Characterization of Acinetobacter baumannii Isolated from Raw Milk

Simple Summary Although milk is a significant nutrient source for humans, it can be associated with various bacterial infections. Acinetobacter species can be found in milk due to residual water in milking machines, milk pipelines or coolers, the inadequate cleaning of dairy equipment, tainted udders and teats, the improper transport and storage of milk and the inadequate cleaning of dairy equipment, causing diseases. Most members of the genus Acinetobacter are opportunistic commensals with limited virulence and are clinically insignificant. However, Acinetobacter infections have recently increased in severity due to the frequent use of mechanical breathing devices, venous catheters and antibiotics, and they pose significant public health concerns. Acinetobacter baumannii (A. baumannii) is an opportunistic pathogen that causes various nosocomial infections. Studies using animal models and clinical data demonstrated that A. baumannii is a highly virulent species. It is a significant pathogen, especially due to the emergence of multidrug-resistant (MDR) strains and their association with many nosocomial infections and community-acquired infections. Abstract Acinetobacter baumannii (A. baumannii) is an opportunistic pathogen associated with nosocomial infections. In this study, 100 raw milk samples were collected from Qena, Egypt, and subjected to conventional and molecular assays to determine the presence of A. baumannii and investigate their antimicrobial resistance and biofilm formation. Our findings revealed that, among the 100 samples, Acinetobacter spp. were found in 13 samples based on CHROM agar results. We further characterized them using rpoB and 16S-23SrRNA sequencing and gyrB multiplex PCR analysis and confirmed that 9 out of the 13 Acinetobacter spp. isolates were A. baumannii and 4 were other species. The A. baumannii isolates were resistant to β-lactam drugs, including cefotaxime (44%), ampicillin-sulbactam and levofloxacin (33.3% for each), imipenem, meropenem and aztreonam (22.2% for each). We observed different antimicrobial resistance patterns, with a multi-antibiotic resistant (MAR) index ranging from 0.2 to 0.3. According to the PCR results, blaOXA-51 and blaOXA-23 genes were amplified in 100% and 55.5% of the A. baumannii isolates, respectively, while the blaOXA-58 gene was not amplified. Furthermore, the metallo-β-lactamases (MBL) genes blaIMP and blaNDM were found in 11.1% and 22.2% of isolates, respectively, while blaVIM was not amplified. Additionally, eight A. baumannii isolates (88.8%) produced black-colored colonies on Congo red agar, demonstrating their biofilm production capacity. These results showed that, besides other foodborne pathogens, raw milk should also be examined for A. baumannii, which could be a public health concern.


Introduction
Milk is a significant nutrient source for humans, but drinking unsafe milk might cause various bacterial infections [1,2], including those by Acinetobacter. These foodborne Acinetobacter's antimicrobial susceptibility test was conducted using the agar disc diffusion method based on the Clinical and Laboratory Standards Institute (CLSI) guidelines [29]. Mueller-Hinton agar was used (Oxoid, CM0337, Basingstoke, UK). The antibiotics examined included amikacin (30 g/disc), gentamicin (10 g/disc), ampicillin-sulbactam (10 g/disc), piperacillin (100 g/disc), cefotaxime (30 g/disc), cefepime (30 g/disc), imipenem (10 g/disc), meropenem (10 g/disc) and ciprofloxacin (5 µg/disc). The zone diameter breakpoints (mm) were interpreted according to CLSI [26] and the European Committee on Antimicrobial Susceptibility Testing (EUCAST) [30]. The Multi-Antibiotic Resistant (MAR) index is calculated based on the number of antibiotics that an isolate is resistant to (a) divided by the total number of antibiotics utilized in the study (b) [31], using the equation shown below: MAR Index = a/b

