Drug-resistant Acinetobacter baumannii: mortality, emerging treatments, and future pharmacological targets for a WHO priority pathogen

SUMMARY

Acinetobacter baumannii has emerged as a formidable global health concern and is a major contributor to infection-related mortality in critically ill patients worldwide. This versatile Gram-negative bacterium is notorious for its highly plastic genome, which enables the rapid emergence and dissemination of multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains, severely limiting the available treatment options. The genetic flexibility of A. baumannii underpins its arsenal of molecular mechanisms, enabling it to resist a range of antibiotics, from traditional agents to the latest therapeutic advancements available. With the progress made in treatments against Acinetobacter infections and various drugs undergoing clinical trials, the effectiveness of these treatments is often outpaced by the pathogen’s swift evolution of resistance, resulting in alarmingly high rates of treatment failure. In this systematic review of literature spanning 2004–2024, we highlight the high mortality rates associated with infections caused by XDR strains and carbapenem-resistant A. baumannii (CRAB). This review provides a comprehensive examination of the resistance mechanisms deployed by A. baumannii, encompassing both conventional antibiotics and novel agents used in global healthcare settings. In addition, we discuss emerging druggable targets and the inherent challenges in their development, offering strategic insights into next-generation therapeutic programs. A deep profound understanding of the pathogen’s molecular defenses is essential to guide the design of innovative therapies aimed at mitigating the escalating threat posed by A. baumannii.

INTRODUCTION

Acinetobacter baumannii was first described in 1968 (1). Subsequently, this pleomorphic Gram-negative coccobacillus has transformed from a relatively harmless environmental bacterium to a drug-resistant pathogen of global significance (2–4). This step change has been facilitated by a highly plastic genome, which contains the largest pathogenicity islands found in any human microbe (5). The efficient acquisition and regulation of resistance genes results in reduced susceptibility to modern antibiotics and xenobiotics, including heavy metals (5–7). Potential classes of antibiotics for the treatment of Acinetobacter infections include aminoglycosides, carbapenems, fluoroquinolones, β-lactam/β-lactamase inhibitor combinations, extended-spectrum cephalosporins, folate inhibitors, polymyxins, and tetracyclines (8). Unfortunately, resistance to multiple antibiotic classes is increasing (9). A. baumannii is predominantly a nosocomial pathogen, where it causes nosocomial pneumonia, skin and soft tissue infection (SSTI), surgical site infection, burns, device-related infection (e.g., central venous catheter), and less commonly urinary tract infections (UTI) (10–13). Community-acquired A. baumannii infections are largely confined to tropical regions where high population densities, vicinity to livestock, and veterinary use of antimicrobials may be relevant (14). Members of the Acinetobacter calcoaceticus-baumannii (ACB) complex include A. baumannii, A. nosocomialis, A. pittii, A. seifertii, A. dijkshoorniae, and A. calcoaceticus. Among these, A. baumannii is the cause of the majority of cases in global clinical settings, while the other members, especially A. calcoaceticus, are rarely encountered in the healthcare settings (15, 16) (Box 1).

Box 1. Microbiology, virulence, and antimicrobial resistance.

Phylogenetic clustering of the ACB complex reflects a unique combination of evolutionary pressures and genetic adaptations, distinguishing these species from their less pathogenic relatives (Fig. 1). The close genetic relationships within the ACB complex indicate shared adaptive traits that are likely to have been pivotal for survival in clinical settings and their extensive capacity to acquire antibiotic resistance (16). In this analysis, phylogenetic relationships were inferred using A. baumannii genes from a representative locus of the Pasteur MLST scheme. This approach facilitates species identification and broad epidemiological comparisons due to its simplicity, reproducibility, and standardization. However, it offers limited resolution for capturing genome-wide diversity compared to core genome SNP-based analyses. The exposure in soil to naturally occurring antibiotics produced by Streptomyces spp. and Actinomyces spp. may have facilitated the development of the molecular machinery required for adaptation to increasingly antibiotic-laden environments (17–21) (Fig. 2). The persistence and dissemination of A. baumannii in healthcare settings are shaped by a confluence of clinical and microbiological factors that merit greater attention. Selective pressure from widespread and often prolonged use of broad-spectrum antibiotics, particularly in intensive care units, creates an environment in which resistant strains are more likely to emerge and dominate (22). Adding A. baumannii's remarkable ability to acquire resistance genes through horizontal gene transfer, often involving co-circulating multidrug-resistant organisms such as carbapenemase-producing Enterobacterales (16). These exchanges are facilitated by shared ecological niches and mobile genetic elements, including plasmids and integrons. At the same time, the increasing complexity of modern healthcare, characterized by invasive interventions, extended hospital stays, and frequent inter-facility transfers, provides ample opportunity for the dissemination of high-risk clones. (23). The emergence of Acinetobacter spp. as nosocomial pathogens is further accelerated by the ability to resist desiccation and form biofilms (24). Evasion of host immunity, the ability to resist phagocytosis, and virulence pathways are additional traits that ensure Acinetobacter is a challenging nosocomial pathogen (25, 26).

Fig 1.

Phylogenetic clustering of the ACB complex.

Fig 2.

Evolutionary pathway of Acinetobacter baumannii adaptation from soil to clinical settings.

Here, we review the clinical outcomes and treatment options for drug-resistant Acinetobacter spp. We conducted a meta-analysis of Acinetobacter infection cases and treatment outcomes reported between 2004 and 2024. We review the currently available anti-Acinetobacter antibiotics, including those in phase I/II clinical trials. We consider the molecular mechanisms and networks that contribute to the emergence of drug resistance in A. baumannii. Finally, we consider new pharmacological targets that may be exploited in future anti-Acinetobacter drug development programs.

CLINICAL DISEASE AND MORTALITY

Prevalence, global distribution

Accurate estimates of global incidence and prevalence of A. baumannii-associated infections are impaired by incomplete and/or inconsistent surveillance data; however, recent reports suggest the prevalence of A. baumannii infections is rising. Undoubtedly, this is driven by changing clinical case mix and extensive antimicrobial use in humans, animals, and the environment. Some reports also suggest that climate change may also be implicated in the changing prevalence of community infections in the tropics (14, 27).

Drug-resistant disease

In addition to the overall rise in A. baumannii infection, the prevalence of drug-resistant disease is increasing (9). In 2021, an estimated six deaths per 100,000 population globally were caused by A. baumannii infection, of which 90% were associated with drug resistance (28). Hospital-acquired infection is driven by antimicrobial consumption in humans (specifically fluoroquinolones, cephalosporins, and carbapenems) and suboptimal antimicrobial stewardship and infection control measures (9, 14, 29, 30). Globally, the highest rates of resistant Acinetobacter infections are reported in India, central and Southeast Asia, the Mediterranean, the Middle East & North Africa, and South America, where rates of carbapenem resistance are >50% (Fig. 3) (9, 28, 31, 32).

Fig 3.

Map displaying the nation-wise percentage of Acinetobacter baumannii infections resistant to carbapenems in 2021 (post-pandemic). These estimates have been obtained from the Measuring Infectious Causes and Resistance Outcomes for Burden Estimation (MICROBE) platform, produced using data sets from the Global Burden of Disease study (33).

Carbapenems are the antibiotics of choice for the treatment of MDR Acinetobacter spp. CRAB is increasingly a global health concern (Fig. 3) and is classified as critical on the updated World Health Organization (WHO) Priority Pathogen List (PPL) (34). CRAB is defined by non-susceptibility to any anti-pseudomonal carbapenem, including imipenem, meropenem, and doripenem. In addition to carbapenems, CRAB isolates often exhibit resistance to other antibiotic classes, leading to MDR, XDR, or PDR phenotypes. Multidrug-resistant A. baumannii (MDRAB) is defined as isolates that are non-susceptible to one or more agents in three or more of the previously described antibiotic classes. Extensively drug-resistant A. baumannii (XDRAB) refers to isolates susceptible to one or more agents in a maximum of two antibiotic classes. Of increasing concern is the emergence of pan-drug-resistant A. baumannii (PDRAB) characterized by non-susceptibility to all available agents (8).

Mortality

A systematic review and meta-analysis of 121 studies caused by Acinetobacter spp. had an overall weighted pooled mortality of 43% (95% CI 38%–48%) (Fig. 4A). Notably, species-level identification of Acinetobacter spp. was not feasible in clinical practice until the introduction of routine MALDI-TOF and molecular identification. Proportion mortality rates for BSI and LRTI (Fig. 4B C) are similar, at 45% (95% CI 38%–52%) and 48% (95% CI 37%–60%) respectively. The weighted pooled mortality estimate for drug-sensitive Acinetobacter infections is 23% (95% CI 12%–38%) (Fig. 4D), which is considerably lower than the estimates for carbapenem-resistant and pan-drug-resistant infection, which are 52% (95% CI 44%–60%) and 75% (95% CI 65%–83%), respectively (Fig. 4C F). Further details of study populations included in the meta-analysis are provided in Supplementary methods. The meta-analysis does not stratify mortality by severity of underlying infection or time from onset of infection until initiation of appropriate therapy, as this was not detailed in all included studies. Given that Acinetobacter spp. infections are primarily hospital-acquired and typically occur in patients with underlying physiological instability (i.e., in critical care settings), we estimate underlying mortality risk is highly elevated in the cohort. In addition, the impact of delay to appropriate therapy initiation cannot be determined by this meta-analysis. Nonetheless, the findings demonstrate the elevated mortality risk in all Acinetobacter spp. infections, further increased by drug-resistant infections.

Fig 4.

Forest plots depicting the weighted pooled proportion in-hospital mortality (with 95% confidence intervals) for studies of Acinetobacter infection. Plot (A) encompasses all included studies with mixed susceptibility profiles and clinical syndromes (B and C) show mortality data for bloodstream infection (BSI) and lower respiratory tract infection (LRTI) of all susceptibility profiles (D–H) represent Acinetobacter infections that are drug sensitive (D), carbapenem resistant (E), multidrug resistant (F), extensively drug resistant (G), and pan-drug resistant (H). Each plot represents a random effects model, where the size of the squares reflects the weight of each study, and the diamonds indicate the pooled proportion of in-hospital mortality estimate. The heterogeneity (I²) across studies is reported for each subgroup, along with tau², which quantifies the between-study variance. P-values from the Q-test for heterogeneity are also provided, indicating whether the observed heterogeneity is statistically significant. The infection syndrome(s) is described within brackets in (A) and (D–H): ANY = any infection syndrome; BSI = bloodstream infection; COL = colonization; LRTI = lower respiratory tract infection, including community-acquired, hospital-acquired, and ventilator-associated pneumonia; LRTI-BSI = pneumonia with associated bloodstream infection.

