Cathelicidins Have Direct Antiviral Activity against Respiratory Syncytial Virus In Vitro and Protective Function In Vivo in Mice and Humans

Abstract

Respiratory syncytial virus (RSV) is a leading cause of respiratory tract infection in infants, causing significant morbidity and mortality. No vaccine or specific, effective treatment is currently available. A more complete understanding of the key components of effective host response to RSV and novel preventative and therapeutic interventions are urgently required. Cathelicidins are host defense peptides, expressed in the inflamed lung, with key microbicidal and modulatory roles in innate host defense against infection. In this article, we demonstrate that the human cathelicidin LL-37 mediates an antiviral effect on RSV by inducing direct damage to the viral envelope, disrupting viral particles and decreasing virus binding to, and infection of, human epithelial cells in vitro. In addition, exogenously applied LL-37 is protective against RSV-mediated disease in vivo, in a murine model of pulmonary RSV infection, demonstrating maximal efficacy when applied concomitantly with virus. Furthermore, endogenous murine cathelicidin, induced by infection, has a fundamental role in protection against disease in vivo postinfection with RSV. Finally, higher nasal levels of LL-37 are associated with protection in a healthy human adult RSV infection model. These data lead us to propose that cathelicidins are a key, nonredundant component of host defense against pulmonary infection with RSV, functioning as a first point of contact antiviral shield and having additional later-phase roles in minimizing the severity of disease outcome. Consequently, cathelicidins represent an inducible target for preventative strategies against RSV infection and may inform the design of novel therapeutic analogs for use in established infection.

Copyright © 2016 The Authors.

Figures

FIGURE 1.

LL-37 binds directly to RSV. LL-37 (A) and RSV (B) solid-phase ELISAs were performed by coating high-affinity binding plates with BSA, RSV, or cell culture medium (A) or BSA, LL-37, or scrLL-37 (B). Plates were incubated with LL-37 (A) or RSV (B), and were either exposed to anti–LL-37 or anti-RSV Ab, followed by HRP-labeled secondary Abs and detection with TMB. Data show mean absorbance at 450 nm from n = 3 independent experiments, each performed in triplicate, and analyzed by one-way ANOVA with Dunnett posttest (*p < 0.05, **p < 0.01, ***p < 0.001).

FIGURE 2.

Exposure to LL-37 reduces RSV N- and F-protein colocalization. RSV, treated with endotoxin-free water (as vehicle control; A), 25 μg/ml LL-37 (B), or 25 μg/ml scrLL-37 (C), was immediately air-dried on glass slides, fixed, and permeabilized, before incubation with anti-RSV F- and N-protein Abs, followed by labeled secondary Abs and assessment by confocal microscopy. (A–C) Images show representative regions (original magnification ×100) demonstrating F-protein (F AF488; green), N-protein (N AF594; red), overlay of channels, and completely colocalizing particles as determined by the IMARIS. The volume of fully colocalized signal quantified using IMARIS (D) and the number of double-positive particles counted using ImageJ (E) is displayed as a percentage, relative to untreated virus. Data represent mean ± SEM for n = 3 independent experiments, performed in duplicate, assessing at least 10 random fields per condition (100–500 particles per field of view), analyzed by one-way ANOVA with Dunnett posttest (*p < 0.05, **p < 0.01, ***p < 0.001).

FIGURE 3.

Exposure to LL-37 induces damage to RSV membrane integrity. Cryotransmission electron micrographs of control RSV (A) and RSV exposed to 25 μg/ml LL-37, followed by immediate vitrification (B). Micrographs were taken at a nominal original magnification ×50,000 and underwent visual inspection using ImageJ. Micrographs represent observations from at least two independent experiments, with a minimum of two grids analyzed per condition. Black arrowhead indicates intact membrane; white arrowhead indicates membrane damage; and black arrow indicates free capsid material.

FIGURE 4.

LL-37 exposure results in lower level of RSV binding to HEp-2 cells. HEp-2 cells were infected with RSV (MOI = 1) at 4°C (A and B) or 37°C (C), delivered concomitantly with 25 μg/ml LL-37, 25 μg/ml scrLL-37, or carrier control, for 1 (A and B) or 2 h (C). Cells were then either fixed, labeled with anti-RSV F-protein Ab and corresponding secondary Ab and analyzed by flow cytometry (A), protein collected and examined by Western immunoblot for F-protein and β-actin, as a loading control (B), or washed and incubated for a further 24 h before collection of total cell RNA for TaqMan-based Q-RT-PCR (C). (A) Mean fluorescence intensity (MFI) normalized to control infected cells for n = 3 independent experiments, performed in duplicate and analyzed by one-way ANOVA with Dunnett posttest (***p < 0.001). (B) Western blot for F-protein in total infected HEp-2 cell protein, representative of n = 3 independent experiments. (C) Transcription of IFN-β, IL-28, and IL-29 in RSV-infected cells treated concomitantly with LL-37, as a percentage of the matched RSV-infected cells treated concomitantly with scrLL-37, showing mean ± SEM for n = 3 independent experiments, performed in duplicate.

FIGURE 5.

