Portal de Periódicos da CAPES

Serum cathelicidin, nasopharyngeal microbiota, and disease severity among infants hospitalized with bronchiolitis

2016; Elsevier BV; Volume: 139; Issue: 4 Linguagem: Inglês

10.1016/j.jaci.2016.09.037

1097-6825

Kohei Hasegawa, Jonathan M. Mansbach, Nadim J. Ajami, Joseph F. Petrosino, Robert J. Freishtat, Stephen J. Teach, Pedro A. Piedra, Carlos A. Camargo,

Tracheal and airway disorders

Bronchiolitis is a common acute respiratory infection and the leading cause of hospitalizations in US infants.1Ralston S.L. Lieberthal A.S. Meissner H.C. Alverson B.K. Baley J.E. Gadomski A.M. et al.Clinical practice guideline: the diagnosis, management, and prevention of bronchiolitis.Pediatrics. 2014; 134: e1474-e1502Crossref PubMed Scopus (167) Google Scholar Although bronchiolitis has been considered virus-induced inflammation of small airways,1Ralston S.L. Lieberthal A.S. Meissner H.C. Alverson B.K. Baley J.E. Gadomski A.M. et al.Clinical practice guideline: the diagnosis, management, and prevention of bronchiolitis.Pediatrics. 2014; 134: e1474-e1502Crossref PubMed Scopus (167) Google Scholar emerging evidence indicates that the pathogenesis involves a complex interplay among viral agents, airway microbiota, and the innate immune system.2de Steenhuijsen Piters W.A. Heinonen S. Hasrat R. Bunsow E. Smith B. Suarez-Arrabal M.C. et al.Nasopharyngeal microbiota, host transcriptome and disease severity in children with respiratory syncytial virus infection.Am J Respir Crit Care Med. 2016; 194: 1104-1115Crossref PubMed Scopus (245) Google Scholar, 3Mansbach J.M. Hasegawa K. Henke D.M. Ajami N.J. Petrosino J.F. Shaw C.A. et al.Respiratory syncytial virus and rhinovirus severe bronchiolitis are associated with distinct nasopharyngeal microbiota.J Allergy Clin Immunol. 2016; 137: 1909-1913Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar, 4Hasegawa K. Mansbach J.M. Ajami N.J. Espinola J.A. Henke D.M. Petrosino J.F. et al.Association of nasopharyngeal microbiota profiles with bronchiolitis severity in infants hospitalized for bronchiolitis.Eur Respir J. 2016; 48: 1329-1339Crossref PubMed Google Scholar Among the multiple components of the innate immune system, cathelicidins are a family of host defense peptides with both direct microbicidal and immunomodulatory properties.5Hilchie A.L. Wuerth K. Hancock R.E. Immune modulation by multifaceted cationic host defense (antimicrobial) peptides.Nat Chem Biol. 2013; 9: 761-768Crossref PubMed Scopus (139) Google Scholar In previous studies, serum cathelicidin level was inversely associated with disease severity in children with bronchiolitis,6Mansbach J.M. Piedra P.A. Borregaard N. Martineau A.R. Neuman M.I. Espinola J.A. et al.Serum cathelicidin level is associated with viral etiology and severity of bronchiolitis.J Allergy Clin Immunol. 2012; 130: 1007-1008.e1Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar and administration of LL-37 (the main active form of human cathelicidin) altered the microbiota in animal models.7Pound L.D. Patrick C. Eberhard C.E. Mottawea W. Wang G.S. Abujamel T. et al.Cathelicidin antimicrobial peptide: a novel regulator of islet function, islet regeneration, and selected gut bacteria.Diabetes. 