Background

The emergence of multidrug-resistant (MDR) bacterial species is a hot global public health challenge. Among these challenging MDR bacteria, WHO has identified six opportunistic pathogens known as ESKAPE (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species) for research and development of new antibiotics [1]. The abbreviation ESKAPE indicates the bacteria`s ability to “escape” the killing of antibiotics and resist eradication by conventional therapies as well, leading to extensive morbidity and mortality among admitted patients within healthcare settings [2].

A. baumannii is a non-fermentative, non-motile, non-fastidious, catalase-positive, oxidase-negative, aerobic Gram-negative coccobacilli opportunistic pathogen responsible for different infections including pneumonia, bloodstream, wound, and urinary tract infections [3], and meningitis among patients in intensive care units [4]. It is also known as the “Iraqibacter” because of its emergence during the Iraq and Afghanistan war among US soldiers causing severe infections [5]; appears as a real challenging superbug for clinicians [6]. Because of its MDR features, the high mortality rate (up to 23% for hospital-admitted patients and up to 43% among ICU patients [7]), endurance on inanimate objects [8], biofilm-assisted survival in harsh environments [9, 10]; A. baumannii has got due attention globally.

Despite the efforts of the scientific community to develop new effective medications against the MDR pathogen, the number of antibiotics joining the market is running dry, and humanity is threatened [11]. So far, 300 million people are expected to die globally for the next 30 years, and a 60 to 100 trillion USD loss is expected if antimicrobial resistance (AMR) continues to be uncontrolled [12]. The strike due to AMR pathogens would be worse, particularly in poor nations like Ethiopia. Recent literature redirected the focus to an emerging eco-friendly, bio-control strategy as an alternative to/with antibiotics against resistant bacteria [13]. Among the options, bacteriophages, or phages in short, are emerging as new specific and rapid killing machines against MDR pathogens [14, 15]. Sharing the concern, this study aimed to investigate the role of phages against MDR A. baumannii (MDRAB) and biofilm-producing isolates recovered from inanimate objects at JMC, Ethiopia.

Materials and methods

Study setting

A health facility-based cross-sectional study was conducted from June to November 2019 in JMC where the facility is found in Jimma town, southwest of Ethiopia. Recently JMC has been providing different medical specialty services and has a bed capacity of 650 for over 15 million populations in the catchment area [16].

Data collection and processing

The number of rooms in the health facility was sampled based on the CDC, 2010 guideline for evaluating environmental cleaning; where sampling 15% of the rooms is considered reasonably representative for hospitals with ≥ 150 beds [17]. A total of 309 swab samples were collected randomly from high-touch surfaces in 37 rooms in five wards (13 surgical, 8 pediatrics, 7 medical, 6 gynecology and obstetrics rooms and 3 ICU rooms (surgical ICU, medical ICU, and pediatrics ICU)) in the morning between 8:30 − 9:00 A.M after routine morning cleaning. The number of swabs collected from each of the wards’ rooms i.e. surgical, pediatrics, medical, gynecology and obstetrics, and ICU wards were 99, 66, 58, 47, and 39, respectively. Each sampled swab was properly homogenized in 1mL sterile normal saline. One hundred μL of the sample was aseptically inoculated onto MacConkey agar (Oxoid, Ltd, Hampshire, England) and incubated aerobically at 37 °C for 24 h [18]. The mean colony forming unit per square centimeter (CFU/cm2) area was calculated and compared with the standard for high-touch surfaces which is ≤ 5CFU/cm2 [19]. Identification of bacteria was done using different characteristics including colony morphology, Gram stain, and biochemical profiling such as catalase, oxidase, citrate utilization test, Kligler iron agar (KIA), Sulfide Indole Motility test, oxidation-fermentation test, growth at 440 C, and inoculated on blood agar to check for hemolysis [20, 21]. A. baumannii produces colorless non-lactose fermenting shiny mucoid colonies on MacConkey agar and is the only group member that is capable to grow at 44ºC from the genus [22,23,24].

