Synergistic Combination of AS101 and Azidothymidine against Clinical Isolates of Carbapenem-Resistant Klebsiella pneumoniae

Owing to the over usage of carbapenems, carbapenem resistance has become a vital threat worldwide, and, thus, the World Health Organization announced the carbapenem-resistant Enterobacteriaceae (CRE) as the critical priority for antibiotic development in 2017. In the current situation, combination therapy would be one solution against CRE. Azidothymidine (AZT), a thymidine analog, has demonstrated its synergistically antibacterial activities with other antibiotics. The unexpected antimicrobial activity of the immunomodulator ammonium trichloro(dioxoethylene-o,o’)tellurate (AS101) has been reported against carbapenem-resistant Klebsiella pneumoniae (CRKP). Here, we sought to investigate the synergistic activity between AS101 and AZT against 12 CRKP clinical isolates. According to the gene detection results, the blaOXA-1 (7/12, 58.3%), blaDHA (7/12, 58.3%), and blaKPC (7/12, 58.3%) genes were the most prevalent ESBL, AmpC, and carbapenemase genes, respectively. The checkerboard analysis demonstrated the remarkable synergism between AS101 and AZT, with the observable decrease in the MIC value for two agents and the fractional inhibitory concentration (FIC) index ≤0.5 in all strains. Hence, the combination of AS101 and azidothymidine could be a potential treatment option against CRKP for drug development.


Introduction
Carbapenem-resistant Enterobacteriaceae (CRE) was announced as the critical priority for antibiotic development in 2017 [1]. Recently, the COVID-19 pandemic promoted the spread of carbapenem-resistant Klebsiella pneumoniae (CRKP) [2], highlighting an urgent need for novel treatment options. The high carbapenem-resistant rates were revealed in some parts of Europe [3], especially Southern Europe. According to the European Centre for Disease Prevention and Control EARS-Net, 58.3% of 312 Greek K. pneumoniae isolates collected in 2019 were resistant to either imipenem or meropenem [4]. In Romania, 32.3% of 470 K. pneumoniae were reported as carbapenem-resistant isolates; in Italy, 28.5% of 7327 K. pneumoniae were defined as carbapenem-resistant isolates. A regional resistance surveillance program in Asia-Pacific reported 25% carbapenem-resistant rate in K. pneumoniae for India and 5% for both the Philippines and Thailand [5]. A more recent surveillance conducted in Taiwan, 2018, described a 7.2% carbapenem-resistant rate in 346 K. pneumoniae isolates [6]. In view of the aforementioned reports, the development of novel therapeutic options to address CRKP is urgently needed.
In recent years, some novel β-lactam-β-lactamase inhibitor combination therapies have been launched in clinical settings, for instance, meropenem-vaborbactam, imipenemrelebactam, and ceftazidime-avibactam [7][8][9][10]. Moreover, azidothymidine was reported to show synergistic activities with some clinical antimicrobial agents such as colistin and fosfomycin. [11][12][13]. Azidothymidine (AZT), a thymidine analog, possesses an antiretroviral activity and clinically treats patients with human immunodeficiency virus (HIV) infections, the acquired immunodeficiency syndrome (AIDS) [14]. Lately, studies have reported that azidothymidine combined with antibiotics produced synergistic activities against antibiotic-resistant gram-negative bacteria both in vitro and in vivo [11]. A previous study also reported that a combination of colistin and azidothymidine revealed synergistic activity against colistin-resistant CRKP clinical isolates [12]. In a phase I clinical trial for the combination of AZT and colistin, the results revealed that a dosage of 2 million IU CMS plus 100 mg AZT twice a day might be sufficient for urinary tract infections (UTIs) [15]. Although the phase I clinical trial of the combination of AZT and colistin is in progress, colistin could be a temporary option for CRKP treatment but could not be permanent because of its nephrotoxicity and neurotoxicity [16].
AS101 is a fully-synthesized, tellurium-containing, organic compound with a small molecular weight of 312 Daltons [17,18]. With its characteristic of immunomodulation, AS101 was used to treat autoimmune diseases, inflammatory bowel disease (IBD), multiple sclerosis (MS), and psoriasis [19], and some of its applications are in ongoing clinical trials. Moreover, AS101 demonstrated anti-inflammatory activity [19], antiviral activities (HIV-1 and WNV) [20,21], and antimicrobial activities (carbapenem-resistant Acinetobacter baumannii, K. pneumoniae, Enterobacter cloacae) [22][23][24][25]. In a recent study, Yang et al. demonstrated its in vitro and in vivo antimicrobial activity against carbapenem-resistant Acinetobacter baumannii [22]. Although previous efforts posed that AS101 could be a potential option to treat CRKP, our study described that the MIC values of AS101 against CRKP was up to 32 µg/mL [23], which is not far away from its 50% cytotoxicity level (145 µg/mL) [21]. Hence, it might be unsafe for high dose. To this end, some improvements might be needed to increase the antibacterial activity of AS101. Accordingly, finding an alternative agent to combine with AZT for CRKP treatment is of vital importance. Given the previous efforts on AZT combinations and AS101, we sought to evaluate the synergistic activities of AZT plus AS101 against CRKP in this study.

