6 Human Defensins in HIV Infection : Friends or Foes ?

Defensins are antimicrobial peptides (AMPs) and play an important role in both innate and adaptive immune response (Ganz, 2002; Yang et al., 2004). Defensins display broad antimicrobial activities against bacteria, fungi and viruses (Ganz, 2003; Lehrer, 2004; Zanetti, 2004),(Yang et al., 2004). Importantly, they also play a role in inflammation, tissue repair and angiogenesis (Kruse and Kristensen, 2008; Rehaume and Hancock, 2008). Increasing evidence indicates that the AMPs can act as a double-edged sword by providing protection against invading pathogens but at the same time causing potentially harmful inflammation or facilitating pathogen invasion. This review focuses on the role of human defensins in HIV infection. We will summarize the complex mechanisms by which defensins inhibit or enhance HIV infection in vitro, clinical evidence and studies in macaques with respect to the role of defensins in HIV transmission and pathogenesis.

detected in various epithelial tissues (Fellermann and Stange, 2001;Garcia et al., 2001;Harder et al., 2001).Both human and -defensins have been found in breast milk (Armogida et al., 2004;Jia et al., 2001), suggesting a role for defensins in protecting infants from infection.The mechanisms of induction of HBD1, HBD2 and HBD3 have been shown to be distinct from each other (Pazgier et al., 2006).HBD2 can be induced by TLR2, TLR3, TLR4, TLR7, NOD1 and NOD2 signaling in various epithelial cells and keratinocytes (Hertz et al., 2003;Nagy et al., 2005;Pivarcsi et al., 2005;Uehara et al., 2007;Vora et al., 2004).Stimulation of TLR3 has been shown to induce HBD1 and HBD2 expression in uterine epithelial cells (Schaefer et al., 2005).Induction of HBD2 and HBD3 but not HBD1 in bronchial epithelial cells in response to human rhinovirus infection is mediated by activation of nuclear factor-B (NF-B) but not of IL-1 (Proud et al., 2004).As TLR3 activation also induced HBD2 and HBD3, it is possible that intracellular double-stranded RNA generated during replication of rhinovirus may be involved in the regulation of HBDs (Duits et al., 2003;Proud et al., 2004).Similarly, HBD2 and HBD3 are induced in normal human oral epithelium cells, even in the absence of HIV-1 replication (Quinones-Mateu et al., 2003).Interestingly, a recent study reported that both X4-and R5-tropic viruses cannot induce HBD2 gene expression in the MatTeck oral tissue model nor primary gingival epithelial cells (Nittayananta et al., 2009).Additionally, high concentrations of X4 virus HIV-1 Lai block HBD2 gene expression by 50% (Nittayananta et al., 2009).In oral epithelium, TLR2 and NOD1/2 ligands synergistically activate NF-B and induce HBD2 gene expression (Uehara and Takada, 2008).Cytokines such as IL-1 and IL-17 also play important roles in the regulation of HBD2 expression.Induction of HBD2 by IL-17A is mediated by PI3K pathway and MAPK pathway to activate NF-B in airway epithelial cells, whereas regulation of HBD2 by the activation of NF-B is not dependent on PI3K pathway in bronchial epithelial cell, (Huang et al., 2007;Jang et al., 2007;Kao et al., 2008), indicating that specific pathways involved in regulation of HBDs are cell type dependent.