Approaches to addressing drug-resistant infections

The increasing global incidence of drug-resistant Acinetobacter infections combined with the excessive mortality drives the requirement to develop mitigating strategies. Three overarching themes are frequently highlighted and are indeed common to the WHO AMR Research Priorities and the United Nations General Assembly Political Declaration following the High-Level Meeting on AMR in 2024 (35, 36). Preventative measures to minimize the spread of AMR and incidence of disease can be subdivided into (i) water, sanitation, and hygiene (WASH), (ii) infection prevention and control (IPC), (iii) immunization, and (iv) stewardship. In the case of Acinetobacter spp., there is no available vaccine at present. WASH and IPC have a role in minimizing the spread of drug-resistant infections, particularly in hospital settings. Estimates suggest that AMR-associated mortality in LMICs (for all pathogens) could be reduced by 7.8% through IPC and 5.2% through WASH interventions (37). Evidence for the success of antimicrobial stewardship interventions in hospital settings, where the main burden of A. baumannii infections lies, is limited (37). Diagnostics play a role in the early recognition of drug-resistant Acinetobacter spp. Given the limited laboratory capabilities for bacteriology, and furthermore scarce antibiotic sensitivity testing in LMICs (settings where there is a significant incidence of drug-resistant Acinetobacter infections [Fig. 3]), there remains significant progress to be made in this theme. This review focuses on the therapeutic options to combat drug-resistant Acinetobacter spp. While there is ongoing research into non-traditional agents, for example, monoclonal antibodies, we focus on existing antibacterial classes and small-molecule agents in development (38).

CURRENT THERAPEUTIC OPTIONS AND MOLECULAR RESISTANCE MECHANISMS

Carbapenems

Carbapenems are a critically important subclass of β-lactam antibiotics. Their mechanism of action is the disruption of transpeptidation that weakens peptidoglycan, which ultimately leads to cell lysis (39). The carbapenems resist hydrolysis by common β-lactamases via the trans-configuration of their C5-C6 bond and the hydroxyethyl substitution compared with penicillin (40). Carbapenems with anti-Acinetobacter activity include imipenem (combined with cilastatin to limit renal excretion), meropenem, and doripenem (41). Treatment of carbapenem-susceptible Acinetobacter spp. with imipenem/cilastatin in VAP and BSI has been reported to have clinical success rates of 83% and 56%, respectively (42, 43). Discordant sensitivities between different carbapenem agents are seen not infrequently (44). Even where there is susceptibility to carbapenems, underlying mortality remains high, with a pooled estimated >20% through meta-analysis (Fig. 4D).

Carbapenems access the periplasmic space via outer membrane proteins (OMPs) (45). Bacterial peptidoglycan synthesis is driven by a series of glycosylation and transpeptidation steps within the periplasmic space. These processes are orchestrated by a cascade of enzymes, glycosylases, transpeptidases, carboxypeptidases, and endopeptidases, collectively known as penicillin-binding proteins (PBPs), which are the molecular targets of β-lactam antibiotics (46). The genome of A. baumannii encodes eight distinct PBPs: Class A PBPs (PBP1a and PBP1b), Class B PBPs (PBP2 and PBP3), four low molecular weight (LMW) PBPs (PBP4-6 and PBP7/8), and an additional mstA endopeptidase (47). While all carbapenems target one of these PBPs, the binding affinity of different carbapenems to specific PBPs in A. baumannii remains less well understood compared with Escherichia coli. Imipenem, for example, has a high binding affinity for PBP2—reduced expression or deletion of PBP2 in clinical isolates is associated with reduced susceptibility. Differences in the expression levels of PBP1, PBP2, PBP5, and PBP7/8 have been observed between carbapenem-resistant clinical isolates and carbapenem-sensitive ATCC lab strains, suggesting a role in carbapenem resistance (48–50).

The resistance caused by β-lactamases is extremely concerning. The rapid global dissemination of genes encoding carbapenemases is largely due to their flanking by insertion sequences (51). The innate resistance to penicillin in A. baumannii is facilitated by the presence of OXA-51-like enzymes (e.g., OXA-66, OXA-69, OXA-70, OXA-71, OXA-72) (52). Similarly, resistance to the first-, second-, and third-generation cephalosporins is mediated by Acinetobacter-derived cephalosporinase (ADC) (16, 53). While OXA-51-like enzymes are relatively weak carbapenemases, OXA-51 is often flanked by ISAba1 elements, which enhances their expression, this contributes to reduced susceptibility to carbapenems, especially in the presence of other resistance-conferring mechanisms (54).

An amino acid substitution from leucine to isoleucine at position 129 improves the binding and hydrolytic characteristics of OXA-51 for imipenem and doripenem (55). The most widespread carbapenemase in A. baumannii, especially in the United States and Southeast Asia, is OXA-23 and its less common variants (e.g., OXA-27, OXA-49, OXA-133) (53). It can be either plasmid-borne or chromosomally encoded, and the latter is typically regulated by the strong ISAba1 promoter, which aids its global spread (56, 57). Recent evidence suggests that OXA-23 originated in A. radioresistens before spreading to other Acinetobacter species, including A. baumannii (57–62). Other prevalent oxacillinases in A. baumannii include OXA-24/40-like (e.g., OXA-25, OXA-26, OXA-160) and OXA-58-like (e.g., OXA-96, OXA-97, OXA-164), which originated in different Acinetobacter species and transferred to A. baumannii via mobile plasmids flanked by ISAba3 (53). In A. baumannii, the co-occurrence of multiple OXA-type β-lactamases within a single strain is well documented, with combinations such as OXA-23-like and OXA-51-like, or OXA-24/40-like coexisting with OXA-58-like enzymes, frequently reported. This multiplicity can enhance carbapenem resistance through additive effects and may reflect the organism’s adaptability via horizontal gene transfer and chromosomal integration (53).

In numerous clinical isolates of carbapenem-resistant A. baumannii, insertion sequence-mediated disruptions of one or more outer membrane proteins (OMPs) have been implicated to account for reduced susceptibility. Notably, key OMPs such as OmpA, CarO, and OprD are frequently affected. OmpA, the most abundant OMP in A. baumannii, plays a crucial role in regulating normal physiology and virulence (63). Evidence suggests that strains expressing OmpA exhibit enhanced virulence, with a higher capacity for adhesion to and invasion of human cells (64). CarO, identified as a carbapenem susceptibility protein, represents a highly diverse class of OMPs within A. baumannii, with more than six polymorphic variants coexisting within the species. While CarO is primarily involved in the transport of basic amino acids such as ornithine and glycine, its disruption by various insertion sequences (ISs), including ISAb1, IS10, IS15, IS36, and IS125, has been strongly correlated with carbapenem resistance, particularly to imipenem (53, 55–57, 65–69). Various mechanisms, including regulatory mutations and environmental pressures, can downregulate porin genes, leading to decreased expression levels. This reduction in porin-mediated antibiotic uptake, when combined with other resistance mechanisms such as the production of carbapenem-hydrolyzing enzymes (e.g., OXA-type β-lactamases), can result in high-level carbapenem resistance.

While A. baumannii lacks a direct ortholog of the OprD porin in Pseudomonas aeruginosa, analogous porins such as Omp33-36 are similarly implicated in carbapenem resistance by reducing antibiotic permeability. The inactivation of omp33-36 via insertion sequence-mediated disruption has been repeatedly observed in clinical strains, further contributing to a resistance phenotype. These findings underscore the complex interplay between OMP disruption and the development of carbapenem resistance in A. baumannii (70).

The β-barrels of OMPs in the outer membrane of A. baumannii interact with OXA-23 to achieve an efficient assembly to rapidly degrade carbapenems. OMPs are the main entry mechanism for carbapenems, and the positioning of lysine residues (OmpA: K118 and CarO: K178) provides a periplasmic motif for linkage of OXA-23 (K60) (71). Given the orientation-specific nature of β-barrel structures in outer membranes, the site-specific cross-linking observed between OXA-23 and porin proteins suggests a conformation-dependent interaction of OXA-23 with porins on the periplasmic side of the outer membrane. Since carbapenems exert their antibacterial effect by inhibiting peptidoglycan synthesis, the strategic localization of OXA-23 near the outer membrane, where carbapenems first penetrate, potentially enables the enzyme to hydrolyze these antibiotics before they can diffuse to and inhibit their target penicillin-binding proteins (PBPs) within the periplasmic space (71). This proximity to the initial point of carbapenem entry likely serves to protect crucial peptidoglycan-synthesizing enzymes from inactivation, thereby contributing to the carbapenem resistance observed in A. baumannii (Fig. 5).

Fig 5.

Mechanistic network of carbapenem resistance in A. baumannii. This schematic illustrates the complex interplay between carbapenemases, efflux pumps, and outer membrane proteins in mediating resistance. The upward arrow indicates the upregulation of gene expression driven by strong promoters such as ISAba1. Specific lysine residues K60 on OXA-23 and K178 on CarO are highlighted, indicating critical sites involved in the carbapenem degradation pathway. Grey dots represent free extracellular zinc ions, while small blue dots denote calprotectin molecules secreted by host neutrophils, which sequester zinc and create a challenging environment for A. baumannii.

An increasingly important carbapenem resistance mechanism is metallo-β-lactamases (MBLs). These enzymes (Class B in the Ambler classification system) are characterized by the presence of two zinc ions at the active site that hydrolyze the β-lactam ring. MBLs inactivate all β-lactam antibiotics except for monobactams. The development of MBL inhibitors is an urgent, unmet medical need (72, 73).

The spread of MBLs is facilitated by colocalization on mobile genetic elements such as plasmids and integrons, which facilitate horizontal transfer between bacteria. New Delhi metallo-β-lactamase-1 (NDM-1) was discovered in 2010 and is the most prevalent and widely disseminated MBL in A. baumannii (74). Another significant MBL is imipenemase (IMP) that was originally detected in P. aeruginosa, but is now commonly found in A. baumannii isolates, especially from the Asia-Pacific region (75). Verona Integron-encoded Metallo-β-lactamase (VIM) is also frequently encountered, VIM-2 is the most reported variant globally (76). Other MBLs present in A. baumannii include Seoul Imipenemase (SIM), German Imipenemase (GIM), São Paulo Metallo-β-lactamase (SPM), and Adelaide Imipenemase (AIM), although these are less common globally compared with NDM, IMP, and VIM (77). Tripoli metallo-β-lactamase (TMB-1) has been identified in A. baumannii isolates from Libya (78). An overview of A. baumannii-associated MBLs and oxacillinases is presented in Table 1.

TABLE 1.

