The human host defense peptide LL-37 interacts with Neisseria meningitidis capsular polysaccharides and inhibits inflammatory mediators release

Abstract

Capsular polysaccharides (CPS) are a major virulence factor in meningococcal infections and form the basis for serogroup designation and protective vaccines. Our work has identified meningococcal CPS as a pro-inflammatory ligand that functions through TLR2 and TLR4-MD2-dependent activation. We hypothesized that human cationic host defense peptides interact with CPS and influence its biologic activity. Accordingly, the interaction of meningococcal CPS with the human-derived cationic peptide LL-37, which is expressed by phagocytic and epithelial cells that interface with meningococci during infection, was investigated. LL-37 neutralized the pro-inflammatory activity of endotoxin-free CPS as assessed by TLR2 and TLR4-MD-2-dependent release of TNFα, IL-6 and IL-8 from human and murine macrophages. The cationic and hydrophobic properties of LL-37 were crucial for this inhibition, which was due to binding of LL-37 to CPS. LL-37 also inhibited the ability of meningococcal CPS to induce nitric oxide release, as well as TNFα and CXCL10 (IP-10) release from TLR4-sufficient and TLR4-deficient murine macrophages. Truncated LL-37 analogs, especially those that retained the antibacterial domain, inhibited vaccine grade CPS and meningococcal CPS prepared from the major serogroups (A, B C, Y and W135). Thus, LL-37 interaction with CPS was independent of specific glucan structure. We conclude that the capacity of meningococcal CPS to activate macrophages via TLR2 and TLR4-MD-2 can be inhibited by the human cationic host defense peptide LL-37 and propose that this impacts CPS-based vaccine responses.

Figure 1. LL-37 neutralized meningococcal CPS bioactivity and inhibited cytokine release from human and murine macrophages.

CPS polymers were purified from the endotoxin-deficient serogroup B meningococcal NMB-lpxA mutant designated CPS-lpxA. A: TNFα release from human macrophage-like THP-1 cells induced overnight with CPS-lpxA polymers pre-incubated with or without 2 µg/ml of LL-37 for 30 min at 37°C (Materials and Methods section). B. IL-6 release from THP-1 cells induced as in panel A. C. IL-8 release from HEK-TLR2/6 stably transfected cells induced with CPS-lpxA as in panel A. TNFα,IL-6 and IL-8 release was measured by ELISA. D. Nitric oxide release from murine RAW 264.7 macrophages induced with CPS-lpxA polymers as in panel A and measured by the Griess method as nitrite accumulation after 24 h of incubation at 37°C with 5% CO₂. Error bars represent ±SD from the mean of quadruplicate measurements. This experiment is representative of three independent experiments.

Figure 2. Cationic charge is important for LL-37 interaction with CPS polymers.

A: IL-8 release from HEK-TLR2/6 stably transfected cells induced with serogroup A meningococcal CPS polymers pre-incubated with 4 µg/ml of either LL-37 parent analog or the negatively charged LL-37R/D-K/E (LL-37-ve) analog used as a control. B: IL-8 release from HEK/TLR2/6 cells induced with a vaccine grade CPS purified from N. meningitidis serogroup A (MAPS) pre-incubated with or without 4 µg/ml of synthetic LL-37 parent analog. TLR2/6-ve is unstimulated HEK-TLR2/6 stably transfected cells. IL-8 measured by ELISA. Error bars represent ±SD from the mean of quadruplicate measurements. This experiment is representative of two independent experiments.

Figure 3. LL-37 active domain is required for interaction with meningococcal CPS polymers.

Acylation and charge of truncated LL-37 analogs enhanced interaction with meningococcal CPS polymers and inhibited TNFα(A) and IL-6 (B) release in a dose-dependent manner. THP-1 cells stimulated with meningococcal serogroup B, endotoxin-free CPS-lpxA doses ranging from 100 µg to 50 ng/ml with or without 10 µg/ml synthetic LL-37 analogs. LL-37-ve is negatively charged analog LL-37 K/E-R/D used as a control. TNFα and IL-6 release was quantified by ELISA. Error bars represent ±SD from the mean of quadruplicate readouts. The results are representative of 4 independent experiments.