Effect of defensins on HIV infection in vitro: Mechanism(s) of action
In contrary to the traditional role of defensins to defense host against pathogens, recent studies indicate that specific defensins can inhibit or enhance HIV infection.With respect to anti-HIV activities of defensins, these peptides have a dual role in antiviral activity.One aspect of antiviral activities involves direct interaction with viral envelopes possibly by disrupting virus envelopes similar to their antibacterial activity or by preventing viral entry.However, in contrast to anti-bacterial activities of defensins, there is no direct evidence supporting that defensins directly inactivate HIV virion by membrane disruption.The other involves indirect antiviral activity through interactions with potential target cells.These defensin-cell interactions are complex and at least in part mediated by interacting with cell surface glycoproteins and/or interfering with cell-signaling pathways that are required for viral replication.HD5 and HD6, induced in cervicovaginal epithelial cells in response to GC infection, enhance HIV infectivity (Klotman et al., 2008).The enhancing effect of HD5 and HD6 was more pronounced with R5 virus compared with X4 virus, indicating a potential clinical relevance as R5 virus is preferentially transmitted during primary infection.The specific mechanism is discussed as the following and activities of defensins and other antimicrobial peptides on HIV replication is summarized in Table 1.The in vitro functions of defensins appear to be affected by factors such as the source of defensins, serum and salt.Different antiviral mechanisms of defensins may be operative in mucosal surfaces versus blood depending on the salt concentration or the presence of serum.This appears to be the case with the direct antiviral effect.Serum and salt conditions did alter the direct effect of defensins on the virion (Daher et al., 1986) , (Chang et al., 2005;Quinones-Mateu et al., 2003).Some defensins (e.g.HNPs but not HD5 or HD6) at high concentrations are known to cause cytotoxicity in the absence of serum, which is associated with changes in cell membrane permeability, similar to their anti-bacterial activity.This cytotoxicity can be abolished by the presence of serum (Okrent et al., 1990 ;Van Wetering et al., 1997) and defensin-mediated cytotoxicity may partially account for the antiviral effect (Mackewicz et al., 2003).While most defensins display potent direct antibacterial activities in conditions of low salt (Lehrer et al., 1993), neither a low concentration of salt nor the absence of serum are required for the chemotactic effects of defensins (Chertov et al., 1996;Yang et al., 1999).It is not clear whether other functions of defensins are altered depending on the environment.

Table 1. Effect of defensins on HIV infection
Inhibition of HIV replication by synthetic guinea-pig, rabbit and rat -defensins was first reported in 1993 (Nakashima et al., 1993), when it was shown that these peptides could inhibit HIV-1 infection in vitro following viral entry into transformed CD4 + T cells in the presence of serum (Nakashima et al., 1993).HNPs1-3 block HIV infection through multiple mechanisms (Furci et al., 2007;Zhang et al., 2002) (Chang et al., 2003;Mackewicz et al., 2003).HNPs1 -3 all have similar activities against HIV primary isolates (Wu et al., 2005), in contrast to their differential chemotactic activities on monocytes, where HNP3 has no effect (Territo et al., 1989).They can inhibit HIV-1 replication by a direct interaction with the virus as well by affecting multiple steps of HIV life cycle (Chang et al., 2003;Chang et al., 2005;Furci et al., 2007;Mackewicz et al., 2003;Wang et al., 2004).In the absence of serum, HNP1 has a direct effect on the virus prior to infection of a cell (Chang et al., 2005).In the presence of serum and at non-cytotoxic concentrations (low dose), HNP1 acts on primary CD4 + T cells and blocks HIV-1 infection at the steps of nuclear import and transcription by interfering with PKC signaling (Chang et al., 2005).The post-entry inhibitory effect of HIV infection occurs in primary CD4 + T cells and macrophages but not in several transformed T-cell lines (Chang et al., 2003;Chang et al., 2005).In the presence of serum, HNP1 did not affect expression of cellsurface CD4 and HIV-coreceptors on primary CD4+ T cells (Chang et al., 2005), whereas HNP2 down-regulates CD4 expression in the absence of serum (Furci et al., 2007).HNPs block HIVmediated cell-cell fusion and the early steps of HIV infection by interacting with HIVgp120 and CD4 through their lectin-like properties (Furci et al., 2007).In macrophages, HNP1 and HNP2 upregulate the expression of CC-chemokines, which could contribute to