Introduction

Acinetobacter baumannii, a Gram-negative coccobacillus, has emerged as an aggressive pathogen due to its ability to adapt resistance against various antibiotics rapidly and hence declared as a priority pathogen by WHO (Howard et al. 2012; Jung et al. 2017). Its ability to persist in the hospital environment encourages nosocomial infections, including ventilator-associated pneumonia, meningitis, bacteremia, urinary tract infection, and wound infections in immune-compromised patients (Greene et al. 2016a, b; Wong et al. 2017; Chapartegui-González et al. 2018). Biofilm formation by A. baumannii is a multifactorial complex process comprising initial adhesion by csuABCDE operon encoded chaperon usher (csu) pili and outer membrane protein A (OmpA); further maturation is facilitated by secretion of matrix components such as Biofilm-associated protein (Bap), and polysaccharides like poly N-acetyl glucosamine and alginate encoded by pgaABCD operon and algC, respectively (Gedefie et al. 2021). Among the A. baumannii population, these factors are reported to be conserved among isolates belonging to different clonal complexes; however, heterogeneity persists in the genes encoding for OmpA and Bap (Viale and Evans 2020; De Gregorio et al. 2015). Bap, a hydrophobic cell surface protein, plays a crucial role in the adhesion of A. baumannii on human cells and medically relevant substances such as polystyrene, polypropylene, and titanium (Brossard and Campagnari 2012; Goh et al. 2013).

A. baumannii encodes for large multi-domain Bap with varying molecular weight of up to 879.60 kDa, secreted extracellularly and was identified as a homolog of Bap produced by Staphylococcus aureus (Loehfelm et al. 2008). In A. baumannii strain 307 − 0294, Bap was verified to facilitate the formation of multi-dimensional structures and water channels within biofilms, leading to mature biofilm. Conversely, the multi-domain mutant A. baumannii strain 307 − 0294 failed to produce mature biofilms and remained in a single layer, indicating its role in biofilm maturation rather than the initial stage of adherence (Loehfelm et al. 2008; Brossard and Campagnari 2012).

Bap consists of tandemly arranged repeated domains. The first half comprises A to C modules arranged in a multi-domain fashion. The second half comprises 28 direct tandem repeats of module D. Each module belongs to an immunoglobulin-like fold superfamily (Loehfelm et al. 2008). A previous study has shown that genome sequences of many A. baumannii isolates available in the National Center for Biotechnology Information (NCBI) database harbor mutations within the Bap gene, resulting in the production of disrupted Bap; this might be attributed to the repetitive regions of this gene, which may hinder accurate assembly (Goh et al. 2013). Despite the conservation of the majority of components involved in biofilm formation, the ability of A. baumannii to form biofilms varies within its population. This variation in phenotype variation may be attributed to various extrinsic and intrinsic factors including the genomic composition that describes the clonal complexity of the strain (Eze et al. 2018). Variations in the number and type of tandem repeats comprising Bap have been observed. Besides clonal complexity, variations in number and type of tandem repeats comprising Bap have been observed. The different Bap variants are categorized into four types based on their unique Ig-like domains (Loehfelm et al. 2008; De Gregorio et al. 2015).

It has been now recognized that different strains of A. baumannii elicit different immune responses in mice (De Breij et al. 2012; Kale et al. 2017; Morris et al. 2019). A study previously demonstrated that A. baumannii AB5075 interacts with carcinoembryonic antigen-related cell adhesion molecules (CEACAM) -1, -5, and − 6 present on the surface A549 cells, leading to the internalization of A. baumannii in a vacuole bound by the cellular membrane (Ambrosi et al. 2020). CEACAMs are expressed as transmembrane receptors on the surface of leukocytes, epithelial cells, and endothelial cells of blood vessels and play a vital role in adhesion, cell cycle regulation, insulin action, angiogenesis, and cancer (Singer et al. 2002; Öbrink 1997; Hammarström 1999; Bamberger et al. 2000; Najjar 2002). The role of adhesins and outer membrane proteins in binding to CEACAMs has been found in various bacterial pathogens including Helicobacter pylori (HopQ), Neisseria meningitis and Neisseria gonorrhoeae (colony opacity-associated, Opa proteins), Haemophilus influenza (Omp P1 protein), Streptococcus agalactiae (β-protein) and Streptococcus pyogenes (R28) (Baranov and Hammarström. 2004; Königer et al. 2016; Kuespert et al. 2011; Mix et al. 2021; Catton et al. 2023). While various studies have confirmed that A. baumannii can attach to human epithelial cells in the lungs and skin, the insights into the host-pathogen interaction involving adhesins like Bap that may facilitate A. baumannii colonization and internalization are not yet well established (Chen et al. 2022; Maure et al. 2023). Contradictory to these studies, an in vitro study showed that five A. baumannii and six A. pittii isolates of clinical origin failed to adhere and induce cytotoxicity to lung epithelial cells (Lázaro-Díez et al. 2016).