Overview of clinically relevant MBLs and OXA-type β-lactamases in A. baumannii

Parameters	Metallo-β-lactamases	Oxacillinases	Ref
Common examples and common locations	NDM-1 (plasmid), VIM-2 (plasmid), IMP (Plasmid)	OXA-23 (chromosome/plasmid), OXA-24/40 (plasmid), OXA-58 (plasmid), OXA-51-like (chromosome)	(79–83)
Hydrolytic efficiency kcat/km (µM⁻¹·s⁻¹)	Greater (approximately 0.6–5)	Comparatively lower (in the range 0.05–0.3)	(53, 84)
Inhibitor in the clinical drug development pipeline	Xeruborbactam	Durlobactam	(85, 86)
Clinical impact	Confer higher resistance due to high hydrolytic activity against carbapenems	Due to weaker hydrolytic activity, to attain high resistance, other resistance mechanisms are required (efflux pumps, porins, etc)	(53, 87, 88)

NDM-1 mediates carbapenem resistance through a zinc-dependent hydrolysis mechanism (89). The active site features two zinc ions that are essential for catalysis. These metal ions facilitate the binding of the carbapenem’s β-lactam ring, positioning it for attack by an activated water molecule, which the zinc ions also help to polarize. This nucleophilic water molecule attacks the carbonyl carbon of the β-lactam ring, forming a transient tetrahedral intermediate. The subsequent breakdown of this intermediate results in the cleavage of the β-lactam ring, thereby inactivating the carbapenem. The hydrolyzed antibiotic is then released from the enzyme, allowing NDM-1 to continue hydrolyzing additional β-lactam molecules, effectively inactivating the antibiotic (73).

A. baumannii periplasmic Zn(II) concentrations are not intrinsically regulated and are dependent on extracellular zinc availability. During infection, this dependence becomes problematic when host neutrophils release calprotectin, a protein that sequesters zinc, leading to zinc limitation (90). Under these conditions, metallo-β-lactamase (MBL) apoenzymes fail to properly fold into their active forms, rendering the pathogen susceptible to carbapenems. While most MBLs are prone to peptide degradation in the periplasm under zinc-deficient conditions, NDM-1 has evolved a remarkable adaptation (91). This strategy not only ensures its survival in zinc-limited environments but also facilitates the extensive dissemination of its gene, independent of plasmid carriage or single gene structures, enhancing its role in antibiotic resistance. Lipidation of NDM-1, specifically through N-myristoylation, is a critical post-translational modification where a 14-carbon saturated fatty acid, myristic acid, is covalently attached to the N-terminal cysteine residue of the enzyme (91, 92). This process, catalyzed by N-myristoyltransferase (NMT), involves the recognition of a consensus sequence at the N-terminus of the protein, typically MGKSH, where glycine at position 2 is essential for lipid attachment; however, the involvement of the exact sequence for NDM-1 lipidation is lacking (92). The myristic acid anchors NDM-1 to the inner leaflet of the bacterial outer membrane, positioning the enzyme in the periplasmic space to effectively hydrolyze β-lactam antibiotics, including carbapenems, before they reach their targets (91). In addition, membrane anchoring facilitates the secretion of NDM-1 in outer membrane vesicles (OMVs), which are involved in bacterial communication and pathogenesis. These OMVs, carrying both NDM-1 protein and the blaNDM gene, exhibit carbapenemase activity and can protect nearby populations of carbapenem-susceptible bacteria, thereby enhancing the spread of resistance (91). Although NDM-1 remains the most prevalent and widely distributed NDM variant among clinical A. baumannii isolates, other variants such as NDM-2, NDM-5, and NDM-9 have also been reported in clinical settings across Africa, Europe, and the Middle East (93, 94).

Given that the major mechanisms of carbapenem resistance in A. baumannii are the production of OXA-type and metallo-β-lactamase (MBL) enzymes, recent clinical efforts have focused on developing combinations of carbapenems with β-lactamase inhibitors to preserve the efficacy of this critical antibiotic class. Specifically, meropenem, in combination with several promising β-lactamase inhibitors, is being evaluated in various phases of clinical trials for the treatment of A. baumannii infections (Table 2).

TABLE 2.

List of carbapenem-inhibitor combinations in clinical trials against A. baumannii infections

Carbapenem combination(s)	Clinical trial phase	Mechanism of action	References
Meropenem-ANT3310	Phase 1 (ongoing)	Antabio’s ANT3310 is a novel serine-β-lactamase inhibitor that has demonstrated promising preclinical efficacy against Enterobacteriaceae and Acinetobacter baumannii. In a Phase 1 clinical trial involving 72 healthy individuals, ascending doses of ANT3310 were well-tolerated, with no reported adverse effects or toxicities. In addition, pharmacokinetic assessments revealed no significant interactions between ANT3310 and meropenem, supporting its compatibility as part of a combination therapy. Future studies are expected to further elucidate the pharmacodynamics, spectrum of activity, and safety profile of this combination in diverse patient populations, paving the way for subsequent phases of clinical evaluation.	(95)
Meropenem-Zidebactam	Preclinical (ongoing)	Zidebactam, a β-lactamase inhibitor, targets both serine and metallo-β-lactamases while also exhibiting direct inhibitory activity against penicillin-binding protein 2 (PBP2). Preclinical studies have demonstrated promising in vitro and in vivo results, effectively restoring the activity of meropenem against carbapenem-resistant A. baumannii (CRAb). In addition, the zidebactam-cefepime combination is under development for A. baumannii infections and is currently in Phase 3 clinical trials.	(96, 97)
Meropenem-InC58-avibactam	Preclinical (ongoing)	InC58 is a potent metallo-β-lactamase (MBL) inhibitor that, when combined with avibactam, extends protection across a broad range of β-lactamases, enhancing the bactericidal efficacy of meropenem. Currently in the preclinical research phase, this combination has not yet progressed to clinical trials.	(98)

Colistin

Colistin, a polycationic lipopeptide, has re-emerged as a crucial last-resort antibiotic in the treatment of MDR Gram-negative bacterial infections, particularly when first- and second-line therapies fail (99). This bactericidal agent exhibits broad-spectrum activity against Gram-negative pathogens by leveraging its amphipathic structure, which includes both hydrophilic and lipophilic moieties. Colistin exerts its antibacterial effects by binding to the negatively charged phosphate groups of lipopolysaccharides (LPS) in the bacterial outer membrane, displacing essential divalent cations such as calcium and magnesium (99, 100). The structural integrity of the outer membrane is disrupted, allowing the hydrophobic tail of colistin to insert into the phospholipid bilayer, leading to membrane destabilization and eventual bacterial cell lysis. This membrane disruption process is repeated in the inner membrane, culminating in cell death (99–101).

The therapeutic window of colistin is narrow, with significant risks of neurotoxicity and nephrotoxicity, necessitating careful therapeutic drug monitoring (TDM) to maintain effective and safe plasma concentrations, particularly in critically ill patients (102). The regimen of colistin varies depending on the formulation used, whether administered as colistin base activity (CBA) or the prodrug colistimethate sodium (CMS) (103). For intravenous administration of CMS, treatment typically begins with a loading dose of 9 million international units (MIU) over 1 hour, followed by a maintenance dose of 3 MIU every 8–12 hours. In patients with renal impairment, dose adjustments are critical, with reductions to 4.5–6 MIU/day for CrCl of 20–50 mL/min, and 3 MIU/day for CrCl below 20 mL/min, to mitigate the risk of further nephrotoxicity (104).

The rise in colistin resistance is increasingly concerning and may be related to agricultural use. In many bacterial pathogens, colistin resistance is associated with the loss of LPS due to mutations that inactivate the lipid A biosynthesis pathway, specifically through alterations in key genes such as lpxA, lpxC, and lpxD. However, the absence of LPS significantly compromises the structural integrity of the bacterial outer membrane, resulting in a fitness cost that impairs the bacterium’s ability to survive in vivo (105). These resistance mutants often exhibit reduced virulence and are more susceptible to immunological clearance, which can also render them vulnerable to antibiotics that would otherwise be ineffective (106, 107).

A. baumannii possesses a sophisticated mechanism that involves two-component systems (TCSs) as an adaptive strategy for colistin resistance. The most well-characterized TCS, PmrCAB, includes a sensor kinase and a response regulator that modulate the expression of the phosphoethanolamine (pEtN) transferase, encoded by pmrC. This modification reduces the negative charge of LPS, thereby decreasing colistin binding and conferring resistance (108). Unlike other pathogens, A. baumannii efficiently employs this mechanism, with numerous clinical isolates harboring multiple pEtN transferase homologs, such as eptA, which are regulated by the strong promoter of the insertion sequence ISAba1 rather than a TCS (109). Remarkably, a single A. baumannii clinical isolate can possess up to three distinct eptA variants, all contributing to colistin resistance. While plasmid-associated mcr-type phosphoethanolamine (pEtN) transferases have been predominantly identified in Enterobacterales, recent reports have highlighted their increasing prevalence in A. baumannii isolates (110, 111). This emerging trend raises significant concerns about the potential for rapid global dissemination of colistin resistance, exacerbating the challenge of controlling multidrug-resistant infections in healthcare settings worldwide and underscoring the importance of surveillance, molecular epidemiology, and infection control measures.

The ability to deploy multiple phosphoethanolamine transferases underscores the resilience of Acinetobacter spp. and highlights the complexity of addressing colistin resistance (112). These developments underscore the urgent need for continued research into novel therapeutic strategies and combination therapies that can overcome these resistance mechanisms, ensuring the sustained efficacy of colistin and other vital antibiotics in the clinical arsenal.

Cefiderocol

Cefiderocol is a catechol-type siderophore cephalosporin (113). Cefiderocol exploits the iron-scavenging mechanisms of Gram-negative pathogens by mimicking natural siderophores, which form complexes with ferric iron (Fe³⁺). This complex facilitates entry into the bacterial periplasm through iron transporter channels in the outer membrane (114, 115). The structure of cefiderocol is designed to enhance access to the periplasmic space, forming stable siderophore-iron complexes while also allowing passive diffusion through porin channels. The cephalosporin core of cefiderocol, akin to cefepime, enhances aqueous solubility, contributing to its efficacy against carbapenem-resistant pathogens, particularly through high-affinity binding to penicillin-binding protein 3 (PBP3) in A. baumannii (113, 114, 116).