Screening of Metallo-β-lactamases Production Using the Disk Method
Indicator strain: E. coli (ATCC 29522) was obtained from the Faculty of Agriculture at Ain Shams University in Giza, Egypt.
A suspension of a tested A. baumannii isolate was made by suspending a full 10 µL inoculation loop of a freshly cultured A. baumannii isolate, taken from a blood agar plate in 400 µL of water. Consequently, the test disc, which contained 10 g of meropenem, was dipped in the suspension and incubated at 35 • C for two hours. The disc was taken out of the suspension using an inoculation loop and put on a Mueller-Hinton agar plate, which was cultured with susceptible E. coli (ATCC 29522) that was prepared as a suspension of 0.5 a McFarland at OD595 and was then incubated at 35 • C. The susceptibility disk's meropenem was rendered inactive, allowing the E. coli to grow unhindered; this means that the isolates were able to produce carbapenemases. A clear inhibitory zone was produced after 24 h with discs that were incubated in phosphate puffer saline (PBS) as a negative control devoid of carbapenemases [32].

Detection of Metallo-β-lactamasesgenes
Multiplex PCR was performed to investigate the presence of carbapenemases, including OXA-type, bla OXA-51-like , bla OXA-23-like and bla OXA-58-like , based on a previous study [33]. Amplification was carried out using a final volume of 25 µL, consisting of 1X PCR buffer (Qiagen, Hilden, Germany), 1U Taq polymerase (Roche, Meylan, France), 2 mMMgCl 2 , 200 mM of Deoxynucleo-tide triphosphate (dNTP) (Biotools, Madrid, Spain), 0.2 mM of each primer (Pharmacia Biotech, Piscataway, NJ, USA) and 1 µL of template DNA. The PCR was carried out using a thermal cycler (Hamburg, Germany) under the following conditions: 94 • C for 5 min, followed by 30 cycles at 94 • C for 30 s, 53 • C for 40 s and 72 • C for 50s, then final extension at 72 • C for 6 min. The PCR products were separated by electrophoresis on 1.5% agarose gels (Synacron) and visualized under a UV gel documentation system after staining with ethidium bromide.
According to Kazi et al. [34] the genes bla IMP , bla VIM and bla NDM were amplified using uniplex PCR. The amplification was conducted using a final volume of 25 µL containing: 12.5 µL of PCR Mastermix (Emerald Amp GT), 1 µL for each primer (Table 1), 1.5 µL of the DNA template and 9 µL of PCR-grade water using a thermal cycler (MJ Research, Inc. Watertown). The optimal cycling conditions were 94 • C for 3 min, followed by 29 cycles of 94 • C for 30 s, 60 • C for 30 s and 72 • C for 30 s, with a final extension of 72 • C for 5 min. The bands were detected on 2% agarose gel electrophoresis.  used to measure biofilm production. Biofilm-producing colonies appear as black colonies and can also stain the medium due to exopolysaccharide synthesis [36,37].

Microtiter Plate Technique
This technique was performed according Stepanović et al. [38] Briefly, a brain-heart infusion broth (BHI) (Oxoid, Waltham, CA, USA) was inoculated with A. baumannii strains and was left to grow for 24 h at 37 • C. OD was adjusted to an OD600 of 1 ± 0.05, and then the bacteria were diluted 1:100 with sterile BHI with 1% glucose (Oxoid) before being put in a 96-well polystyrene microplate containing 200 µL of BHI medium and then being incubated overnight at 37 • C. The negative control wells only used the medium. The P. aeruginosa ATCC 27853 reference strain was the positive control for biofilm formation. After incubation, the media and the majority of the bacteria were quickly removed from the plates after incubation. A total of 200 µL of PBS 1× was introduced into the wells using a pipette, and after that, the plate was tilted to remove the liquid for pipette-based washing. Washing was repeated two more times. Using crystal violet staining (0.5%), which involves pouring 150 µL of dye into each well and letting them sit at room temperature for 5 min, biofilm formation was assessed. The extra dye was washed out using a pipette. After the plates had air-dried, the leftover dye was dissolved with 200 µL of glacial acetic acid (33% v/v) in each well. Using the ELISA auto-reader (Thermo Fisher Multiskan TM FC), staining was measured at OD620. The criteria for categorized biofilm formation were as follows: biofilm was not formed if OD ≤ ODc (negative), it was weak if ODc < OD < 2 × ODc and it was moderate if 2 × ODc < OD < 4 × ODc. A biofilm was considered strong at 4 × ODc < OD.

