Contribution of Acinetobacter-derived cephalosporinase-30 to sulbactam resistance in Acinetobacter baumannii

The sulbactam resistance rate in Acinetobacter baumannii has increased worldwide. Previous reports have shown that the β-lactamase blaTEM-1 confers resistance to sulbactam in A. baumannii. The purpose of this study was to examine whether other β-lactamases, including the Acinetobacter-derived cephalosporinase (ADC), OXA-23, OXA-24/72, and OXA-58 families, also contribute to sulbactam resistance in A. baumannii. The correlation between these β-lactamases and the sulbactam minimal inhibitory concentration (MIC) was determined using A. baumannii clinical isolates from diverse clonality, which were collected in a nationwide surveillance program from 2002 to 2010 in Taiwan. A possible association between the genetic structure of ISAba1-blaADC-30 and sulbactam resistance was observed because this genetic structure was detected in 97% of sulbactam-resistant strains compared with 10% of sulbactam-susceptible strains. Transformation of ISAba1-blaADC-30 into susceptible strains increased the sulbactam MIC from 2 to 32 μg/ml, which required blaADC-30 overexpression using an upstream promoter in ISAba1. Flow cytometry showed that ADC-30 production increased in response to sulbactam, ticarcillin, and ceftazidime treatment. This effect was regulated at the RNA level but not by an increase in the blaADC-30 gene copy number as indicated by quantitative PCR. Purified ADC-30 decreased the inhibitory zone created by sulbactam or ceftazidime, similarly to TEM-1. In conclusion, ADC-30 overexpression conferred resistance to sulbactam in diverse clinical A. baumannii isolates.


Introduction
Acinetobacter baumannii causes various nosocomial infections, and the prevalence of multidrug-resistant (MDR) A. baumannii has been increasing in different countries. This bacterium has intrinsic resistance to multiple drugs and can gain resistance mechanisms from other species (Peleg et al., 2008). The SENTRY program documented non-susceptibility to carbapenems, the last resort of drugs for the treatment of MDR A. baumannii, increased from 34.5% in 2006 to 59.8% in 2009 worldwide (Gales et al., 2011). In Taiwan, the rate of multidrug resistance in Acinetobacter spp. also increased from 1.3% in 2002 to 41.0% in 2010 (Kuo et al., 2012). In severely ill patients, infections with MDR isolates have been associated with high mortality due to the absence of appropriate or effective treatment options (Peleg et al., 2008). Combination therapies or new drugs such as antimicrobial peptides or silver nanoparticles have been proposed as novel modalities to treat MDR A. baumannii (Peleg et al., 2008; Sulbactam is a β-lactamase inhibitor that is typically combined with penicillins because sulbactam lacks antimicrobial activity against most bacterial species (Adnan et al., 2013). However, sulbactam has demonstrated bacteriostatic or bactericidal effects against A. baumannii (Corbella et al., 1998). Combination treatment with sulbactam and carbapenems has shown promising in vivo and in vitro synergistic effects against MDR A. baumannii (Wolff et al., 1999;Ko et al., 2004;Song et al., 2007), and clinical success has been reported (Karageorgopoulos and Falagas, 2008). The addition of sulbactam to other antibiotics has been proposed in the treatment of MDR A. baumannii; however, the resistance rate to ampicillin/sulbactam in Acinetobacter spp. has increased to approximately 60% in certain area (Kuo et al., 2012).
Until recently, the mechanism underlying sulbactam resistance in A. baumannii was less commonly studied. In 2013, Krizova and colleagues demonstrated that the β-lactamase TEM-1 contributes to sulbactam resistance (Krizova et al., 2013), which led us to examine whether other selected β-lactamases found in A. baumannii, including the Acinetobacter-derived cephalosporinase (ADC), OXA-23, OXA-24/72, and OXA-58 families, also contribute to sulbactam resistance. Using clinical isolates collected from a Taiwanese surveillance program, we aimed to identify the β-lactamases associated with sulbactam resistance and to test the role of these β-lactamases in sulbactam resistance in A. baumannii.

