Hospital surfaces may hide virulent multidrug-resistant Acinetobacter baumannii reservoirs

Multidrug-resistant Acinetobacter baumannii has earned its reputation as one of the most formidable hospital pathogens on Earth, and a new study from South India shows why the threat extends well beyond infected patients. Researchers at JSS Medical College and Hospital in Mysuru, working with colleagues at Kasturba Medical College Mangalore, paired every clinical isolate of the bacterium with strains recovered from the patient’s immediate surroundings — bed rails, floors, trolleys, walls, and nursing station surfaces — and then subjected all 110 isolates to both classical virulence assays and whole-genome sequencing. The results, published in Frontiers in Cellular and Infection Microbiology, reveal a stark quantitative divide between bacteria thriving inside the host and bacteria lurking on dry hospital surfaces, while simultaneously confirming that environmental strains carry a substantial arsenal of virulence genes that could reignite infection at any moment.

The study design was elegantly simple but technically rigorous. Between August 2023 and October 2024, the team collected 55 non-duplicate clinical isolates of multidrug-resistant (MDR) A. baumannii from blood, urine, sputum, endotracheal aspirates, pus, wound swabs, and body fluids. For each infected patient, five environmental samples were collected from frequently touched surfaces, generating 275 environmental specimens from which 55 non-duplicate environmental isolates were derived after excluding strains with identical antimicrobial susceptibility profiles. All 110 isolates were confirmed as MDR by international consensus criteria, meaning resistance to at least one antimicrobial agent in three or more drug classes. The clinical isolates were overwhelmingly recovered from critically ill inpatients — 82.4 percent came from intensive care unit patients — and endotracheal aspirates accounted for 67.3 percent of clinical specimens, reflecting the pathogen’s tight association with ventilator-associated pneumonia.

Antimicrobial susceptibility testing painted a grim picture of the resistance landscape. Every clinical isolate was resistant to imipenem, meropenem, ciprofloxacin, ceftazidime, ceftriaxone, and piperacillin-tazobactam, with 100 percent resistance also observed for these agents among environmental isolates. Tigecycline retained the highest in vitro activity, with 78.2 percent of clinical isolates and 44.8 percent of environmental isolates remaining susceptible, followed by minocycline. All clinical isolates were reported as intermediate to colistin by the VITEK 2 system, underscoring how narrow the therapeutic options have become for this WHO-designated critical priority pathogen.

But the real story of this research lies not in resistance alone — it lies in virulence. The team evaluated four major phenotypic virulence traits using standard laboratory assays: biofilm formation via the tissue culture plate method with crystal violet staining and optical density quantification at 570 nanometers, haemolytic activity on 5 percent sheep blood agar, proteolytic activity on skim milk agar through casein hydrolysis, and siderophore production on Chrome Azurol S agar detected by characteristic orange or yellow halos. The phenotypic divide between clinical and environmental isolates was dramatic. Strong biofilm formation was observed in 61.8 percent of clinical isolates compared with just 10.9 percent of environmental isolates — a difference so pronounced it yielded a P value below 0.0001. Haemolytic activity was detected in 52.7 percent of clinical isolates versus 23.6 percent of environmental isolates, and siderophore production in 56.4 percent versus 30.9 percent, both statistically significant differences.

Proteolytic activity, by contrast, was remarkably conserved across both groups, with 65.5 percent of clinical isolates and 60.0 percent of environmental isolates showing casein-degrading capacity — a difference that did not reach statistical significance. This suggests that proteolytic capacity, mediated largely by the secreted zinc-metalloprotease CpaA and the autotransporter adhesin Ata, may represent a relatively stable trait within this bacterial population, one that persists regardless of whether the organism is colonising a host or a hospital surface. CpaA has been shown in prior research to interfere with host coagulation through degradation of coagulation-associated proteins, adding a layer of pathogenic sophistication to this conserved capability.