In this study, we attempt to identify differences among types of Bap to understand the evolutionary changes among A. baumannii that may favor biofilm formation and, subsequently, the pathogenesis. Further, using in silico approaches, we predicted the partial structure of the Bap protein, comprising multiple domains, and studied their interaction with human CEACAM-1 present on the epithelial cells and polymeric immunoglobulin receptor (PIgR) expressed on the mucosal surface of epithelial cells of pharynx and gut region. These interactions may induce a chain of signaling cascades that may benefit the pathogen by allowing it to survive during adhesion and evade immune response.

Methods

Retrieving genome sequences and classification based on bap type

Genome sequences of 7338 A. baumannii isolates available on the NCBI database until Feb 2023 were downloaded. tBlastn was performed to elucidate the distribution of different types of BAPs among the A. baumannii population. Query sequences for different types of Bap were obtained from a previous study (De Gregorio et al. 2015). An identity score of 90% or higher was used to select Bap sequences for analysis. The isolation source of isolates was also recorded from the NCBI database. Further, the Sequence Type (ST) of strains was determined according to the Pasteur scheme using the genome sequence as input on online available tool Public databases for molecular typing and microbial genome diversity (https://pubmlst.org/organisms/acinetobacter-baumannii).

Physicochemical analysis of types of Bap

Physicochemical characterization of different types of Bap was performed using ExPASy ProtParam online tool (https://web.expasy.org/protparam/) (Gasteiger et al. 2005). Different types of Baps were compared on the basis of a number of amino acids, molecular mass, isoelectric point, aliphatic index, grand average of hydrophobicity (GRAVY), instability index, extinction coefficient, and Type-1, Type-2, and Type-3 Bap were compared based on these factors. The amino acid composition of Bap was also recorded.

Phylogenetic tree construction

The phylogenetic tree was constructed using the Bap amino acid sequence of 232 isolates, including strain AB5075 (with complete level genome sequence available at NCBI database) using the EMBL-EBI simple phylogeny online tool (https://www.ebi.ac.uk/Tools/phylogeny/simple_phylogeny). The data generated by the EMBL-EBI simple phylogeny tool was used to generate a diagram for a phylogenetic tree using iTol (https://itol.embl.de/).

Prediction of domain and motif function

Bap domains were predicted using amino acid sequences using the InterPro tool (https://www.ebi.ac.uk/interpro/), which predicts the family and superfamily of a particular domain. The predicted domains were further analyzed to identify the conserved sequences and evolutionary relationships. Motif Scan (https://myhits.sib.swiss/cgi-bin/motif_scan) was used to identify all the motifs occurring in Bap with the primary amino acid sequence as the input. Pfam database and Hidden Markov Model-based profiles were employed for scanning motifs in the Bap sequence.

In silico tertiary structure modelling using protein threading

The three-dimensional structure of types of Bap was predicted by protein threading using the Iterative Threading ASSEmbly Refinement (I-TASSER) tool, an online software that predicts structure based on multiple threading alignments, Local Meta- Threading Server (LOMETS). The amino acid sequence of a single copy of each multiple repeat sequence representing an immunoglobulin-like domain comprising the protein was used as input to generate the predicted structure (Wollacott et al. 2007). Five closest stereochemical structures were obtained, and the confidence of each model was quantitatively measured by confidence (C)-score. A higher C-score signifies higher confidence in the model. Subsequently, the quality of the structure was estimated by the C, template modelling (TM), and root mean deviation score (RMDS) scores.