Materials and Methods
Association of Selected β-Lactamases with Sulbactam Resistance in A. baumannii A. baumannii clinical isolates were randomly selected from the Taiwan Surveillance of Antimicrobial Resistance (TSAR) program, which contains 1640 Acinetobacter isolates collected from 2002 to 2010 (Kuo et al., 2012). A. baumannii was identified at the species level using multiplex PCR targeting the specific 16-23S rDNA intergenic spacer region (Chen et al., 2007). Pulsed-field gel electrophoresis was performed to determine clonality as previously described (Kuo et al., 2013). Isolates with similarity of >80% was designated as a single clone. The sulbactam-resistant and sulbactam-susceptible isolates were randomly selected and subjected to PCR testing for the presence of genes encoding ADC, OXA-23, OXA-24/72, and OXA-58 βlactamases (Table S1). The PCR program (Krizova et al., 2013;Kuo et al., 2013) for genes encoding OXA was as followed: 94 • C for 1 min, and 30 cycles of 25 s at 94 • C, 40 s at 52 • C and 50 s at 72 • C; for PCR of bla ADC , 30 cycles of 60 s at 94 • C, 60 s at 58 • C, and 120 s at 72 • C. GoTaq Flexi DNA polymerase (Promega, Madison, WI) was used for PCR assays performed in the GeneAmp PCR System 2700 (Applied Biosystems, Foster City, CA). Amplified DNA product was resolved by electrophoresis in agarose 2% w/v gels, stained with ethidium bromide, and purified according to the manufacturer's instruction (Geneaid Biotech Ltd, Taipei, Taiwan).

Transformation of Plasmids Carrying Different β-Lactamase Genes
Representative β-lactamase genes and their associated promoters, including bla TEM−1 , bla ADC and bla OXA−23 with their upstream insertion sequence ISAba1 (ISAba1-bla ADC and ISAba1-bla OXA−23 , respectively), bla OXA−24/72 , and bla OXA−58 with its upstream ISAba3 that was truncated with IS1008 (IS1008-ISAba3-bla OXA−58 ) (Chen et al., 2010), were PCRamplified using the forward and reverse primers shown in Table S1. The PCR products were amplified with a proofreading DNA polymerase (Phusion High-Fidelity DNA Polymerase, Finnzymes, Espoo, Finland), cloned into the pCRII-TOPO vector (Invitrogen, Carlsbad, CA, USA) and subjected to sequencing (Mission Biotech, Taipei, Taiwan). The digested fragments were cloned into the XbaI and XhoI sites of the Escherichia coli-A. baumannii shuttle vector pYMAb2 (Kuo et al., 2013), which contains a kanamycin-resistant determinant. The fragment was cloned in-frame with a polyhistidine (His) tag, causing the resulting protein to be His-tagged. The recombinant plasmid and a control plasmid (pYMAb2 without β-lactamase genes) were transformed into the kanamycinsusceptible A. baumannii strain ATCC15151. ATCC15151 already contained bla OXA−51 ; therefore, bla OXA−51 was not included in the experiment. Electroporation was performed with a gene pulser electroporator (Bio-Rad, Hercules, CA, USA) and 2-mm electrode gap cuvettes (Kuo et al., 2013). Transformants were selected based on kanamycin resistance, and sequencing was performed to confirm the presence of each β-lactamase gene.

Antimicrobial Susceptibility
Minimal inhibitory concentrations (MICs) of sulbactam, ceftazidime, ampicillin, imipenem, meropenem, and ticarcillin were determined by the agar dilution method according to the guidelines provided by the Clinical and Laboratory Standards Institute (CLSI) (Clinical and Laboratory Standards Institute, 2012). Sulbactam susceptibility parameters were adopted from the previous CLSI guidelines in which sulbactam MICs of less than 4 or more than 16 µg/ml were defined as susceptible or resistant, respectively.

