Haemophilus parainfluenzae in bronchiectasis

Abstract

INTRODUCTION: Bronchiectasis is a chronic airway condition with permanently dilated airways, dysfunctional cilia and overproduction of mucus as main clinical symptoms, and in bronchiectasis there is a continuous cycle involving bacterial infection, host immune response and airway damage (Cole, 1986). More than 70% of bronchiectasis patients are chronically colonised with potential pathogens (Angrill et al., 2002). The Haemophilus species is the most commonly found bacteria in bronchiectasis (Chalmers et al., 2012), with Haemophilus influenzae (H. influenzae) being considered a pathogen and Haemophilus parainfluenzae (H. parainfluenzae) a commensal bacteria. However, H. parainfluenzae have been found more and more frequently associated with lower airway diseases, with further data suggesting it can also play a pathogenic role. Up to date, there is limited research about the role of H. parainfluenzae in bronchiectasis. This study will investigate the potential pathogenicity of H. parainfluenzae in bronchiectasis on its own and when combined with Pseudomonas aeruginosa (P. aeruginosa), the second most frequent organism isolated in bronchiectasis. We hypothesised that: H. parainfluenzae are common in bronchiectasis, and in high bacterial loads, it can play a similar pathogenic role as H. influenzae; H. parainfluenzae can induce a stronger inflammatory response when combined with P. aeruginosa.

MAIN RESULTS: 1. H. parainfluenzae are commonly found in bronchiectasis patient’s lower airways, but it is seldom reported Of the 140 patient sputum samples that were previously found to be infected with Haemophilus species, 40% of the patients presented H. parainfluenzae and 47% H. influenzae. Another 51 were unselected bronchiectasis patients’ sputum samples were processed both in our laboratory and an NHS microbiology laboratory. 18% of these patients presented H. parainfluenzae in their samples according to our laboratory, but none according to the NHS laboratory. 2. H. parainfluenzae have similar clinical relevance as H. influenzae in patients with clinically stable bronchiectasis.

From those previously mentioned 140 bronchiectasis patients, 63 had their serum and sputum inflammatory markers compared. 24 of them had H. parainfluenzae and 39 H. influenzae. There was a similar bronchiectasis disease severity index (BSI) score between these two groups. There was similar systemic inflammation, with no significant differences in white cell counts (WCC), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), intercellular adhesion molecule 1 (ICAM-1) or Myeloperoxidase (MPO). These results suggest that H. parainfluenzae and H. influenzae have similar clinical relevance in bronchiectasis. 3. H. parainfluenzae in high bacterial loads can stimulate significant inflammatory responses from epithelial cells H. parainfluenzae (at a concentration of 107 cfu/mL) can stimulate a range of inflammatory responses comparable to those of H. influenzae (107 cfu/mL) and P. aeruginosa (107 cfu/mL) on 16HBE cells, NHBE cells and bronchiectasis patients' nasal epithelial cells.

3.1 Inflammatory responses from human bronchial epithelial cell line (16HBE) H. parainfluenzae (107 cfu/mL) can trigger significant interleukin 8 (CXCL8, p=0.001); chemokine ligand 1 (CXCL1, p=0.004); ICAM-1(p=0.004) and lipocalin 2 (LCN2, p=0.04) production from 16HBE cells compared to the control group after 8 hours. There was no significantly difference between 107 cfu/ml H. parainfluenzae, H. influenzae and P. aeruginosa.

3.2 Inflammatory responses from normal human bronchial epithelial (NHBE) cells and bronchiectasis patients’ nasal epithelial cells H. parainfluenzae (107 cfu/mL) can trigger significantly more CXCL8, CXCL1 and ICAM-1 production in NHBE cells (p-values are 0.02, 0.033 and 0.02 respectively) and patients’ nasal epithelial cells (p-values are 0.038, 0.0005 and 0.0001 respectively) compared to the control group. 4. Poly-bacterial infection reduces 16HBE cells’ inflammatory response compared to single bacterial infection When 16HBE cells were co-cultured with H. parainfluenzae (105 cfu/mL) and P. aeruginosa (105 cfu/mL) for 8 hours, there was a significant decrease in CXCL8, CXCL1 and LCN2 production by 16HBE cells compared to cells that were cultured with only P. aeruginosa (105 cfu/mL), p-values are 0.0007, 0.0032 and 0.04 respectively.

5. Indirect interaction between H. parainfluenzae and P. aeruginosa may lead to the change observed in inflammatory response from previous result When 16HBE cells were co-cultured with heat-treated H. parainfluenzae (105 cfu/mL) and P. aeruginosa (105 cfu/mL) for 8 hours, there was no significant decrease in the co-infection group (CXCL8 p=0.16, CXCL1 p=0.2), which suggests that the previously observed reductive effect requires living bacteria. Therefore, this reduction is likely to be due to bacterial interaction between H. parainfluenzae and P. aeruginosa.

When 16HBE cells and H. parainfluenzae (105 cfu/mL) (n=6) were cultured in filtered media that had cultured P. aeruginosa (n=6) for 8 hours, there was a significant decrease of CXCL8 and CXCL1 levels compared to cells cultured with H. parainfluenzae in media that had not cultured P. aeruginosa (CXCL8 p=0.002, CXCL1 p=0.027. This suggests that P. aeruginosa induced media conditions may play a role in this observed inhibition effect. Later experiments revealed that P. aeruginosa can promote the growth of H. parainfluenzae.

CONCLUSION: H. parainfluenzae are commonly found in bronchiectasis patient’s lower airways, and it is under-reported in bronchiectasis. Patients with H. parainfluenzae in their lower airways have similar sputum and serum inflammation compared to patients with H. influenzae. H. parainfluenzae can stimulate significant inflammatory responses from epithelial cells.

Poly-microbial (H. parainfluenzae with P. aeruginosa) infection reduces 16HBE cell inflammatory response (CXCL8, CXCL1 and LCN2) compared to single bacterial infection. The media cultured with P. aeruginosa can reduce cellular inflammatory response when co-cultured with H. parainfluenzae. The underlying mechanism underpinning this merit further study.