2015; 64: 4135-4147Crossref PubMed Scopus (41) Google Scholar We have previously demonstrated that, in a cohort of infants with severe bronchiolitis (bronchiolitis requiring hospitalization),4Hasegawa K. Mansbach J.M. Ajami N.J. Espinola J.A. Henke D.M. Petrosino J.F. et al.Association of nasopharyngeal microbiota profiles with bronchiolitis severity in infants hospitalized for bronchiolitis.Eur Respir J. 2016; 48: 1329-1339Crossref PubMed Google Scholar infants with a Haemophilus-dominant nasopharyngeal microbiota profile had higher severity; however, host immune responses were not examined. In the present study, we sought to determine interactions between serum LL-37 levels and nasopharyngeal microbiota profiles with regard to disease severity by using data from a prospective cohort of infants with severe bronchiolitis. Details of the study design, setting, population, testing, and analysis may be found in the Online Repository (see this article's Methods section at www.jacionline.org). Briefly, this prospective cohort study, the 35th Multicenter Airway Research Collaboration,4Hasegawa K. Mansbach J.M. Ajami N.J. Espinola J.A. Henke D.M. Petrosino J.F. et al.Association of nasopharyngeal microbiota profiles with bronchiolitis severity in infants hospitalized for bronchiolitis.Eur Respir J. 2016; 48: 1329-1339Crossref PubMed Google Scholar enrolled infants (age 46 ng/mL). The outcome of interest was intensive care use, defined as admission to intensive care unit and/or use of mechanical ventilation. To determine heterogeneity of microbiota-outcome associations by LL-37 status, we fit a random-effects model adjusting for 12 potential confounders (age, sex, race/ethnicity, gestational age, history of breathing problems, daycare attendance, siblings at home, breast-feeding, lifetime history of antibiotic and corticosteroid use, use of antibiotics during prehospitalization visit, and respiratory viruses) for each stratum (low and high LL-37 strata). Of 1016 enrolled infants, 1005 (99%) met the 16S rRNA gene sequence quality control requirements for analysis. The median age was 3.2 months (interquartile range [IQR], 1.6-5.9 months), 60% were males, and 43% were non-Hispanic white. The median LL-37 level was 46 ng/mL (IQR, 34-60 ng/mL) and that of serum 25-hydroxyvitamin D (25(OH)D) was 27 ng/mL (IQR, 18-33 ng/mL). Several patient characteristics differed by LL-37 status (Table I). For example, infants with lower LL-37 levels were younger and more likely to have respiratory syncytial virus infection (both P < .05), compared with those with higher LL-37 levels. In contrast, there were no significant differences in the microbiota profiles (Table I) or abundance of major bacteria genera (see Table E2 in this article's Online Repository at www.jacionline.org) between the groups (all P ≥ .40). Likewise, 25(OH)D and LL-37 levels were not significantly correlated (P = .33; see Fig E2 in this article's Online Repository at www.jacionline.org).Table ICharacteristics of 1005 infants hospitalized for bronchiolitis, according to serum LL-37 levelVariablesLL-37 ≤46 ng/mLn = 515 (51%)LL-37 >46 ng/mLn = 490 (49%)P valueCharacteristics Age (mo), median, (IQR)2.7 (1.5-5.4)3.8 (2.1-6.3)<.001 Male sex312 (61)291 (59).75 Race/ethnicityNon-Hispanic white231 (45)197 (40).23Non-Hispanic black112 (22)121 (25)Hispanic149 (29)157 (32)Other23 (5)15 (3) Parental history of asthma182 (35)159 (33).38 Maternal smoking during pregnancy85 (17)59 (12).06 C-section delivery178 (35)165 (34).88 Prematurity (32-37 wk)99 (19)84 (17).44 Previous breathing problems before the index hospitalization∗Defined as a child having cough that wakes him or her at night and/or causes emesis, or when the child has wheezing or shortness of breath without cough.87 (17)116 (24).009 History of eczema69 (13)77 (16).33 Ever attended daycare98 (19)132 (27).004 Sibling in the household414 (80)385 (78).53 Mostly breast-fed for the first 3 mo of age211 (47)208 (49).73 Smoke exposure at home80 (16)73 (15).85 Antibiotic use before the index hospitalization149 (29)163 (33).16 Corticosteroid use before the index hospitalization70 (14)76 (16).44Clinical presentation Weight at presentation (kg), median (IQR)5.7 (4.5-7.4)6.5 (5.1-7.9)<.001 Respiratory rate at presentation (per minute), median (IQR)48 (40-60)48 (40-60).85 Oxygen saturation at presentation.25 .05). These findings are concordant with our previous single-center study of 82 children with bronchiolitis reporting that serum 25(OH)D levels were not correlated with cathelicidin levels in the setting of acute infection, but that cathelicidin was associated with viral etiology.6Mansbach J.M. Piedra P.A. Borregaard N. Martineau A.R. Neuman M.I. Espinola J.A. et al.Serum cathelicidin level is associated with viral etiology and severity of bronchiolitis.J Allergy Clin Immunol. 