Evaluation of Synergistic Effects
The MIC values for AS101 against the 12 CRKP isolates ranged from 2 to 512 µg/mL (Table 2), with the MIC 50 , MIC 75 , and MIC 90 of 128, 256, and 512 µg/mL, respectively. The MIC range of azidothymidine (AZT) against the 12 CRKP was from 0.5 to 4 µg/mL, and the MIC 50 , MIC 75 , and MIC 90 were revealed as 1, 2, and 2 µg/mL, respectively. With the combination of AS101 and AZT, noticeable decreases for MIC 50 , MIC 75 , and MIC 90 were observed in both AS101 (from 128, 256, and 512, respectively, to 8, 16, and 16 µg/mL) and AZT (from 1, 2, and 2, respectively, to 0.25, 0.25, and 0.5 µg/mL) ( Table 2). The MIC distributions of AS101 and AZT alone or in combination are visualized in Figure 2. Significant changes in MIC distributions were noted for AS101 and AZT (both p < 0.0001), with decreased MIC ranges for AS101 (from 2-12 to 0.5-32 µg/mL) and AZT (from 0.5-4 to 0.0625-1 µg/mL) ( Table 2). The fractional inhibitory concentration (FIC) indexes of the 12 isolates were all ≤0.5 (Table 3), suggesting the synergistic interaction between AS101 and azidothymidine. Additionally, the synergistic effect was observed in the growth curve ( Figure S1.) Supported by the aforementioned results, the combination therapy of AS101 and AZT is a potential treatment option for CRKP infections, needing further investigation in future study. All of the checkerboard methods for checking the synergistic effect were performed and repeated three times, and all results were reproducible.

Evaluation of Synergistic Effects
The MIC values for AS101 against the 12 CRKP isolates ranged from 2 to 512 μg/mL (Table 2), with the MIC50, MIC75, and MIC90 of 128, 256, and 512 μg/mL, respectively. The MIC range of azidothymidine (AZT) against the 12 CRKP was from 0.5 to 4 μg/mL, and the MIC50, MIC75, and MIC90 were revealed as 1, 2, and 2 μg/mL, respectively. With the combination of AS101 and AZT, noticeable decreases for MIC50, MIC75, and MIC90 were observed in both AS101 (from 128, 256, and 512, respectively, to 8, 16, and 16 μg/mL) and AZT (from 1, 2, and 2, respectively, to 0.25, 0.25, and 0.5 μg/mL) ( Table 2). The MIC distributions of AS101 and AZT alone or in combination are visualized in Figure 2. Significant changes in MIC distributions were noted for AS101 and AZT (both p < 0.0001), with decreased MIC ranges for AS101 (from 2-12 to 0.5-32 μg/mL) and AZT (from 0.5-4 to 0.0625-1 μg/mL) ( Table 2). The fractional inhibitory concentration (FIC) indexes of the 12 isolates were all ≤0.5 (Table 3), suggesting the synergistic interaction between AS101 and azidothymidine. Additionally, the synergistic effect was observed in the growth curve ( Figure S1.) Supported by the aforementioned results, the combination therapy of AS101 and AZT is a potential treatment option for CRKP infections, needing further investigation in future study. All of the checkerboard methods for checking the synergistic effect were performed and repeated three times, and all results were reproducible.
In the present work, among the 12 CRKP isolates we examined, 7 isolates harbored the bla KPC gene and 1 isolate carried the bla OXA-48 gene. Our observations were in agreement with the previous work conducted in Taiwan, with also bla KPC as the dominant gene [28].