Validation of predicted structures

The structures predicted by I-TASSER were further validated by using the online available SAVEs tool, which comprised of programs such as ERRAT (Colovos C and Yeates TO. 1993), VERIFY 3D (Bowie et al. 1991; Lüthy et al. 1992); and the stereochemical quality of protein was determined by analyzing the orientation of dihedral angles (torsion) angles phi (Φ) and psi (ψ) by generating Ramachandran Plot using PROCHECK analysis tool (Laskowski et al. 1998).

Prediction of protein-protein interaction of Bap and host carcinoembryonic antigen cell adhesion molecules (CEACAM-1) and polymeric immunoglobulin receptor (PIgR)

The interaction of Bap with CEACAM-1 and PIgR was performed using HDock online software (https://hdock.phys.hust.edu.cn/). The modelled structure of Bap generated by I-TASSER was used as a receptor, and the structure of CEACAM-1 (PDB ID: 4QXW) and PIgR (PDB ID: 1xed) available on RCSB Protein Data Bank was used. The top 10 structures predicting the interaction of host protein and Bap were displayed. The structure with the correct interaction stereochemistry was selected for further analyzing the interacting residues by LigPlot+ v.2.2.

Statistical analysis

GraphPad Prism version 5.1 was used for all the statistical analysis and plotting graphs. One-tailed t-test was used to analyze the differences in physicochemical properties among different Bap Types.

Result and discussion

Baps correlation with isolation source and sequence type (ST) in A. baumannii population

The biofilm-associated protein of A. baumannii includes a diverse range of Bacterial Immunoglobulin (BIg) domains that facilitate biofilm formation on abiotic and biotic surfaces. Bap contributes to A. baumannii pathogenesis by adhering to host epithelial cells. Initially, we determined the distribution of Bap among 7,338 A. baumannii genome sequences available on NCBI database until Feb 2023. The presence and classification of Bap were determined using tBlastn with query sequence reported previously by De Gregorio et al. 2015 on the extensive characterization of the genetic organization of Bap (De Gregorio et al. 2015). tBlastn results revealed that 87.52% (6,422) of analyzed sequences harbor heterogeneous Bap. Based on the amino acid composition described previously, A. baumannii genomes featuring Type-1, Type-2, and Type-3 Bap were classified and encoded by 12.12%, 68.44%, and 6.91%, A. baumannii strains, respectively (Fig. 1a). Type-6 or SDF-Type was found in 0.054% of A. baumannii strains, while 12.51% lacked the Bap or biofilm-like proteins (BLP) genes (Fig. 1a). Type-2 Bap differed from Type-1 due to the presence of Ig-like repeat, Z, distinct from D repeat identified in Type-1, while Type-3 Bap diverged from Type-1 and Type-2 with only 30–40% similarity in the amino acid sequence of the carboxyl terminal, at the end of G domain. (De Gregorio et al. 2015).

Further, we analyzed the isolation source of A. baumannii to determine if the presence of a particular Bap type is associated with specific clinical manifestations. The majority of A. baumannii strains carrying Type-1 Bap were retrieved from respiratory specimen (23.172%) and bloodstream (15.80%). Similarly, A. baumannii isolates encoding for Type-2 Bap were also predominantly obtained from respiratory specimens (34.42%) and bloodstream (13.97%). A significant proportion of A. baumannii carrying Type-3 Bap were primarily isolated from respiratory specimens (25.83%) and environment (12.82%). However, the isolation source of A. baumannii harboring Type-1, Type-2, and Type-3 of 22.94%, 33.07%, and 9.8% isolates, respectively, was not defined. These findings suggest that A. baumannii is more likely to infect upper respiratory tissues and organs, regardless of the Bap type, which is consistent with previous studies (Alsan and Klompas 2010; Howard et al. 2012; Zilberberg et al. 2016). The second significant infection site in the case of strains with Type-1 and Type-2 Bap was found to be blood or catheter-associated infections (Alsan and Klompas 2010). In contrast, many isolates (12.82%) harboring Type-3 Bap were also isolated from the environment (Fig. 1b).