2012; 130: 1007-1008.e1Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar The current multicenter study of 1005 infants with severe bronchiolitis corroborated these earlier findings and extended them by demonstrating interactions between cathelicidin and airway microbiota with regard to disease severity. The mechanism underlying the observed relationships —that is, an association between Haemophilus-dominant microbiota profile and higher severity only among infants with low LL-37 levels—is beyond the scope of our data. However, the association may be causal—that is, LL-37 contributes, through altercations in airway microbiota, to the bronchiolitis morbidity. Alternatively, LL-37 cathelicidin in conjunction with specific microbiota, potentially via facilitating neutrophil extracellular traps formations,8Neumann A. Berends E.T. Nerlich A. Molhoek E.M. Gallo R.L. Meerloo T. et al.The antimicrobial peptide LL-37 facilitates the formation of neutrophil extracellular traps.Biochem J. 2014; 464: 3-11Crossref PubMed Scopus (91) Google Scholar might have contributed to host antiviral response. In addition, another mechanism—that is, specific composition and function of the microbiota upregulates/downregulates the expression of LL-37 cathelicidin locally and/or systemically, thereby contributing to bronchiolitis severity—is also possible. Furthermore, these possibilities are not mutually exclusive. Although the clinical significance is not yet clear, our findings are of scientific significance with regard to the apparent interrelation between the innate immune response and microbiota. Recent studies are beginning to delineate the mechanisms by which innate immune systems interact with microbes in nonrespiratory niches, such as the gastrointestinal tract.7Pound L.D. Patrick C. Eberhard C.E. Mottawea W. Wang G.S. Abujamel T. et al.Cathelicidin antimicrobial peptide: a novel regulator of islet function, islet regeneration, and selected gut bacteria.Diabetes. 2015; 64: 4135-4147Crossref PubMed Scopus (41) Google Scholar The present study demonstrates, for the first time, an integrated role of LL-37 antimicrobial peptides (an important component of the innate immune system) and the airway microbiota in the pathogenesis of airway disease. Our data should facilitate further mechanistic investigations into this complex interplay. We acknowledge several potential limitations. First, our data were based on serum LL-37 levels and nasopharyngeal microbiota, which may not reflect the cathelicidin activity or microbiota in the lung. Nonetheless, studies have shown that cathelicidin derived from bone marrow rather than airway epithelial cells is responsible for antimicrobial effects in the lung,9Kovach M.A. Ballinger M.N. Newstead M.W. Zeng X. Bhan U. Yu F.S. et al.Cathelicidin-related antimicrobial peptide is required for effective lung mucosal immunity in Gram-negative bacterial pneumonia.J Immunol. 2012; 189: 304-311Crossref PubMed Scopus (87) Google Scholar and that upper airway microbiota is a reliable representation of that of lower airway in children.10Marsh R.L. Kaestli M. Chang A.B. Binks M.J. Pope C.E. Hoffman L.R. et al.The microbiota in bronchoalveolar lavage from young children with chronic lung disease includes taxa present in both the oropharynx and nasopharynx.Microbiome. 