Immunofluorescent Staining and Enumeration by Flow Cytometry for Protein Expression
Immunofluorescent staining was performed as previously described with several modifications (Moe et al., 1999). Specifically, bacterial cultures were diluted in phosphate buffered saline (PBS) to ∼10 8 CFU/mL, and 0.5-ml samples were transferred into 1.5-ml Eppendorf microtubes. The samples were centrifuged at 5000 × g for 5 min, washed with 0.1% NaN 3 PBS, and centrifuged for 5 min. The resulting bacterial pellet was resuspended in fixative (4% paraformaldehyde in PBS) for 20 min. The fixed samples were washed twice with quenching solution (100 mM NaCl, 50 mM Tris-HCl, pH 8.0) and resuspended in permeable buffer (1% Triton X-100, 0.1% NaN 3 in PBS) for 5 min. After centrifugation, the samples were resuspended in 500 µL blocking buffer (1% BSA and 0.1% NaN 3 in PBS).
To identify His-tagged ADC-30 expressed by ATCC15151 (pYMAb2::ISAba1-bla ADC−30 ), each 100-µL sample was incubated with 2 µL of mouse anti-His-6-tag antibody (Sigma-Aldrich, St. Louis, MO, USA) at 4 • C for 2 h. After washing by blocking buffer, the samples were stained for 1 h at 4 • C with 5 µL of phycoerythrin (PE)-conjugated anti-mouse IgG antibody (Sigma-Aldrich). The stained samples were spun at 5000 × g for 5 min, and the cell pellet was resuspended in 500 µL of 1% paraformaldehyde buffer and stored at 4 • C overnight. Cytometry samples were resuspended in PBS and analyzed in a flow Quantitative PCR (qPCR) to Determine the bla ADC−30 Gene Copy Number after Challenging with Different Antimicrobials ATCC15151 (pYMAb2::ISAba1-bla ADC−30 ) strains at mid-log phase were incubated in Luria-Bertani (LB) broth with different antimicrobial agents (25% of MIC) for 6 h. The bla ADC−30 copy number in these bacteria was estimated by qPCR using primers targeting bla ADC−30 , and the housekeeping gene, recA, was used as an internal control. Each qPCR reaction contained a total volume of 10 µL with 2 ng of genomic DNA as template, 100 nmol/L of each primer, and 1 × SYBR Green R PCR Master Mix (Applied Biosystems, Carlsbad, CA, USA) with ROX ™ (Kapa Biosystems, Woburn, MA, USA). The relative bla ADC−30 copy number in the bacteria treated with different antimicrobial agents was normalized to the number found in the bacteria treated with LB broth without antimicrobial agents. The qPCR conditions included 2 min at 50 • C (UNG activation), 10 min at 95 • C, followed by 45 cycles of 15 s at 95 • C and 1 min at 60 • C. At the end, a dissociation stage was added: 15 s at 95 • C, 15 s at 60 • C, and 15 s at 95 • C. All experiments were conducted using the ABI 7500 Fast Real-time PCR system (Applied Biosystems, Inc., Carlsbad, CA, USA) and were performed in triplicate.

Quantitative Reverse Transcription PCR (qRT-PCR) to Assess mRNA Expression after Challenging with Different Antimicrobials
After incubation with different antimicrobial agents, the ADC-30 mRNA levels in ATCC15151 (pYMAb2::ISAba1-bla ADC−30 ) were compared using qRT-PCR (Chen et al., 2010). Briefly, around 2 µg of RNA was extracted with RNAprotect Bacteria Reagent and an RNeasy mini-kit (Qiagen, Valencia, CA, USA). Residual genomic DNA was removed using RNase-free DNase. The RNA was reverse transcribed into single-stranded cDNA with random hexamers and Moloney Murine Leukemia Virus reverse transcriptase (Epicenter, Madison, WI, USA). The cDNAs were subsequently quantified by real-time PCR amplification with conditions mentioned above. Expression level results were standardized to the transcription levels of rpoB gene for each strain, but relatively to the culture in LB (2 delta-delta Ct method). Negative controls without reverse transcription were performed to detect DNA contamination in the purified RNA.

ADC-30 Purification
His-tagged ADC-30 (in which the stop codon was deleted, and the proteins were fused with 6 His amino acids) expressed by ATCC15151 (pYMAb2::ISAba1-bla ADC−30 ) was purified with Ni-NTA Superflow column (Qiagen). Briefly, the bacteria equal to ∼10 7 CFU/ml were centrifuged, resuspended in lysis buffer and sonicated. The lysate was diluted in binding buffer (25 mM Tris, 150 mM NaCl, 10 mM imidazole, pH 7.5) and loaded onto the column. The column was washed with five column volumes of wash buffer, and the protein was eluted with five column volumes of elution buffer (25 mM Tris, 150 mM NaCl, 300 mM imidazole). The protein solution was then dialyzed and concentrated by ultrafiltration on a 10 KDa-cutoff Amicon membrane (Millipore). The purity was assessed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting as greater than 95%.