inhibition of HIV through competition for receptors (Guo et al., 2004).CC-chemokines can also induce the release of HNPs from neutrophils by degranulation (Jan et al., 2006).Both effects could play a role in vivo in an innate immune response to HIV.At the mucosal surface, HNPs might work to directly inactivate the virions in the absence of serum; however, in the presence of serum, their inhibitory effect would largely be on the infected cell.HNPs1-3 have been reported to act as lectins and bind to HIV envelope glycoprotein gp120 and to CD4 with high affinity (Wang et al., 2004).The binding to gp120 is strongly attenuated by serum, thus accounting for the loss of the direct virion effect in the presence of serum.Interestingly, in contrast to HNPs1-3, HNP4 acts in a lectin-independent manner and does not bind to CD4 or HIV gp120 (Wang et al., 2004;Wu et al., 2005).However, HNP4 inhibits HIV replication more effectively than HNP1, -2 and -3 (Wu et al., 2005).Other -defensins, including HD5 and HD6, mouse paneth cell cryptdin-3 and cryptdin-4, and rhesus macaque myeloid -defensin-3 (RMAD3) and RMAD4 have been tested for their ability to block HIV infection (Klotman et al., 2008;Tanabe et al., 2004).While HD5 did not exhibit any effect on X4 HIV-1 LAI infection of transformed CD4+ T cell lines (Tanabe et al., 2004), HD5 and HD6 significantly enhanced infectivity of HIV-1 R5 strains (Klotman et al., 2008).At high concentrations associated with cytotoxicity, RMAD4 blocks HIV replication, whereas, cryptdin-3 enhances viral replication.Studies on the molecular mechanism of the HIV enhancing effect of HD5 and HD6 indicate that defensins enhance HIV infection through promoting HIV attachment (Rapista et al., 2011).In addition, HD5 but not HD6 competes with heparan for binding to HIV.Importantly, these defensins have been shown to block in vitro anti-HIV activity of polyanionic microbicides, which have failed to protect women against HIV infection, and to interfere with anti-HIV activity of HIV entry and fusion inhibitors under specific conditions (Ding et al., 2011;Rapista et al., 2011).The anti-HIV activities of HBD2 and HBD3 have been demonstrated under different conditions (Quinones-Mateu et al., 2003;Sun et al., 2005).Similar to HNP1 (Chang et al., 2005), HBD2 and HBD3 have dual anti-HIV activities through direct interactions with the virus and by altering the target cell.The binding of defensins to cellular membranes and HIV virions has been demonstrated by electron microscopy, although membrane disruption is not apparent (Quinones-Mateu et al., 2003).HBD2 does not affect viral fusion but inhibits the formation of early reverse transcribed HIV DNA products (Sun et al., 2005).There are conflicting reports on the downregulation of expression of HIV co-receptors by -defensins.In studies reported by Sun et al. (Sun et al., 2005), HBD1 and HBD2 did not modulate cellsurface HIV co-receptor expression by primary CD4+ T cells, whereas Quinones-Mateu et al. (Quinones-Mateu et al., 2003) showed HBD2-and HBD3-mediated downregulation of surface CXCR4 but not CCR5 expression by peripheral blood mononuclear cells (PBMCs) at high salt conditions and in the absence of serum.Interestingly, HBD2 is constitutively expressed in healthy adult oral mucosa but the level seems to be diminished in HIV-infected individuals (Sun et al., 2005).Retrocyclins, and RTD1, -2 and -3 act as lectins and can inhibit HIV entry (Cole et al., 2002;Munk et al., 2003;Wang et al., 2003;Wang et al., 2004).Retrocyclin and RTD1,-2 and -3 inhibit several HIV-1 X4 and R5 viruses including primary isolates (Munk et al., 2003;Wang et al., 2003;Wang et al., 2004).Unlike and -defensins, retrocyclin does not appear to directly inactivate the HIV virion although it is not clear whether the experiments reported to date were performed under serum-free condition (Cole et al., 2002).Retrocyclin does however bind to HIV gp120 as well as CD4 with high affinity, which is consistent with inhibition of viral entry (Cole et al., 2002) , (Munk et al., 2003).This high-binding affinity to glycosylated gp120 and CD4 is mediated through interactions with their O-linked and Nlinked sugars (Wang et al., 1998).Serum strongly reduces their binding to gp120 (Wang et al., 2004).RTD1 binds directly to the C-terminal heptad repeat of HIV envelope protein gp41, blocking formation of the six helix bundle required for fusion (Gallo et al., 2006).Recently, high concentrations of RTD1 and HNP1 have been shown to down-regulate CXCR4 in PBMCs in the absence of serum (Nittayananta et al., 2009).Studies on retrocyclin-1 analogues indicate that modification of this peptide can enhance its potency against HIV in vitro (Owen et al., 2004), suggesting their potential use as a microbicide.

