Antifungal properties of (2S, 4R)-Ketoconazole sulfonamide analogs

Invasive candidiasis remains a significant health concern, as it is associated with a high mortality risk. In addition, the risk of infection is significantly elevated in immunocompromised patients such as those with HIV, cancer, or those taking imcmunosuppressive drugs as a result of organ transplantation. The majority of these cases are caused by C. albicans, and C. glabrata is the second most common cause. These infections are typically treated using approved antifungal agents, but the rise of drug-resistant fungi is a serious concern. As part of our on-going effort to identify novel antifungal agents, we have studied the in vitro antifungal properties of a series of sulfonamide analogs of (2S, 4R)-Ketoconazole. Herein we report on the in vitro activity against the key fungal pathogens C. albicans, and C. glabrata.


Introduction
According to the U.S. Center for Disease Control (CDC), invasive candidiasis remains a significant health issue. The average incidence rate of candidemia in 2013-2017 was 9 per 100,000, and it is estimated that~25,000 new cases occur annually nationwide (Tsay et al., 2018). The annual global incidence of invasive candidiasis is~750,000 (Bongomin et al., 2017). Candida is one of the most common causes of bloodstream infections in the U.S (Magill et al., 2018). In addition, candidemia is associated with a high mortality risk. The U.S. CDC reported that between 2012 and 2016,~25% of candidemia patients died while hospitalized (US CDC website 2022b). Though less common, invasive candidiasis can also occur in various organs in the absence of bloodstream infection. The danger of fungal infection is significantly elevated when the immune system is compromised by viral infection (e.g. Human Immunodeficiency Virus (HIV)/Acquired Immunodeficiency Syndrome (AIDS)), immunosuppressive agents designed to support organ transplants, anticancer agents, and antibody therapies used to treat autoimmune disorders. Risks are also elevated when natural barriers are disrupted (e.g., surgery, burns) (Monk et al., 2020).
Importantly, nearly 95% of all forms of invasive candida infection in the U.S. are caused by 5 candida species (C. albicans, C. glabrata, C. parapsilosis, C. tropicalis, and C. krusei.). C. albicans is the leading cause of candida infection in the U.S., but non-C albicans infections account for nearly 2/3 of candidemia case in the U.S. (US CDC website (2022b). C. glabrata is the second most common invasive fungal species, accounting for 15-25% of clinical isolates in the U.S. (Pfaller and Diekema, 2007;McCarty and Pappas 2016). This species is associated with high morbidity and mortality (40-60%) (Timmermans et al., 2018). In addition, there has also been a steady increase in fungal drug resistance (Roemer and Krysan, 2014). Much like antibacterial agents, prolonged and/or repeated clinical use of antifungal agents induces resistance pathways and the emergence of drug-resistant fungi (Sanglard et al., 1999;Vermitsky and Edlind, 2004;Chau et al., 2006). The broad use of antifungal agents in the agrochemical sector has also been linked to the development of cross-resistance to antifungal agents used in medicine (Snelders et al., 2008;Snelders et al. 2011;Snelders et al. 2015). Drug resistance is increasingly problematic with C. albicans as well as non-albicans Candida species such as C. glabrata (Vallabhaneni et al., 2015;Kolaczkowska and Kolaczkowski. 2016).
As part of our effort to address the need for new antifungal agents, we have been studying the chemical space surround (±)-Ketoconazole (2). We were surprised to discover that since its initial discovery (Backx et al., 1979), there have been very few reports of direct analogs of (±)-Ketoconazole (2) or its individual enantiomers. In 1992, D. M. Rotstein et al. reported on the ability of the 4 stereoisomers for (±)-Ketoconazole (2) to inhibit lanosterol 14α-demthylase (Cytochrome P450 51a, Cyp51a), the biochemical target that gives rise to the azole's antifungal properties. They found that (2S, 4R)-Ketoconazole was the most potent inhibitor of the four (Rotstein et al., 1992). Separately, we have reported on the synthesis and evaluation of a series of (2S, 4R)-Ketoconazole sulfonamide analogs (3) (Figure 1) as potential treatments for metabolic syndrome (Blass et al., 2016). Herein, we report the in vitro antifungal properties of these compounds against C. albicans and C. glabrata.
Broth microdilution susceptibility test: The minimum inhibitory concentrations (MIC 75 ) were determined by the broth microdilution method according to the Clinical and Laboratory Standards Institute (CLSI) M27-A4 standard (CLSI, 2017)). Briefly, a suspension of C. albicans or C. glabrata culture (~0.5-2.5 × 10 3 cells/ml) was resuspended in RPMI 1640 media (Fisher Catalog number 11875093). The resuspended culture was further incubated with 100 μl of twofold serially diluted test compounds (100-0.8 µM and 1.0 µM to 7 nM) at 30°C for 24-48 h. Positive and negative control (wells without antifungal agents and wells without yeast) were conducted. Increased optical density (OD) corresponds to the Candida growth and was quantified by comparison with untreated Candida control samples. The MIC 75 of test compounds was determined visually and spectrophotometrically (OD600) after 24-48 h. Experiments were run in triplicate, and data was reported as a mean.
Cyp3A4 inhibition assay: Human Cyp3A4 Inhibition assay: An IC 50 for inhibition human Cyp3A4 metabolism of midazolam to 1-OH-midazolam was determined with 10 concentrations of test compounds (half-log serial dilutions; duplicate points). Assays were conducted in 1 ml 96 well polypropylene plates containing 100 µL of 100 mM potassium phosphate (pH 7.4), 3 mM MgCl 2 , 2 µM midazolam, 1 mM NADPH (tetra Na salt), Corning Gentest supersomes (product 456202: 3 pmol/ml human Cyp3A4, 47 nmol/min human P450 reductase activity, 14 pmol human cytochrome b5; and product 456201: insect control microsome protein; total final microsome protein concentration 0.15 mg/ml) and test compound. All components except NADPH were added to a prewarmed plate and reactions were initiated by adding NADPH. After 30 min at 37°C, reactions were terminated with 200 µL acetonitrile containing 30 µM prednisone. After centrifugation for 10 min at 2200 x g, 165 µL of supernatants were transferred to analysis plate. Samples were analyzed for 1-OH-midazalam concentration, using prednisone as the internal standard, using a Waters Aquity UPLC in tandem with a Xevo TQ MS mass spectrometer (BEH C18 column; 5-95% acetonitrile gradient with 0.1% formic acid; ESP + mode). IC 50 's were determined using GraphPad's Prism v 5.04 nonlinear curve fitting program.