SDS-PAGE and Western Blotting to Detect Purified β-Lactamases
For SDS-PAGE and Western blotting (Blair et al., 2009), the purified protein was first separated by SDS-PAGE on 12% acrylamide gel using Mini-PROTEAN system (Bio-Rad) and transferred to a nitrocellulose membrane (PerkinElmer, Boston, MA, USA) in transfer buffer (25 mM Tris-HCl, 190 mM glycine, 20% methanol, 0.1% SDS, pH 8.3) at 350 mA for 1 h. The membranes were blocked with 10% non-fat dried milk in Tris-bufferred saline with 0.5% Tween-20. After hybridization with anti-His antibodies (Sigma-Aldrich) and peroxidase-conjugated goat anti-mouse secondary antibodies (Millipore, Temecula, CA, USA), the band was visualized with an ECL Western blot kit (PerkinElmer, Boston, MA, USA).

Bioassays to Confirm the Activity of Purified ADC-30 against Ceftazidime and Sulbactam
The first bioassay involved the mixture of 10 µL of purified ADC-30 or extracts of sulbactam-resistant or -susceptible strains with  10 µL of ceftazidime for 30 min at 37 • C. Each 20-µL mixture was loaded onto a blank disk (Becton Dickinson, Sparks, MD, USA) that was placed in an agar plate containing a lawn of ceftazidime-susceptible A. baumannii ATCC 15151. Inhibitory zones were measured after incubating the plates overnight at 37 • C. The second bioassay conducted was similar to the modified Hodge test for carbapenemase detection. A 0.5 McFarland standard suspension of the sulbactam-susceptible A. baumannii ATCC 15151 strain was diluted to 1:10 and inoculated on an Mueller-Hinton agar plate for routine disk diffusion test. The 30-µg sulbactam disk was placed in the center of the plate, and the purified ADC-30, TEM-1, or PBS samples were drawn in a straight line out from the edge of the disk. The phenotype was evaluated after an overnight incubation at 37 • C.
The presence of one resistant strain without ISAba1-bla ADC nor bla TEM−1 indicates that another mechanism may be involved in sulbactam resistance. Various combinations of resistance mechanisms, including β-lactamase overexpression, the up-regulation of the efflux pump and the inactivation or downregulation of porin, are often required for the development of Frontiers in Microbiology | www.frontiersin.org β-lactam resistance in Pseudomonas aeruginosa and A. baumannii (Quale et al., 2006;Peleg et al., 2008;Tiwari et al., 2012;Tiwari and Moganty, 2014), which may explain the presence of ISAba1-bla ADC−30 in one susceptible strain. However, the high rate of ISAba1-bla ADC−30 in resistant strains and its low rate in susceptible strains (97% vs. 10%, Chi-square test p < 0.001) indicate the importance of ISAba1-bla ADC−30 for the development of sulbactam resistance.

Contribution of ADC-30 and ISAba1 to Sulbactam Resistance
To further confirm the association observed in the epidemiological survey, ISAba1-bla ADC−30 , bla TEM−1, ISAba1-bla OXA−23 , bla OXA−24/72 (and its promoter), and IS1008-ISAba3bla OXA−58 were cloned and transformed into a sulbactamsusceptible reference strain, respectively. Shuttle vectors were also transformed, and changes in the MICs were measured ( Table 1). ATCC15151 (pYMAb2::ISAba1-bla ADC−30 ) exhibited the highest increases of sulbactam MIC (16-fold). Other β-lactamases, including OXA-23, OXA-72, and OXA-58, contributed to the increase in the sulbactam MICs, although at lower levels. ISAba1-bla ADC−30 was used in the study due to the high level of sulbactam resistance and high prevalence of ISAba1-bla ADC−30 , which were comparable to the sulbactam resistance levels and prevalence of bla TEM−1 .
The regulatory mechanism for ADC expression in Acinetobacter spp. may be different from the mechanism in many Enterobacteriaceae because Acinetobacter spp. lack the AmpR gene (Jacoby, 2009). The presence of ISAba1 upstream of bla AmpC is essential for ceftazidime resistance due to the AmpC overexpression (Heritier et al., 2006). To confirm the role of the promoter located within ISAba1 in mediating sulbactam resistance, plasmids harboring bla ADC−30 without upstream -35 (within ISAba1) or -35/-10 promoters were transformed into ATCC15151. The MICs increased by only 2-fold compared with the control (Table 1). Therefore, as previous studies have indicated (Heritier et al., 2006), these promoters in the ISAba1 are required for ADCmediated sulbactam resistance. Similarly, deletion of the promoter upstream of bla TEM−1 decreased the sulbactam MIC.