Antimicrobial susceptibility test and biofilm detection

Three to five pure colonies of A. baumannii isolates from overnight grown culture were suspended in sterile normal saline. The turbidity of the suspension was checked against 0.5 McFarland standard. Antimicrobial susceptibility testing (AST) was performed using the Kirby Bauer disk diffusion technique on Muller Hinton Agar (Oxoid, Ltd, Hampshire, England). The following antimicrobials were tested: ceftriaxone, 30 μg; ciprofloxacin, 5 μg; gentamicin, 10 μg; ceftazidime, 30 μg; cefepime, 30 μg; imipenem, 10 μg; meropenem, 10 μg; amikacin, 30 μg; doxycyciline, 30 μg; and trimethoprim/sulfamethoxazole, 1.25/23.75 μg) (Liofilchem srl, Italy). Antibiotic discs were placed firmly and incubated at 37ºC for 24 h. The zone of inhibition was measured and interpreted according to the CLSI 2018 recommendations [25]. Reference strains including P. aeruginosa ATCC 27853, K. pneumoniae ATCC 700603, S. aureus ATCC 25923, and E. coli ATCC 25922 were used for antimicrobial susceptibility and phage host range testing.

Microtiter plate assay (96 wells) was used to determine biofilm production following the protocol used by Sanchez et al. [26]. The bacterial suspension was added to freshly prepared Trypticase soya broth (TSB) (Oxoid, Ltd, Hampshire, UK) supplemented with 1% glucose and diluted to 0.5 McFarland turbidity standard. Then, 200 μL was added in each microtiter well for each isolate in triplicate and incubated at 37ºC for 48 h. Following incubation, the content of each well was aspirated and washed 3 times gently with sterile phosphate buffer saline (PH 7.2) to remove planktonic bacterial cells. The attached bacteria were fixed with 200 μL of methanol in each well. Then, 250 μl of 0.1% crystal violet solution was added to each well and was kept for 10 min at room temperature. Each microtiter well was washed with PBS saline to remove the staining solution. After plates were allowed to air-dry, 250 μl of 95% ethanol was added to solubilize the crystal violet dye by incubating for 15 min at room temperature. The solubilized content of each well was aspirated and transferred into a new microtiter plate well. The optical density (OD) of each well was measured at 595 nm by using an automatic ELISA Reader (Elisys Uno, Human Germany). The measured optical density (OD) from the triplicate wells was then averaged and the standard deviation was calculated. The cut-off optical density (ODc) was calculated and defined as three standard deviations above the mean OD of the negative control (Trypticase soya broth without bacteria). Based on the average OD produced by bacterial films at a wavelength of 595 nm; A. baumannii isolates were classified as bacterial OD < ODc = biofilm non-producer; OD > ODc, but < 2 ODc = weak biofilm producer; OD > 2 ODc but < 4 ODc = moderate biofilm producer and > 4 ODc = strong biofilm producer as described by Stepanovic and his team [27].

Bacteriophage isolation, enrichment, and characterization

Bacteriophage specific against A. baumannii was isolated from sewage samples collected at four different collection sites in JMC following the standard phage isolation protocol stated by Clokie et al., [28]. 50 ml of each sample was centrifuged at 10,000 rpm for 10 min to remove particulate materials. The supernatants were filtered by a 0.45-micrometer membrane filter. Then 20 ml of filtrate was mixed with an equal volume of double-strength broth containing 5mM MgSo4 along with 2 ml of log phase growth of A. baumannii and incubated at 370 C by shaking every 2–4 h. After 24 h incubation the content of the flask was centrifuged at 10,000 rpm at 40 C for 15 min. The supernatant containing phage was passed through a 0.45-micrometer pore membrane filter under aseptic conditions and the filtrate was used for amplification of phage. The spot assay was used to check for the phage activity against A. baumannii [29, 30]. The host bacterial cell suspension (0.1ml) was added to sterile soft agar (0.8%) maintained in a molten state at 45 °C in a water bath and quickly mixed. Then, the mixture was poured into previously prepared nutrient agar plates and two drops (10 μl) of the amplified filtrate were spotted on the plate at two different places. The plates were examined the next day for clearance at the spotted area after incubation at 37 °C for 24 h. Phage activity was examined against known control strains of P. aeruginosa ATCC 27853, K. pneumoniae ATCC 700603, S. aureus ATCC 25923, and E. coli ATCC 25922. The temperature stability of bacteriophages was evaluated by incubating the phage suspension at 10, 25, 37, 44, 50, 60, and 65°C for 1 h before overnight incubation with A. baumannii to determine if the phage retains its lytic activity against the host bacteria using spot assay [31]. Similarly, the phage-biofilm degradation was assessed [32] along with ciprofloxacin (30 μg/ml) to compare anti-biofilm activity [33], and normal saline was used as a control [27]. To evaluate the biofilm eradication activity of phage, 100 μl of A. baumannii culture in the log phase was inoculated into 200ml of Brain heart infusion (BHI) broth. The inoculated broth was aseptically poured into a tip box containing cover glass leaving liquid air interphase for growth of biofilm and incubated for 36 h. After the growth, the cover glass was aseptically removed and washed with sterile phosphate buffer saline (pH 7.2). Then, the biofilm developed on cover glass was treated with bacteriophage or ciprofloxacin (30 μg/ml) or normal saline and incubated for 3 and 36 h [32]. After the respective treatment, biofilm grown on a cover glass was washed gently with sterile phosphate buffer saline (pH 7.2) and stained with (0.1%) crystal violet for 10 min. The stained biofilm was rinsed with sterile distilled water allowed to air dry and put on a clean microscope slide for microscopic examination. The cover glass treated with normal saline was used as a control [27, 33].