Figure 4. LL-37 active domain is required for interaction with meningococcal CPS polymer and inhibited immune responses in murine macrophages.

A: Mouse TNFα release from murine RAW 264 macrophages induced with 20 µg /ml serogroup B meningococcal CPS-lpxA polymers pre-incubated with 2 µg/ml synthetic LL-37 analogs for 30 min at 37°C prior to cellular induction. B: Nitric oxide release from stimulated RAW 264.7 used in panel A was quantified by the Greiss method. Error bars represent the ±SD from the mean of 4 independent wells. The results are representative of 2 independent experiments.

Figure 5. LL-37 interaction with meningococcal CPS polymer is TLR4-independent.

Nitric oxide release from murine ScCr (TLR4-deficient) cells induced with with 20 µg /ml serogroup B meningococcal CPS-lpxA polymers pre-incubated with 2 µg/ml synthetic LL-37 analogs for 30 min at 37°C prior to cellular induction. PMX: polymyxin B; Pam3CSK4: synthetic lipopeptide and a TLR2 ligand used as a control at 2 µg/ml final concentration. Error bars represent the ±SD from the mean of 4 independent wells. The results are representative of 2 independent experiments.

Figure 6. Meningococcal CPS serogroups A, B, C, Y and W135 interact with LL-37 analogs.

Laboratory prepared meningococcal CPS polymers (20 µg/ml) from invasive meningococcal serogroups were pre-incubated with 4 µg/ml of LL-37 analogs and used to stimulate ScCr (TLR4-deficient) murine macrophages overnight. A- NMA CPS is meningococcal serogroup A CPS; B- NMB CPS is meningococcal serogroup B; C- NMC CPS is meningococcal serogroup C CPS; D- NMY CPS is meningococcal serogroup Y CPS and E- W135 CPS meningococcal serogroup W135 CPS. Nitric oxide release was measured as nitrite accumulation in the supernatants using the Greiss method. A non-cationic antibiotics ampicillin and erythromycin were used at 20 µg/ml as controls. Error bars represent the ±SD from the mean of 4 independent wells. This is representative of three independent experiments.

Figure 7. Synthetic LL-37 inhibited NFκB activation and TNFα release by meningococcal CPS-lpxA polymers.

A: Inducible NFκB dual luciferase reporter transfected into HEK-TLR2/6 stably transfected cells then stimulated with serogroup B meningococcal CPS-lpxA with or without LL-37 analogs. CPS-lpxA polymer (100 µg) was pre-incubated with 10 µg of LL-37, C12-LL17-32 or polymyxin B (PMX) prior to cellular induction for 5 h. Inducible NFκB firefly luciferase activity was normalized to constitutively expressing Renilla luciferase reporter. Negative control is transfected with non-inducible firefly luciferase reporter. Positive control is a mixture of constitutively expressing GFP and firefly luciferase construct. B: TNFα release from human THP-1 cells induced with CPS-lpxA-LL-37 analog prepared complexes used in panel A. TNFα protein was quantified by ELISA and error bars represent ±SD from the mean of 4 independent wells. This experiment is representative of three independent determinations.

Figure 8. Electrophoretic mobility gel shift assay of meningococcal CPS-lpxA polymers in the presence of LL-37, LBP and CD14.

CPS-lpxA complexes were prepared as described in the Methods section, were run on 8% native PAGE, and detected by Western blot using biotinylated LBP antibody. Lane 1 (MWt): molecular weight marker; lane 2 CPS-lpxA alone; lane 3: CPS-lpxA -CD14-LBP; lane 4: CD14-LBP alone; lane 5: CPS-lpxA -LL-37-CD14-LBP; lane 6: LL-37-CD14-LBP; lane 7: CPS-lpxA -LL-37 alone.