Cefiderocol has been evaluated in phase III clinical trials, including CREDIBLE-CR and APEKS-cUTI/NP, targeting serious infections such as complicated urinary tract infections (cUTIs), nosocomial pneumonia, and Gram-negative bacteremia. The trials demonstrated comparable efficacy to existing treatments for carbapenem-resistant Enterobacteriaceae (CRE) and carbapenem-resistant A. baumannii (CRAb) infections, with cefiderocol showing better tolerability and stability against hydrolysis by class A, B, C, and D β-lactamases (117–120). While data have demonstrated that cefiderocol is an effective antibiotic against Gram-negative pathogens, a notably higher mortality rate was observed in the CREDIBLE-CR and GAMECHANGER clinical trials. In the CREDIBLE-CR trial, the 28-day all-cause mortality rate was significantly higher in the cefiderocol treatment arm (34%) compared to the best available therapy (18%). Subsequent post hoc analyses have not conclusively linked cefiderocol to the elevated mortality rate; however, they suggest caution when treating infections caused by A. baumannii, as the mortality difference was particularly pronounced in patients infected with this pathogen (118, 121, 122).

A. baumannii responds to iron depletion by upregulating phospholipase activity and siderophore production. The bacterium synthesizes three types of siderophores: acinetobactin, baumannoferrin, and fimsbactin. Phospholipase C genes (plc1 and plc2) enable erythrocyte lysis, facilitating heme acquisition by the pathogen (123, 124). The liberated iron is then sequestered by the siderophores and transported into the bacterial cell via TonB-dependent outer membrane receptors, in conjunction with an ABC transporter system. Notably, PirA and PiuA serve as major TonB receptors, while BauABC mediates the import of Fe²⁺ into the cell (125, 126).

Previous efficacy studies have suggested that downregulation of TonB receptors contributes to reduced uptake of cefiderocol in A. baumannii cells. However, recent findings have revealed that this reduced uptake is more accurately attributed to functional loss of TonB receptors. This loss of function is driven by two point mutations in the pirA gene, resulting in Leu275Phe and Ile277Val substitutions in the β-strand. These mutations, involving the substitution of hydrophobic with aromatic amino acids, impair the functionality of these receptors (127). In addition, emerging evidence implicates efflux in the development of cefiderocol resistance. Notably, the BaeSR two-component system (TCS), a widely distributed sensor-receptor assembly in A. baumannii that detects physical and chemical stress, has been linked to this resistance. Genomic and expression analyses of cefiderocol-resistant isolates have identified a point mutation in the response regulator BaeR (Ser104Asn), which leads to the upregulation of the RND efflux pump MacAB-TolC (128) (Fig. 6). This upregulation correlates with a fourfold increase in the MIC for cefiderocol.

Fig 6.

Molecular mechanisms underlying cefiderocol treatment failure in A. baumannii. This illustration depicts the factors contributing to cefiderocol treatment failure, including point mutations in TonB iron receptors, two-component system (TCS)-mediated upregulation of the MacAB-TolC efflux pump, and β-lactamase-mediated hydrolysis of the antibiotic. Upward arrows indicate the upregulation of phospholipase C and the MacAB-TolC efflux pump genes, respectively.

Previous in vitro studies investigating the IC₅₀ of cefiderocol against various β-lactamases have demonstrated that the drug exhibits considerable stability against hydrolysis by enzymes such as NDMs, KPC, and OXA-23 (129). These studies indicated that the catalytic efficiency of these β-lactamases for cefiderocol is approximately 400 times lower than for meropenem. However, despite this relative stability, these β-lactamase enzymes can still hydrolyze cefiderocol to a minor extent. The role of β-lactamases in reducing the efficacy of cefiderocol becomes particularly significant when combined with other resistance mechanisms, such as loss-of-function mutations in the TonB receptor and/or the upregulation of efflux pumps (127, 128). These factors collectively contribute to decreased cefiderocol uptake into the periplasmic space of A. baumannii. The small amounts of the drug that do enter through passive diffusion are further compromised by hydrolysis mediated by β-lactamases such as NDM-1 or ADCs. The clinical development of boron β-lactamase inhibitor xeruborbactam in combination with cefiderocol is a potent way to rescue the latter (85).

Target site mutations have also been implicated in the failure of cefiderocol treatment in A. baumannii. Although resistance to cefiderocol has been associated with various mutations in the PBP3 gene, a particularly significant mutation involves the substitution of isoleucine with asparagine at position 236 (Ile236Asn) (127). This alteration changes the hydrophobic environment of the active site, which is believed to play a crucial role in conferring resistance in numerous clinical isolates of A. baumannii. The emergence of this mutation highlights the adaptive mechanisms employed by A. baumannii.

Although stable antibiotic resistance remains a major obstacle in the treatment of bacterial infections, heteroresistance, an elusive and dynamic phenotype, also plays a critical role in clinical treatment failure (130). Defined by the presence of resistant subpopulations within an otherwise susceptible isogenic strain, heteroresistance often escapes detection by standard MIC testing, thereby undermining treatment efficacy (131). This phenotype can arise through diverse mechanisms (e.g., replication slippage, gene amplification, and epigenetic modulation), allowing bacteria to switch between susceptible and resistant states. Depending on selective pressure and associated fitness costs, heteroresistance may be transient or stable (132, 133). It is widespread among both Gram-positive and Gram-negative pathogens and becomes particularly problematic in infections with high bacterial burden, where resistant subpopulations are more likely to expand (134, 135). Notably, heteroresistance has been implicated in the inconsistent clinical efficacy of cefiderocol against CRAB. A recent study reported that 59% of CRAB isolates from Italy and the United States exhibited cefiderocol heteroresistance, with whole-genome sequencing revealing disruption of the TonB-dependent receptor gene (piuA) as a key mechanism. Importantly, isolates with cefiderocol MICs > 1 mg/L were associated with higher rates of treatment failure compared to those with MICs ≤ 0.5 mg/L (136).

Sulbactam-durlobactam

Sulbactam is a β-lactamase inhibitor that inhibits Amber Class A enzymes as well as possessing intrinsic bactericidal activity via targeting penicillin-binding proteins PBP1 and PBP3, which are crucial for the survival of A. baumannii (137). Despite its dual mechanism of action, the clinical efficacy of sulbactam alone is significantly compromised by the ubiquitous presence of OXA-51-type and AmpC β-lactamases, resulting in a high minimum inhibitory concentration (MIC) of 32-64 mg/L. Durlobactam is a novel diazabicyclooctane (DBO) derivative that extends the activity of sulbactam by inhibiting a broad range of β-lactamases, including Class A, C, and D enzymes, with the important exception of metallo-β-lactamases (MBLs) (86). When combined with durlobactam, the MIC of sulbactam is 2–8 mg/L, restoring its clinical utility (138).

Sulbactam-durlobactam is a promising alternative to carbapenems and colistin for the treatment of ventilator-associated pneumonia (VAP), hospital-acquired pneumonia (HAP), and bacteremia caused by A. calcoaceticus-baumannii complex (ABC) in the multinational ATTACK trial (138, 139). In the phase 3 ATTACK trial, both the sulbactam-durlobactam and colistin treatment groups received imipenem-cilastatin (1 g every 6 hours) as background therapy (140). Although preclinical studies using murine models and hollow-fiber infection systems suggested that sulbactam-durlobactam alone is effective against CRAB isolates with MICs of 4 mg/L, more recent in vitro studies employing molecular approaches indicate that co-administration with imipenem may provide additional therapeutic benefit. This enhanced activity likely results from the complementary binding profiles of the agents: sulbactam targets PBP1a, PBP1b, and PBP3, while imipenem primarily targets PBP2. Their combined use may therefore improve bacterial eradication by simultaneously inhibiting multiple PBPs in CRAB isolates (141, 142). The combination of sulbactam-durlobactam effectively circumvents carbapenem resistance mechanisms mediated by Class A, C, and D β-lactamases in A. baumannii. The recommended adult dosage for treating severe A. baumannii infections, particularly in cases with limited treatment options, is 1 g of sulbactam combined with 1 g of durlobactam, administered intravenously over 2 hours every 6 hours.

Recent reports of target site mutations, particularly those associated with PBP3, have raised significant concerns regarding the clinical longevity of sulbactam-durlobactam. Among the various mutations identified, T526S and A515V have garnered particular attention. The T526S mutation has been shown to elevate the MIC of sulbactam, which is further amplified in the presence of durlobactam, potentially compromising the efficacy of the combination (143). Conversely, clinical isolates harboring the A515V mutation have not demonstrated a significant change in susceptibility to sulbactam-durlobactam. However, a majority of A515V-bearing isolates also carry MBLs, which effectively render the combination inactive (143, 144).

RND efflux pumps significantly contribute to the reduced efficacy of the sulbactam-durlobactam combination in clinical Acinetobacter baumannii isolates. While the AdeABC efflux system is well-recognized for mediating resistance to β-lactam antibiotics, it appears to play a limited role in the resistance to sulbactam-durlobactam. Instead, gene disruption studies have revealed that the AdeIJK efflux pump is critically important, capable of increasing the MIC of the combination by up to fourfold. This increase is attributed to the active efflux of durlobactam, which results in lowered intracellular concentrations, thereby diminishing the bactericidal efficacy of sulbactam (145).

Zosurabalpin

Lipopolysaccharides (LPS) have long been recognized as an attractive target for antibiotic development due to their essential role in the structural integrity and defense mechanisms of Gram-negative bacteria (146). However, earlier drug development efforts, which primarily focused on screening small-molecule libraries, often failed to yield viable candidates. This was largely due to challenges such as the high hydrophobicity of potential compounds, resulting in poor solubility and bioavailability (147). Consequently, for many years, antibacterial drug development has concentrated on more traditional targets, such as cell wall synthesis, and DNA, RNA, and protein synthesis. Unfortunately, the clinical efficacy of new antibiotics has often been compromised by the ability of A. baumannii to rapidly develop resistance.

Zosurabalpin is a first-in-class macrocyclic peptide (MCP) that is currently in Phase 1 clinical trials (RG6006) (other potential anti-Acinetobacter drugs in early phase of clinical trial are listed in Box 2). Zosurabalpin targets the transport of lipopolysaccharide (LPS) from the inner to the outer membrane, thereby compromising the integrity of the outer membrane and rendering the pathogen more vulnerable to immune defenses and antibiotic treatments (148). Zosurabalpin binds competitively with the transmembrane domain of LptC component within the LptB2FGC complex, a key ABC transporter responsible for the translocation of LPS. This disruption is particularly detrimental to A. baumannii, as the loss of outer membrane integrity makes the bacterium susceptible to cellular lysis under hostile conditions encountered during infection.

Box 2. Next-gen antibiotics for Acinetobacter: early clinical trial landscape.