LL-37 is protective against RSV infection in vivo. Six- to eight-week-old female BALB/c mice in groups of four littermates were housed in infected and uninfected sister pair cages and intranasally inoculated with 3–6 × 105 PFU RSV or PBS, concomitant with 10 μg LL-37 (given as 100 μl of 100 μg/ml solution) or scrLL-37 on day 0. On days 1–6, all mice received further inoculations of LL-37 or scrLL-37 intranasally, and weight was monitored and assessed as a percentage of starting weight. (A–D) Data represent n = 9 animals per group, performed in three independent experimental blocks, shown as mean ± SEM, analyzed by (A) general linear model ANOVA (***p < 0.001, *p < 0.05) and (B–D) Wilcoxon matched-pairs signed rank test (**p < 0.01, *p < 0.05). (E) Pulmonary RSV l-gene transcription was assessed at 4 and 7 d postinfection and expressed as fold change relative to the mean value for scrLL-37–treated infected mice at virus peak on day 4. Data show mean ± SEM from n = 6 mice/condition, analyzed by two-way ANOVA with Bonferroni posttest (***p < 0.0001). (F) Pulmonary Ifn-β expression was assessed on day 1 postinfection and expressed as fold change relative to the mean value for scrLL-37–treated uninfected. Data show mean ± SEM from n = 6 mice/condition, analyzed by two-way ANOVA with Bonferroni posttest (***p < 0.001).

FIGURE 6.

Prophylactic or therapeutic LL-37 administration in RSV infection. Six- to eight-week-old female BALB/c mice in groups of four cohoused littermates were intranasally inoculated on day 0 with 3–6 × 105 PFU RSV and 10 μg LL-37 (given as 100 μl of 100 μg/ml solution), applied intranasally either 1 h before, 1 h after, or concomitantly with infection, or with concomitant delivery of scrLL-37. On days 1–6, all mice received further inoculations of LL-37 or scrLL-37 as appropriate, and weight was monitored and assessed as a percentage of starting weight. (A) Data represent n = 8 animals per group, performed in three independent experimental blocks, shown as mean ± SEM, analyzed by one-way ANOVA with Dunnett posttests, *p < 0.05 (comparing concomitant LL-37 with control infected), ψp < 0.05 (comparing LL-37 pretreatment with control infected). (B) Pulmonary RSV L-gene transcription was assessed at 4 d postinfection and expressed as fold change relative to the mean value for scrLL-37–treated infected mice. Data show mean ± SEM from n = 4–6 mice/condition, analyzed by one-way ANOVA with Dunnett posttests (*p < 0.05).

FIGURE 7.

Murine cathelicidin is antiviral against RSV and induced by infection. (A) HEp-2 cells were infected with RSV administered concomitantly with a concentration range of mCRAMP (1, 10, 25, 50 μg/ml), 25 μg/ml LL-37, or carrier control for 2 h, washed, and incubated for 24 h in fresh medium. Immunoplaque assay was used to quantify the number of infected cells relative to the infected carrier control. Data show mean ± SEM for n = 3 independent experiments, performed in triplicate, analyzed by one-way ANOVA with Dunnett multiple comparison posttest (**p < 0.01, ***p < 0.001). (B–D) Wild type mice were intranasally inoculated with 3–6 × 105 PFU RSV or PBS, concomitant with 10 μg LL-37 (given as 100 μl of 100 μg/ml solution) or scrLL-37, and culled 24 h later. Uninfected Camp−/− mice were used as a negative control. Total protein and RNA were prepared from lung lobes and assessed for mCRAMP and β-actin (as a loading control) by Western immunoblot (B) quantified by Li-Cor Odyssey (C and D), and assessed for Camp expression by TaqMan-based Q-RT-PCR (E). Data show mean ± SEM expression levels, relative to the mean level in scr-LL-37–treated uninfected controls, from n = 6 per group, analyzed by two-way ANOVA with Bonferroni posttest (*p < 0.05, ***p < 0.0001). n.d., not detected.

FIGURE 8.

Endogenous mCRAMP is protective against RSV infection in vivo. Eight- to twelve-week-old female littermate pairs of Camp−/− or wild-type (wt) C57BL/6J OlaHsd mice were split between infected and matched uninfected cages, cohousing both genotypes. Mice were infected with 3–6 × 105 PFU RSV. Weight was monitored and assessed as a percentage of starting weight. (A–D) Data represent n = 6 animals per group, performed in three independent experimental blocks, shown as mean ± SEM, analyzed by Wilcoxon matched-pairs signed rank test, comparing genotypes (*p < 0.01). (E) Light micrographic images of pulmonary histology of H&E-stained lungs collected at day 6 postinfection, shown at original magnification ×100, representative of n = 3 per genotype. (F) Pulmonary RSV L-gene transcription was assessed at 3 d postinfection and expressed as fold change relative to the mean value for wt infected mice. Data show mean ± SEM from n = 8 mice per condition, analyzed by Wilcoxon matched-pairs signed rank test.

FIGURE 9.

Higher nasal LL-37 levels are associated with protection in an experimental human RSV infection model. Nasal lavage (A), bronchoalveolar lavage fluid (B), and plasma (C) were collected from a cohort of healthy, nonsmoking adults (age 18–33 y), before intranasal inoculation with 104 PFU RSV A Memphis 37. Individuals were monitored and categorized as RSV infected or RSV uninfected based on detection of nasal RSV by Q-RT-PCR on at least 2 d between days 2 and 10 postinfection. Nasal and plasma samples from 10 uninfected and 9 infected individuals, and bronchoalveolar lavage fluid samples from 7 uninfected and 5 infected individuals were quantified for LL-37/hCAP-18 levels by ELISA, in duplicate, in blinded fashion. Results are displayed as mean ± SEM and analyzed by t test, *p < 0.05.