Multilocus sequence typing (MLST) was analyzed at A. baumannii MLST website using the Pasteur scheme. In total, 889 isolates harboring Type-1 Bap belong to 113 different sequence types, and we found that ST1 and ST295 exclusively harbored Type-1 Bap with 20.80% and 11.58% isolates, respectively. Similarly, isolates presenting Type-2 Bap belong to 316 different STs, but 55.47% of the population belongs to ST2. In the case of isolates harboring Type-3 Bap, ST25 was more predominant with 33.92% of isolates (Fig. 1c). A. baumannii strains belonging to ST92, 296, 307, 318, 328, and 499 were mutual among all three Bap types. However, it was observed that isolates belonging to a particular ST were associated with a particular type of Bap, suggesting a correlation between the genetic background of bacterial strain, as characterized by its ST and the Bap variant it carries. Instances where ST has less than 10 isolates and conceals a single type of Bap are not shown in Fig. 1c. A previous study showed that strains belonging to ST2, ST25, and ST78 have a higher ability to form a biofilm (Giannouli et al. 2013). A separate study observed that sporadic strains had a significantly higher ability to form biofilms than epidemic isolates associated with ST2 (Hu et al. 2016).

Fig. 1
figure 1

(a) Distribution of Type-1, Type-2, and Type-3 BAP among various sequence types of A. baumannii. (b) Isolation source of A. baumannii harboring Type-1, Type-2, and Type-3 Bap. (c) Number of A. baumannii isolates with Type-1, Type-2, and Type-3 Bap belonging to different ST. Majority of isolates with Type-1, Type-2 and Type-3 Bap belong to ST1, ST2 and ST25, respectively

Full size image

Bap plays a crucial role in the evolution of virulent A. baumannii

Neighbor-Joining (NJ) tree was constructed using Bap amino acid sequences of isolates belonging to Type-1 (n = 68), Type-2 (n = 92), Type-3 (n = 70), and SDF-Type (n = 2). Type-1 Bap was the most ancient, followed by Type-2 and Type-3. SDF shared the common ancestors with Type-1 (Fig. 2a). Domain analysis of Bap protein among A. baumannii using InterPro server revealed five major domains, VCBS (Vibrio, Colwellia, Bradyrhizobium, and Shewanella), BapA, Bacterial immunoglobulin-like domains 6, 12, and 13 (BIg6, BIg12, and BIg13). BIg12 (Pfam ID-19078) and BIg13 (Pfam ID-19077) domains are also present in multiple tandem copies and are associated with bacterial surface proteins. SDF Type Bap exclusively comprised of only the BIg6 domain (Pfam ID-17936), reported in several extracellular proteins produced by bacteria and is found in multiple consecutive repeats. These bacterial domains have been reported in various extracellular bacterial proteins and are crucial for adhesion (Chatterjee et al. 2021). Further, we analyzed the conserved amino acid residues among different domains using the NCBI COBALT tool (Fig. 2b, c). The amino acid sequence “DADGTVGTGTVDADGTFS” and QANGE-TLSVTA—TDAAGNVSPA” among BIg6, BIg12, BIg13, BapA, and VCBS domain; amino acids SDDGTTLTGTGEAG, ATVTVT, DADGTVGTGTVDADGTFS, FTPTAAVAGAT-TLTVTA, and TDLAGNAGTG among BIg6, BIg12, and BIg13 were found to be conserved (Fig. 2c). Upon analyzing the neighborhood trees of these domains, it was found that BIg6 and BIg12 exhibited closer relationship, than BIg13 and VCBS (Fig. 2b).

The presence of sialate O-acetylesterase domain was analysed among 232 Bap sequences. Several sequences of Type-1 (n = 39/68; 57.35% of Type-1), Type-2 (n = 34/92; 36.95% of Type-2) and Type-3 (n = 4/70; 5.71% of Type-3) Bap were found, including the most recent strain ACIN00156 containing sialate O-acetylesterase domain (Pfam ID- IPR039329). Bacteria possessing O-acetyl sialate esterase(s) employ acetylated sialic acids as a source for growth, potentially securing a metabolic advantage over rivals that do not exhibit this enzymatic activity (Rangel et al. 2016). The sialidase activity is also responsible for the degradation of the mucosal barrier in the respiratory tract to promote pneumonia as reported during S. pneumoniae infection (Kahya et al. 2017). It can be hypothesized that the evolution of A. baumannii Bap could also have an immunomodulatory role during human infection.