1. Apramycin: Originally developed in the 1970s for veterinary use, apramycin was not advanced into human clinical development due to the availability and effectiveness of other aminoglycosides already in clinical practice at the time (149). However, renewed interest has emerged due to its potent activity against multidrug-resistant pathogens and its favorable safety profile. Preclinical studies have demonstrated that apramycin exhibits significantly lower nephrotoxicity and ototoxicity compared to conventional aminoglycosides (149). Resistance to aminoglycosides in A. baumannii, particularly carbapenem-resistant strains (CRAB), is frequently mediated by 16S rRNA methyltransferases such as ArmA and Rmt (150). These genes are often co-located with carbapenemase genes on the same plasmids or resistance islands, facilitating co-selection and spread (151). Apramycin’s unique chemical structure, a 4-monosubstituted 2-deoxystreptamine ring, distinguishes it from clinically used aminoglycosides (which typically have 4,5- or 4,6-disubstitution patterns). This structural difference contributes to a distinct ribosomal binding mode that allows apramycin to evade common resistance mechanisms, particularly rRNA methylation at the aminoglycoside binding site of the 30S ribosomal subunit and hence, low cross-resistance has been observed (152). Apramycin is being developed for human use under the investigational name EBL-1003 and has passed Phase 1 clinical evaluation (ClinicalTrial.gov NCT04105205), where it was considered safe, well-tolerated, and pharmacokinetics following intravenous administration in healthy individuals.

2. BWC0977: A novel bacterial topoisomerase inhibitor (NBTI), originated from phenotypic screening followed by Structure Activity Relationship (SAR) to optimize efficacy against Gram-negative pathogens and attain desirable pharmacokinetic parameters. Developed by Bugworks Research Inc., India, this NBTI has a dual mode of action as it blocks DNA replication by inhibiting DNA gyrase and topoisomerase IV (153). Preclinical studies showed this candidate drug has broad-spectrum bactericidal activity against Gram-positive, Gram-negative, and anaerobic bacteria, irrespective of the presence of fluoroquinolone, colistin, and carbapenem resistance (MIC: 0.5-2 mg/L) (153). The phase 1 trial on healthy individuals showed that the intravenous dose of BWC0977 was safe and well tolerated, achieving the proposed dose exposure. Plans for the phase 2 trial are underway by the Global Antibiotic Research and Development Partnership (GARDP), where the focus patient population will be individuals with hospital and ventilator-associated pneumonia, particularly caused by CRAB.

3. RG6436 (GDC- 0829): This investigational drug (Genentech, USA) is an arylomycin derivative that showed between penetration of Gram-negative outer membrane via a porin-independent manner (154). Arylomycin targets bacterial type I signal peptidase, facilitating the essential function of bacterial peptide transportation. RG6436 blocks signal peptidase LepB, accumulating peptides and causing cell death. Preclinical data showed promising in vivo efficacy against Enterobacterales and other Gram-negative priority pathogens, including A. baumannii (MIC: 0.5-2 mg/L) (148, 155). RG6436 is currently in Phase 1 clinical trial and is focused on the future potential treatment of complicated UTIs and other infections (ISRCTN18049481) (156).

4. APL-2301: A novel nitroxoline derivative (or MET-102) is a first-in-class antibacterial small molecule developed by Asieris Pharmaceuticals, China. In vitro studies showed that APL-2301 is effective in the infection control of A. baumannii strains, particularly carbapenem-resistant isolates (157). Asieris Pharmaceuticals classified APL-2301 as a “novel mode of action” and has neither been disclosed by the company nor its structure. In 2023, the Australian Therapeutic Goods Administration approved APL-2301 for a phase 1 clinical trial to assess the molecule's safety, tolerability, and pharmacokinetics in healthy adult individuals (158).

5. TP-6076: A novel tetracycline class derivative (synthetic fluorocycline) developed by Tetraphase Pharmaceuticals, USA, for treating MDR Gram-negative pathogens, including CRAB. Similar to tetracycline, it binds to the 30S subunit of the ribosome, thereby obstructing the attachment of aminoacyl-tRNA to messenger RNA and ribosome complex. It also showed binding and inhibition of the AdeJ efflux pump, which further potentiates its efficacy against efflux pumps. Development of TP-6076 is supported by CARB-X and is currently progressing in a Phase 1 clinical trial in healthy adult individuals for its safety, tolerability, and pharmacokinetics profile.

6. SPR741: A polymyxin derivative, developed by Spero Therapeutics, USA, as an antibiotic adjunct to potentiate the efficacy of older antibiotics, thus reducing the increasing need for new antibiotics in clinics for the treatment of Gram-negative pathogens, including CRAB. Preclinical studies were promising as SPR741 successfully revived the activity of rifampicin, minocycline, erythromycin, and clarithromycin. Phase 1 clinical trial showed that SPR741 is well-tolerated in healthy subjects, with a favorable safety profile and minimal nephrotoxicity, distinguishing it from its polymyxin predecessors.

Mutational frequency analysis indicates that A. baumannii exhibits spontaneous mutation frequencies ranging from 10⁻⁶ to 10⁻⁹ when exposed to zosurabalpin at 2×, 4×, 8×, and 16× the MIC, across multiple clinical strains (159). The mutants selected under these conditions demonstrated a fourfold increase in MIC relative to the wild-type parent strain. Genomic analyses of these mutants identified single nucleotide polymorphisms (SNPs) in the LptF and LptG components of the LptB2FGC complex, which is essential for lipopolysaccharide (LPS) transport. In addition, mutations were found in lpxM, a gene involved in the final acylation step of LPS biosynthesis. These mutations led to the production of penta-acylated LPS, instead of the typical hepta-acylated form, resulting in reduced binding affinity of zosurabalpin (159) (Fig. 7). Furthermore, resistance to zosurabalpin has also been linked to the upregulation of RND efflux pumps. Specifically, loss-of-function mutations in the two-component system (TCS) regulator AdeR, such as a short insertion-deletion (indel) of AGTGTGGAGTA—were identified, leading to the upregulation of the AdeABC efflux system. This upregulation results in the active expulsion of zosurabalpin, thereby reducing its intracellular concentration and efficacy (159).

Fig 7.

Impact of loss-of-function mutations on zosurabalpin activity in Acinetobacter baumannii. This illustration depicts the potential SNPs in the LptF and LptG domains of the LPS transporter, loss-of-function mutations in the two-component system (TCS) BaeSR, and the production of penta-acylated LPS by LpxM instead of the typical hepta-acylated form. CoA: Coenzyme A; ACP: Acyl Carrier Protein; AdeSR: Sensor kinase and response regulator elements of the TCS.

Rifabutin (BV-100)

Rifabutin, a rifamycin analog, belongs to the ansamycin class of antibiotics and functions as an RNA polymerase inhibitor. Rifabutin exerts its antibacterial effect by binding to the RpoB subunit of bacterial RNA polymerase, thereby disrupting transcription. Structurally, it differs from rifampicin and other analogs due to a distinctive tail attached at C3 and C4 (160). This structural modification enhances rifabutin’s ability to inhibit the synthesis of the first RNA phosphodiester bond and facilitates interactions with sigma factors. Initially approved by the U.S. FDA in 1992 for the treatment of Mycobacterium avium complex infections, rifabutin’s potential as a potent anti-Acinetobacter agent remained unrecognized for nearly three decades (161, 162). Its potent activity against A. baumannii and other members of the ABC complex was evident much later when it was demonstrated that rifabutin exhibited significantly enhanced antibacterial activity in nutrient-limited conditions that mimic the human physiological environment.

A pivotal study demonstrated that rifabutin’s antibacterial activity was up to 1,000-fold higher in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with fetal calf serum compared to standard Mueller-Hinton Broth (MHB). These findings were consistent in in vivo studies that further corroborated these findings, as rifabutin outperformed rifampicin in murine pneumonia and sepsis models of A. baumannii infection. Notably, this activity was specific to A. baumannii and was not observed in other Gram-negative bacteria due to the lack of rifabutin-specific siderophore transported in A. baumannii (162).

The potent activity of rifabutin against A. baumannii is attributable to uptake via the TonB-dependent siderophore transporter, FhuE. Under iron-limiting conditions, FhuE expression is upregulated, leading to its overexpression in the outer membrane. Rifabutin exploits this transporter to facilitate intracellular transportation to achieve concentrations sufficient for bacterial killing. By contrast, the addition of exogenous iron suppresses FhuE expression via regulation by the ferric uptake regulator (Fur) transcription factor. This iron-dependent modulation potentially explains why the anti-Acinetobacter activity of rifabutin was overlooked for so long. The difference in the type of siderophore transporter involved in rifabutin and cefiderocol intake, mutation is one; the latter’s transporter does not contribute toward rifabutin resistance in A. baumannii (163).

Resistance to rifabutin in A. baumannii primarily arises through mutations in the fhuE gene, which leads to reduced drug uptake (164). In addition, the pathogen can express an alternative siderophore transporter, FhuE^Lac-4, which facilitates ferric ion acquisition while limiting rifabutin entry. Other mechanisms conferring reduced susceptibility include mutations in rpoB, which decrease binding of rifabutin to its target, and the presence of the arr-2 gene encoding an ADP-ribosyltransferase, which inactivates rifabutin. However, rpoB mutations and arr-2-mediated modification alone or in combination do not confer complete resistance, highlighting the additional pivotal role of FhuE-mediated drug uptake for fully understanding susceptibility to rifabutin (163).

Rifabutin has successfully completed a Phase 1 clinical trial, demonstrating a favorable safety profile and tolerability in patients undergoing single or multiple ascending dose studies. In mid-2023, BV-100 entered Phase II clinical trials to assess its efficacy and safety in adult patients with ventilator-associated pneumonia (VAP) caused by carbapenem-resistant A. baumannii (CRAB) (165).

OMN6

OMN6 is a novel bioengineered antimicrobial peptide (AMP), derived from Cecropin A, which is a 29–42 amino acid peptide found in Diptera and Lepidoptera within the Arthropoda phylum (166, 167). Cecropins are characterized by two α-helices that facilitate bacterial membrane penetration. OMN6 has been developed by Omnix Medical for the treatment of MDR, XDR, and PDR Gram-negative pathogens, with a specific focus on CRAB (167).

OMN6 has successfully completed a Phase I clinical trial in healthy volunteers, demonstrating a favorable safety profile and desirable pharmacokinetics. In late 2023, the U.S. Food and Drug Administration (FDA) granted approval for Phase II clinical trials to assess its safety and PK profile in patients with ventilator-associated pneumonia (VAP) and hospital-acquired pneumonia (HAP) caused by A. baumannii complex (168).

