Ethics statement

This work was carried out under the supervision of the institutional research ethics committee of Islamic Azad University, Science and Research branch, Tehran, Iran (Approval ID: IR.IAU.SRB. REC. 1398. 064).

Preparation of OmpA-derived peptide as an immunogen

Based on previous in-silico design and bioinformatics analysis, a 27 amino acid peptide (VTVTPLLLGYTFQDSQHNNGGKDGNLT) at N-terminal region located at 24–50 position derived from OmpA of A. baumannii was designed and used as a safe and suitable immunogen for mice immunization [20]. As previously described, the OmpA antigenic epitopes were predicted using different tools with the highest score and based on hydrophobicity, antigenicity, flexibility, mobility, accessibility, polarity, exposed surface, and coils. Then, among five OmpA consensus epitopes, one of the novel synthetic peptides was selected [peptide 1 (amino acids located in the 24–50 position of the OmpA protein)] that had been elicited higher immune responses [20].

Peptide conjugation to carrier protein

The synthetic peptide was conjugated to bovine serum albumin (BSA) and keyhole limpet hemocyanin (KLH) (Sigma, St. Louis MO, USA) using glutaraldehyde 1% as linker separately and simultaneously utilizing the same buffer systems and methods, as described [6]. As, the KLH is a large aggregating protein and has a high molecular weight (4.5 × 105 to 1.3 × 107 Da) it does not penetrate well into very fine particles of polyacrylamide gel in the SDS PAGE [21]. Indeed, high molecular mass complexes remain at the top of the electrophoresis gel. For this purpose, a carrier protein such as BSA with lower molecular weight (about 67 KDa) was prepared simultaneously using the same method and used to evaluate the accuracy of conjugation. In this way, 1 mg (400 µl) of immunogen peptide (First BASE laboratories, 604944-X, Malaysia) and 1 mg of dissolved KLH (500 µl) were mixed together. Then 70 μl of distilled water and 30 μl of glutaraldehyde 1% were added to the mixture to obtain a final concentration of 1 mg/ml for the peptide-KLH. Similarly, to prepare the peptide-BSA, 0.5 mg of dissolved BSA (250 µl) was mixed with 0.5 mg of the peptide (200 µl) along with 50 μl of distilled water and 15 μl of glutaraldehyde. The peptide–KLH conjugate was used to mice immunization and the peptide-BSA was used for conjugation assessment.

Mice immunization procedure

In this study, which is a new experience in the production of monoclonal antibody (mAb) against drug-resistant Acinetobacter baumannii, three female BALB/c mice were used for an immunization model. The mice were healthy, young and free of infection and were purchased from the Royan Institute, Iranian academic center for education, culture and research (ACECR). Mice were about 6 weeks old and were injected 5 times intraperitoneally with 25 μg of the OmpA-peptide on days 0, 21, 35, 49, and 63 [22]. Each mouse was injected with a total of 125 μg of peptide antigen. In addition, to induce an acceptable immune response, the first injection was carried out using complete Freund’s adjuvant (Sigma-Aldrich, Saint Louis, USA), while the other five injections were performed using incomplete Freund’s adjuvant (Sigma-Aldrich, Saint Louis, USA). Ten days after the last injection, mice were tail-bled and the sera were assayed for antibody molecules against antigenic peptide in an indirect ELISA test. One mouse with a highest titer of anti OmpA-peptide was selected for further steps. Finally, three days before the cell fusion, 20 μg of peptide-KLH in PBS were injected intravenously to the well immunized mouse [6].

Evaluation of mice immune response by enzyme-linked immunosorbent assay (ELISA)

Wells of the ELISA plate (Greiner-Bio-One, Italy) were coated with 50 µl OmpA-derived peptide (10 µg/ml) overnight at 4 °C. After washing three times with PBS-Tween 20 (0.05%) (Sigma-Aldrich, St Louis, Mo, USA), the plate was blocked with 2.5% BSA at 37 °C for one hour and also at 4 °C for three hours. Wells were then washed three times. To evaluate the immune response of mice, different dilutions of mice serum taken from the tail vein were added to the coated wells (50 µl/well), starting from titer of 1:500 and at 37 °C for one hour [23]. After washing, 50 µl/well of HRP-conjugated rabbit anti-mouse Ig (1:1000; Avicenna Research Institute, Tehran, Iran) was added and incubated for one hour at 37 °C. After washing, 50 µl of tetramethylbenzidine (TMB) (DNAbiotech Co, Iran, Cat No: DB9510) was added to each well and the plates were incubated at room temperature (RT) in the dark. After about 12 min, 15 µl of stopping solution (20% H2SO4) was added to each well. The optical densities (OD) of the reactions were measured at 450 nm by an ELISA reader system (Anthos 2020, Salzburg, Austria).

In order to evaluate the amount of produced antibodies, ELISA tests were performed in duplicate at each stage of blood sampling. The mouse with the highest titer of specific antibody was considered for cell fusion and hybridoma generation. To screen for the antibody production by the hybridoma cells, the same procedure was performed on the culture supernatants.