Antimicrobial Susceptibility
Susceptibilities of 19 antimicrobial agents against the 12 isolates were determined with a standard broth microdilution method following CLSI guidelines [36]. Briefly, two-fold serial dilutions of 19 antimicrobial agents were prepared. The bacterial suspension containing 2 × 10 8 CFU/mL of cells was prepared in brain heart infusion (BHI) broth (Becton Dickinson (BD), Sunnyvale, CA, USA) and diluted for 200 folds. The well-prepared drug solution and the diluted bacterial suspension were added into 96-well plate, respectively. The final bacterial density was 5 × 10 5 CFU/mL. After 16-18 h incubation, the result was detected by SpectraMax Absorbance Reader (CMax Plus, Molecular Devices, LCC, Sunnyvale, CA, US) at 600 nm. Among the 19 agents examined in this study, five classes of antibiotics were involved in the examination, including β-lactams and monobactam (ampicillin, ceftazidime, cefazolin, cefepime, cefoxitin, imipenem, ceftriaxone, meropenem, doripenem, ertapenem, cefotaxime, piperacillin-tazobactam, and aztreonam), aminoglycosides (amikacin and gentamicin), quinolones (ciprofloxacin and levofloxacin), inhibitors for folate synthesis (trimethoprim/sulfamethoxazole), and tetracycline (tigecycline). The results were also interpreted according to the breakpoints indicated by CLSI [36], and the interpretation for tigecycline was based on the guideline recommended by EUCAST [37]. Furthermore, the MICs for AS101 and azidothymidine (AZT) were also determined using the aforementioned broth microdilution method, and the results were collected for further evaluating the combination therapy of AS101 and AZT.

Synergistic Analysis
To investigate the synergistic effects between AZT (Toronto Research Chemicals, Toronto, Canada) and AS101 (Development Center for Biotechnology, Taipei, Taiwan), a checkerboard method was applied as described in a previous study [12,38]. In short, bacterial suspension containing 2 × 10 8 CFU/mL of cells was prepared in brain heart infusion (BHI) broth (Becton Dickinson (BD), Sunnyvale, CA, USA) and subsequently diluted to an appropriate concentration. Two-fold serial dilution of AZT and AS101 were also prepared in the appropriate ranges of concentrations, and the well-prepared AZT and AS101 solutions (50 µL each) were added into a 96-well plate. One hundred µL of the diluted bacterial suspension was also dispensed into wells at a final bacterial concentration of 5 × 10 5 CFU/mL. The result was also detected by SpectraMax Absorbance Reader (CMax Plus, Molecular Devices, LCC, Sunnyvale, CA, US) at 600 nm. The fractional inhibitory concentration (FIC) index was calculated to determine the synergistic effects as the following formula FIC = MIC AS101 in combination MIC AS101 in sin gle + MIC ZDV in combination MIC ZDV in sin gle Generally, the interaction between two agents was defined as synergism if FIC index ≤ 0.5; it was interpreted as interaction while the FIC index was between 0.5 and 4; antagonism was defined as an FIC index > 4.

Statistical Analyses
The MIC distribution graphs of AS101 and AZT, in combination or alone, were constructed by GraphPad Prism (v9.0, CA, USA) and analyzed with paired Student's t-test.

Conclusions
In this study, the noticeable changes in the MIC values after combining two drugs and fractional inhibitory concentration (FIC) index ≤ 0.5 in all strains indicated the synergistic activities between AS101 and azidothymidine against 12 clinical carbapenem-resistant K. pneumoniae (CRKP) isolates, harboring the carbapenemase genes or ESBL genes. Taken together, our efforts provided a new insight to develop a novel therapeutic option. Even though the sample size was small, our work still provided a proof of concept for the combination of AZT and AS101. Further in vivo studies and large-scale evaluations are needed for AS101-AZT combination therapy in future study.