Results of the Isolation and Identification of A. baumannii
In this study, out of 100 milk samples collected, 13 were suspected to be contaminated with Acinetobacter spp. based on the CHROM agar observations. Microscopic and biochemical examination showed the presence of gram-negative cocci that were catalase-positive and negative for oxidase and nitrate reduction tests. Moreover, negative results for hemolysis and motility. The isolates showed alkaline reaction in triple sugar iron (K/K) and variable results with citrate, arginine hydrolysis and the glucose fermentation test were demonstrated (Supplementary Table S2).
Furthermore, we used molecular identification to identify different Acinetobacter species by amplifying and sequencing genes, including rpoB, 16S-23SrRNA, and gyrB. The results showed that certain rpoB genes were amplified in the suspected 13 isolates. Finally, these isolates were classified as A. baumannii (nine isolates), A.pitti (two isolates), A. rudis (one isolate) and A. oryzae (one isolate) ( Table 1). We observed 99% similarity in the rpoB gene sequences from all strains using pairwise comparison. Further, the intergenic spacer region of the 16S-23SrDNA was sequenced to identify the A. calcoaceticus-A. baumannii complex, and the results showed that, of the 13 Acinetobacter isolates, 11 belonged to the A. calcoaceticus-A. baumannii complex, 9 were A. baumannii and 2 were A. pitti (Table 1).
Multiplex PCR was performed using three primers for the gyrB gene for both A. baumannii and genomic sp. 13TU, and in 11 isolates, a 294-base-pair (bp) amplicon (sp4F to sp4R) was observed, whereas 9 A. baumannii isolates produced the second amplicon of 490 bp (sp2F to sp4R) ( Figure 1).
Multiplex PCR was performed using three primers for the gyrB gene for both A. baumannii and genomic sp. 13TU, and in 11 isolates, a 294-base-pair (bp) amplicon (sp4F to sp4R) was observed, whereas 9 A. baumannii isolates produced the second amplicon of 490 bp (sp2F to sp4R) ( Figure 1).

PCR results of the Carbapenem Resistance Genes
The PCR results showed that blaOXA-51 and blaOXA-23 genes were amplified in nine (100%) and five (55.5%) of the A. baumannii isolates, respectively, while the blaOXA-58 gene was not amplified in any isolate ( Figure 3). Additionally, the MBL genes, blaIMP, and blaNDM, were found in one (11.1%), and two (22.2%) isolates, respectively (Figures 3-5), while blaVIM was not amplified in any of them.

PCR Results of the Carbapenem Resistance Genes
The PCR results showed that bla OXA-51 and bla OXA-23 genes were amplified in nine (100%) and five (55.5%) of the A. baumannii isolates, respectively, while the bla OXA-58 gene was not amplified in any isolate ( Figure 3). Additionally, the MBL genes, bla IMP, and bla NDM , were found in one (11.1%), and two (22.2%) isolates, respectively (Figures 3-5), while bla VIM was not amplified in any of them.  Even though the data clearly indicated some resistance of A. baumannii isolates to β-lactam antibiotics, including penicillin, cephalosporin and carbapenem, we focused on carbapenem resistance because 44.4% (4\9) of isolates showed resistance for imipenem and meropenem. The phenotypic results of the MBL test confirmed the production of carbapenemase in these four out of nine isolates (Figure 2).

PCR results of the Carbapenem Resistance Genes
The PCR results showed that blaOXA-51 and blaOXA-23 genes were amplified in nine (100%) and five (55.5%) of the A. baumannii isolates, respectively, while the blaOXA-58 gene was not amplified in any isolate ( Figure 3). Additionally, the MBL genes, blaIMP, and blaNDM, were found in one (11.1%), and two (22.2%) isolates, respectively (Figures 3-5), while blaVIM was not amplified in any of them.