Cell fusion process

One of the mice with the highest titer of specific antibody was chosen and used for fusion process. The mouse was euthanized by exposure to CO2 and placed on a clean portable plate. Using sterile instruments such as scissors and forceps, the spleen was removed and the spleen cells were isolated. In order to develop suitable hybridoma cells, mouse myeloma SP2/0 cells and splenocytes of the immunized mouse were washed with pre-warmed RPMI medium (pH: 7.2) and then mixed in a ratio of 1:5 respectively. After rinsing the cell mixture, pre-warmed (37 °C) 50% polyethylene glycol (PEG) 1500 (HybriMax, Sigma-Aldrich, USA), as a non-ionic hydrophilic polymer, was added to the cell pellet slowly with continuous mixing [24]. After pouring 20 ml of the wash medium, cells were centrifuged at 20 °C for 5 min at 1000 rpm (or 500 G). Cell pellets containing fused cells were then cultured in the RPMI-1640 medium (Gibco by life technologies, USA). Also, some factors for cell culture such as sterility, pH, nutrients, and proper temperature were considered. As a selective medium, hypoxanthine-aminopterin-thymidine (HAT) supplemented medium (Sigma, USA) was added to the wells and only hybridoma cells were allowed to grow [24]. For better growth of hybridomas, fetal bovine serum (FBS) (Gibco, USA) was used at concentrations of 10% or 20% (v/v) as an enrichment in the culture medium [25]. The cell culture supernatants were then tested for the mAb evaluation by ELISA as a mentioned above. Hence, among the grown hybrid cells, the mono clones capable of producing mAbs were screened. Then, a mono clone with freshly condition and without possible contamination was selected as 3F10-C9 clone.

Isotype determination of mAb by capture ELISA

Goat anti-mouse IgG1, IgG2a, IgG2b, IgG3, IgA, and IgM subclass specific antibodies including Lambda (λ) and Kappa (κ) at 1/1000 dilution (Sigma-Aldrich-ISO2 MSDS, USA), with 50 µl/well, were adsorbed on to the wells of a microtitre ELISA plate (SPL, Korea). After blocking and washing similar to the above explanation, the supernatant of the 3F10-C9 clone was added in the amount of 100 µl to each well. Also, PBS 1× was added instead of 3F10-C9 mAb to each well as a negative control. After performing the ELISA steps similar to the previous description, the isotype of 3F10-C9 mAb was determined.

Determination of affinity constant

The affinity constant (Kaff) of mAb from the clone 3F10-C9 was determined by non-competitive enzyme immunoassay method as previously described [26]. All reactions were done in a sealed (to prevent evaporation) microtitre polystyrene plate (Maxisorp, Nunc, Denmark) with all reaction volumes of 50 µl. Briefly, an ELISA plate, pre-coated with five different concentrations of peptide (5000, 2500, 1250, 625, 312.5 ng/ml), were then separately incubated with serial dilutions of the 3F10-C9 mAb. After performing the ELISA test steps as described above, the amount of mAb adherent to the peptide antigen in each well was reflected by the measured OD450. The use of serial dilutions of mAb resulted in sigmoid curves of OD versus antibody concentrations. Sigmoid curves were plotted using the OD450 values against the logarithmic antibody concentrations. The half of maximum OD (OD-50) on each curve was determined at 450 nm. Then, antibody concentration was assigned on the X-axis corresponding to OD-50 of each antigen curve. [Ab]t and [Ab′]t were the measurable total antibody concentrations in the wells at OD-50 and OD-50' for plates coated with [Ag] and [Ag′], respectively. K is the antigen–antibody affinity constant in l/mol (M−1). The Kaff was then calculated using the following equation [26]:

$$Kaff=\left(n-1\right)/2 \left(n \left[Ab^\right]t-\left[Ab\right]t\right)$$

where,

$$n=\left[\mathrm\right]t/\left[\mathrm^\right]t$$

Affinity purification of generated mAb

3F10-C9 mAb was affinity purified using a column of pre-activated resin, CNBr-activated sepharose 4B (GE Healthcare, Uppsala, Sweden), conjugated to the OmpA-peptide. The elution was performed using 0.1 M glycine–HCL (pH: 2.7). The eluted antibody was dialyzed overnight against PBS 1× (pH: 7.2) at 4 °C, and the antibody reactivity was measured by ELISA as described above.

Preparation of bacterial samples

Bacterial samples were collected from carbapenem-resistant A. baumannii isolates in hospitals that have been previously studied [27]. In the previous published work, these isolates harbored resistance genes such as OXA-23, OXA-24 and OXA-58 and subsequently were analyzed for possible presence of resistance indices including Ambler class A, metallo-β-lactamases (MBLs), carbapenem-hydrolysing class D β-lactamases, extended spectrum β-lactamases like (blaTEM, blaPER, blaGES) and insertion sequence of ISAba1. The genetic relatedness between the isolates was previously analyzed using the pulsed-field gel electrophoresis (PFGE) method [27]. Among different clonal pulsotypes of Acinetobacter baumannii, we randomly received five different isolates including pulsotypes A, B, C, D and E (kind gifts from Dr. Zohreh Ghalavand) and subjected them to various experiments.