Statistical analysis

Data were checked and cleared for completeness and exported to SPSS for analysis. The chi-square (χ2) test was used to determine the association between variables. A P-value of < 0.05 was considered statistically significant for association.

Results

Isolation and enumeration ofA. baumannii.

From a total of 309 health facility high-touch surfaces bacteriological samples, 184 (59.5%) showed Gram-negative bacterial growth on MacConkey agar plates. However, the recovery rate of A. baumannii was 6.5% (n = 20) or about 11% from among Gram-negative. The distribution of A. baumannii from the inanimate objects was seven from the floor, four from tables, three from bed frame, two from the door handle, and one each from the oxygen control valve, wash sink, ventilator screen, and circuit of mechanical ventilation. The number of A. baumannii recovered varies significantly with inanimate objects (P < 0.005), but not with wards (Table 1).

Table 1 Distribution of A. baumannii on inanimate objects at JMC, June-November, 2019
Full size table

Evaluation of the environment for cleaning and disinfection process of frequently hand contact surfaces indicated the possibility of an increased risk of infection for patients from the environment whatever the type of organism is. The colony count in all the rooms was above standard for high touch surface ≤ 5CFU/cm2.

Antibiotic resistance profile and biofilm production of A. baumannii

The antimicrobial resistance patterns of A. baumannii demonstrated an increased level of resistance to imipenem (100%), followed by ceftriaxone and ceftazidime (95%), Cefepime (80%), Cotrimoxazole (70%), and Meropenem (60%). However, they were sensitive to Doxycycline (100%) followed by Amikacin (95%) (Table 2). Of the 20 isolates recovered, 85% (n = 17) of them were biofilm producers, and recovery of biofilm-forming and MDR A. baumannii was shown to vary with admission wards (Table 3).

Table 2 Antibiotic resistance profiling of A. baumannii isolates at JMC, June-November, 2019
Full size table
Table 3 Biofilm production and level of MDR profiling of A. baumannii isolates at JMC, June-November, 2019
Full size table

60% (12/20) of the isolates were MDR A. baumannii. Similarly, nearly 59% of (10/17) the biofilm producer A. baumannii isolates were MDR. Although the association is indeterminate, being MDR isolate was shown to vary with inanimate objects too; where 5 out of the 12 MDRAB isolates were recovered from the facility floor. The isolates’ OD value was cross-tabulated against the number of antibiotics resisted (Supplementary Table 1). The Spearman correlation of OD values of biofilm assay had shown a significant association with the number of A. baumannii isolates resistant to antibiotics (r = 0.635, p-value = 0.027).

Activity of phages against biofilm-forming and MDR A. baumannii isolates

Of the four sewage samples processed, one lytic phage specific against MDR A. baumannii was isolated. Seven of the 20 A. baumannii isolates (biofilm producer, MDR, or both biofilm producer and MDR) tested have shown lysis by the phage fully or partially (Table 4). In this study, 40% of both biofilm producers and MDRAB were potentially affected by the virulent lytic pages. The phage has specific lytic activity against A. baumannii isolates (Fig. 1) but non-lytic against other control reference bacterial strains including P. aeruginosa ATCC 27853, K. pneumoniae ATCC 700603, S. aureus ATCC 25922, and E. coli ATCC 25922.

