A member of the cathelicidin family of antimicrobial peptides is produced in the upper airway of the chinchilla and its mRNA expression is altered by common viral and bacterial co-pathogens of otitis media

Abstract

Cationic antimicrobial peptides (AMPs), a component of the innate immune system, play a major role in defense of mucosal surfaces against a wide spectrum of microorganisms such as viral and bacterial co-pathogens of the polymicrobial disease otitis media (OM). To further understand the role of AMPs in OM, we cloned a cDNA encoding a cathelicidin homolog (cCRAMP) from upper respiratory tract (URT) mucosae of the chinchilla, the predominant host used to model experimental OM. Recombinant cCRAMP exhibited alpha-helical secondary structure and killed the three main bacterial pathogens of OM. In situ hybridization showed cCRAMP mRNA production in epithelium of the chinchilla Eustachian tube and RT-PCR was used to amplify cCRAMP mRNA from several other tissues of the chinchilla URT. Quantitative RT-PCR analysis of chinchilla middle ear epithelial cells (CMEEs) incubated with either viral (influenza A virus, adenovirus, or RSV) or bacterial (nontypeable H. influenzae, M. catarrhalis, or S. pneumoniae) pathogens associated with OM demonstrated distinct microbe-specific patterns of altered expression. Collectively, these data showed that viruses and bacteria modulate AMP messages in the URT, which likely contributes to the disease course of OM.

Fig. 1

In silico analysis of cCRAMP. (A) DNASTAR Clustal W alignment of cCRAMP with rhesus monkey RL-37 (accession no. AF181954), human LL-37 (accession no. NM004345), mouse CRAMP (accession no. NM009921), rat CRAMP (accession no. AF484553), and guinea pig CAP-11 (accession no. D87405). Shaded residues indicate amino acids which are identical between cCRAMP and the respective cathelicidin. Numbers on the left represent residue numbers of the corresponding cathelicidin in the labeled row, and the percent similarity of an AMP with cCRAMP is shown on the right. (B) Three views of the predicted 3-D structural model of cCRAMP. In the top (1) and bottom (3) views of this computer-generated model, the N-terminus is to the left and in the middle (2) view, the N-terminus is facing the viewer. The orientation of the β-sheet, as folded against the α-helix, may be an artifact due to performing the molecular mechanics and dynamics simulations with the model isolated in aqueous solution rather than in a complex environment, such as in a membrane, in which it is naturally found. However, the predicted amphipathic nature of both the α-helix and β-sheet can be clearly seen in this model.

Fig. 2

The pGM-5 plasmid in E. coli BL21 STAR (DE3) expresses recombinant cCRAMP. (A) Maps of vectors constructed to express (r)cCRAMP as detailed in experimental procedures. (B) E. coli-derived total cell lysates separated in a SDS-PAGE gel containing cCRAMP expressed from pGM-5 (arrow indicates recombinant cCRAMP fusion protein containing a His-tag). (C) A SDS-PAGE gel stained with Coomassie brilliant blue showing purification of cCRAMP from E. coli BL21 STAR (DE3) harboring pGM-5, as detailed in experimental procedures. Protein samples were loaded as follows: Protein bound to the first Ni²⁺-NTA column (lane 2), Ni²⁺-NTA column bound protein digested with thrombin (lane 3), a control protein (Novagen) digested with thrombin (lane 4), thrombin digested protein that did not bind to a second Ni²⁺-NTA column (lane 5), and bound protein from the same second column (lane 6). Arrows show (r)cCRAMP. (D) Further purification of (r)cCRAMP. Thrombin digested protein retained on a 5-kDa size-exclusion column (Lane 2) and (r)CRAMP isolated as a single band after retention on a 3-kDa size-exclusion column (Lane 3). Lane 1 in all gels contains a molecular weight standard.

Fig. 3

(r)cCRAMP is bactericidal at concentrations comparable to other antimicrobial peptides. (A) An E. coli strain sensitive to killing by antimicrobial peptides and NTHI were grown to mid-log phase, diluted to 1 x10⁸ colony-forming units, incubated with increasing concentrations of (r)cCRAMP for 1 h at 37°C, and plated for determination of number of surviving colony forming units. (B) Same as in (A) except that M. catarrhalis, and S. pneumoniae were used as the test microorganisms.

Fig. 4

CD spectra of a 25 μM solution of (r)cCRAMP. Unprocessed data for all scans are shown as either filled or unfilled shapes whereas the corresponding averaged traces (with background subtracted) are depicted as lines. Increasing the pH to 8.6 or 9.9 resulted in a spectral shift to a more disordered conformation (see dashed blue and thin solid red lines). SDS rescued and intensified the structure seen in aqueous solution at pH 7.2 (see bold green and red lines).

Fig. 5

cCRAMP mRNA analysis. (A) RT-PCR analysis of cCRAMP transcripts in several chinchilla tissues. Lanes containing amplicons generated with reverse transcriptase (+) and without reverse transcriptase (−) are shown. cCRAMP transcripts were detected in every tissue evaluated. (B) in situ hybridization of chinchilla Eustachian tube using cCRAMP sense and antisense riboprobes. Images were obtained from the mid-portion of the Eustachian tube and cCRAMP mRNA was detected in the ciliated columnar epithelium of this tubal organ.

Fig. 6

Quantitative real-time PCR analysis of cCRAMP and cBD-1 mRNA levels. Chinchilla middle ear epithelial cells were grown to confluency and were either untreated or inoculated with (A) adenovirus, influenza A virus, RSV or (B) M. catarrhalis, NTHI, or S. pneumoniae. The abundance of mRNA was normalized to GAPDH levels and values are reported as the ratio of normalized AMP mRNA levels from infected to uninfected cell cultures. Values greater than 1.0 represent samples in which mRNA levels increased after CMEEs were incubated with microorganisms, whereas values less than 1.0 represent samples in which the AMP mRNA abundance decreased after inoculated with a bacterium or virus. Data reported in (A) are the average of at least 3 experiments each performed in triplicate, whereas those in (B) are from two experiments each performed in duplicate. Asterisk denote statistically significant reduction (p ≤ 0.05) in either cBD-1 or cCRAMP mRNA levels obtained when CMEEs were infected with virus or bacteria relative to baseline, with standard error of the mean values shown.