had induced membrane hemifusion. Although this method was proven against influenza viruses, it is a broad-spectrum antiviral mechanism targeting fusion proteins based on the ability of certain endogenous lectins, such as RC2, to crosslink viral glycoproteins. Subsequently, the viruses are prevented from forming fusogenic structures with host cell receptors.

Another peptide-based inhibitor (entry blocker [EB] peptide) was active against influenza and blocked virus attachment to the host cell, probably by binding to the HA protein [27]. This peptide provided significant protection against numerous avian influenza virus subtypes including the highly pathogenic H5N1 viruses in vivo and in vitro (50% inhibitory concentration =4.5 μM), demonstrating a broad-spectrum antiviral activity. However, the post-infection treatment was less effective than the pre-infection treatment, possibly because of low dosage and or treatment schedule. The scrambled version of the peptide did not possess any anti-influenza activity, which shows that the antiviral activity of the EB peptide is sequence-specific. However, the detailed mechanism of action how the peptide prevents the virus attachment is still being investigated.

Although the surface glycoprotein HA has been the most preferred target for the peptide-based anti-influenza drug discovery, other proteins such as RNA-dependent RNA polymerase have also been studied. The polymerase enzyme consists of three subunits: PA, PB1 and PB2. In one study [28], a PB1-derived peptide was identified to inhibit the complex formation of these subunits. The N-terminal 1–25 amino acids of PB1 polymerase subunit of influenza A virus has previously been demonstrated to bind the PA subunit [29,30]. On the basis of these data, a short peptide (15 amino acids) was designed to specifically prevent the association of PB1 with PA. This peptide was ineffective against influenza B virus polymerase complex formation. This peptide was further improved as a chimeric peptide that actively inhibits both types of viruses [31]. This chimeric peptide was actually a derivative of the PB1 peptide of influenza A virus with the same amino acid sequence identified earlier [28] but with a single influenza B PB1-specific amino acid substitution, which recognizes PA of both virus types. Wunderlich et al. [31] used phylogenetic analysis and a novel ELISA-based screening approach to identify that particular amino acid.

Peptide-based viral inhibitors can also act on the host cells themselves. Certain peptide molecules modulate cellular pathways that are important for virus propagation. Human α-defensin peptides inhibit the influenza A virus replication [32]. Because α-defensins are some of the most potent inhibitors of the protein kinase C (PKC) pathway and PKC activation is essential for influenza virus replication, Salvatore et al. [32] hypothesized that α-defensins inhibit influenza virus replication by modulating PKC activation. This was confirmed by demonstrating that PKC activation did not occur in infected cells after the peptide treatment. The identification of such host cell modulating effectors could reveal the mechanisms behind the host-pathogen interaction, which in turn might lead to further novel antiviral targets.

Recent work by our group [33,34] identified a peptide (P1) from a phage display library that inhibits viral replication in vitro and in ovo. This peptide P1 inhibits binding of the HA to host cell receptors either in a linear or cyclic form as well as in a fusion phage form displaying this peptide, with a 50% inhibitory concentration of 71 μM, 48 μM and 5×10¹¹ plaque-forming units/100 ml, respectively. Among them, the cyclic peptide, perhaps due to conformation stabilization, was the most active form. To our knowledge, this is the only antiviral peptide <10 amino acids in length; however, as this peptide was tested only against the H9N2 subtype, its efficiency against other subtypes and related viruses has yet to be studied.

In another study, a pentadecapeptide (s2), identified from a phage display library against HA protein was shown to inhibit the replication of H1N1 and H3N2 subtypes of avian influenza virus [35]. This peptide seems to mimic the sialic acids, which are the natural receptors for the HA of influenza viruses. The peptide identified at the first cycle was further improved to increase the affinity towards the HA protein with secondary and tertiary selection from sub-libraries. Based on computational docking simulation, it was shown that the peptide, in particular the N-terminal five amino acids, binds with the receptor-binding site of HA. On the basis of a number of experiments, which showed that the peptide inhibited the interaction between the HA protein and sialic acid containing glycolipid GM3, it was concluded that these peptides bind with the target receptor-binding site of HA. The peptides showed higher affinity for H1N1 and H3N2 than anti-GM3 antibody and wheat germ agglutinin. Based on alanine scanning, the N-terminal five amino acids (R2, L3, R5, M7 and K11) were identified as essential for binding of HAs, that is, the HA binding motif. By fragmentation and subsequent inhibitory analysis, it has been shown that the N-terminal amino acids are very important in binding with the HA, which is consistent with alanine scanning.

Other than acting as the effectors in the viral inhibition, the peptides also serve as a backbone for sugar-based antiviral development [36]. In this study, a series of glycopolypeptides were developed by joining the glycans together with the polypeptide backbone through spacer molecules. The molecular weight of the backbone played a crucial role by enhancing the human influenza virus binding in a molecular-weight-dependent manner.

Conclusions and future perspectives

Many cationic antimicrobial peptides have been shown to possess a wide spectrum antimicrobial activity against various viral, bacterial and fungal pathogens. Peptides are being developed and studied against different targets ranging from the replication machinery to viral attachment proteins. Although only a few peptide-based virus inhibitors targeting fusion proteins have been studied so far, the much acclaimed, successfully commercialized HIV fusion inhibitor enfuvirtide (T-20) demonstrates the effectiveness of synthetic peptides as an antiviral agent and peptide fusion as an excellent target for antiviral drug development. New technologies are constantly being developed with notable successes solving certain difficulties associated with peptide therapeutics, offering a promising future for peptide-based drug development [37]. We face a constant threat from yearly influenza epidemics and pandemics, and so developing drugs to block influenza viral infection is one of the top public health priorities.