Biofilm formation of A. baumannii Isolates
Furthermore, eight A. baumannii isolates (88.8%) produced black colonies on Congo red agar, demonstrating their biofilm-producing capacity, while just six isolates (66.6%) were able to form biofilm (one strong, three moderate and two weak biofilm formations) on the microtiter plate method (Figure 6 and Supplementary Table S3).

Biofilm formation of A. baumannii Isolates
Furthermore, eight A. baumannii isolates (88.8%) produced black colonies on Congo red agar, demonstrating their biofilm-producing capacity, while just six isolates (66.6%) were able to form biofilm (one strong, three moderate and two weak biofilm formations) on the microtiter plate method (Figure 6 and Supplementary Table S3).

Biofilm Formation of A. baumannii Isolates
Furthermore, eight A. baumannii isolates (88.8%) produced black colonies on Congo red agar, demonstrating their biofilm-producing capacity, while just six isolates (66.6%) were able to form biofilm (one strong, three moderate and two weak biofilm formations) on the microtiter plate method (Figure 6 and Supplementary Table S3).

Discussion
Recently, A. baumannii has been significantly linked to several infections, including catheter-related infections, meningitis, bacteremia, soft-tissue infections, peritonitis and endocarditis, in intensive care units (ICU) due to its extraordinary capacity to obtain or enhance the antimicrobial resistance factors. Therefore, scientists are currently interested in discovering this novel, antibiotic-resistant strain [39].
We identified Acinetobacter spp. in 13 milk samples on CHROM agar media. Several authors have suggested using CHROM agar as a quick and easy medium for detecting Acinetobacter, as it contains chromogenic substrates cleaved by Acinetobacter spp. enzymes, resulting in unique, red-colored Acinetobacter colonies. This increases the efficiency of infection control procedures, decreases the time required to administer the right antibiotic therapy to the infected patients and, hence, lowers mortality [2,40].
However, further information about the interspecies relationships among the genus Acinetobacter is needed. It is challenging to distinguish between the Acinetobacter species, as they share several phonotypical traits. There are several methods for identifying the pathogenic Acinetobacter strains. Even though PCR is the preferred method for identifying Acinetobacter species, the sequencing of several genes is required for species-level identification [26] In this study, due to the close relationship between Acinetobacter genomic species, especially those belonging to the A. calcoaceticus-baumannii complex, we used different primers to amplify and sequence target genes, such as rpoB, 16S-23S and gyrB, to identify Acinetobacter spp., especially A. bummanii. Several studies have found that the sequencing of the rpoB gene and the 16S-23S rRNA gene spacer region enables the identification of Acinetobacter isolates at the species level [2,26,[41][42][43], A. bummanii and the 13TU genomic species showed interspecies variability, which could be discriminated by amplifying the gyrB gene [2,28,44].
We performed molecular identification to confirm the presence of Acinetobacter spp. DNA in 13 isolates. Among them, eight were identified as A. bummanii. A previous study [45] illustrated that 18 isolates of Acinetobacter species were isolated from 120 raw milk samples; among them, 12 were A. bummanii. Additionally, Gurung et al. [2] found that out of 2287 bulk milk samples, Acinetobacter spp. were isolated from 176 bulk samples. Among them, 57 were A. bummanii. Moreover, Jayarao et al. [46] identified Acinetobacter spp. in 28 out of 205 isolates from bulk milk samples. Additionally, Ndegwa et al. [47]