OMN6 is a 40-amino acid cyclic cationic peptide with intrinsic antibacterial properties mediated by its interaction with bacterial membranes. Gram-negative bacterial membranes, including those of A. baumannii, are enriched with phosphatidylglycerol and cardiolipin, which contribute to membrane fluidity and a net negative charge. This charge facilitates OMN6 binding via electrostatic attraction, allowing the AMP to interact with lipopolysaccharides and insert into the outer membrane (169).

Upon insertion, OMN6 induces localized bending of the lipid bilayer, promoting the formation of toroidal pores, where both inner and outer membrane leaflets curve inward. These pores, lined by OMN6 molecules and phospholipids, allow the uncontrolled efflux of ions and solvents, resulting in membrane depolarization, osmotic imbalance, and bacterial cell lysis. Notably, OMN6’s mechanism of action is independent of classical resistance pathways, such as carbapenemase production or porin mutations, ensuring sustained efficacy against drug-resistant A. baumannii (169).

Although no reports of OMN6 resistance have emerged to date, potential adaptive mechanisms could conceivably arise through alterations in outer membrane architecture. A reduction in phosphatidylglycerol and cardiolipin content would diminish the net negative charge and decrease membrane fluidity, thereby reducing OMN6’s binding efficiency and impairing toroidal pore formation. Such structural modifications could ultimately confer reduced susceptibility to OMN6 (169).

The introduction of antimicrobial peptides (AMPs) such as OMN6 represents a promising advance to address unmet medical needs related to drug-resistant A. baumannii. Given its novel mechanism of action and minimal susceptibility to existing resistance pathways, OMN6 holds significant potential as a future therapeutic option. However, ongoing surveillance of resistance evolution and further clinical validation will be required to ensure long-term efficacy in combating MDR, XDR, and PDR infections (167).

POTENTIAL THERAPEUTIC TARGETS AND CHALLENGES FOR NEXT-GENERATION ANTI-ACINETOBACTER DRUG DEVELOPMENT

Recent breakthroughs in the fields of cancer and cardiovascular research in 2024 have led to the approval of 53 novel drugs across the United States, European Union, and Japan, excluding vaccines and gene therapies. Notably, 28 of these drugs possess novel mechanisms of action, underscoring the innovative strides in these therapeutic areas (170). In stark contrast, antibacterial drug development has lagged, with zosurabalpin emerging as the only anti-Acinetobacter agent in the last 40 years to introduce a novel mechanism of action (148). This disparity highlights a critical need for the identification and development of new drug targets to combat the rising threat of antibiotic resistance in A. baumannii and other multidrug-resistant pathogens (Fig. 8).

Fig 8.

Potential drug targets and associated challenges for anti-Acinetobacter drug development. The bar chart represents the challenges associated with targeting each protein, such as issues related to specificity, essentiality, structural accessibility, resistance development, and potential off-target effects. ROS defense proteins related to A. baumannii oxidative stress, AbaI/R, Wza/b/c, and T6SS represent targets related to pathogenicity, Hfq, BamA, DsbA/B, and LpxC are involved in processes critical to cell stability and integrity, AdeABC is related to resistance, and FtsZ is essential for cell division. The figure emphasizes the need for a multifaceted approach to target selection with the feasibility of drug development.

Oxidative defense systems

Most antibacterial drugs, beyond their primary modes of action, also exert bactericidal effects through the generation of reactive oxygen species (ROS) (171). When ROS accumulate beyond a threshold, they become detrimental, leading to the oxidation of critical macromolecules such as DNA, RNA, and lipids (172). In addition, ROS can extract iron from Fe-S complexes, producing hydrogen peroxide, which further exacerbates oxidative stress. A. baumannii, like other bacteria, is equipped with intracellular defense mechanisms to mitigate this oxidative damage (173, 174).

The primary defense against ROS in major Gram-negative bacteria, including A. baumannii, is mediated by enzymes such as superoxide dismutase (Sod), catalase, and peroxidase (175, 176). A. baumannii is equipped with two superoxide dismutase enzymes, that is, sodB and sodC that rely on either manganese or copper/zinc as cofactor. The SodB/C enzymes are crucial in converting superoxide radicals into hydrogen peroxide and oxygen, playing a vital role in protecting A. baumannii from the host immune system and facilitating persistent infections (175). Over the years, there has been debate over the potential of targeting SodB as a novel approach in anti-Acinetobacter therapy. Suppressing SodB with small molecules could render A. baumannii more susceptible to eradication by the host immune system, and combining this approach with traditional antibiotics might enhance treatment efficacy against A. baumannii infections (175, 177, 178). Similar studies have also demonstrated the same outcome in other Gram-negative pathogens and Mycobacterium spp. However, despite the promise of SodB as a drug target, it remains relatively underexplored compared to conventional targets such as cell wall synthesis and protein synthesis. Major challenges include the functional redundancy among Sod proteins, where the loss of SodB might be compensated for by other Sod enzymes. In addition, the structural similarities between bacterial SodB and human superoxide dismutases raise concerns about potential host-related toxicity, underscoring the need for selective targeting strategies (179–181). This leads to the urgency to look for a better candidate for drug development.

In A. baumannii, the SoxR/S regulatory system plays a dual role in controlling oxidative stress responses and modulating the activity of multiple efflux pumps. Targeting SoxR/S through structure-based drug design could potentially enhance bacterial susceptibility to a wide range of antibiotics by inhibiting efflux pump function (182). Moreover, given the conservation of SoxR/S across various bacterial species, such an approach could lead to the development of broad-spectrum antibacterial agents. However, a significant challenge in targeting SoxR/S is the potential off-target effects on the gut microbiota, which could result in unintended dysbiosis as in the case of any broad-spectrum antibiotics (e.g. carbapenems) (183–185). Therefore, future studies should focus on the selective modulation of the A. baumannii ROS pathway, which could offer a future opportunity for the development of microbiome-sparing antibiotic scaffold design. This underscores the need for careful consideration in the drug development process to minimize disruption of beneficial microbial communities while effectively targeting pathogens.

HFQ (host factor for phage Qβ replication)

Recent advances in RNA biology have significantly expanded our understanding of post-transcriptional regulation in various bacterial pathogens. This mode of regulation allows bacteria to rapidly adjust gene expression in response to environmental stresses, offering a flexible and dynamic approach to survival. The RNA-binding protein Hfq has emerged as a critical component of this process and is increasingly recognized as a potential drug target for combating A. baumannii (186). Hfq is a homo-hexameric protein that facilitates the post-transcriptional regulation of gene expression by promoting the interaction between small RNAs (sRNAs) and their target messenger RNAs (mRNAs), thereby modulating gene expression (187).

Notably, A. baumannii Hfq possesses an extended C-terminal region compared to its homologs in other bacteria. This extended C-terminus enhances the ability to interact with a broader range of RNA molecules, contributing to the stabilization of Hfq-RNA complexes, particularly under prolonged stress conditions (186–188). In A. baumannii, Hfq is integral to the regulation of numerous virulence-associated genes. These include the csuA/BABCDE operon, which is involved in pilus formation, motility, adhesion to abiotic surfaces, and biofilm formation; the abaI gene, which controls quorum sensing and biofilm production; the bauA gene, which encodes a siderophore receptor essential for iron acquisition; components of the Type VI Secretion System (T6SS), which deliver toxic effectors to host cells and competing bacteria; and ompA, which facilitates bacterial adhesion to host tissues (189, 190). In addition, Hfq influences genes responsible for LPS modifications, which are crucial for evading the host immune response (191, 192). Given its central role in the regulation of these critical virulence factors, Hfq represents an attractive target for novel antimicrobial strategies. Inhibiting Hfq could significantly impair the ability of A. baumannii to adapt to hostile environments and reduce its virulence, thereby decreasing its capacity to cause disease.

The multifaceted gene regulation mechanisms in A. baumannii play a crucial role in facilitating the expression of antibiotic resistance genes. One notable example is the small regulatory RNA (sRNA) AbsR25, which regulates the expression of the major facilitator superfamily (MFS) efflux pump, AbaF (193, 194). This efflux pump contributes to the intrinsic resistance of A. baumannii to fosfomycin, highlighting the complexity of resistance mechanisms in this pathogen. Consequently, targeting the RNA-binding protein Hfq, which is essential for the function of sRNAs like AbsR25, could enhance the efficacy of traditional antibiotics against A. baumannii infections.

Despite the promising potential of Hfq as a drug target, several challenges complicate the development of effective inhibitors. The primary challenge lies in the extensive interface between sRNAs and their target mRNAs, which is too large for small molecules to disrupt without also affecting host cellular processes. In addition, RNA-binding proteins are highly conserved across many organisms, including humans, raising concerns about off-target effects and toxicity in the host. Moreover, the high genomic plasticity of A. baumannii could lead to rapid mutations that either prevent inhibitor binding to Hfq or activate compensatory pathways that bypass the need for this protein (195–198). Overcoming these hurdles is critical to making Hfq a viable druggable target for combating A. baumannii.

BamA (β-barrel assembly machine A)

BamA (β-barrel assembly machine A) is indispensable in A. baumannii and serves as the central component of the BAM (β-barrel assembly machine) complex, which is essential for the proper folding and insertion of β-barrel proteins into the outer membrane (199, 200). This process is critical not only for maintaining the structural integrity of the outer membrane but also for facilitating nutrient intake and enabling interactions with host cells, positioning BamA as a key virulence factor (201). Disrupting BamA function could weaken the outer membrane, rendering A. baumannii more susceptible to environmental stresses and eventual death (202). In addition, the high conservation of BamA across bacterial species suggests that such inhibitors could serve as broad-spectrum antibiotics (201, 203). Darobactin and dynobactin are ribosomally synthesized, post-translationally modified peptide antibiotics that target BamA in Gram-negative pathogens (204). Darobactin binds to the lateral gate of BamA of E. coli, preventing its insertion into the OMP, leading to cell death. A chemical analog of Darobactin, D22, displays improved efficacy against A. baumannii, reducing its MIC from 64 to 8 mg/L (205, 206).

However, the complex structure of the BamA β-barrel poses significant challenges for drug design. BamA interacts extensively and variably with a wide array of cellular proteins, making it difficult for small molecules to specifically inhibit its essential functions, as interacting proteins might affect the activity of the drug molecule (203, 207). Moreover, mutations in specific amino acids could reduce the binding efficacy of BamA inhibitors, necessitating the design of drugs that target multiple motifs within the protein (208–210). Despite these challenges, BamA’s large structure encapsulates unique bacterial motifs, offering opportunities for designing inhibitors with minimal host toxicity. Additionally, A. baumannii’s robust defense mechanisms, including a robust efflux system and an impermeable outer membrane, present further challenges for effective drug delivery to BamA (211, 212). Addressing these hurdles is crucial for developing BamA-targeted therapies against A. baumannii.

