Respiratory tract antimicrobial peptides more effectively killed multiple methicillin-resistant Staphylococcus aureus and nontypeable Haemophilus influenzae isolates after disruption from biofilm residence

Abstract

Bacteria newly released (NRel) from biofilm residence via multiple methodologies are commonly significantly more sensitive to antibiotics. We've induced NRel with this phenotype after incubation of biofilms formed by diverse human pathogens with an epitope-targeted monoclonal antibody directed at protective domains within bacterial DNABII proteins that provide structural support to the eDNA-dependent biofilm matrix. The observed heightened sensitivity was due, in part, to increased NRel membrane permeability. In three animal models of human biofilm-mediated infections, this monoclonal induced biofilm disruption with rapid concomitant bacterial clearance and disease resolution in the absence of any co-delivered antibiotic, which suggested a key role of innate immune effectors. Recently, we showed that NRel of the respiratory pathogen nontypeable Haemophilus influenzae (NTHI), as mediated by the DNABII-directed monoclonal, are also highly vulnerable to killing by human polymorphonuclear neutrophils (PMNs). Here, we extended these observations to show that the transient, yet highly vulnerable anti-DNABII NRel phenotype of three isolates of both NTHI and methicillin-resistant Staphylococcus aureus (MRSA) included significant sensitivity to killing by three antimicrobial peptides commonly expressed within the respiratory tract or by PMNs (e.g., human β-defensins 1 and 3 as well as the cathelicidin, LL-37). We envision induction of the NRel phenotype by delivery of this monoclonal antibody to patients with recalcitrant biofilm-mediated diseases to provide greatly improved medical management. Ideally, clearance of NRel will be mediated by innate immune effectors of an immunocompetent host or, if needed, by co-delivered traditional antibiotics, which are canonically ineffective against biofilm-resident bacteria but would be highly effective against NRel.

Importance: Pathogenesis of most common chronic and/or recurrent bacterial diseases (e.g., middle ear infections, urinary tract infections, rhinosinusitis, among others) can be attributed to biofilms that are canonically highly resistant to both immune effectors and antibiotics. If we treat biofilms formed by diverse human pathogens with a targeted monoclonal antibody directed at protective domains of bacterial DNA-binding proteins integral to the structural stability of the eDNA-rich biofilm matrix, they are rapidly disrupted with concomitant release of the resident bacteria. These newly released (NRel) bacteria are transiently significantly more sensitive to killing by both traditional antibiotics and human PMNs, and herein, we showed that they are also more readily killed by antimicrobial peptides. Clinically, we hope to leverage this understanding of the NRel phenotype for better medical management of these challenging infections, as well as perhaps even limit or eliminate further contribution to the global antimicrobial resistance 'pandemic'.

Keywords: DNABII; LL-37; MRSA; NRel; beta defensins; cathelicidin; nontypeable Haemophilus influenzae.

Fig 1

Relative killing of MRSA subpopulations by hBD-1, hBD-3, or LL-37. MRSA that was either biofilm-resident; grown planktonically to mid-log phase, or newly released from biofilm residence (NRel) by incubation with MsTipMab for 2 h, then incubated for 1.5 h with either (A, D, and G) 10 µg hBD-1; (B, E, and H) 5 µg hBD-3, or (C, F, and I) 0.3 µg LL-37 per mL of DIS. Relative killing of MRSA clinical isolates recovered from a person with cystic fibrosis (A, B, and C), from a bone abscess (SAS130, panels D, E, and F) and an invasive blood isolate (SAS135, panels G, H, andI). AMPs were used at physiologically relevant concentrations (16–18) and at a concentration pre-determined to restrict killing of planktonically grown MRSA to ≤20% to allow detection of any enhanced ability to kill NRel. Note significantly increased ability of all three tested AMPs to kill anti-DNABII MRSA NRel (P ≤ 0.0001).

Fig 2

Relative killing of NTHI isolates by hBD-1, hBD-3, or LL-37. AMPs were used at physiologically relevant concentrations (16–18) and at a concentration pre-determined to restrict killing of planktonically grown NTHI ≤20% to allow detection of any enhanced ability to kill NRel. NTHI #1128 that was either biofilm-resident, incubated planktonically to mid-log phase or been newly released from biofilm residence (NRel) by incubation with MsTipMab were then incubated for 1 h with either (A) 10 µg hBD-1, (B) 5 µg hBD-3, or (C) 0.3 µg LL-37. The same three populations of NTHI #1728 were treated similarly with either (D) 10 µg hBD-1, (E) 5 µg hBD-3, or (F) 0.3 µg LL-37. Finally, we assessed the same three populations of NTHI #86-028NP following treatment with either (G) 10 µg hBD-1, (H) 5 µg hBD-3, or (I) 0.3 µg LL-37. Note significantly increased ability of all three tested AMPs to kill all three anti-DNABII NTHI NRel populations (P ≤ 0.0001). To begin to investigate the role of surface charge in sensitivity to killing by LL-37, we tested an isogenic ∆licD mutant of NTHI strain #86-028NP both with and without the addition of exogenous ChoP (I). Note significantly greater killing of the ∆licD mutant by LL-37 than the planktonically grown parental isolate (P ≤ 0.001), which was significantly reduced (P ≤ 0.005) and also restored back to that of the parental isolate by the addition of exogenous ChoP. Interestingly, anti-DNABII NTHI NRel were significantly more sensitive to LL-37 killing than the ∆licD mutant (P ≤ 0.0005). Collectively, these data suggested a role of outer membrane charge, as well as likely other factors, in the ability of LL-37 to kill anti-DNABII NTHI NRel.