Discussion
Recently, A. baumannii has been significantly linked to several infections, including catheter-related infections, meningitis, bacteremia, soft-tissue infections, peritonitis and endocarditis, in intensive care units (ICU) due to its extraordinary capacity to obtain or enhance the antimicrobial resistance factors. Therefore, scientists are currently interested in discovering this novel, antibiotic-resistant strain [39].
We identified Acinetobacter spp. in 13 milk samples on CHROM agar media. Several authors have suggested using CHROM agar as a quick and easy medium for detecting Acinetobacter, as it contains chromogenic substrates cleaved by Acinetobacter spp. enzymes, resulting in unique, red-colored Acinetobacter colonies. This increases the efficiency of infection control procedures, decreases the time required to administer the right antibiotic therapy to the infected patients and, hence, lowers mortality [2,40].
However, further information about the interspecies relationships among the genus Acinetobacter is needed. It is challenging to distinguish between the Acinetobacter species, as they share several phonotypical traits. There are several methods for identifying the pathogenic Acinetobacter strains. Even though PCR is the preferred method for identifying Acinetobacter species, the sequencing of several genes is required for species-level identification [26] In this study, due to the close relationship between Acinetobacter genomic species, especially those belonging to the A. calcoaceticus-baumannii complex, we used different primers to amplify and sequence target genes, such as rpoB, 16S-23S and gyrB, to identify Acinetobacter spp., especially A. bummanii. Several studies have found that the sequencing of the rpoB gene and the 16S-23S rRNA gene spacer region enables the identification of Acinetobacter isolates at the species level [2,26,[41][42][43], A. bummanii and the 13TU genomic species showed interspecies variability, which could be discriminated by amplifying the gyrB gene [2,28,44].
We performed molecular identification to confirm the presence of Acinetobacter spp. DNA in 13 isolates. Among them, eight were identified as A. bummanii. A previous study [45] illustrated that 18 isolates of Acinetobacter species were isolated from 120 raw milk samples; among them, 12 were A. bummanii. Additionally, Gurung et al. [2] found that out of 2287 bulk milk samples, Acinetobacter spp. were isolated from 176 bulk samples. Among them, 57 were A. bummanii. Moreover, Jayarao et al. [46] identified Acinetobacter spp. in 28 out of 205 isolates from bulk milk samples. Additionally, Ndegwa et al. [47] found that 10 isolates of Acinetobacter species were identified from 21 goat milk samples in Kenya. Due to these varied results, we used various selective methods to isolate Acinetobacter spp. and to clarify the prevalence of Acinetobacter spp. as compared with these previous studies.
Infections by antibiotic-resistant bacteria, especially multi-resistant bacteria, are challenging to treat, resulting in serious health issues and even death due to extended hospital stays and unsuccessful treatment attempts [48].
In our study, some A. baumannii isolates showed resistance to different antibiotics. This is consistent with previous studies [49,50] which found that clinical A. bumannii isolates from humans were highly resistant to fluoroquinolone, amino-glycosides, cephalosporin and carbapenem. The resistance of some A. bumannii isolates is attributed to the AdeR-AdeS complex, encoded by the adeRS operon, which contributes to acquired antibiotic resistance against different antibiotics [51].
Another factor that might contribute to the resistance found in this study is the selection pressure exerted by using these antimicrobial drugs to treat sick cattle [49]. However, Gurung et al. [2] observed that most Acinetobacter isolates obtained from bulk milk tank samples were susceptible to several antibiotics. They hypothesized that this was because milk is generally extracted from healthy cattle, not those treated with antimicrobial agents.
Antibiotic resistance patterns varied among the nine A. bummannii isolates with an MAR index higher than 0.2. This is considered an indicator of serious contamination in milk, which is linked to the excessive use of antibiotics in animal treatments, which resulted in the creation of strains resistant to a number of antibiotics [52].
The carbapenem resistance observed in these isolates is a significant issue because carbapenem is generally used as an alternative for other β-lactam drugs, including penicillin and cephalosporins [53]. This resistance was mediated by several carbapenemase genes. OXA-type genes, which mainly cause carbapenem resistance in A. baumannii, are intrinsic and can be located on chromosomes and plasmids. The gene blaoxa encoded and expressed OXA-like enzymes and significantly increased carbapenem resistance [54,55].