T6SS (Type VI secretion system)

The Type VI secretion system (T6SS) in A. baumannii represents a promising drug target due to its critical role in the bacterium’s survival and pathogenicity. This sophisticated syringe-like apparatus injects toxic effector molecules into neighboring host cells and competing bacteria, facilitating A. baumannii’s persistent survival in both environmental and hospital settings (213, 214). In polymicrobial environments, T6SS allows A. baumannii to assert dominance by delivering lethal effectors into rival bacteria, thereby inhibiting their growth or killing them outright, which contributes to its ability to resist antimicrobial treatments (215).

In addition to outcompeting other bacteria, T6SS mediates interactions with the host immune system. By injecting effectors into neutrophils and macrophages, A. baumannii can modulate host immune responses, establishing infections, particularly in immunocompromised patients (215, 216). Furthermore, T6SS is implicated in the enhanced production of biofilms, which are protective layers that contribute to chronic infections and colonization of medical devices, making A. baumannii infections particularly difficult to eradicate (217).

The complexity of the T6SS in A. baumannii presents several compelling drug targets that could be exploited to mitigate the bacterium’s virulence and resistance mechanisms. Key components such as the valine-glycine repeat G protein (VgrG) are integral to the structural integrity of the T6SS and play a crucial role in mediating the transfer of effector molecules into host or competing cells (218). A small-molecule inhibitor that disrupts VgrG function could significantly reduce A. baumannii’s virulence and its ability to evade the immune system. Similarly, the hemolysin-coregulated protein (Hcp) forms the tubular conduit for the delivery of effector molecules. Inhibitors that block Hcp polymerization could prevent the transfer of these toxic effectors, thereby neutralizing the T6SS (219).

Another critical component, TssM (ClpV), functions as an ATPase that powers the contraction of the T6SS sheath, facilitating effector injection. Inhibiting TssM could effectively halt the energy supply necessary for the T6SS’s operation, rendering the system non-functional (220). Furthermore, TssB and TssC, which constitute the sheath surrounding the Hcp tube, play essential roles in the physical mechanism of effector delivery. Targeting the assembly or function of these sheath components would disrupt the entire secretion system, blocking effector transfer and weakening the bacterium’s pathogenic potential (221). By focusing on these essential elements of the T6SS, which have no homology with humans, we can disrupt A. baumannii’s ability to establish and maintain infections, offering a promising new strategy in the fight against this formidable pathogen with any off-target or toxicity issues.

DsbA/DsbB (disulfide bond formation proteins)

In A. baumannii, the DsbA/DsbB system is integral to the proper folding of proteins within the periplasmic space, a process essential for maintaining outer membrane integrity and virulence. In the oxidative environment of the periplasm, disulfide bonds form between cysteine residues of periplasmic and membrane-associated proteins, which are crucial for bacterial virulence, membrane stability, and resistance to antibiotics (222, 223). This makes the DsbA/DsbB system an attractive target for the development of novel therapeutics against A. baumannii.

DsbA, an oxidoreductase, plays a pivotal role in this system by transferring disulfide bonds from its active site cysteine residues to substrate proteins, thereby facilitating their proper folding (223). During this process, DsbA is re-oxidized through electron transfer to the bacterial electron transport chain, a function mediated by DsbB, which re-establishes DsbA’s active state. The active site of DsbA, characterized by the conserved Cys-X-X-Cys motif, is critical for disulfide bond formation. Inhibiting this active site with small molecules could result in the misfolding and loss of function of numerous virulence and resistance proteins, potentially rendering A. baumannii avirulent, more susceptible to antibiotics, or even non-viable (224–228).

In addition, strategies targeting the DsbA-DsbB interface or the quinone-binding site of DsbB could prevent DsbB from reactivating DsbA, leading to the accumulation of inactive DsbA and ultimately bacterial cell death. However, the development of inhibitors against this system faces significant challenges. The redundancy in bacterial protein folding pathways may allow A. baumannii to compensate for the loss of DsbA/DsbB activity by activating alternative oxidative folding mechanisms (229, 230). Furthermore, the process of disulfide bond-mediated protein folding is highly conserved across all organisms, from gut microbiota to humans. Therefore, drug development efforts must ensure that inhibitors are highly specific to the DsbA/DsbB system of A. baumannii to avoid off-target effects and minimize toxicity.

FtsZ (filamenting temperature-sensitive mutant Z)

FtsZ has emerged as a highly promising drug target for the development of novel therapies against A. baumannii and other MDR pathogens (231). As a central player in bacterial cell division, FtsZ orchestrates the formation of the cytokinetic ring, known as the Z-ring, which is essential for the initiation and progression of cell division (232). During cytokinesis, FtsZ is recruited to the future site of septum formation, where it serves as a scaffold for the assembly of other critical cell division proteins, including FtsA, ZipA, FtsK, FtsQ, FtsL, FtsB, FtsW, FtsI (PBP3), FtsN, EnvC, and NlpD (233–235). These proteins work in concert to mediate cell wall remodeling and the successful segregation of daughter cells.

Structure-based alignment of the FtsZ protein from A. baumannii (SDF; AF-B0VNZ4-F1-v4) and E. coli (AF-P0A9A6-F1-v4) revealed key structural differences, as highlighted by DSSP annotations (Fig. S2). The N-terminal region (residues 1–60) is broadly conserved between the two species, maintaining the overall fold. However, several surface-exposed residue variations introduce local mismatches, despite preserving the core secondary structure. This is followed by the highly conserved β–α–β motifs that constitute the core of the GTPase active site. Minor loop differences in this region, particularly around the GTP-binding pocket, may subtly influence its conformation and dynamics. The conservation of this pocket, the inhibitor PC190723, along with its derivatives, has shown promising inhibitory activity by targeting the GTP-binding site and blocking Z-ring formation, leading to cell division arrest (170).

The central helical domain (residues 180–260) displays a dense helical arrangement in both proteins. Nevertheless, minor insertions and deletions, particularly around residue 233, lead to localized shifts in helical content. Notably, a glycine-to-serine substitution in A. baumannii may alter flexibility in this region, potentially affecting GTPase function. The interhelical loops often form flexible pockets that may represent promising sites for species-specific inhibitor targeting.

By contrast, the C-terminal region (residues 260–340) shows striking divergence in both sequence and structure. A. baumannii harbors multiple insertions, especially between residues 320–340, resulting in a pronounced shift from helix to loop conformations. These extended loops may present unique binding motifs absent in E. coli, offering a potential platform for the design of A. baumannii-selective FtsZ inhibitors. Furthermore, the extreme C-terminus of FtsZ is more compact in E. coli, whereas A. baumannii exhibits extended, intrinsically disordered regions. Given the critical role of this domain in mediating interactions with key division partners such as FtsA and ZipA, these disordered regions represent attractive targets for small molecules aimed at disrupting protein–protein interactions unique to A. baumannii.

Despite the ~20% homology between bacterial FtsZ and human tubulin, the design of inhibitors must prioritize specificity to avoid off-target effects while maintaining potent antibacterial activity. This specificity will be key to developing effective, safe therapies against A. baumannii and other MDR pathogens (236).

RND efflux pump AdeABC

Among the various AMR mechanisms in A. baumannii, RND efflux pumps play a central role due to their broad substrate specificity (7). Most clinically relevant A. baumannii isolates encode multiple RND systems, with nine identified in the ACB complex. Of these, AdeABC, AdeIJK, and AdeFGH are the most significant from a clinical standpoint, as they confer resistance to several major antibiotic classes, including β-lactams, aminoglycosides, fluoroquinolones, and tetracyclines (237). Each of these efflux systems functions as a tripartite complex spanning the bacterial cell envelope, comprising an inner membrane transporter (e.g., AdeB, AdeJ, AdeK), a periplasmic membrane fusion protein (e.g., AdeA, AdeI, AdeF), and an outer membrane channel (e.g., AdeC, AdeK, AdeH) (237).

Overexpression of AdeABC, particularly the AdeAB components, is frequently observed in MDR A. baumannii isolates. The component AdeC is often chromosomally encoded in MDR clinical isolates only (238). The expression of AdeABC is tightly regulated by multiple factors. Two key two-component systems (TCSs), AdeSR and BaeSR, alongside the transcriptional regulator DksA and a histone-like nucleoid-structuring protein (H-NS), contribute to this regulation (239–241). AdeS, a sensor histidine kinase of the AdeSR system, undergoes autophosphorylation upon detecting environmental stimuli such as exposure to bactericidal antibiotics (239). The phosphate group is then transferred to the response regulator AdeR, which subsequently binds to the intercistronic region between adeR and adeA, likely facilitating the recruitment of RNA polymerase and sigma factors to initiate transcription of the adeABC operon (239).

Interestingly, AdeABC can also be expressed at basal levels independent of AdeR phosphorylation, as the unphosphorylated form of AdeR is capable of binding, albeit at a lower level. The BaeSR system, typically activated in response to envelope stress induced by agents like tigecycline, overexpresses AdeABC. This stress often arises from impaired lipoprotein trafficking via the Sec and Lol pathways, leading to the accumulation of misfolded proteins in the periplasm (242). While BaeSR primarily responds to such envelope perturbations, AdeSR is more sensitive to disruptions like peptidoglycan instability and membrane damage induced by antibiotics such as aminoglycosides and β-lactams. Cross-talk between these TCSs allows for finely tuned expression of AdeABC in response to diverse stress conditions (243).

AdeB contains a large hydrophobic substrate-binding pocket enriched in aromatic and polar residues, which may allow for dynamic interactions with various drugs. This structural feature makes AdeB a potential target for drug development (244). However, targeting the AdeABC system remains challenging. Functional redundancy among RND pumps allows compensatory overexpression of alternative systems in the absence of AdeABC, undermining therapeutic efficacy (245). Although efflux pump inhibitors (EPIs) have been explored as adjuncts to restore antibiotic susceptibility, most available EPIs act non-specifically by disrupting the proton motive force (PMF), raising concerns about host toxicity (246). Moreover, the evolutionary capacity of A. baumannii to modulate efflux pump expression via complex regulatory networks complicates efforts to target TCSs (247). Therefore, further mechanistic studies are critical to overcoming the current limitations in targeting efflux-mediated resistance in A. baumannii.

Quorum sensing regulators AbaI and AbaR

A. baumannii, like many bacterial pathogens, employs quorum sensing (QS) to mediate population-level communication and coordinate collective behaviors (248). This process involves the synthesis and accumulation of signaling molecules known as autoinducers (AIs), primarily N-acyl homoserine lactones (AHLs). These are produced by the enzyme AbaI, an autoinducer synthase that catalyzes the reaction between S-adenosyl methionine (SAM) and acyl chains derived from acyl carrier protein (ACP) (249).

