Ethics statement

Animal procedures were approved by the Research Ethics Committee of the Faculty of Pharmacy, Cairo University, Approval No. MI (2510), following the Guide for the Care and Use of Laboratory Animals published by the Institute of Laboratory Animal Research, USA.

Bacterial strains and culture conditions

M. catarrhalis O35E was kindly provided by Dr. Eric J. Hansen [24] and it was used as the wild type (WT). The derivatives were all generated in its background. M. catarrhalis strains were grown on Tryptic Soy Agar (TSA) or Columbia blood agar at 37 °C and 5% CO2 in a carbon dioxide incubator (Binder, Germany), or in Tryptic Soy Broth (TSB) at 37 °C with shaking at 180 rpm under aerobic conditions [25]. Escherichia coli DH5-α, used as a cloning host, was grown at 37 °C in TSB with shaking at 180 rpm, or on TSA. When needed, media were supplemented with kanamycin at a final concentration of 15 µg/mL, streptomycin at a final concentration of 250 µg/mL, and ampicillin at a final concentration of 100 μg/mL.

Bioinformatics analyses

The protein sequence of the EstD of N. meningitidis (NCBI protein id CAM08666.1) served as a template for a BlastP analysis [26], to identify homologs of this esterase in M. catarrhalis O35E. To survey the conservation of the protein, the BlastP search was also extended to all the strains available in M. catarrhalis taxid 480. Sequences of previously studied FghA homologs were retrieved from the NCBI (Table S1), and the same was performed with the AdhC homologs (Table S2). Then, multiple sequences’ alignment with the Blast-retrieved FghA of M. catarrhalis O35E (NCBI Protein id EGE27440.1) as a query sequence was carried on using Clustal Omega [27] applyling the default parameters. Phylogenetic trees were constructed via NGphylogeny.fr web tool, which employs multiple alignment using fast Fourier transform (MAFFT) for multiple sequence alignment, Block Mapping and Gathering with Entropy (BMGE) for alignment curation, Phylogeny software for the maximum-likelihood principle (PhyML) for tree inference, and finally Newick display for tree rendering [28,29,30,31,32]. To calculate the percent similarity and identity, EMBOSS Needle tool was used with the default parameters [33]. To further confirm the interaction between AdhC and FghA and show possible interactions with other proteins, the protein–protein functional interaction analysis tool STRING [34] was used. To investigate if the adhC and fghA genes form an operonic pair as consistently reported for the system, the operon prediction tool Operon Mapper [35] was used applying the default parameters, and the whole M. catarrhalis O35E genome (AERL00000000.1) as a query sequence.

Construction of M. catarrhalis ΔfghA and ΔadhC–fghA deletion mutants and rescue strains

Using the M. catarrhalis O35E chromosomal DNA as a template, the primer pairs DO001–DO002 and DO003–DO004 (Table S3 and Fig. 1) were used to amplify a 1002 and a 1006 base pair fragments upstream and downstream of the fghA open-reading frame (ORF), respectively. The fragments were then digested using XmaI and ligated together. The ligation product was used as a template for a second PCR reaction using primer pair DO001–DO004. The product was then ligated into the rapid cloning vector pJET1.2/blunt (Thermo Fisher Scientific, Lithuania), to yield plasmid pJET-U + D. A non-polar kanamycin resistance cassette was obtained by digesting plasmid pUC18K [36] with XmaI, followed by gel purification. The product was then ligated with plasmid pJET-U + D digested with the same restriction enzyme, then transformed into E. coli DH5-α and plated on TSA containing ampicillin and kanamycin. The resultant plasmid was designated pJET-UkanD. This plasmid was used to amplify a ~ 3 kb construct consisting of the kanamycin cassette flanked by the fghA upstream and downstream fragments, and the product was transformed into M. catarrhalis O35E as previously described [37]. The homologous recombination of the mutant construct into the chromosome and the replacement of the WT fghA gene to yield ΔfghA mutant was confirmed by a series of PCR reactions using primers within and outside the mutant construct. To construct the ΔadhC-fghA double mutant, a similar approach was adopted but using primer pairs DO008–DO009 and DO003–DO004 to amplify a 329 and a 1002 base pair fragments upstream and downstream of adhC and fghA, respectively. To repair the ΔfghA mutant, the fragment amplified with primer pair DO001–DO004 from the WT O35E M. catarrhalis strain together with the mutated rpsL amplicon [38] were transformed into the ΔfghA in a congression experiment as previously described [39]. Isolated colonies that could grow on streptomycin, and failed to grow on kanamycin, were selected as potential complemented mutants, confirmed using PCR, and designated ΔfghA/R. To repair ΔadhC–fghA mutant, a similar approach was used, but using primer pair DO008–DO004 to amplify the WT amplicon. The confirmed rescue mutant was designated ΔadhC–fghA/R. The sequences of all the oligonucleotides used in this study are listed in Table S3.