ADC-30 Production in Response to Treatment with Sulbactam and Other Substrates
The production of β-lactamase usually increases in response to its substrates; therefore, we asked whether the addition of sulbactam increases ADC-30 production. Flow cytometry (Figure 2) showed that ADC-30 protein expression increased significantly after the addition of sulbactam or ceftazidime compared with the control. ADC-30 production was also induced by ticarcillin but not by ciprofloxacin or imipenem. The results indicate that the co-selection of sulbactam resistance by other antimicrobial agents occurs.
The qRT-PCR results (Figure 3) showed that ADC-30 mRNA expression increased in response to its substrates (ceftazidime and sulbactam) but not in response to other antimicrobial agents (ciprofloxacin). Carbapenemase expression has been related to the high bla OXA−58 plasmid copy number (Bertini et al., 2007;Chen et al., 2008). Therefore, we asked whether the increased ADC-30 protein and mRNA expression levels in the presence of sulbactam or ceftazidime are attributed to an increase in the bla ADC−30 copy number. The qPCR results (Figure 3) showed no differences in the bla ADC−30 copy numbers of cells treated with sulbactam or ceftazidime compared with the negative controls. In contrast, the gene copy number increased in response to kanamycin. The results indicate that the increase in ADC-30 was regulated at the RNA level.
How the addition of sulbactam increased ADC-30 protein and mRNA expression level in A. baumannii is unknown. The induction of AmpC by β-lactams has been commonly described in P. aeruginosa and many Enterobacteriaceae (Jacoby, 2009). After the treatment of β-lactams, altered peptidoglycan synthesis leads to increased expression of AmpC through AmpG-AmpR-AmpC pathway (Zeng and Lin, 2013). Clavulanate, another βlactamase inhibitor with structure similar to β-lactams, also induces the expression of AmpC in many Enterobacteriaceae (Drawz and Bonomo, 2010). Although sulbactam may interfere the wall synthesis by interacting with penicillin-binding protein 1 and 3 (Penwell et al., 2015), lack of AmpR in A. baumannii indicated other mechanisms, rather than AmpG-AmpR-AmpC pathway, are responsible. The two-component system has been proposed to be involved in the induction of AmpC and other chromosomally encoded β-lactamases (Jacoby, 2009;Zeng and Lin, 2013). However, the mechanism regarding the induction of ADC in response to sulbactam in A. baumannii requires further investigation.
Contribution of Purified ADC-30 to Sulbactam Resistance ADC-30 was purified ( Figure S1), and the addition of purified ADC-30 to the disk decreased the inhibitory zone caused by ceftazidime (Figure 4A), confirming the activity of the purified protein. The bioassay resembling the modified Hodge test showed that purified ADC-30 promoted the inward growth of sulbactam-susceptible A. baumannii similarly to purified TEM-1 ( Figure 4B). Therefore, ADC-30 may directly interact with sulbactam and confer sulbactam resistance.
In conclusion, bla ADC−30 overexpression contributes to sulbactam resistance in A. baumannii, which is prevalent in clinical isolates of different clones in Taiwan over a long period of time. The resistance mechanisms are induced at the mRNA and protein levels by other antimicrobial agents in addition to sulbactam, supporting the cautious use of these antibiotics to avoid the selection of sulbactam-resistant A. baumannii isolates.

Funding
This work was supported by grants from the National Health Research Institute and the National Science Council (MOST 103-2314-B-400 -020 -MY2 and 101-2314-B-010-027-MY3). The sponsors did not have a role in the study design, in the collection, analysis, or interpretation of data, in the writing of the report, or in the decision to submit the article for publication.

Ethical Approval
This study was approved by the Institutional Review Board of National Health Research Institutes.