Role of Defensins in HIV pathogenesis and transmission
Depending on the preparation of samples and analytical methods, the levels of defensins can be varied from one report to another.In addition, defensins have been found to interact with other cellular proteins in plasma (Higazi et al., 1996;Panyutich and Ganz, 1991;Panyutich et al., 1994), which may affect the measurement of defenisn levels by ELISA.In healthy donors, the plasma concentration of HNPs1-3 is ranging from ~150-500 ng/ml (Mukae et al., 2002).The levels of defensins in the plasma or at the mucosa are frequently elevated in patients with infections or diseases (Coffelt and Scandurro, 2008;Ihi et al., 1997).For examples, defensin levels in plasma from patients with sepsis reach 900-170,000 ng/ml (Panyutich et al., 1993).Using liquid chromatography-tandem mass spectrometry, the levels of HNPs in the saliva from healthy donors range from 1 to 10 ug/ml, whereas the level of HBDs 1-2 range from undetectable to 33 ng/ml (Gardner et al., 2009).The level of HNPs in cervicovaginal fluid from healthy women ranges from 250 ng/ml to 5 g/ml depending on the laboratories (Levinson et al., 2009;Simhan et al., 2007).HNPs1-3 The role of HNPs in HIV pathogenesis was first suggested that HNPs 1-3 were to account for the soluble anti-HIV activity of CD8 + T cells (CAF) isolated from patients infected with HIV but remaining free of AIDS for a prolonged period (long-term nonprogressors, LTNPs) (Zhang et al., 2002).These peptides were detected in the media of stimulated CD8 + T cells from normal healthy controls and LTNPs but not from HIV progressors.Subsequent studies on the cell source of defensins revealed that HNPs were probably produced by co-cultured monocytes and residual granulocytes of allogenic normal donor irradiated PBMCs that were used as feeder cells, but they were not produced by the CD8 + T cells themselves (Mackewicz et al., 2003;Zaharatos et al., 2004).Using similar coculture systems, levels of HNPs1-3 were measured in CD8 + T-cell supernatants and cervicalvaginal mononuclear cells derived from HIV-exposed seronegative individuals, HIV-infected patients, and normal controls (Trabattoni et al., 2004).Higher levels of HNPs were found in CD8+ T cells from HIV-exposed seronegative individuals and HIV patients compared to normal controls.D'Agostino et al. recently found that HIV-infected patients have a higher level of HNPs in plasma than healthy donors (D'Agostino et al., 2009).Using a co-culture system with radiated PBMCs, higher levels of HNPs in CD8+ T cells were found in patients with HIV infection compared to the healthy donors, and the intracellular HNP levels were further increased in stimulated CD8+ T cells.The intracellular level of HNPs in neutrophils is higher in HIV-infected patients than healthy donors.There is no significant difference in the plasma level of HNPs in HIV-infected patients with or without antiviral treatment (ART).However, reduction of HNPs in CD8+ T cell was found in HIV-infected patients on ART.Interestingly, this reduction in the HNP level was not found in HIV-infected patients on ART with virologic failure.In contrast to the report by D'Agostino et al., Rodriguez-Garcia et al. did not observe any association between plasma levels of HNPs and immunologic or virolgoic parameters (Rodriguez- Garcia et al., 2010).This report also described an increase in HNPs1-3 in dendritic cells, differentiated in vitro, in HIV controllers but not noncontrollers compared to healthy controls.While it was suggested that increased HNPs1-3 production by dendritic cells in HIV-infected patients is associated with slower disease progression, analysis of specific immune cell subsets without further manipulation is needed to clarify the role of HNPs in HIV disease progression.The association between production of HNPs1-3 in breast milk and transmission of HIV has also been investigated (Kuhn et al., 2005).In a case-controlled study of HIV-positive women, levels of HNPs in breast milk correlated with HIV RNA copy number in breast milk, which was a strong predictor of transmission.However, after adjusting for breast milk HIV copy number, higher levels of HNPs in breast milk were associated with a decreased incidence of intrapartum or postnatal HIV transmission.Bosire and colleagues performed similar studies to determine the correlation between the level of HNPs in breast milk and transmission risk in a cohort of 260 HIV-1-infected pregnant women in Nairobi followed for 12 months postpartum with their infants (Bosire et al., 2007).Analysis of breast milk from these women at month 1 postpartum demonstrated that women with detectable alpha-defensins and significantly higher mean breast milk HIV-1 RNA levels than women with undetectable alpha-defensins.Increased alpha-defensins concentrations in breast milk were also associated with subclinical mastitis and increased CC-chemokines in breast milk.Interestingly, in contrast to the report by Kuhn et al. (Kuhn et al., 2005), the level of defensins are not associated with vertical transmission, indicating