Western blot analysis

A single colony of each bacterial isolates was harvested from the bacterial culture and then centrifuged [5000 rpm for 5 min]. Bacterial lysates were prepared as previously described with modifications [28, 29]. Briefly, bacterial pellets were suspended in 1 ml lysis buffer (lysis buffer: 150 mM Sodium chloride, 2 mM EDTA, 1 mM NaF, 40 mM Na4P2O4, 0.1% SDS, 50 mM Tris pH 7.4, 1% glycerol, and 1% Triton X-100). The suspension was boiled for 5 min and sonicated three times for 20 s (Ultrasonic Apparatus XO-650, Xianou, Nanjing, China). The supernatant of sonicated proteins from bacteria were considered as the native OmpA lysates, which were prepared by breaking Acinetobacter pulsotypes.

As stated, OmpA of A. baumannii is closely similar to Escherichia coli OmpA and OprF of Pseudomonas aeruginosa [30]. The supernatants from sonicated E. coli (ATCC 25922) and P. aeruginosa (ATCC 27853) lysates were used as quality control to evaluate antibody reactivity with the samples loaded in the gel. Accordingly, bacterial lysates were loaded at 15 μg on each well of 10% bis–tris gel with SDS running buffer. After electrophoresis, resolved proteins were transferred from the gel onto Immobilon-PVDF blotting membrane (Millipore, Billerica, Massachusetts, USA). The membrane was blocked overnight at 4 °C with 5% non-fat dry milk (Sigma, Millipore, USA) in PBS-Tween 20 (0.05%). After gentle washing with PBS-Tween, 3F10-C9 mAb (20 μg/ml) was added to the membrane and incubated for 90 min at RT. The membrane was washed extensively with PBS-Tween and incubated with HRP-conjugated rabbit anti-mouse IgG (Avicenna research institute, Tehran, Iran) (1:2000) for 1 h at RT, followed by washing and developing with DAB (3,3′-Diaminobenzidine) detection system (Thermo Scientific, PI34002, USA).

Indirect immunofluorescence assay (IFA)

To detect surface binding of A. baumannii by 3F10-C9 mAb, an indirect immunofluorescence assay (IFA) was developed as previously described with modifications [11]. About 1.5 × 106 of each Acinetobacter pulsotype (isolates A to E) and also E. coli were prepared and washed twice with PBS pH 7.2, and then centrifuged at 5000 rpm for 5 min. After washing, bacterial cells were incubated with 100 µl of purified 3F10-C9 mAb (20 µg/ml) as the primary antibody for one hour at 37 °C. After washing twice with PBS-Tween, the subject microtubes were incubated with 100 µl of fluorescein isothiocyanate (FITC) conjugated sheep anti-mouse immunoglobulin as secondary antibody (Dilution 1:500; Avicenna research institute, Tehran, Iran) for 45 min. After washing with PBS-Tween, 50 µl of each sample was placed on the slide and covered with coverslip to observe microscopically under ultraviolet light.

Opsonophagocytic killing assay

The opsonization assay was performed to examine the antibody which act to coat the bacterial cell wall and prepare it for ingestion [11, 16]. Five pulsotypes of A. baumannii were cultured overnight in Luria–Bertani broth medium at 37 °C, then passaged to mid-log growth, rinsed, and suspended in a sterile PBS 1×. Murine macrophage RAW 264.7 cells were cultured at 37 °C in 5% CO2 in DMEM/F-12 growth medium (Gibco, USA) with 10% FBS. The RAW 264.7 cells were activated by four days of exposure to 1 µg/ml lipopolysaccharide (LPS) (Sigma-Aldrich, L2880, USA). LPS-differentiated RAW cells were transferred from the culture flask to a microtiter plate (~ 2 × 106 cells/ml, 80 µl/well). Then, five isolates of A. baumannii were added distinctly to each well of the plate (~ 1 × 106 CFU/ml, 10 µl/well). The 3F10-C9 mAb at a concentration of 50 µg/ml were then added to each well (10 µl/well) for 6 h at 37 °C. For complement studies, non-immune mouse serum was added to the wells (10 µl/well) in both of active and heat-inactivated forms in order to create conditions with complement and without complement, respectively. For quality control, Escherichia coli ATCC 25922 strain was used as a gram-negative bacterium that has a relative structural similarity to Acinetobacter [31]. (Notably, if available, the A. baumannii knock-out of the OmpA gene can be used for the quality control). In addition, as a negative control, a non-specific monoclonal antibody with the IgG isotype was used instead of the 3F10-C9 mAb in the presence and absence of the serum complement. After 6 h incubation of mixtures with gentle shaking, the supernatant of each well was removed and quantitatively plated in Mueller–Hinton agar. Then, the number of bacteria colony forming units (CFU) of each well were counted after an overnight culture and followed by calculation the approximate rate of bacterial mortality [32].

Statistical analysis

All statistics analyses were run using GraphPad Prism 8 software. Evaluation of the relationship between bacterial killing and type of opsonin substances (mAb or Isotype control antibody), in two groups with complement and without complement, was studied using the chi-square test separately for different pulsotypes of bacteria. Differences were considered significant if the p-value was < 0.05.