2016; 4: 37Crossref PubMed Scopus (114) Google Scholar Second, 16S rRNA gene sequencing precluded us from examining the function of microbiome. We hope to address this important issue in future work using metatranscriptomic approaches. Third, this study did not have the information of a "control" group. Yet, the study objective is not to assess the role of microbiome and cathelicidin on the development of bronchiolitis but to determine their relationship with disease severity among infants who have bronchiolitis. Finally, although the study cohorts consisted of racially/ethnically diverse US sample of severe bronchiolitis, our inferences might not be generalizable to those with mild-to-moderate illness. In sum, on the basis of data from a multicenter cohort of 1005 infants with severe bronchiolitis, we found an interaction between serum cathelicidin (LL-37) and nasopharyngeal microbiota with regard to higher disease severity. Specifically, we observed significant associations between Haemophilus-dominant profile and higher risks of intensive care use only in infants with lower LL-37 levels. Our findings should facilitate research into understanding the complex interplay between the innate immunity, airway microbiome, and bronchiolitis pathogenesis in infants. We thank the 35th Multicenter Airway Research Collaboration study hospitals and research personnel for their ongoing dedication to bronchiolitis and asthma research (see Table E1). We also thank Ashley F. Sullivan, MS, MPH, and Janice A. Espinola, MPH (Massachusetts General Hospital, Boston, Mass), and Alkis Togias, MD (National Institute of Allergy and Infectious Diseases) for their contributions to the study. This is an analysis of data from a multicenter prospective cohort study of infants (age 46 ng/mL). We compared the patients' characteristics, clinical presentation, virology, nasopharyngeal microbiota profiles, and laboratory data, by LL-37 status, using the chi-square test or the Wilcoxon-Mann-Whitney test as appropriate. Next, we tested for an interaction between LL-37 status and microbiota profiles with regard to risks of intensive care use by fitting a random-effects model adjusting for 12 patient-level variables (ie, age, sex, race/ethnicity, gestational age, history of breathing problems, daycare attendance, siblings at home, breast-feeding, lifetime history of antibiotic use, history of corticosteroid use, use of antibiotics during the prehospitalization visit, and respiratory viruses detected by PCR). The model also accounted for patient clustering at the hospital level. We chose these potential confounders on the basis of clinical plausibility and a priori knowledge.E22Jartti T. Hasegawa K. Mansbach J.M. Piedra P.A. Camargo Jr., C.A. Rhinovirus-induced bronchiolitis: lack of association between virus genomic load and short-term outcomes.J Allergy Clin Immunol. 2015; 136: 509-512.e11Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar, E23Mansbach J.M. Piedra P.A. Stevenson M.D. Sullivan A.F. Forgey T.F. Clark S. et al.Prospective multicenter study of children with bronchiolitis requiring mechanical ventilation.Pediatrics. 2012; 130: e492-e500Crossref PubMed Scopus (108) Google Scholar, E24Hasegawa K. Mansbach J.M. Camargo Jr., C.A. Infectious pathogens and bronchiolitis outcomes.Expert Rev Anti Infect Ther. 2014; 12: 817-828Crossref PubMed Scopus (57) Google Scholar, E25Hasegawa K. Linnemann R.W. Mansbach J.M. Ajami N.J. Espinola J.A. Petrosino J.F. et al.The fecal microbiota profile and bronchiolitis in infants.Pediatrics. 