Nontypeable H. influenzae ATCC49766 was obtained from the American Type Culture Collection and was grown at 37°C in brain heart infusion broth (Merck Company, Germany) supplemented with Hemin-L-Histidine (Sigma-Aldrich, Germany) and NAD (Mast Group Ltd, Liverpool, UK) ⁽¹⁶⁾. Escherichia coli (E. coli) strains DH5α and BL21 (DE3) and pET24a plasmid were provided by the national gene bank, Pasteur Institute of Iran.

Genomic DNA of nontypeable H. influenzae ATCC49766 was extracted using Phenol-Chloroform extraction method as described by Sambrook et al ⁽¹⁷⁾. CBD of HapS (1000 bp) was PCR-amplified using HindIII-tailed forward (TAAGAAGCTTCGCGTTCA GATTGGACAGG) and XhoI-tailed reverse (A T A C T C G A G CA G G C T T T G A T C A G G C A GG), primer pairs and through the application of the following thermal program. It was conducted in 94°C for 7 min and followed by 35 cycles of 94°C for 1 min, 64.5°C for 45 s, 72°C for 1 min and a final extension time of 72°C for 10 min.

PCR reaction (20 µl) contained 1X Pfu buffer with MgCl₂, 1 mM dNTPs (Fermentas, St. Leon-Rot, Germany), 10 pmole of each primers, 1.25 U Pfu polymerase (Thermo, USA) and 100 ng of genomic DNA. The final product was purified by GeneJET™ Gel Extraction Kit (Fermentas, St. Leon-Rot, Germany).

CBD PCR product was double digested by HindIII and XhoI and then ligated into pET24a, so that the C-terminus of recombinant protein could be expressed in fusion with 6xHis-tag. The integrity of the obtained construct, pET24a-cbd, was confirmed by restriction endonuclease analysis and bidirectional automated sequencing using BigDye reagent (Applied Biosystems, CA, USA) in Sequencing Center, Pasteur Institute of Iran. The nucleotide sequence of cbd gene fragment was deposited at the GenBank database under the accession number of KC261489.

To express the recombinant C-terminal fragment of HapS, pET24a-cbd recombinant plasmid was transformed into competent E. coli BL21 (DE3) cells and bacteria were grown on kanamycin-enriched (50 µg/ml) LB broth up to the optical density of 0.5. Following a 3 hr expression induction with 1.0 mM IPTG, the cell pellets were analyzed for protein expression, using 12% SDS-PAGE and CBD expression levels were determined using gel densitometry by Quantity One software (Bio-Rad), version 4.4.1.

In order to express the highest possible level of CBD protein, several experimental parameters affecting the rate of recombinant protein expression including individual expressing single colonies, culture media (LB and 2xTY), induction stage (OD_{600 nm}: 0.2, 0.4, 0.6, 0.8, and 1.0), inducer (IPTG) concentration (0.2, 0.5, and 1.0 mM), induction duration (2, 3, 5 hrs) and induction temperature (30, 35, and 37°C) were analyzed by one factor at a time method. The protein expression levels in each experiment were quantified by scanning of SDS-PAGE gel and densitometric analysis of CBD protein band, as mentioned above.

The best expression parameters obtained at the flask level (50 ml of bacterial culture) were scaled up to 1 L of culture. The pellet of E. coli BL21 cells expressing CBD potein was resuspended in lysis buffer (100 mM NaH2P04, 10 mM Tris-HCl, 8 M Urea, pH = 8) by gentle vortexing and then centrifuged for 10 min at 11000 rpm. The obtained supernatant was loaded onto a nickel-nitrilotriacetic acid (Ni-NTA) column (Qiagen, Germany) for immobilized metal affinity chromatography (IMAC) purification under denaturing conditions based on the protocol provided by the manufacturer. After washing steps with wash buffer (100 mM NaH2P04, 10 mM Tris-HCl, 8 M Urea, pH = 6.4), fractions of pure CBD protein were elued in optimized elution buffer (100 mM NaH2P04, 10 mM Tris-HCl, 8 M Urea, pH = 4.5). Protein concentration was determined by Nanodrop analyzer (Thermo scientific, nanodrop1000 spectrophotometer) and the purity was determined by SDS-PAGE and Coomassie blue staining.

Proteins separated by SDS-PAGE were blotted onto a 0.45 µm pore size polyvinylidene difluoride (PVDF) membrane (Amersham Biosciences, Pittsburgh, USA) by using a semi-dry blotter unit (Bio-Rad Laboratories, California, USA). The membrane was probed, first with a Anti-6xHis rabbit antibody (Qiagen, Hilden, Germany), diluted to 1:1000 in TBS-Tween 20 and then with HRP-conjugated sheep anti-rabbit IgG (Sigma, Saint Louis, MO, USA) diluted to 1:3000 in PBS-Tween 20 for one hr at room temperature. Detection of CBD protein was achieved upon the development bands with DAB substrate (Sigma, Saint Louis, MO, USA) ⁽¹⁸⁾.

The amino acid sequences of three NTHi strains (N187, TN106, P860295) retrieved from GenBank were aligned with the sequence of CBD protein expressed in this study (ATCC49766) using Mega 4 software to assess the degree of amino acid conservation among them. An antigenicity plot of the studied CBD protein was also generated using the algorithm proposed by Kolaskar and Tongaonkar (http://imed.med.ucm.es/Tools/anti-genic.pl) ⁽¹⁹⁾.