a complex interplay between innate effectors, inflammation and HIV transmission.There is a correlation between the abundance of several anti-HIV proteins, including HNPs1-3 and cell-associated HIV replication in lymphoid follicles compared with extrafollicular lymphoid tissue (Folkvord et al., 2005).Expression of these antiviral proteins is significantly lower in the follicular region, where HIV replication is concentrated, compared with the extrafollicular regions in lymph nodes from HIV-positive individuals.Cationic peptides including defensins are required for anti-HIV activity of vaginal fluid from healthy women (Venkataraman et al., 2005).While it is well established that sexual transmitted infections (STIs) significantly increase the likelihood of HIV transmission (Chesson and Pinkerton, 2000;Cohen et al., 1997;Galvin and Cohen, 2004;Mabey, 2000;Plummer, 1998) and that levels of defensins including HNPs, HBDs and HD5 in genital fluid, are elevated in patients with STIs (Porter et al., 2005;Simhan et al., 2007;Valore et al., 2006;Wiesenfeld et al., 2002), the role of defensins in HIV transmission seems to be complex.Studies using a cohort of HIV uninfected sex workers in Kenyan demonstrated the association between an increase in HNPs and LL-37 levels in the IgA-depleted cervicovaginal secretions from women with bacterial STIs and increase in HIV acquisition, despite that cervicovaginal secretions with high levels of HNPs and LL-37 exhibited anti-HIV activity in vitro (Levinson et al., 2009).This study underscores the complex role of defensins in HIV transmission at the vaginal mucosa and the urgent need to define the effect of elevated innate effectors on immune responses that contribute to enhanced HIV acquisition.HBDs Significant correlations between the single-nucleotide polymorphism (SNPs) -44C/G and -20G/A in 5' untranslated region of DEFB1 (coding for HBD1) and a risk of perinatal transmission of HIV-1 in Italian and Brazilian populations, respectively (Braida et al., 2004;Milanese et al., 2006).The SNP -52G/G genotype is associated with reduced HIV-1 RNA in breast milk, but not in plasma in Mozabican HIV-infected women (Baroncelli et al., 2008).Interestingly, the functional analysis of promoter indicates that these SNPs suppress expression (Milanese et al., 2007).Studies on the role of HBD1 in mother-to-child transmission of HIV indicated that the -52G/G genotype and the -44/-52G haplotype exhibited a protective role against HIV infection in children, whereas the -52G/G genotype and the -44G/-52G haplotype were associated with low levels of HIV plasma viremia and a lower risk of maternal HIV transmission in mothers (Ricci et al., 2009).Although HBD1 does not exhibit any effect on HIV infection in vitro, the presence of SNP may affect HIV transmission by modulating immune response.The role of defensins in protection against HIV infection has been studied in HIV-exposed seronegative (ESN) individuals.ESN expressed significantly greater mRNA copy numbers of HBD2 and 3 in oral mucosa than healthy controls, while no difference in mRNA copy numbers of HBD-1, 2 and 3 in vaginal and endocervical mucosa was observed between ESN and controls (Zapata et al., 2008).In addition, homozygosity for the A692G polymorphism is significantly more frequent in ESN than in seropositive individuals (Zapata et al., 2008).Sequence analysis of -defensin pseudogenes in ESN female sex-workers from Thailand revealed that all subjects had premature stop codons (Yang et al., 2005).Therefore, restoration of endogenous -defensin production does not account for the resistance to HIV-1 infection in these women.

Conclusion
Defensins play an important role in innate immune response.These peptides display versatile functions in modulating various immunological and biological aspects.Aberrant defensin expression has been associated with many human diseases (de Leeuw and Lu, 2007), although studies on the role of defensins in HIV pathogenesis and transmission in humans just began to reveal the complex functions of defensins in modulating HIV infection.While the innate immune system is evolutionarily conserved among multicellular organisms, it is challenging to find a suitable animal model to study the role of defensins in HIV pathogenesis and transmission due to complex diversity of defensins in mammals as well as apparent differences in mechanisms of action.Recently, increased expression of rhesus enteric a-defensins (REDs) in response to SIV infection was reported (Zaragoza et al., 2011).Additionally, decreased RED protein levels correlate with enteric opportunistic infection and advanced SIV disease.However, the primary sequences of RED and HD5 differ and it is not clear whether REDs could represent HD5.Future studies focusing on the development of a better animal model for studying innate immunity in HIV transmission and pathogenesis as well as careful assessments of immune responses in patients with reduced or elevated levels of defensins will shed light on the development of better strategies for HIV therapeutics.