Growth kinetics of multiple Acinetobacter baumannii resistotype after meropenem-based antibiotic combination exposure

Background: Carbapenems are the treatment of choice for multidrug-resistant (MDR) and extensively drug-resistant (XDR)  Acinetobacter baumannii infections, but the emergence of carbapenem-resistant  A. baumannii (CRAB) has rendered it ineffective in the vast majority of cases. Combination therapy has grown in popularity over the last decade; this study aims to analyze  A.baumannii growth kinetics after exposure to meropenem and ampicillin-sulbactam compared with meropenem and amikacin antibiotic combinations in clinically relevant concentrations. Methods: This experimental laboratory study was conducted on the  A. baumannii ATCC 19606 isolate and three clinical isolates that were intermediate or resistant to tested antibiotics. Meropenem and ampicillin-sulbactam, as well as meropenem and amikacin, were tested at four different concentrations against isolates. Turbidity measurements were taken at predetermined time points of 0, 1, 2, 4, 6, 8, and 24 hours following exposure; bacterial concentration was enumerated using the agar plate method, with the results plotted in a time-kill curve.  Results: A bactericidal effect was achieved in isolates that were intermediate to ampicillin-sulbactam and resistant to meropenem after the administration of meropenem and ampicillin-sulbactam combination with a concentration of 4 µg/ml and 16/8 µg/ml, respectively. The combination of meropenem and ampicillin-sulbactam demonstrated bacteriostatic activity against isolates that were resistant to both antibiotics. Isolates treated with resistant antibiotics showed an increased growth rate compared to the growth control. Conclusion: The combination of meropenem and ampicillin-sulbactam could be a promising combination therapy in treating CRAB infections. The mechanism and degree of antibiotic resistance in the isolates affect the efficacy of antibiotic combinations; further research is needed to corroborate the findings of this study.


Introduction
Acinetobacter baumannii is a Gram-negative rod that garners attention due to its role as a primary pathogen in healthcareassociated infections with a broad spectrum of antibiotic resistance 1,2 . Carbapenems are the preferred treatment for multidrug-resistant (MDR) A. baumannii infections. However, treatment options have dwindled due to high isolation rates of extensively drug-resistant (XDR) A. baumannii with concurrent carbapenem resistance 3,4 .
The discovery of new antibiotics is critical for treating MDR and XDR A. baumannii infections. Nevertheless, antibiotic studies take a long time to complete and are difficult to implement in developing countries with limited access to the latest antibiotics. The alternative strategy that has gathered the most interest is antibiotic combination therapy, which is theoretically supposed to boost antibiotic effectiveness compared to single antibiotics [5][6][7] .
In studies evaluating antibiotic combinations, isolates that are susceptible to at least one of the regimens are frequently used, whereas many A. baumannii clinical isolates frequently lack susceptibility to any antibiotic 5,8 . Additionally, because the antibiotic concentrations used in studies are typically multiple times of minimum inhibitory concentration (MIC) and are difficult to achieve during the administration of therapeutic antibiotic doses, the clinical application of study results is complicated 5,[9][10][11] .
Meropenem is one of the few remaining low-toxicity treatment options for MDR and XDR A. baumannii infections 12,13 . Sulbactam is a beta-lactamase inhibitor with intrinsic activity against A. baumannii, whilst amikacin is an aminoglycoside with relatively maintained efficacy against multidrug-resistant Gram-negative bacteria, including A. baumannii [14][15][16][17] . Ampicillinsulbactam and amikacin are two antibiotics that are available and easy to obtain in Indonesia. A sole sulbactam regimen is not available; it is marketed in conjunction with ampicillin or cefoperazone. Ampicillin-sulbactam formulations were chosen because of the availability of breakpoints in CLSI M100 2022 and technical considerations such as affordability and convenience of access to the antibiotics.
Numerous in vitro studies have demonstrated synergy between meropenem and ampicillin-sulbactam as well as meropenem and amikacin; thus, this study aimed to compare the growth kinetics of various A. baumannii strains exposed to these two antibiotic combinations at clinically relevant concentrations 18-23 .