Frontiers in Drug Discovery
frontiersin.org

Results
The in vitro antifungal activity of our of (2S, 4R)-Ketoconazole sulfonamide analogs (3a) -(3l) against C. albicans and C. glabrata is described in Table 1. Replacing the acetamide of (2S, 4R)-Ketoconazole (4) with small, linear suflonamides (3a-3c) produced compounds whose in vitro potency against the test fungi was maintained (C. albicans MIC 75 = 125 nM, C. glabrata MIC 75 = 500 nM). Insertion of a branch point as seen in the isopropyl analog (3d), however, led to a complete loss of in vitro antifungal activity. Constraining the branch point through the formation of the corresponding cyclopropane ring (3e) restore in vitro antifungal activity against both species (C. albicans MIC 75 = 800 nM, C. glabrata MIC 75 = 800 nM), but not to the levels observed in (3a) -(3c). Insertion of an additional methylene unit between the cyclopropane ring and the sulfonamide moiety (3f) produced a significant loss in vitro activity against C. glabrata (MIC 75 = 25,000 nM), but activity against C. albicans (MIC 75 = 800 nM) was unchanged in comparison to (3d).
The addition of polar groups to the sulfonamide moiety produced mixed results. Appending a cyano group on the sulfonamide moiety (3g) provided a compound with in vitro antifungal activity comparable to that of (±)-Ketoconazole (2) (C. albicans MIC 75 = 250 nM, C. glabrata MIC 75 = 500 nM), but adding a methyl sulphone unit (3h) caused in vitro antifungal activity to drop substantially (C. albicans MIC 75 = 800 nM, C. glabrata MIC 75 = 6250 nM).

Discussion
As noted above, C. albicans and C. glabrata are responsible for a large portion of Candida infection in the US and represent a serious health threat (US CDC website, 2022a). Blood stream infection of either of these organisms are associated with high mortality (Timmermans et al., 2018). Immunocompromised patients are at an even greater risk should they be unfortunate enough to experience infections with C. albicans, C. glabrata, or other candida species (Monk et al., 2020). In addition, the rise of drug resistant fungi (Sanglard et al., 1999;Vermitsky and Edlind, 2004;Chau et al., 2006) suggests that the current staple of antifungal agents may eventually be insufficient to meet patient needs. Similar to antibiotics and antimicrobial agents, the appearance of clinically relevant drug resistant fungi has been driven by prolonged use of antifungal agent in clinical setting as well as in the agricultural sector (Snelders et al., 2008;Snelders et al. 2011;Snelders et al. 2015).
Our studies focused on the further exploration of the chemical space surrounding (±)-Ketoconazole (2) began with the observation that the initial disclosure of this antifungal agent (Beckx 1979) described only a handful of analogs. This initial disclosure predates the advent of both high through put chemistry and high throughput screening, which may explain the limited nature of the original disclosure. Surprisingly, we found that the literature contained very few additional disclosures of (±)-Ketoconazole analogs, and we viewed this as an opportunity to explore the chemical space surrounding this antifungal agent. Specifically, we have demonstrated that the acetamide of (2S, 4R)-Ketoconazole (4) can be replaced with sulfonamides (3a-3l) to produce compounds that have in vitro antifungal properties with respect to C. albicans and C. glabrata. We have also found that similar to (±)-Ketoconazole (2), these compounds inhibit human Cyp3A4 (IC 50 = 75.6-450 nM), a key metabolic enzyme that has been linked to drug/drug interactions (Molenaar-Kuijsten et al., 2021). Interestingly, we noted a divergence between antifungal activity and Cyp3A4 inhibition, Specifically, we observed that our most potent in vitro antifungal agent (3l) demonstrated a wider window between Cyp3A4 inhibition and its activity against C. albicans than (±)-Ketoconazole (2) (2.6-fold versus 0.5-fold). This 5-fold improvement in selectivity suggests that it may be possible to identify analogs with a wide enough separation to avoid the risk of Cyp3A4 mediated drug/drug interaction. Additional studies will be required to determine the true potential of this opportunity.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.