Fig. 2
figure 2

a.Phylogenetic tree analysis of Bap types. The protein sequence code in green belongs to Bap Type-1, Type-2 is depicted with yellow, and Type-3 and SDF type are shown in violet and red, respectively. b and c. Highly conserved and less conserved amino acid positions based on relative entropy threshold among identified Bap domains are highlighted in red and blue, respectively. The neighbor-joining tree of identified Bap domains showed that VCBS and BIg13 are more closely related. c. and BIg6, and BIg13, were more closely and distantly related to BIg12

Full size image

Variations in the physicochemical properties of baps

The physicochemical analysis of different types of Bap encoded by 201 representative strains (57 Type-1, 69 Type-2, 75 Type-3) using Expasy ProtParam revealed the heterogeneity of this protein. Significant differences in molecular weight, isoelectric point, and instability index were observed among Type-1, Type-2, and Type-3 Bap (Fig. 3). Bap among A. baumannii was found in molecular weights ranging from 73.930 kDa to 872.555 kDa of strain XH960 and 9120, respectively. The isoelectric point (pI) of the protein ranged from 2.74 to 4, indicating the overall negative charge and acidic nature of the protein. The Type-1 Bap proteins were highly stable with an instability index ranging from 3.58 to 6.73, Type-2 ranging from 5.1 to 10.56, Type-3 ranged from 12.04 to 32.4, and the Type 6 Bap was unstable with an instability index of 59.65 to 67.62 for strains SDF, AB046, AB048 and AB053 (not shown). Type-1 Baps encoded by A. baumannii were found to have the highest average molecular weight (670935.88 ± 163028.54 Da), lowest isoelectric point (3.15 ± 0.06), and higher stability (Instability index = 4.83 ± 0.90). In comparison, Type-2 and Type-3 Bap had significantly lower average molecular weight (480149.94 ± 190419.45 Da and 249574.84 ± 134585.79 Da, respectively), higher isoelectric point (3.42 ± 0.18 and 3.57 ± 0.16) and were less stable (with instability index of 8.12 ± 1.64 and 15.46 ± 4.80), respectively (Fig. 3a, b, c). Grand average of hydropathicity (GRAVY) of Type-1 BAP protein was 0.071 to 0.168. Type-2 0.079 to 0.193, Type-3 -0.188 to 0.271 (Fig. 3d). The wide-ranging differences in GRAVY of Bap among A. baumannii show that the protein can be either hydrophilic or hydrophobic isolates. Although most Bap variants, including Type-1 and Type-2, were hydrophobic, some strains encoded for hydrophilic Type-3 and Type-6 Bap.

Fig. 3
figure 3

Differences in physicochemical properties of Type-1, Type-2, and Type-3 Bap on the basis of their (a) Molecular weight, (b) Isoelectric point, (c) Instability index and (d) Grand average of hydropathicity (***, p value < 0.001)

Full size image

Analysis of the amino acid composition of Baps showed that threonine, a hydrophilic amino acid containing the hydroxyl group in its side chain, was the most abundant amino acid in Type-1, -2 and − 3 Bap accounting for 12.4–18.5% of amino acids within the analyzed protein sequences (Fig. 4a). This observation aligns with previous findings, although the specific significance of the abundance of threonine remains undefined (Yousef et al., 2007). Later, it was postulated that phosphorylation on the hydroxyl group of threonine may play a role in pathogenesis during adherence on host cells, however, further investigation is required (Ribet and Cossart 2010). The percentage of threonine in Bap Type-3 was significantly less than that of Type-1 (p-value = 0.0333) and Type-2 (p-value = 0.0261) (Fig. 4b). Serine, (containing hydroxyl group) was found to have significantly higher levels in Type-3 (p-value = 0.0040) and Type-2 (p-value = 0.0172) as compared to Type-1 Bap (Fig. 4c). Leucine, a non-polar, aliphatic amino acid, was also significantly abundant in Type-3 Bap as compared to Type-1 (p-value = 0.0082) and Type-2 (p-value = 0.0133) (4d). Since we found increased levels of leucine in Type-2 and Type-3 Bap, we screened Bap sequences for leucine-rich repeats (LRR) and found that Type-2 and Type-3 Bap contained 4–5 LRR motifs at the carboxyl-terminal region of Bap. These LRR might play role in mediating protein-protein interactions. Further studying the role of these amino acids among different Baps would provide valuable insights into their role during host-pathogen interaction. Besides that, hydrophobic amino acids glycine, alanine, and valine were found to be conserved among these three Baps. Negatively charged aspartate and glutamate constituted 9-11.5% and 1.4–3.2% of the protein, respectively. Sulphur containing amino acid cysteine was completely absent, but methionine was found in traces constituting 0.1–0.5% of the protein.