Procedure
Drug concentrations were selected based on the CLSI breakpoint value for the susceptible category of tested antibiotics as it represents clinically achievable concentrations of drugs in human plasma following standard dosing. Fresh stocks of each antibacterial were prepared on the day of the experiment to achieve 0.5 MIC + 0.5 MIC, 1 MIC + 1 MIC, 2 MIC + 2 MIC, and 2 MIC + 0.5 MIC of meropenem + ampicillin-sulbactam and meropenem + amikacin (Sigma). Prior to the time-kill assay experiment, strains were subcultured onto blood agar (Oxoid CM0055 Blood Agar Base supplemented with 5% sheep blood) and incubated for 24 hours at 35°C. Mid-log phase growth suspension was obtained by inoculating isolated colony into cation-adjusted Mueller-Hinton broth (Oxoid CM0405 Mueller-Hinton Broth base) followed by 4 hours of incubation at 35°C. Static time-kill experiments were performed in sextuplicates on separate days at an initial inoculum of 6×10 5 CFU/ml with the combined antibiotic concentrations in the glass tube, incubated at 35°C. Samples were collected at 0, 1, 2, 4, 6, 8, and 24 h, measured for turbidity by nephelometer (BD PhoenixSpec TM Nephelometer), serially diluted in saline, plated on Mueller-Hinton agar (Oxoid CM 0337 Muelle-Hinton Agar base), and counted after 24 h of incubation for viable-cell counting. Enumeration was performed manually after 24 hours of incubation at 35°C. The limit of detection (LOD) was 10 2 CFU/ml. In the meantime, a control experiment was carried out simultaneously with the same procedure without antibiotic addition. Bactericidal activity was assessed as a ≥ 3 log 10 reduction in a colony-forming unit (CFU)/mL over the period measured. Regrowth was defined as an initial decrease of turbidity or colony count followed by an escalation in the subsequent measurement hour.

Results
The turbidity and colony count data did not follow a normal distribution (Shapiro-Wilk value 0.000). There were significant differences in mean turbidity between isolates of ATCC 19606, MDR-1, MDR-2, and XDR at 2, 4, 6, 8, and 24 hours

Amendments from Version 1
In accordance with the reviewer's recommendations, we made several adjustments. The Figure and Table have been altered to convey the content better. On the underlying data in FigShare, tables have been newly constructed. The writing of the manuscript has undergone a few minor adjustments.
Any further responses from the reviewers can be found at the end of the article REVISED following antibiotic exposure (p<0.05; Wilcoxon; CI 95%). There were significant differences in the mean colony count between isolates of ATCC 19606, MDR-1, MDR-2, and XDR at 6, 8, and 24 hours following exposure, (p = 0.001, p = 0.01, and p = 0.000; Wilcoxon; CI 95%). The full turbidity and colony count data can be found under Underlying Data 24 .
Exposures to meropenem and ampicillin-sulbactam yield encouraging results. In the MDR-1 isolate, which was resistant to carbapenem and intermediate to ampicillin-sulbactam, the bactericidal effect of meropenem and ampicillin-sulbactam was achieved at a 2 MIC + 2 MIC concentration, respectively ( Figure 1). During 0-24 hours, concentrations of  MDR-2 isolate (isolate resistant to meropenem and ampicillinsulbactam) treated with meropenem and ampicillin-sulbactam : Bactericidal: ≥ 3 log 10 reduction in a colony-forming unit (CFU)/ml over the period measured. Bacteriostatic: < 3 log 10 reduction in a colonyforming unit (CFU)/mL over the period measured (compared to initial measurement of tested isolate) c : Δ Log 10: Log 10 of the total colony-forming unit (CFU/ml) reduction over the measurement time (compared to initial measurement of tested isolate) d : Regrowth: initial decrease of turbidity or colony count followed by an escalation in the subsequent measurement hour e : Comparison of the colony count between the treatment group and growth control group of isolate. Growth control: isolate without antibiotic combination exposure combination at concentration equal to or less than the MIC demonstrated higher turbidity compared to positive growth control after 24 and 48 hours. At a concentration twice the MIC, there is a reduction in colony count after four hours, followed by regrowth. During post-exposure monitoring, XDR isolate exposed to meropenem and ampicillin-sulbactam did not show any signs of regrowth, except at a concentration of 1 MIC + 1 MIC, where regrowth occurred at 8 and 24 hours (Table 1).