Fig. 1

Genetic organization and functional relation between the M. catarrhalis adhC and fghA. A A schematic diagram showing the organization of the neighboring ORFs in the genetic loci of the adhC and fghA in the M. catarrhalis WT O35E genome. The binding position of the primers used in this study and the loci tags are indicated. The direction of the ORF arrows indicates the direction of transcription. The ruler above the arrows indicates the size of DNA fragment in base pairs; bp. The map was generated by Ankh diagram v1.1tool by HITS Solutions Co. (Bioinformatics Department, Cairo, Egypt). B A schematic diagram representing the interaction between AdhC (encoded by EA1_02212) with FghA and other M. catarrhalis proteins including Gdsl-like lipase (encoded by EA1_06621), MsrAB, and GdhA. The figure was generated using STRING database and the lines drawn between the functional pairs represent the predicted functional relationship (neighborhood; green, gene fusion; red, co-occurrence; dark purple, co-expression; black, databases; teal, text mining; yellow, experimentally determined interactions; pink line, and protein homology; light blue)

Growth curve analysis

Colonies of the WT O35E, ΔfghA, ΔfghA/R, ΔadhC-fghA, and ΔadhC-fghA/R grown overnight on Columbia blood agar were suspended in TSB to an optical density at 600 nm (OD600) ~ 1.0 then diluted 1:50 in 15 mL TSB broth. The cultures were incubated in a shaking incubator at 37 °C and 180 rpm and the OD600 was measured each hour for 8 h using a visible spectrophotometer Jenway 6300 (Jenway, United Kingdom). Growth curves were constructed by plotting OD600 versus time.

Formaldehyde sensitivity assay

Formaldehyde susceptibility was assessed using the disc diffusion susceptibility assay as previously described [20]. Briefly, cells of the WT O35E, ΔfghA, ΔfghA/R, ΔadhC-fghA, and ΔadhC-fghA/R freshly streaked on Columbia blood agar were suspended in TSB to an OD600 ~ 0.4. The adjusted cell suspensions were spread over TSA plates (supplemented with kanamycin for ΔfghA and ΔadhC-fghA mutants, streptomycin for ΔfghA/R and ΔadhC-fghA/R complemented mutants) using sterile cotton swabs. Sterile single Whatmann no. 1 filter paper discs were saturated with 5 μL of a 5% formaldehyde solution (Piochem, Egypt) and placed onto the agar surface. The diameter of the inhibition zone around the disc was measured after an overnight incubation at 37 °C in a CO2 incubator with the petri-dishes inverted.

Formaldehyde sensitivity was also tested by another assay [20]. Briefly, a bacterial suspension was prepared as in the sensitivity assay detailed above, and seven tenfold serial dilutions were prepared in a 96-well microplate in TSB. Five microliters of each dilution were spotted on TSA plates, supplemented with 0-, 0.8-, or 1-mM formaldehyde. The spots were left to dry then survival was determined by counting the number of visible colonies after an overnight incubation at 37 °C in a CO2 incubator with the petri-dishes inverted.

Determination of the minimum inhibitory concentration (MIC)

To obtain a more quantitative assessment of the susceptibility of the five strains under investigation to formaldehyde, we performed an MIC experiment using the standard microdilution method [40]. Briefly, a 0.5 McFarland standard suspension of each of the five strains was prepared using freshly grown bacterial cells, and then, it was diluted 1:10 and 10 µL were used to inoculate 12 wells containing 190 µL of TSB containing twofold dilutions of formaldehyde from 1 mM.to 0.5 µM.

The wells incubated for 24 h at 37 °C then inspected for visual growth. The formaldehyde concentration in the first clear well was considered the corresponding MIC value for the respective strain.

Protein profiling using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)

Cells of the five tested strains were grown overnight in TSB in the presence and absence of 1 mM formaldehyde in a shaking incubator, harvested by centrifugation, and resuspended in sterile saline to an OD600 ~ 1. The cell suspensions were mixed with a 3 × reducing Laemmli buffer [41]. These samples were then heated at 95 °C for 10 min in a thermal cycler (Boeco, Germany) before loading on a 10% SDS-PAGE gel. Gels were afterwards stained with Coomassie blue [41], visualized, and photographed using a gel documentation system (UVP, Germany).

Murine pulmonary clearance model

The pulmonary clearance model in mice was carried out as previously described [42]. Briefly, five groups (n = 6) of 6–8-week-old female BALB/C mice were infected intranasally by injecting 40 μl of a bacterial suspension (∼5 × 106 CFU) of each of the WT O35E, ΔfghA, ΔfghA/R, ΔadhC-fghA, and ΔadhC-fghA/R into the nostrils under anesthesia using 250 µL of 25 µg/mL 2,2,2-tribromoethanol. Four-and-a-half-hour post-inoculation, mice were sacrificed by an overdose of the anesthesia (750 µL of 25 µg/mL 2,2,2-tribromoethanol), followed by cervical dislocation. The lungs were excised, homogenized, serially diluted, and plated on TSB agar. Plates were incubated for 48 h followed by colony counting.