Specific egg yolk immunoglobulin as a promising non-antibiotic biotherapeutic product against Acinetobacter baumannii pneumonia infection

Acinetobacter baumannii is a serious health threat with a high mortality rate. We have already reported prophylactic effects of IgYs raised against OmpA and Omp34 as well as against inactivated whole-cell (IWC) of A. baumannii in a murine pneumonia model. However, the infection was exacerbated in the mice group that received IgYs raised against the combination of OmpA and Omp34. The current study was conducted to propose reasons for the observed antibody-dependent enhancement (ADE) in addition to the therapeutic effect of specific IgYs in the murine pneumonia model. This phenomenon was hypothetically attributed to topologically inaccessible similar epitopes of OmpA and Omp34 sharing similarity with peptides of mice proteins. In silico analyses revealed that some inaccessible peptides of OmpA shared similarity with peptides of Omp34 and Mus musculus. Specific anti-OmpA and anti-Omp34 IgYs cross-reacted with Omp34 and OmpA respectively. Specific IgYs showed different protectivity against A. baumannii AbI101 in the murine pneumonia model. IgYs triggered against OmpA or IWC of A. baumannii were the most protective antibodies. IgY triggered against Omp34 is ranked next after those against OmpA. The lowest protection was observed in mice received IgYs raised against the combination of rOmpA and rOmp34. In conclusion, specific IgYs against OmpA, Omp34, and IWC of A. baumannii could serve as novel biotherapeutics against A. baumannii pneumonia.

www.nature.com/scientificreports/ is needed. Moreover, a high amount of specific IgY could be recovered from the egg yolk of immunized hens 13,14 .
Anti-Pseudomonas IgY has been successfully applied in clinical trials 15 . This specific IgY could modify bacterial fitness, increase bacterial hydrophobicity, and facilitate the formation of immobilized bacteria in aggregates. It could enhance bacterial killing via Polymorphonuclear neutrophils (PMN) 16,17 . The anti-P. aeruginosa IgY could also significantly reduce the bacterial burden and inflammatory cytokines in the lung of a murine (BALB/c) pneumonia model 18 . Protective effect of specific anti-Acinetobacter IgYs raised against OmpA, Omp34 (also known as Omp34kDa or Omp33-36), or inactivated the whole-cell of A. baumannii ATCC 19606 was demonstrated in our recent study 19 . OmpA and Omp34 are major virulence factors of A. baumannii involved in the bacterial adhesion to the human lung epithelial cell line i.e. A549 cells 20 . These virulence factors are highly immunogenic proteins 10,[21][22][23] . OmpA, one of the most abundant OMP of the pathogen 24 , is the most promising A. baumannii immunogen amongst introduced protein antigens 25 . Omp34 is a specific antigen suggested to be appropriate for diagnostic kits development 23 . This OMP is a 14-stranded barrel 26 harboring loops sharing similarity with human peptides 27 . Despite higher absorbance in ELISA, specific IgY triggered against the combination of OmpA and Omp34 unexpectedly exacerbated disease in comparison to other IgYs 19 . In addition to the study, which was the first report on antibody-dependent enhancement (ADE) of A. baumannii infection, recently, this phenomenon has also been reported on a mAb developed against K2 capsular polysaccharide of A. baumannii 28 . The study suggests that the capsule could act as decoys interacting with the mAb 28 . Administration of anti-Acinetobacter IgYs before the challenge with viable pathogens could cause protection. However, the post-challenge administration of the IgY needs to be investigated. The occurrence in the murine pneumonia model of ADE is not clear. This phenomenon was attributed to the existence of similar epitopes within OmpA and Omp34 topologically are inaccessible 19 . As a complementary hypothesis, some of these inaccessible epitopes of OmpA and Omp34 are similar to the epitopes of mice proteins. The present study is designed to evaluate the therapeutic effects of specific IgYs against A. baumannii infection in a murine pneumonia model. The abovementioned hypothesis is also analyzed by an in silico approach. Also, cross-reaction of the specific IgYs with each other was assessed by immunoassay tests. The results of this study could elaborate the role of IgYs as novel efficient treatment options to apply against A. baumannii infections. To the best of our knowledge, this is the first report on the therapeutic effects of the specific IgYs against A. baumannii.