Fig. 4
figure 4

(a) Variations in amino acid composition among Bap/s. (b) Significant reduction of threonine, with the increase in (c) serine and (d) leucine percentage among Baps with its evolution. (**, p value < 0.01; *, p value < 0.05)

Full size image

In silico structure modelling of Baps

The experimental three-dimensional structure of Bap of A. baumannii is yet to be reported. For predicting the 3-D structure of each type of Bap, the repeats from the same type of Bap constituting different types of domains were merged into a single sequence, and the structure was predicted using I-TASSER. For each sequence submitted, I-TASSER generated the five most favorable structures based on protein threading. Structures with the highest Ramachandran favored score were selected (Fig. 5). The estimated TM score, estimated RMSD, and Ramachandran score of each selected structure are mentioned in Table 1. The predicted structure revealed that the multi-domain structure comprised antiparallel β-sheets constituting the β-sandwich domain. The multiple subunits were linked with the alpha-helical structures constituting the multi-domain structure of the protein.

Fig. 5
figure 5

Predicted 3-Dimentional structures comprising single unit of multiple repeats: (a) AYE (Type-1), (b) ACICU (Type-2), (c) AC1633 (Type-3), (d) SDF (Type-6) and their Ramachandran plots (e, f, g and h, respectively)

Full size image
Table 1 Validation scores of predicted 3D structures of types of Bap
Full size table

Evolution of Bap facilitates improved interaction with CEACAM-1 and PIgR

Previous studies have shown that Bap facilitates A. baumannii adhesion to human epithelial cells, and another study found that A. baumannii strain 5075 that contains Type-1 Bap (sequence ID: AKA30645.1) interacts with the CEACAMs receptors (Brossard and Campagnari 2012; Ambrosi et al. 2020). BAP sequence of the A. baumannii isolate AYE (GenBank ID: CAM85746.1) was used as the representative sequence for investigating the interaction of Type-1 BAP and CEACAM. Bap of this isolate exhibits 95% sequence similarity with that of strain AB5075. CEACAMs have been found as target molecules by various pathogens, such as capsule-producing N. meningitidis, H. influenzae, and Moraxella catarrhalis, that infect the nasopharynx (Villullas et al. 2007). Interaction of opportunistic pathogens with CEACAMs has led to tissue invasion and migration across the epithelial layer (Hill et al. 2005; Sheikh and Fleckenstein 2023). The BIg13 domain and BIg6 domain of Type-1, -2, -3, and SDF-Type Bap, respectively, were found to target the N-terminal dimer of CEACAM-1 receptor by hydrophobic interactions and ionic interactions. The BIg 13 domain of AYE was 100% similar to that of strain AB5057, that was previously shown to interact with CEACAMs. This finding corroborated with the earlier observations where the BIg13 domain of the R28 receptor expressed on the surface of Streptococcus pyogenes, which causes puerperal sepsis and other Gram-positive bacteria expressing BIg13 interacts with dimeric CEACAM-1 receptor (Catton et al. 2023; van Sorge et al. 2021). Three residues of both the N-terminal chains of dimeric CEACAM-1 (Asn23, Glu37, Asn81of chain B and Asn81, Arg43, and Asp40 of Chain C) interacted with Gly221, Val169, Asp162, Asn200, Asp164 and Thr167 of BIg13 domain of AYE Bap along with several other hydrophobic interactions (Supplementary Material, Fig. S1). In the case of docking between ACICU Bap and CEACAM, interactions between Asp125 and Arg38 of chain B and Asp125, Ser123, Asp99, Thr119, and Leu50 interacted with Arg38, Arg43, Thr83, Val106, and Gln103 of chain B of CEACAM-1 (Supplementary Material, Fig. S2). Interaction between Type-3 Bap of AC1633 and CEACAMs involved the highest number of residues among all the Bap types. Ser93, Ala2, Asn88, and Ser90 interacted with Arg38, Gln103, and Glu37 of Bap and CEACAM-1, respectively (Fig. 6a). Since A. baumannii majorly infects the respiratory tract, specifically the nasopharynx region, it was of interest to understand the interaction of Bap with PIgR that mediates the secretion of antibodies on the mucosal lining of the nasopharynx region. Using the HDock tool online, in silico docking of Bap with PIgR showed that Bap interacts with different binding energies. It was observed that all types of Bap interact with the CEACAM-1 receptor and PIgR but with varying binding energies and interacting residues. Table 2 summarizes the binding energies of interaction studies and docking parameters. A previous study also showed that Streptococcus pneumoniae interacts with PIgR, specifically with its choline-binding protein A (Brock et al. 2002). Another study showed that a pneumococcal surface protein, pspC, interacts with PIgR and facilitates the invasion of S. pneumoniae into host cells (Agarwal et al. 2010). The amino acid residues Thr, Asp, Ala, and valine of Bap involved in interaction with CEACAM-1 and PIgR reflect the reason for their abundance in the protein (Fig. 6b). Analyzing the differences in the interaction of Baps with CEACAM-1 and PIgR, it can be hypothesized that A. baumannii Bap is evolving to facilitate the enhanced virulence of this pathogen. Previous studies have shown that A. baumannii Bap facilitates adhesion to human bronchial epithelial and neonatal keratinocyte cells but inhibits intracellular invasion of A. baumannii in these cells (Brossard and Campagnari 2012). To advance our understanding of A. baumannii’s invasion of host cells, further research on the potential of different Bap types is crucial. Identifying the specific domains responsible for this mechanism and studying the role of different Bap types in eliciting varied immune responses would pave the way for developing targeted therapeutics. Such studies would significantly contribute to the development of effective treatment options for A. baumannii infections.

