Artificial intelligence (AI) is a new technology that has been employed to screen and discover new drugs. Using AI, an anti-diabetic treatment (Halicin) was nominated and proven to have a unique antibacterial activity against several harmful bacterial strains, including multidrug-resistant bacteria. This study aims to explore the antibacterial effect of halicin and microbial susceptibility using the zone of inhibition and the minimum inhibition concentration (MIC) values while assessing the stability of stored halicin over a period of time with cost-effective and straightforward methods. Linear regression graphs were constructed, and the correlation coefficient was calculated. The new antibacterial agent was able to inhibit all tested gram-positive and gram-negative bacterial strains, but in different concentrations-including the A. baumannii multidrug-resistant (MDR) isolate. The MIC of halicin was found to be 16 μg/mL for S. aureus (ATCC BAA-977), 32 μg/mL for E. coli (ATCC 25922), 128 μg/mL for A. baumannii (ATCC BAA-747), and 256 μg/mL for MDR A. baumannii. Upon storage, the MICs were increased, suggesting instability of the drug after approximately a week of storage at 4 °C. MICs and zones of inhibition were found to be high (R = 0.90 to 0.98), suggesting that halicin has a promising antimicrobial activity and may be used as a wide-spectrum antibacterial drug. However, the drug's pharmacokinetics have not been investigated, and further elucidation is needed.
Keywords: disc diffusion; gram-negative bacterial strains; gram-positive; halicin; minimum inhibitory concentration; multidrug-resistant bacteria; zone of inhibition.