Whole-genome sequencing on the Illumina MiSeq platform, followed by de novo assembly with SPAdes and annotation with Prokka, allowed the researchers to screen all genomes against the Virulence Factor Database using ABRicate. Among strong biofilm-producing clinical isolates, the quorum-sensing genes abaI and abaR, the outer membrane protein ompA, and the pgaB, pgaC, and pgaD genes of the poly-β-1,6-N-acetylglucosamine synthesis operon were detected in 100 percent of isolates. The biofilm-associated protein gene bap was present in 88.2 percent, and the csu pilus assembly genes ranged from 64.7 to 91.2 percent carriage. Among haemolytic clinical isolates, hemO was universally present, followed by plc2 in 89.7 percent, plc1 in 65.5 percent, and plcD in 55.2 percent. Siderophore-producing clinical isolates carried the bar, bas, bau, and entE gene clusters at near-universal frequencies, consistent with the acinetobactin-mediated iron acquisition system that A. baumannii deploys to overcome nutritional immunity — the host’s strategy of sequestering free iron to starve invading bacteria.

Yet the genomic analysis revealed something more unsettling than the phenotypic results alone. Several virulence-associated genes were detected in isolates that showed no corresponding phenotypic expression — a phenomenon the authors describe as phenotype-genotype discordance. Non-biofilm-producing isolates carried abaI, ompA, pgaB, pgaC, and pgaD at low but detectable frequencies, and one non-proteolytic environmental isolate still harboured cpaA. Gene carriage, the findings make clear, represents genetic potential rather than guaranteed functional output. Phenotypic expression depends on regulatory mechanisms, environmental conditions, and interactions among multiple determinants, meaning that the presence of a virulence gene alone cannot reliably predict the behaviour of a given strain under real-world conditions.

The environmental isolates, despite their lower phenotypic virulence, retained a substantial repertoire of virulence-associated determinants. All six strong biofilm-producing environmental isolates carried the full complement of abaI, abaR, ata, csuA/B, csuB, csuC, ompA, pgaB, pgaC, and pgaD genes. Environmental haemolytic isolates carried hemO in 84.6 percent of cases, and environmental siderophore producers harboured the bas and bau clusters at high frequencies. The lower phenotypic expression observed among environmental strains may reflect differences in ecological conditions or regulatory states rather than complete loss of virulence capacity — meaning these surface-dwelling organisms possess the genetic blueprint to become highly pathogenic should conditions favour expression.

Environmental surveillance identified MDR Acinetobacter in 24.4 percent of the 275 samples collected from patient surroundings, with bed rails accounting for 41.8 percent of recoveries and floor surfaces for 36.4 percent. Although environmental recovery does not establish the directionality of transmission — the cross-sectional design cannot determine whether patients contaminated their surroundings or vice versa — the findings support the established role of contaminated hospital surfaces as reservoirs for this pathogen. Prior research has demonstrated that biofilm formation enhances A. baumannii survival on dry surfaces, and the current study adds a critical dimension by showing that these environmental reservoirs are not merely passive persistence sites but potentially active sources of virulent, drug-resistant organisms capable of causing severe infections in vulnerable patients.

The integrated approach — combining phenotypic virulence assays with whole-genome sequencing — provides a more comprehensive assessment of A. baumannii virulence potential than either method alone. Whole-genome sequencing identifies the full repertoire of potential virulence determinants, whereas phenotypic assays provide evidence of functional expression under specific laboratory conditions. The coexistence of extensive antimicrobial resistance, environmental persistence, and virulence potential in both clinical and environmental isolates underscores the importance of continuous environmental surveillance, rigorous surface cleaning, hand hygiene, and appropriate disinfection strategies. As the authors conclude, hospital environmental reservoirs may harbour organisms with considerable pathogenic potential, and controlling MDR A. baumannii in healthcare settings requires attention to both patient-associated and environmental sources — a lesson that grows more urgent as therapeutic options continue to narrow.

Subject of Research: Comparative phenotypic and genomic analysis of virulence determinants in multidrug-resistant Acinetobacter baumannii clinical and patient-environmental isolates

Article Title: Virulence determinants in multidrug-resistant Acinetobacter baumannii: comparative phenotypic and genomic analysis of clinical and patient-environmental isolates

Article References: Basavaraju, P., Mahale, R. P., Rao, M. R., & Shenoy M, S. (2026). Virulence determinants in multidrug-resistant Acinetobacter baumannii: comparative phenotypic and genomic analysis of clinical and patient-environmental isolates. Frontiers in Cellular and Infection Microbiology, 16, Article 1969523. https://doi.org/10.3389/fcimb.2026.1969523

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1969523

Keywords: Acinetobacter baumannii, multidrug resistance, virulence factors, biofilm formation, whole-genome sequencing, hospital environment, healthcare-associated infections, siderophore production, haemolytic activity, proteolytic activity, phenotype-genotype discordance, infection control