Materials and methods
In silico analyses. In silico analyses were performed on the probable similarity of some of the inaccessible epitopes of OmpA and Omp34 with the epitopes of mice proteins. The sequences of OmpA and Omp34 26,27,29 were aligned by PRALINE 30 at http://www.ibi.vu.nl/progr ams/prali newww /. The primary sequence similarity was carried out as follows: the mature Omp34 sequence was spliced to 12-residue peptides offset by 3 amino acids; the obtained peptides were searched within the OmpA sequence via Epitope Conservancy Analysis tool at http://tools .iedb.org/conse rvanc y/. The peptides (from Omp34 and OmpA) sharing > 40% identity were subjected to a BLAST search (with ≥ 70% identity) at http://www.iedb.org/ to find similar linear epitopes in Mus musculus. The topology of the peptides was elucidated from our previous in silico studies 26,27,29 . Immunoassay tests. The IgYs were prepared as described earlier 19 . Western blotting was carried out as follows: the recombinant proteins, rOmpA, and rOmp34 were loaded on 8% SDS-PAGE. The proteins were then transferred onto a nitrocellulose membrane via electro-blotting at 75 V for 2 h followed by blocking of the membrane with 5% skimmed milk in phosphate-buffered saline with Tween-20 (PBST) with gentle shaking overnight at 4 °C. The membrane was washed thrice with PBST and was then incubated with specific IgYs raised against rOmpA, rOmp34, or Inactivated Whole Cell, with gentle shaking for 2 h. A dilution of IgYs at OD 450 of ~ 1.5 that was already observed in indirect ELISA 19 was used. The specific IgYs were diluted in PBST. The washing step was repeated and the membrane was then submerged in HRP-conjugated rabbit anti-chicken IgG antibody (Sigma-Aldrich) (1:2000 diluted in PBST) with gentle shaking for 2 h.
Sets of indirect ELISA were harnessed to assess the immunoreactivity and cross-reactivity of IgYs. The reactivity of the IgYs was initially evaluated against the whole cell of A. baumannii ATCC 19,606. The wells were coated with A. baumannii at a concentration of 3.6 × 10 7 CFU/well suspended in 50 mM sodium carbonate buffer, pH 9.6, and were then incubated at 4 °C overnight. The wells were washed 3 times with PBS containing 0.05% Tween 20 (PBST). Blocking buffer (100 μL of PBST containing 5% skim milk) was added to the wells followed by incubation at 37 °C for 1 h. The wells were washed thrice with PBST followed by the addition of 100 μl specific IgY solutions prepared in PBS (5 and 2.5 µg/well). The plate was then incubated for 1 h at 37 °C. The washing procedure was repeated and 100 μL of HRP-conjugated rabbit anti-chicken IgG antibody (Sigma-Aldrich) diluted to 1:1500 in PBST was added to the wells. The plate was incubated at 37 °C for 1 h, washed with PBST 3 times followed by the addition of 100 μL of TMB solution. The plate was kept in dark at room temperature for 15 min to develop a minor color change in control wells. 100 μl of 3 M H 2 SO 4 was added to each well to stop the reaction. The optical density was measured at 450 nm on an ELISA reader. Another ELISA test was designed to assess the reactivity of specific IgY-IWC (IgY raised against Inactivated Whole-Cell) against recombinant OmpA and Omp34 proteins. The ELISA was performed as mentioned above except for the antigen coating step. 2 µg of the recombinant OmpA or Omp34 suspended in 50 mM sodium carbonate buffer, pH 9.6 was coated in each well. To validate the synergic effect of the combination of rOmpA and rOmp34 as antigens administered for hen immunization, an ELISA test was performed. In this ELISA, rOmpA or rOmp34 (2 µg/well) were coated as antigens. IgYs (0.625, 0.312 or 0.157 µg/well) added to the wells were as follow: IgY-A (IgY raised against rOmpA), IgY-34 (IgY raised against rOmp34), IgY-A + 34 (IgY raised against the combination of rOmpA and rOmp34) Clinical strain. For a challenge with clinical strain, AbI101 nominated as the most fatal among the six evaluated strains 19 , was selected. The bacterial LD 50 was determined in the murine pneumonia model. In this regard, five groups (n = 6 mice/group) of neutropenic BALB/c mice were studied. 20 µl of bacterial suspensions in PBS ranging from 1.18 × l0 6 to 6 × l0 8  Two control groups only received bacteria or IgY. All administrations were carried out via the intranasal route. The mice were monitored for 8 days. The survivors were euthanized and their spleens and lungs were aseptically excised, weighed, and homogenized in 0.9% sterile NaCl. 100 µl from each tissue dilutions was plated on LB agar medium, incubated at 37 °C overnight to determine CFU/ Tissue. In addition to survival monitoring, clinical symptoms (weight loss, piloerection, clustering, hypothermia and tachypnea) as well as eye condition (a range between no sign and closed septic eyes) of the mice were also considered.

Statistical analyses.
Data were subjected to analysis of variance and Kruskal-Wallis test using SPSS25.