Table 2 Summary of docking models
Full size table
Fig. 6
figure 6

(a)The figure illustrates the interaction between predicted 3-D structure of Type-3 Bap (chain A ) of A. baumannii isolate AC1633 and N-terminal of human CEACAM-1 (chain B and C) via its BIg13 domain (b) Predicted 3-D structure of Type-3 Bap (Chain Z) of A. baumannii isolate AC1633 shows interaction with E domain of PIgR (Chain E) using hybrid docking strategy. The interface residues predominantly showed hydrophobic and ionic interactions.

Full size image

Conclusion

A. baumannii, once considered non-pathogenic, has rapidly transformed into a significant contributor to healthcare-associated and community-acquired infections. The biofilm-forming ability has played a significant role in transforming A. baumannii into a pathogen and in the development of multidrug resistance. However, our understanding of mechanisms underlying A. baumannii infection and interactions with the host contributing to its pathogenicity is inadequate. In this study, using in silico approaches, we illustrated the structural aspects of biofilm-associated protein that play a crucial role in establishing and forming mature biofilms during infection. This study reports the differences in Bap types by predicting their partial structure, amino acid composition, physicochemical properties, and domain analysis. Furthermore, we speculated the differences in their ability to interact with the host CEACAM-1 and PIgR receptors and found that the recently evolved Bap Type-3 showed more efficient interaction, thus advancing our knowledge about the pathogenesis of this rapidly evolving pathogen. A. baumannii strain AB5075 has been shown to interact with the CEACAMs, previously (Ambrosi et al. 2020). Our analysis revealed that strain AB5075 carried Type-1 Bap with sequence ID AKA30645.1, depicted in the phylogenetic tree. However, future studies should focus on the role of Bap mediating this interaction and internalization of A. baumannii. Investigating the impact of elevated levels of serine and leucine, precisely leucine-rich repeats identified in the carboxyl region of some Type-2 and Type-3 Bap will offer insights into their regulatory mechanisms or their function in adhesion. Since the BIg13 domain played a major role in facilitating Bap-CEACAM-1, constructing BIg13 mutants and its impact on biofilm formation ability on biotic and abiotic surfaces would be important. It would be helpful to study the interaction between Bap and PIgR to understand the strategies used by A. baumannii to evade the immune system, which can contribute to its ability to cause disease. By examining the role of Bap domains in invading host cells and evading the immune system, we can gain insights into developing targeted therapies. Further research is needed to understand the signaling cascade resulting from this interaction, which could give the pathogen an advantage by facilitating its survival during initial adhesion and allowing it to evade the host’s immune responses.

Discovering the evolutionary and pathogenic role of A. baumannii Bap is a critical step in developing targeted therapies against this dangerous pathogen. Using in silico approaches, this study successfully characterizes Bap and unravels its crucial role through detailed structure modelling. The findings of this study have significant implications for the development of more effective treatments against A. baumannii infections.