The PCR results showed that bla OXA-51-like was amplified in all A. baumannii isolates, while bla OXA-23-like was amplified only in 55.5% of isolates. These findings were almost identical to previous studies [39,56], showing that 100% of A. baumannii had bla OXA-51-like genes, while 90% were positive for bla OXA-23-like genes. [57] The ubiquitous nature of OXA-51 in A. baumannii has made it a critical genetic marker for identifying Acinetobacter at the species level. Turton et al. [58] found that the bla OXA51-like genes played an important role in the identification and differentiation of A. bummannii from other Acinetobacter spp., and this explains the high incidence of the bla OXA-51-like genes compared to other genes in our study. Another study [59] discovered that the ability of OXA-51-like proteins to hydrolyze β-Lactam antibiotics, including penicillins (benzylpenicillin, ampicillin, ticarcillin and piperacillin) and carbapenems (imipenem and meropenem), was supported by the expression of other genes. However, bla OXA-23-like proteins were primarily responsible for imipenem resistance, and its plasmid location enhanced the possibility of the horizontal transfer of resistance.
Our findings also revealed that the bla IMP and bla NDM genes were amplified in one and two isolates, respectively, while the bla VIM gene was not amplified. However, previous studies reported that the bla IMP and bla VIM genes were not amplified in the isolates [34]. A study showed that the Verona integron-encoded MBL (VIM) genes were found to be extremely infrequent in Enterobacteriaceae [60], while another [61] discovered that MBL oxacillinases and bla NDM1 are the primary causes of carbapenem resistance.
Based on the MBL phenotypic screening results, 44.4% of A. baumannii isolates produced MBL. Despite phenotypic testing, only one amplicon for the bla imp gene was found among the studied MBL genes. This might be due to the presence of unidentified hereditary determinants, which PCR might not detect, as the PCR primers were only created for specific known genetic positions. Contrastingly, these results might be true-positive for other MBL genes [62].
A. baumannii can efficiently produce biofilms that enable it to tenaciously proliferate under challenging circumstances and habitats. A. baumannii can produce biofilms on biotic surfaces, including epithelial cells, and on abiotic surfaces, including glass and the medical equipment used in ICUs. Most of the A. baumannii isolates from our study were consistent with the findings Malta et al. [63], who found that 100% of A. baumannii isolated from goat milk formed biofilm. Several studies have also emphasized the severity of A. baumannii biofilm-associated infections, such as ventilator-associated pneumonia and catheter-related infections, both of which are resistant to antibiotic therapy and are most frequently caused by A. baumannii biofilms. [21,64,65].
The biofilm matrix around the bacterial cells enables them to withstand harsh conditions and promotes the propagation of antibiotic-resistance genes. Therefore, the current drugs that are available for treating infections caused by A. baumannii biofilms are ineffective [66]. Here, a microtiter plate test overcame the problem of false-positive results in two isolates on Congo red agar; these results are reinforced by Melo et al. [67], who recorded that that sensitivity of the Congo red agar test was 88.9%, while that of the microtiter plate test was 100%. Gaddy and Actis [68] found that A. baumannii can efficiently produce biofilms due to the presence of pili, outer membrane proteins and macromolecular secretions. When pili adhere to abiotic surfaces, they initiate the formation of microcolonies, followed by the formation of fully developed biofilms.

Conclusions
In conclusion, the detection of A. baumannii and its antimicrobial resistance is critical for controlling infections. Efficient molecular techniques, such as DNA sequencing, enable faster identification than other traditional techniques and can readily differentiate between closely related species, facilitating immediate prevention. Therefore, these techniques are necessary for the early detection of these microbes. A. baumannii isolates showed resistance to several antibiotics with MAR indices higher than 0.2, indicating that these are potentially hazardous to consumers. The antibiotic resistance in Acinetobacter spp. still needs further studies, especially the bla OXA51 gene, because the role of this gene in antibiotic resistance is linked to the mechanism of other genes. In addition to that, it has a major role in differentiating A. baumannii from other species.