Duncan's multiple range and non-parametric log-rank tests were performed for comparison of means at P ≤ 0.05. The graph pad prism software version 8 was used for drawing the graphs.

Results
Similarity sharing of inaccessible peptides of OmpA with the peptides of Omp34 and Mus musculus. The protein sequences of OmpA and Omp34 shared a 10% sequence identity. Peptides of Omp34 were aligned with OmpA to find potential epitope similarity. Twenty-one 12-meric peptides shared > 40% identity with OmpA peptides. To find similar linear epitopes in Mus musculus, these peptides served as queries in BLAST searches. None of the Omp34-derived peptides matched with linear epitopes of Mus musculus. However, two topologically inaccessible peptides (LSLRTEARATYN and PVEPTPVAPQPQ) of OmpA shared similarity with 2 and 18 linear epitope identifiers (IDs) of Mus musculus respectively (Table 1). These murine epitopes were of CD180 antigen, Myelin-associated neurite-outgrowth inhibitor, Hematological and neurological expressed 1 protein and Tyrosine-protein phosphatase non-receptor type 23. Table 2 shows the topology of OmpA-and Omp34-derived peptides.
Cross reaction of the specific anti-OmpA and anti-Omp34 IgYs with Omp34 and OmpA. The reactivity of the specific IgYs was evaluated by western blotting and various ELISAs. In the performed western blotting, IgY-A and IgY-34 showed significant reactivity with their corresponding antigens (rOmpA and rOmp34 respectively). IgY-IWC recognized both recombinant OMPs and developed sharp bands. IgY-A and IgY-34 showed cross-reactivity with rOmp34 and rOmpA respectively (see Supplementary Fig. S1).

Discussion
Our previous publication was the first report on the protective effects of nasal administration of specific IgYs against A. baumannii 19 . However, the therapeutic (post-challenge) effects of these antibodies are yet to be addressed. Synergic effect of OmpA and Omp34 had been observed regarding their ability to trigger antibody responses. In contrast, mice received specific IgYs, raised against combination of OmpA and Omp34, showed  www.nature.com/scientificreports/ ADE of A. baumannii infection. Although a hypothesis had been suggested for these phenomena, the reason behind these observations remains to be elucidated. Since no additional studies had been carried out to explore these contradictory phenomena (synergic effect and ADE), the current study was conducted. In addition to evaluation of therapeutic (post-challenge) potential of specific IgYs (developed against rOmpA, rOmp34, and inactivated whole-cell of A. baumannii in a murine pneumonia model), the in silico analyses, immunoassays and a part of challenges are designed to explore details and to propose reasons for the observed phenomena. The synergic effect of OmpA and Omp34 from antibody triggering point of view was attributed to the existence of similar epitopes in OmpA and Omp34 19 . Similar peptides within OmpA and Omp34 sequences, revealed by in silico analyses, justify the synergic effect of these antigens in the elicitation of antibodies. The cross-reactivity of IgY-34 with OmpA, and IgY-A with Omp34 in the ELISA assay confirmed the existence of similar epitopes within the two OMPs. The higher cross-reactivity of IgY-34 with rOmpA in comparison to cross-reactivity of IgY-A with rOmp34 implies that similar epitopes in the context of the Omp34 sequence are probably more immunodominant than those presented in the context of OmpA sequence. ELISA results of IgY-A + 34 and IgY-mix proved obvious synergic effects of rOmpA and rOmp34 in terms of triggering specific antibodies. These results revealed that co-administration of these antigens could result in an increased number of similar epitopes followed by an increase in antibody response against these epitopes. This finding is in agreement with Lin et al. 22 who showed that the escalation of A. baumannii rOmpA antigen dose could alter immune polarization and immunodominant epitopes. The escalation of rOmpA antigen dose could enhance epitope spreading and Type 2 immune response. The reactivity of IgY-IWC with rOmpA and rOmp34 in the western blotting as well as ELISA indicates that these two proteins are expressed under in vitro culture conditions. This result is in line with the previous immuno-proteomic study conducted by Fajardo Bonin et al. 21 . OmpA has higher antigenicity than Omp34 19 . Moreover, OmpA is the most abundant OMP in A. baumannii; and its expression is significantly higher than Omp34 32 . So, the approximately twofold higher absorbance of IgY-IWC against rOmpA than IgY-IWC against rOmp34 in the ELISA could be due to higher antigenicity of rOmpA than rOmp34 as well as its higher number on the bacterial surface. The whole-cell ELISA results with IgY-A and IgY-34 also in support of the above. In addition to higher titers of specific anti-rOmpA and anti-rOmp34 IgYs in IgY-A + 34, the observed higher absorbance of IgY-A + 34 against whole cell antigen could be attributed to co-existence of exposed OmpA and Omp34 epitopes on the A. baumannii surface.