2016 Jul; 138Crossref PubMed Scopus (45) Google Scholar, E4Hasegawa K. Jartti T. Mansbach J.M. Laham F.R. Jewell A.M. Espinola J.A. et al.Respiratory syncytial virus genomic load and disease severity among children hospitalized with bronchiolitis: multicenter cohort studies in the US and Finland.J Infect Dis. 2015; 211: 1550-1559Crossref PubMed Scopus (103) Google Scholar, E5Mansbach J.M. Piedra P.A. Teach S.J. Sullivan A.F. Forgey T. Clark S. et al.Prospective multicenter study of viral etiology and hospital length of stay in children with severe bronchiolitis.Arch Pediatr Adolesc Med. 2012; 166: 700-706Crossref PubMed Scopus (249) Google Scholar Because the model indicated a significant LL-37– × –Microbiota interaction (Pinteraction = .02), we repeated the analysis with stratification by LL-37 status. Analyses used R version 3.2 (R Foundation, Vienna, Austria). All P values were 2-tailed, with P < .05 considered statistically significant.Fig E2Scatterplot of serum LL-37 and 25(OH)D levels. There was no significant correlation between serum LL-37 and 25(OH)D levels (r = 0.03; P = .33).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table E1Principal investigators at the 17 participating sites in MARC-35Amy D. Thompson, MDAlfred I. duPont Hospital for Children, Wilmington, DelFederico R. Laham, MD, MSArnold Palmer Hospital for Children, Orlando, FlaJonathan M. Mansbach, MD, MPHBoston Children's Hospital, Boston, MassVincent J. Wang, MD, MHAChildren's Hospital of Los Angeles, Los Angeles, CalifMichelle B. Dunn, MDChildren's Hospital of Philadelphia, Philadelphia, PaJuan C. Celedon, MD, DrPHChildren's Hospital of Pittsburgh, Pittsburgh, PaMichael Gomez, MD, MS-HCA, and Nancy Inhofe, MDThe Children's Hospital at St Francis, Tulsa, OklaBrian M. Pate, MD, and Henry T. Puls, MDThe Children's Mercy Hospital & Clinics, Kansas City, MoStephen J. Teach, MD, MPHChildren's National Medical Center, Washington, DCRichard T. Strait, MDCincinnati Children's Hospital and Medical Center, Cincinnati, OhioIlana Waynik, MDConnecticut Children's Medical Center, Hartford, ConnSujit Iyer, MDDell Children's Medical Center of Central Texas, Austin, TexMichelle D. Stevenson, MD, MSKosair Children's Hospital, Louisville, KyWayne G. Schreffler, MD, PhD, and Ari R. Cohen, MDMassachusetts General Hospital, Boston, MassAnne K. Beasley, MDPhoenix Children's Hospital, Phoenix, ArizThida Ong, MDSeattle Children's Hospital, Seattle, WashCharles G. Macias, MD, MPHTexas Children's Hospital, Houston, Tex Open table in a new tab Table E2Richness, alpha diversity, and relative abundance by serum LL-37 levelIndicesLL-37 ≤46 ng/mLn = 515 (51%)LL-37 >46 ng/mLn = 490 (49%)P valueRichness No. of genera, median (IQR)17 (10-24)15 (8-24).06Alpha diversity Shannon index, median (IQR)0.99 (0.57-1.45)0.89 (0.52-1.36).07Relative abundance of 10 most common genera, mean ± SD Streptococcus0.31 ± 0.290.31 ± 0.30.99∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Moraxella0.29 ± 0.340.31 ± 0.34.99∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Haemophilus0.19 ± 0.300.21 ± 0.31.99∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Prevotella0.02 ± 0.060.02 ± 0.06.99∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Staphylococcus0.02 ± 0.080.02 ± 0.10.99∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Neisseria0.03 ± 0.080.02 ± 0.06.96∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Corynebacterium0.02 ± 0.080.01 ± 0.04.40∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Alloprevotella0.01 ± 0.050.01 ± 0.04.99∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Veillonella0.01 ± 0.030.01 ± 0.03.99∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Gemella0.01 ± 0.030.01 ± 0.03.99∗Benjamini-Hochberg–adjusted P value accounting for multiple comparisons.IQR, Interquartile range.∗ Benjamini-Hochberg–adjusted P value accounting for multiple comparisons. Open table in a new tab IQR, Interquartile range.