Table 4 Lytic activity of phage against MDR and/or biofilm producer A. baumannii isolates at JMC June-November, 2019
Full size table
Fig. 1
figure 1

Spot assay showing complete clearance of spotted area which indicates lytic phage activity

Full size image

Temperature stability of the phage at 106 PFU/ml was tested by incubating the phage suspension at 10, 25, 37, 44, 50, 60, or 65°C for 1 h before overnight incubation of phages with the host bacteria. Thus, the phage was active in lysing the host bacteria at 10 to 50 °C but did not at 60°C and beyond. The phage biofilm degradation was examined relative to ciprofloxacin (30 μg/ml) and normal saline as control. Accordingly, phages were effective in eradicating biofilm producers and MDRAB isolates more efficiently than ciprofloxacin (30 μg/ml) as evaluated microscopically in this study (Fig. 2).

Fig. 2
figure 2

Illustrating biofilm eradication by phage isolates compared to ciprofloxacin and normal saline

N.B: NS-1, Ci-1 and Ph-1 cover glass were treated with normal saline, ciprofloxacin, and phage respectively for 3 h whereas NS-2, Ci-2 and ph-2 cover glass were treated with normal saline, ciprofloxacin, and phage for 36 h, respectively

Biofilm containg cover glass treated in phage and ciprofloxacillin showed variable eradiction in different time frame. The phage treated showed more eradication. The normal saline treated cover glass showed growth of the bacteria

Full size image

Discussion

The detection rate of A. baumannii in JMC facility environment including the surface of the floor, table handle, bed frames, and other frequent hand touch inanimate objects was 6.5% with its maximal recovery from ICU surfaces. The finding is numerically comparable with studies reported from Algeria (7.7%) [34], Brazil (9.5%) [35], and France (4.9%) [36]. But, it was quite lower relative to findings from Iran (17%) [37] and Jordan (49.7%) [38]. This epidemiologic variation could be attributed by the difference in health facilities, the neatness of settings, and adherence to infection prevention and control implementation strategies. Though the medical center rooms were disinfected with a 1:10 concentration of 5% bleach three times a day, A. baumannii was isolated from inanimate surface samples. This is an indication of a potential outbreak of A. baumannii infections as all the sampled wards were contaminated [34, 35, 37].

The feature of A. baumannii being MDR and its capability to remain viable in soil, and environmental contamination of health facilities leads to a global public health challenge [39]. Briefly in this study, 60% of the isolates were MDR (and 85% biofilm producers). Higher prevalence of the bacterium in health settings has been known in different studies such as in Brazil (98.8%) [35], and China (65% and more) [3] whereas less prevalently in Maryland, USA (9.8%) [40]. This could be due to its high adaptability of harsh environmental conditions [4, 9, 10, 39]. As a result, different scientific reports suggest the use of alternative antimicrobial agents and in this regard, the application of phages was endorsed as a newly emerged potential therapeutic option against MDR pathogens [14, 41,42,43,44]. With these underlying reasons, the current work investigated the use of specific phage against MDR, and biofilm producer A. baumannii isolates recovered from the hospital inanimate objects.

In this pilot study, the phage isolated against one of the MDR bacterium, A. baumannii had shown full or partial lytic activities against 35 to 42% of biofilm producers, MDRAB or both biofilm producers and MDRAB isolates, (Table 4). The phage isolated was able to lyse only A. baumannii isolates in contrast to other ATCC reference strains. The thermal stability ranged from 10 to 50 °C for 1 h [45] with possible variation with the host bacterial strain [46]. This finding is supported by previous studies [47], and even with different bacterial species [32, 41] and animal models used [42, 44, 46]. In addition, phages were more active at deterring bacterial resistance (42%) than degrading biofilms (35.3%) as demonstrated in the previous study [41]. As a limitation, our study merely depends on the phenotypic characterization of phages, and phage-biofilm clearance. In line with other literatures, this study can decipher the most abundant biological entities that have the potential to be used on inanimate objects and environments in health facilities as a biological control as well as a therapeutic candidate against multidrug-resistant and biofilm-producing A. baumannii isolates.

Conclusion

The detection of substantial MDRAB isolates in inanimate objects and environments of the medical center is an indication of the potential occurrence of MDRAB-associated outbreaks in the study setting unless proper decontamination strategies are in place. The good sensitivity of MDRAB isolates, biofilm degradation, thermal stability, and host specificity of phages in our study aspired to potentially identify them as a biocontrol or decontaminating agent from sewage sources in Jimma Medical Center besides the routine cleansing agents. Therefore, we recommend further efforts to characterize phages against emerging MDRAB isolates in detail.