A mutation in the sap operon attenuates survival of nontypeable Haemophilus influenzae in a chinchilla model of otitis media

Abstract

Bacteria have evolved strategies to resist killing by antimicrobial peptides (APs), important effectors of innate immunity. The sap (sensitivity to antimicrobial peptides) operon confers resistance to AP-mediated killing of Salmonella. We have recently shown that sapA gene expression is upregulated in the middle ear in a chinchilla model of nontypeable Haemophilus influenzae (NTHI)-induced otitis media. Based on these findings, we constructed an NTHI strain containing a Lux reporter plasmid driven by the sapA promoter and demonstrated early yet transient expression of the sap operon within sites of the chinchilla upper airway upon infection. We hypothesized that the sap operon products mediate NTHI resistance to APs. In order to test this hypothesis, we constructed a nonpolar mutation in the sapA gene of NTHI strain 86-028NP, a low-passage-number clinical isolate. The sapA mutant was approximately eightfold more sensitive than the parent strain to killing by recombinant chinchilla beta-defensin 1. We then assessed the ability of this mutant to both colonize and cause otitis media in chinchillas. The sapA mutant was significantly attenuated compared to the parent strain in its ability to survive in both the nasopharynx and the middle ear of the chinchilla. In addition, the mutant was impaired in its ability to compete with the parent strain in a dual-strain challenge model of infection. Our results indicate that the products of the sap operon are important for resisting the activity of APs and may regulate, in part, the balance between normal carriage and disease caused by NTHI.

FIG. 1.

sap genes are cotranscribed as a single operon. The sap gene cluster contains six genes (sapA to -Z) and is 6.8 kb long. A nonpolar kanamycin cassette (kan) was used to insertionally inactivate the sapA gene where indicated. The RNA was prepared from mid-log-phase broth cultures, and the mRNA was reverse transcribed. Primers spanning the six open reading frames were used to amplify cDNAs in the presence (+) or absence (−) of RT. PCR products were separated by gel electrophoresis.

FIG. 2.

The sapA gene is conserved in NTHI. A sapA gene fragment was PCR amplified and used to probe genomic DNAs isolated from numerous NTHI clinical isolates, as listed in Materials and Methods.

FIG. 3.

Construction of a sap promoter-driven Lux reporter plasmid. (A) The promoterless luxCDABE operon was cloned into the unique KpnI site of pGZRS-39A, generating the promoterless Lux reporter plasmid pKMLN-0. The sapA promoter region (Psap) was PCR amplified and cloned into SalI- and BamHI-digested pKMLN-0, generating the sap promoter-driven Lux reporter plasmid pKMLN-01. (B) NTHI strain 1885MEE was transformed with plasmid pKMLN-01 by electroporation, and a kanamycin-resistant clone was analyzed for bioluminescence by use of a Xenogen imaging system.

FIG. 4.

Analysis of sap promoter activity and localization of expression in a chinchilla model of OM. NTHI 1885MEE/pKMLN-01 was used to inoculate the nasopharynx and middle ear of a chinchilla, which was monitored for bioluminescence 2, 4, and 6 days after the bacterial challenge in prone (A, C, and E, respectively) and left lateral (B, D, and F, respectively) positions for 3 min each by use of an IVIS imaging system. Note the foci of infection present in both tympanic bullae on days 2 and 4 (A and C), with clearance on day 6 (E). Similarly, note the presence of bioluminescent NTHI from the pharynx to the middle ear via the eustachian tube in panel B. The loss of a pharyngeal focus of infection can be seen in panel D (at 4 days postchallenge), with full loss of the biophotonic signal by day 6 (F). The scale of intensity was set based on the bioluminescence signal at 2 days postchallenge. (G) Enumeration of bacteria in middle ear (▴) and nasal lavage (▪) fluids. At 3, 5, and 7 days postchallenge, middle ear and nasal lavage fluids were collected and plated on chocolate agar to determine the numbers of bacteria (CFU per milliliter). Middle ear data are presented as means ± standard deviations for the left and right ears of one chinchilla. (H) Bioluminescence within the nasal cavity. Bioluminescence was assessed at 7 days postchallenge after sacrifice and sagittal bisectioning of the skull. Bioluminescent bacteria were detected in the nares and nasoturbinates and as punctate signals at the opening of the nasopharynx (near the septal window) as well as at the distal end of the nasopharynx.

FIG. 5.

The sapA mutant is more sensitive to killing by recombinant chinchilla β-defensin. The sapA mutant (gray bars) and the parent strain (black bars) were incubated with increasing concentrations of (r)cBD-1 for 1 h and then plated on chocolate agar for determination of the numbers of bacteria. Data represent percentages of killing relative to the bacterial concentration at the start of the assay. The data shown are representative of three separate experiments.

FIG. 6.

A mutation in sapA attenuates the survival of NTHI in vivo. The chinchilla middle ear and nasopharynx were inoculated with the sapA mutant (▴, dotted line) or the parent strain (▪, solid line) and then monitored for OM development and viable bacterial counts. (A) Relative abilities of the strains to colonize the nasopharynx. The sapA mutant was not recovered from nasal lavage fluids beginning 12 days following inoculation. The data are presented as means ± standard deviations for two animals (n = 2). (B) Relative abilities of the strains to survive in the middle ear. The sapA mutant was cleared from the middle ear beginning 8 days after inoculation. The data are presented as means ± standard deviations for the left and right ears of two animals (n = 4).

FIG. 7.

The sapA mutant (▴, dotted line) was unable to compete with the parent strain (▪, solid line) for survival in the middle ear environment. The parent and sapA mutant strain were coinoculated at a 1:1 ratio and monitored for the ability to induce OM and viable bacterial counts over time. The sapA mutant was attenuated for survival in the middle ear, resulting in a 7-log decrease in the amount of the sap A mutant strain recovered from middle ear fluids relative to the parent strain at 8 days postchallenge. The data are presented as means ± standard deviations for the left and right ears from two animals (n = 4).