The C-terminal fragment of HapS harbors the adhesive responsibility was amplified by PCR using specific primers. A 1% agarose gel was run to confirm the existence of proper size of PCR product and only a single band was appeared with an estimated size of ∼1000 base pairs (Figure 1A). Then, the double digested PCR products and vectors were ligated together at a molar ratio of 3:1 and subsequently transformed into competent E. coli DH5α cells. The resultant colonies were evaluated for the true insert size by Colony PCR method. The oligonucleotide insertion into targeted plasmid was verified by double digestion of pET24a-cbd construct with two restriction enzymes, HindIII and XhoI simultaneously. Double digestion of correct oligonucleotide insertion should have produced two bands of 5300 bp, 1000 bp (Figures 1B and C).

Expression of genes linked to T7 promoters in E. coli BL21 (DE3) cells are triggered by adding IPTG. Expressed proteins in these cells along with uninduced ones were then separated using SDS-PAGE. After staining of proteins by Coomassie brilliant blue (R-250) and destaining, along with other bands, a band with an approximate size of ∼37 kDa on lane of the induced samples was expected to be the protein of interest (Figure 2).

Six parameters of cultivation medium composition, growth phase at induction, IPTG concentration, the time of induction and incubation temperature after induction were modified in order to optimize the conditions for the production of recombinant CBD of HapS in E. coli BL21 (DE3) cells. To estimate the amounts of CBD production in each mentioned experiment, the resulting bands on SDS-PAGE gel were scanned and intensities of specific bands, corresponding to the protein of interest were measured.

Since freshly transformed colonies may differ significantly in their expression rates, CBD positive colonies were screened for high-level production of recombinant protein. Among seven CBD positive colonies, one colony showed a more intense band of target protein on SDS-PAGE gel and was selected for further investigations (Figure 3A). In media experiment, it was found that the maximum production of CBD occurred in 2xTY medium. On the other hand, protein band of interest on SDS-PAGE gel obtained from cultivation of E. coli BL21 (DE3) cells harboring pET24a-cbd in 2xTY/kanamycin was more intense in comparison with that obtained from LB medium/kanamycin. Furthermore, the ratio of the target protein band (CBD) to the density of total bacterial protein was higher in the case of 2xTY medium in comparison with LB (57% against 43%, respectively) (Figure 3B). This observation may be explained by the existence of higher amounts of yeast extract in 2xTY medium. As a result, the subsequent optimization experiments were conducted in 2xTY medium supplemented with appropriate antibiotics.

In addition to optimum starting culture, another important step for high-level protein production is determination about the appropriate growth phase at induction. For this purpose, different inoculation times of induction, OD_{600 nm}: 0.2, 0.4, 0.6, 0.8, and 1 were considered. According to densitometric analysis, the suitable inoculation time of induction was at the mid-log phase (OD_{600 nm} equal to 0.6) (Figure 3C). Thereafter, in order to examine the effect of inducer concentration on expression level of cbd gene fragment, different concentrations of IPTG, (0.2, 0.5, and 1.0 mM), were added to cultures at OD_{600 nm} equals to 0.6. Intensity quantification of obtained target protein bands revealed that an increase in the IPTG concentration from 0.2 mM to 1.0 mM resulted in the improvement of cbd expression level. However, an increase in the IPTG concentration from 0.5 to 1.0 mM did not result in any significant improvement in recombinant protein production and the optimum IPTG concentration was considered 1.0 mM (Figure 3D).

In order to determine which time course of induction resulted in higher level of protein production, the same amount of IPTG, 1.0mM, was added to the cultures at OD_{600 nm} equals to 0.6 and at different times of 2, 3, 5 hrs post-induction, production levels of target protein was quantified. The optimal amount of CBD was attained over the period of 5 hrs after induction with IPTG (Figure 3E). The effect of different incubation temperatures (30, 35, and 37°C) on amount of protein production was also investigated under the optimum parameters obtained from the above experiments. Accordingly, 37°C was more suitable for protein expression following addition of IPTG (Figure 3F).

According to above experiments, the high expression level of CBD was obtained when CBD expressing E. coli BL21 (DE3) cells were grown at 37°C in 2xTY supplemented with 50 µg/ml kanamycin and induction of cells were conducted with 1.0 mM IPTG after the OD _{600 nm} reached 0.6. The optimal expression level was achieved 5 hrs after the addition of IPTG to the medium. Therefore, the 6xHis-tagged C-terminal fragment of HapS was overproduced under the mentioned optimal conditions. Both supernatant and the pellet of cell lysates were checked for the presence of recombinant proteins and the majority of the expressed protein was detected in supernatant. Recombinant protein was purified with Ni-NTA affinity chromatography under denaturing conditions. Theoretical molecular weight of target protein was measured to be around ∼37 kDa. The 37 kDa protein band was observed in SDS-PAGE and confirmed as CBD protein by western blot analysis using rabbit anti-His tagged antibody (Figures 4 and 5). The rate of CBD expression was approximately 62% among total bacterial proteins.

Protein sequence alignment of the CBD of HapS from three NTHi strains N187, TN106, P860295 and CBD sequence employed in current study showed more than %97 identity (Figure 6A). The antigenicity plot of the CBD sequence from NTHi ATCC49766 indicated 9 antigenic domains that were similar to antigenic domains in three above strains. This finding propounds the possibility of using CBD (ATCC49766) as a universal antigen for vaccine studies in different geographical areas. The four prominent antigenic determinants were identified in the sequence also corresponded to the residues 13-20, 24-34, 199-210 and 230-239 (Figure 6B, Table 1).