As the concentration of AIs in the environment increases and reaches a critical threshold, these molecules bind to AbaR, a LuxR-type receptor protein (250). Upon binding, the AbaR-AI complex functions as a transcriptional activator, regulating the expression of a range of genes associated with virulence, including those involved in biofilm formation (such as bap), outer membrane proteins (ompA), efflux systems (adeABC), type I pili, polysaccharide biosynthesis (PNAG), and oxidative stress response (251). AbaR also binds to a conserved lux box located upstream of the abaI promoter, thereby enhancing abaI transcription and reinforcing AI production through a positive feedback loop (252).

Although A. baumannii encodes a single type of AbaI synthase, clinical isolates have been shown to produce a diverse array of AHLs, ranging from short-chain (C6–C8, less common) to long-chain (C10–C16, more common) molecules. Among these, 3-hydroxy-C12-homoserine lactone is the most frequently detected (253, 254). More recently, a regulatory protein called AbaM has been identified within the QS locus. AbaM functions as a negative regulator of quorum sensing, acting as a feedback inhibitor that fine-tunes the QS response and prevents its overactivation (255).

Several anti-virulence strategies targeting QS have been explored over the past decade. These include the use of natural compounds such as essential oils (e.g., cinnamaldehyde, eugenol, thymol), flavonoids (e.g., quercetin), and marine-derived furanones, which have been shown to downregulate abaI and abaR expression, thereby reducing biofilm formation and virulence. Enzymatic degradation of AHLs using lactonases (such as AiiA) or acylases represents another approach to disrupt QS signaling (256–259). In addition, gene silencing technologies such as CRISPR interference (CRISPRi) and antisense oligonucleotides have been proposed to suppress transcription of QS-related genes (260).

Despite promising results in vitro, the clinical translation of QS-targeting strategies remains limited. To date, no quorum-sensing inhibitor (QSI) has been approved for therapeutic use, and QS inhibition is still not widely accepted as a viable clinical approach (261). Several challenges contribute to this translational gap, including the redundancy and variability of QS systems among A. baumannii strains. Blocking the AbaI/AbaR circuit alone may be insufficient, as some isolates either do not produce detectable AHLs or rely on alternative regulatory mechanisms. Moreover, many QSI candidates suffer from poor pharmacokinetic profiles, such as hydrophobicity, rapid clearance, low bioavailability, and inadequate tissue penetration, further limiting their efficacy in vivo (260).

Uridine diphosphate-3-O-(hydroxymyristoyl)-N-acetylglucosamine deacetylase (LpxC)

LpxC is a zinc-dependent metalloenzyme that catalyzes the first committed step in the biosynthesis of lipid A, the hydrophobic anchor of LPS in the outer membrane of Gram-negative bacteria (262). Beyond its structural role, lipid A functions as a potent endotoxin, capable of triggering septic shock, which is fatal in approximately 50% of affected individuals (263). In most Gram-negative pathogens, including E. coli, P. aeruginosa, Aquifex aeolicus, and A. baumannii, LpxC coordinates Zn(II) ions in its active site, although the number and coordination geometry can vary (264). In A. baumannii, a single Zn(II) ion is coordinated by His78, His237, and Asp241; however, this molecular arrangement may differ across species (265).

Despite LpxC’s essential role in bacterial viability and virulence, the efficacy of LpxC inhibitors has varied across species. While several inhibitors have demonstrated potent antibacterial activity against E. coli, P. aeruginosa, and Klebsiella pneumoniae, their activity against A. baumannii has been inconsistent or absent, warranting further investigation into species-specific resistance mechanisms (266, 267).

LpxC inhibitors are broadly categorized into hydroxamate and non-hydroxamate derivatives. While potent, hydroxamate-based inhibitors often exhibit host toxicity due to metabolic conversion to hydroxylamine (267). Notably, two inhibitors, ACHN-975 and RC-1, advanced to clinical trials but were discontinued in Phase I due to toxicity concerns (268, 269). In response, efforts have shifted toward the development of non-hydroxamate alternatives with improved safety profiles.

One such compound, TP0586532, is a non-hydroxamate LpxC inhibitor that has demonstrated efficacy against multiple Gram-negative pathogens, including carbapenem-resistant Enterobacteriaceae (CRE), with minimal toxicity. However, limited data are available regarding its activity against A. baumannii (270).

While direct bactericidal activity against A. baumannii remains limited, LpxC inhibitors may still serve as valuable adjuncts in combination therapies. For instance, PF-5081090, a hydroxamate-based LpxC inhibitor, exhibits no intrinsic antibacterial activity at 32 mg/L against A. baumannii isolates. Nonetheless, this concentration significantly potentiates the efficacy of other antibiotics, reducing the MIC of rifampicin by up to 1,000-fold and vancomycin or azithromycin by 8- to 16-fold (271). These findings underscore the potential of LpxC inhibitors to restore or enhance the activity of otherwise ineffective agents. Ongoing efforts in medicinal chemistry, including structural modification, aim to overcome limitations related to solubility and host toxicity, thereby improving the clinical viability of this promising class of antibacterial agents.

Capsule biosynthesis pathway: Wza, Wzb, Wzc, and Wzi

Capsule biosynthesis is a key determinant of A. baumannii’s resistance and persistence, particularly through its role in biofilm formation. The integrity of biofilms depends on extracellular polysaccharides (EPS), with capsular polysaccharides (CPS) contributing significantly to immune evasion (272). In A. baumannii, CPS is synthesized via the Wzy-dependent capsule biosynthesis pathway, which involves four principal components: Wza, Wzb, Wzc, and Wzi (273).

Wza, an outer membrane protein, mediates the translocation of CPS from the periplasm to the bacterial surface. Functional studies have identified Wza as a critical virulence factor—mutants lacking Wza exhibit significantly reduced survival in human ascites and serum (273). Wzb functions as a protein tyrosine phosphatase, while Wzc is a membrane-bound tyrosine kinase that undergoes autophosphorylation. Wzb dephosphorylates Wzc, a step that facilitates the coordinated export of EPS (274). Wzi, also located in the outer membrane, is involved in the final anchoring of CPS to the bacterial surface (275).

Despite its therapeutic potential, targeting the Wza-Wzb-Wzc-Wzi complex poses several challenges. The foremost is the extensive capsule heterogeneity among A. baumannii strains, with over 100 distinct capsular (K) loci identified. This diversity raises concerns about the strain specificity of any potential inhibitor, thereby limiting its clinical applicability (276). Furthermore, while capsule inhibition attenuates virulence, it is unlikely to be bactericidal. A. baumannii can compensate for capsule loss by upregulating alternative virulence mechanisms, such as efflux pumps, thereby maintaining overall fitness. Lastly, the organism’s highly plastic genome contributes to functional redundancy within capsule biosynthesis pathways, meaning not all strains rely on the Wza-Wzb-Wzc-Wzi machinery, further complicating universal targeting strategies (107).

CONCLUSION

The constant battle against A. baumannii signifies one of the great challenges in modern medicine. The pathogen has a marked ability to quickly develop resistance to multiple drugs and can survive well in hostile environments. The incidence of drug-resistant infections is worldwide, and combating carbapenem resistance is a global health priority. Literature over the last two decades demonstrates elevated mortality associated with Acinetobacter spp. infections, particularly for drug-resistant disease. This highlights the pressing demand for new treatments. The current treatment options are compromised by a sophisticated interplay of resistance mechanisms such as various β-lactamase production, mutation-mediated enhanced expression/activity of efflux pumps, and alteration in the target sites. The declining effectiveness of last-resort antibiotics such as carbapenems and colistin highlights the uncertain state of current therapeutic approaches.

Despite these obstacles, there are promising avenues for developing new anti-Acinetobacter agents that combat global AMR. The identification and understanding of drug targets, such as the Type VI secretion system (known as T6SS), the FtsZ protein, and proteins involved in disulfide bond formation (DsbA/Dsb), provide potential paths for overcoming resistance. Each of these targets plays a role in bacterial survival, making them appealing therapeutic targets. The incorporation of cutting-edge techniques in drug discovery, such as AI and structure-based screening, could accelerate the identification of effective compounds.

Looking ahead, the effective management of A. baumannii infections will require a multifaceted approach that combines the development of new drugs with robust antimicrobial stewardship practices. This plan should also emphasize the significance of global cooperation, in monitoring and research initiatives to track the rise of resistance and to ensure global availability of new treatments. Ultimately, though the obstacles are substantial, the advancements in understanding of resistance mechanisms and the identification of novel drug targets provide a hopeful outlook for the future of antimicrobial therapy against this formidable pathogen.

Biographies

Vineet Dubey (BTech, PhD) is a Postdoctoral Research Associate in the Antimicrobial Pharmacology and Therapeutics Group at the University of Liverpool. His research focuses on antimicrobial pharmacodynamics of WHO essential medicines, with a particular emphasis on Gram-negative pathogens. He is actively engaged in UKRI Strength in Places and Wellcome Trust-funded projects aimed at developing novel therapeutic strategies to combat antimicrobial resistance. Dr Dubey completed his PhD as a Department of Biotechnology-India Fellow at the Indian Institute of Technology Roorkee, where his work centered on combination therapies to address persistence and resistance in Acinetobacter baumannii. His areas of research interest include antimicrobial resistance mechanisms, pharmacology, and innovative approaches to overcoming multidrug resistance.

Nada Reza is currently an MRC Clinical Research Training Fellow at the University of Liverpool, undertaking a PhD in the molecular pharmacology of AMR. She is also a Specialist Registrar in Infectious Diseases and Microbiology at the Royal Liverpool University Hospital. Prior to this, Dr Reza completed undergraduate medical training at the University of Dundee and an intercalated BSc in Immunity and Infection at Imperial College London. She then undertook the Academic Foundation Programme in the Scottish Highlands before starting in Liverpool as an NIHR Academic Clinical Fellow. She is also the current Chair of the British Infection Association Early Careers Researchers' Committee.

William Hope (BMBS, FRACP, FRCPA, PhD) is Dame Sally Davies Chair of AMR Research at the University of Liverpool in the UK. Professor Hope is a Fellow of the Royal Australasian College of Physicians and a Fellow of the Royal College of Pathologists of Australasia. Areas of special interest and research are antimicrobial pharmacokinetics and pharmacodynamics, antimicrobial drug development, and individualization of antimicrobial therapy. He is a Fellow of the American Academy of Microbiology and the European Society of Clinical Microbiology and Infectious Diseases.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/cmr.00279-24.

Supplemental material. cmr.00279-24-s0001.docx.

Methods S1 to S3; Fig. S1 and S2.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.