Meropenem and amikacin had no bactericidal impact on intermediate and drug-resistant isolates; hence on all clinical
isolates of A. baumannii in this study. The most significant reduction in the number of bacteria was observed following exposure to 2 MIC and 2 MIC; however, these concentrations had no effect on the number of colonies in XDR isolates when compared to the number of colonies at 0 hours measurement.

Discussion
This investigation discovered regrowth in clinical isolates from nearly all exposure groups. Regrowth is influenced by various factors related to the concentration of antibiotics and bacterial inoculum, as well as the susceptibility of bacteria 25 . Regrowth may occur when bacterial growth is not fully inhibited by exposure to antibiotics (due to insufficient antibiotic concentration or a resistant bacterial strain) 26  Additionally, this study found that isolates treated at sub-MIC concentrations of antibiotics had a higher colony count than the growth control group. This finding merits additional investigation to ascertain the underlying mechanism. Antibiotics have a selection and inducer effect on antibiotic resistance, which demonstrates the importance of using them prudently.

Open Peer Review
My co-authors and I were pleased to receive your response and the opportunity to resubmit a revised version of this manuscript. We attempt to respond to reviewer questions with relevant information obtained from our research.
1. Is there any preliminary examination to ensure that the antibiotic concentration used is as expected at the beginning and end of observation time?
Thank you for the question. ○ There were two preliminary trials conducted before this study. The first preliminary trial was carried out to determine the appropriate colony measurement method for the test isolates exposed to the selected antibiotics for this study. The second preliminary study was carried out to determine the time required by the test isolates to reach the log phase of growth.

○
We did not conduct a preliminary trial to ascertain the concentration of the test antibiotic because antibiotic exposure was performed for 24 hours (additional measurements were taken at 48 hours to collect post-antibiotic exposure data), which is comparable to the duration of antibiotic susceptibility tests conducted in clinical microbiology laboratory with antibiotic powders that were subjected to routine quality control. ○ 2. How did the author confirm that the bacterial isolates tested were in the exponential growth period/log phase? Thank you for drawing attention to this.

○
In the preliminary test, isolates were grown without antibiotic treatment in liquid media. This test is designed to determine the time required for the test isolate to reach the log phase under identical conditions to the actual test. After transferring isolated colonies from solid to liquid media, turbidity measurements and colony growth calculations were undertaken every 30 minutes. The data obtained was therefore plotted on a growth curve. According to the preliminary test results, all isolates entered the log phase after two hours of incubation, and six hours later, they began to reach the stationary phase. Therefore, in the actual experiment, isolated colonies were cultured in liquid media for four hours prior to the time-kill test (at the mid-log phase).  The tests were carried out six times over the course of two days. On the first day of the trial, three replications were performed. On the second day, three additional replications were conducted, bringing the total number of replications to six. This experiment was repeated six times with four test isolates treated with two types of combination antibiotics at four different concentrations on each combination. . We shall attempt to revise the graphic to ensure its meaning is more evident.
whereas it seems to provide ≥ 3 log 10 reductions (bactericidal), compared to the growth control ( Figure 1). "Colony count higher than growth control d " -did the authors mean "Turbidity higher than growth control d "? Because the results seem to be received from Figure 2. If yes, please change the title and the description of this column.
Dear dr. Chusri, My co-authors and I were pleased to receive your response and the opportunity to resubmit a revised version of this manuscript. We would like to thank you for providing your constructive and detailed review comments on our manuscript. We have attempted to fully address comments in the revised manuscript; the reviewer's original comments are listed below, followed by our response to each comment. Edited text in the attached revised manuscript is visible as tracked changes under the markup mode of Microsoft Word that we've sent to Editor. All authors have read and approved the revised manuscript. We hope our resubmission is now suitable for acceptance, and we look forward to hearing from you. In XDR isolates exposed to meropenem and ampicillin-sulbactam, regrowth only occurred at a concentration of 1 MIC + 1 MIC. In XDR isolates exposed to meropenem and amikacin, regrowth did occur in all concentration groups.To clarify this conclusion, we attempt to rearrange the sentences.