The specific anti-rOmpA and anti-rOmp34 IgYs raised against these OMPs, were purified and administered in denatured condition. Owing to perturbation of their structures and consequently their conformational epitopes, only IgYs raised against their exposed linear B-cell epitopes could theoretically recognize OmpA and Omp34 on the surface of A. baumannii (native proteins) and also could confer protection against A. baumannii 19 . In the pilot mice challenge, since no significant difference existed between the death time of mice receiving 40 or 100 µg of IgYs, the lower dose (40 µg) was selected for administration in further challenges. Cyclophosphamideinduced neutropenic mice are a suitable model to study A. baumannii pneumonia since this model is mimicking the opportunistic nature of this pathogen in the hospital setting 33 . OmpA plays pivotal roles in A. baumannii www.nature.com/scientificreports/ pathogenicity and its adherence to the epithelial cells 20 . This is the most abundant OMP in A. buamannii. So, higher protection observed in neutropenic mice receiving IgY-A is reasonable. A surprising result in our recent study was ADE caused by IgY-A + 34 19 which also occurred in the current study. Since both antigens are presented on A. baumannii, and also in concordance with the ELISA results, higher protection was expected as observed for OmpA, OmpC, and OmpF in Salmonella enterica 34 . In addition to the previous study 19 , a recently published report on ADE of A. baumannii infection 28 evaluates the therapeutic effect of a monoclonal antibody (mAb) against the K2 capsular polysaccharide of A. baumannii. Surprisingly, not only no protection was achieved; but also the mAb enhanced the infection of a clinical strain of A. baumannii in the mouse pneumonia model with increased mice mortality and bacterial burden in the lungs, spleens, and blood. The phenomenon was attributed to the interaction of the mAb with capsule shed as decoys. Although some possibilities expressed for the phenomenon, the mechanisms explaining observed ADE of infection for the pathogen have not been completely understood 28 . ADE had also been reported for lethal toxin activity of Bacillus anthracis via a subset of mAbs developed against its protective antigen. It seems that the observed ADE is to some extent epitope-specific 35 and a subset of mAbs developed against this antigen was toxin-neutralizing 35,36 . The observed ADE for the lethal activity of Bacillus anthracis toxin is Fcγ receptor-dependent 36 . This mechanism has also been suggested as a possibility for ADE of A. baumannii infection caused by mAbs directed against its K2 capsular polysaccharide 28 . However, this mechanism does not explain ADE observed in our studies; because IgY could not interact with the mammalian Fc receptor 13,37 . Hence, the existence of murine peptides as decoys for IgYs raised against epitopes, responsible for the observed synergic effect and cross-reactivity in ELISA, is more reasonable. The in silico analyses demonstrated that the majority of similar peptides particularly in OmpA are topologically inaccessible. Hence, if existence of decoys (murine peptides) is not a true assumption, higher doses of IgY-A + 34 is anticipated to compensate for idle specific IgYs generated against topologically inaccessible epitopes of OmpA and Omp34. Contradictorily, death of all mice receiving higher amount (80 µg) of IgY-A + 34 is supporting existence of murine peptide as decoys as shown by in silico analyses. Two epitopes of OmpA shared similarity with some proteins in mice. Some of these proteins are among immune system molecules. Some idle antibodies raised against the peptides topologically inaccessible could bind to some host immune system molecules such as CD180 antigen. One advantage of IgY is its ability to be triggered against highly conserved mammalian epitopes due to the genetic differences between chicken and mammals 38 . Although this advantage is favorable in diagnostic applications, it could be considered as a disadvantage in therapeutics. This issue may be considered from the safety point of view for therapeutic applications of IgY. Although the in silico results revealed existence of similar linear epitopes in mouse, further studies are needed to assess the immune response using the murine pneumonia model. This will unveil existence of any direct link between these epitopes and the antibody-dependent enhancement of the disease. Biophysical studies would add to dimensions of the molecular basis of cross reactivity in term of interaction of IgYs with OmpA and Omp34. In conclusion, the specific IgY antibodies raised against OmpA, Omp34, and inactivated the whole-cell of A. baumannii showed a therapeutic effect in the murine pneumonia model. Amongst, IgYs raised against OmpA or inactivated the whole-cell of A. baumannii had the highest protective effect in neutropenic mice. So, IgY could be employed as a novel effective natural biotherapeutics against pneumonia caused by A. baumannii. However, the ADE observed in mice receiving IgY-A + 34 highlights the importance of determination of epitopes responsible for this phenomenon. Designed antigens possessing no epitope responsible for the ADE could elicit safe protective antibodies. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.