Major Revision
○ Only 1 isolate of XDR was included in the study. Why did the authors write "XDR isolates" in this sentence? Thank you very much for the reminder. We revised the sentence accordingly.

"Meropenem and amikacin had no bactericidal impact on intermediate and drug-resistant isolates; hence on all clinical isolates of A. baumannii in this study."
According to Figure 1 and Table 1, not all concentrations of MEM + AK have no bactericidal effect on the clinical isolates. At twice MIC of this combination against MDR-1, it seems to provide ≥ 3 log 10 reductions, which was considered to have bactericidal activity. Please rephrase this sentence. Thank you for pointing this out. According to the colony count of the MDR-1 isolate (attached in the Underlying Data), the colony count decreased by 2.83 log 10 CFU/ml following exposure to 2 MIC + 2 MIC concentrations of meropenem and amikacin (from 5.78 log 10 CFU/ml to 2.95 log 10 CFU/ml). Because the reduction in colony counts did not surpass 3 log 10 CFU/ml over the period measured, we categorized the activity as bacteriostatic. We will attempt to add column in table with Δ Log information to ensure that the findings are more easily discernible. .48 log 10 CFU/ml). As it explains the growth of the isolates when exposed to antibiotics, the decrease in colony count of the isolates was compared to the time-totime colony count of the isolates rather than to the growth control. We will attempt to add column in table with Δ Log information to ensure that the findings are more easily discernible.
○ Title and description of Table 1 Thank you for the helpful reminder. We have made the necessary adjustments.
introduces additional information regarding something stated earlier, often in the form of a list to remove an ambiguity or supply a word omitted in the preceding text. We shall attempt to rephrase the sentence such that the meaning is more clearly apparent. 6 . Use other codes such as "MDR-1" and "MDR-2" Thank you. Revised accordingly. ○ 8. Page 6-7, Results "… these concentrations had no effect on the number of colonies in XDR isolates when compared to the number of colonies at 0 hours measurement." According to Figure 1, the authors should also specify the type of antibiotic combination, because this phenomenon was only found in XDR against MEM + AK combination, but not XDR against MEM + SAM. We thank you for bringing this to our attention. As stated in Point 1.a, attempts are made to arrange the sentences. ○ 9. Page 7, Discussion "… sulbactam has a high affinity for PBP 1 and 3" . Replaced with "… sulbactam has a high affinity for PBP 1 and PBP 3". Revised accordingly. We have made adjustments in accordance with the revision. ○ 11. Page 4 and 5, Figure 1 and Figure 2 Use another colour and shape to represent the results of "Growth Control" We have made adjustments in accordance with the revision. ○ 12. How many replications did the authors performed for time-kill assay? Experiments were conducted in six replications. ○ 13. According to the description in Figure 1, Figure 2, and Table 1, why did the MICs not represent the exact MICs, but it represents the MIC at the susceptible breakpoint from CLSI guideline? Thank you for the question. The antibiotic concentration was based on the CLSI breakpoint since the susceptible breakpoint value was based on the patient's clinically standard dosing regimen. This study aims to identify clinically relevant, effective antibiotic combinations for patients with MDR and XDR A.baumannii infections; consequently, it is necessary to utilize antibiotic concentrations achieved through a standard dosing regimen. MDR and XDR A.baumannii are frequently resistant to the tested antibiotic, with MIC typically being multiple times that of susceptible isolates, which is difficult to achieve during the administration of therapeutic antibiotic doses, thereby complicating the clinical application of those kinds of studies.