Introduction

Neisseria gonorrhoeae, the causative agent of the sexually transmitted disease, gonorrhea, presents a substantial, global antimicrobial resistance threat1. Due to the increased rates of infection as well as the prevalence of multidrug-resistant N. gonorrhoeae strains worldwide, in 2017 the World Health Organization (WHO) listed N. gonorrhoeae as a “Priority 2” or “high” tier risk to public health1. Dual therapy of azithromycin (AZM) and ceftriaxone has been the standard-of-care for treatment of gonococcal infections. However, due to both increasing resistance to azithromycin (> 33% in some regions), and the potent anti-commensal activity of dual therapy, the CDC removed AZM from the treatment regimen for gonorrhea in 20202,3. Therefore, ceftriaxone remains the only recommended antibiotic for treatment of gonococcal infections4. However, there has been a concerning trend of increasing resistance to this treatment option, leading to the emergence of what is commonly referred to as ‘super gonorrhea’. This form of gonorrhea is characterized by extensive drug resistance, with high-level resistance to the recommended antibiotics, ceftriaxone and azithromycin, in addition to other classes of antibiotics5,6,7. Consequently, the world faces the real possibility of untreatable gonococcal infections5,8, and there is an urgent need to identify novel therapeutics against N. gonorrhoeae9.

Drug repurposing is a popular strategy that explores new therapeutic opportunities for approved drugs with available information on their pharmacokinetic data, dosages, and toxicity10,11,12,13,14,15,16,17,18,19. Salicylic acid is a highly privileged chemical scaffold: its derivatives are reported to exhibit a wide range of analgesic, antioxidant, antiproliferative, and anti-cancer activities20,21. In addition, azo-salicylates such as sulfasalazine and olsalazine are used in the treatment of ulcerative colitis. Salicylic acid derivatives were also reported to exhibit antibacterial activity against Gram-positive bacteria and some Gram-negative bacteria such as Escherichia coli and Enterobacter aerogenes22. Of particular relevance to our work, the use of salicylic acid to treat sexually transmitted diseases (including gonorrhea) was reported as early as the nineteenth century23. Recently, one of us (MNS) reported that salicylic acid (1a) exhibited modest activity against N. gonorrhoeae strains including the AZM-resistant strain (CDC-181) (Fig. 1). Promisingly, this compound also retained activity (MIC = 16 µg/mL) against an N. gonorrhoeae strain with reduced susceptibility to ceftriaxone (CDC-194)24.

Figure 1
figure 1

Minimum inhibitory concentrations (MICs in µg/mL) of azithromycin (AZM), ceftriaxone, salicylic acid 1a and methyl ester 2a against N. gonorrhoeae CDC-181 (AZM-resistant strain)24.

In the same report the methyl ester (2a) was found to be inactive. In this study, we sought to further explore the anti-gonococcal activity of salicylic acid analogs.

Results and discussion

Analog by catalog

For our initial exploration of SAR studies of salicylic acid 1a, we pursued an “analog by catalog” strategy, and purchased 20 structurally related compounds and tested them against three multi-drug resistant N. gonorrhoeae strains (Table 1).

Table 1 MICs (μg/mL) of commercial salicylic acid derivatives and 2-naphthoic acid analogs against select strains of N. gonorrhoeae. See Table 7 for the resistance status of various strains.

Most of these compounds showed lower potency than 1a against the strains tested. Specifically, mono-halogenation at C3–C6 (cf. 1b1h) significantly decreased potency, a point we will return to at the end of this report. Moving onto naphthoic acid derivatives, 3-hydroxy-2-naphthoic acid 3a was slightly less potent than 1a, but addition of a bromine at C7 resulted in the more potent compound 3b (MICs = 8 μg/mL vs 16 μg/mL for 1a). Regioisomer 3c was slightly less potent than 3b, and other substituted 3-hydroxy-2-naphthoic acids 3dh were less potent than 3b. The importance of the hydroxy group of 3b is shown by methyl ether 3i, which is significantly less potent. Finally, 1-hydroxy-2-naphthoic acid 3j, 6-hydroxy-2-naphthoic acid 3k, and amino-substituted 2-naphthoic acids 3l and 3m were less potent than 3b.

Synthesis and evaluation of structural analogs of 3b

Focusing on the favorable activity of 3b, we synthesized several structural additional compounds. Firstly, since 4-bromo-3-hydroxy-2-naphthoic acid 3c was nearly as potent as 3b against N. gonorrhoeae CDC-181 strain, we prepared its chloro analog 3n by electrophilic chlorination (Fig. 2)25. Similarly, since 4,7-dibromo-3-hydroxy-naphthoic acid 3d was nearly as potent as 3b against this strain, 3o was prepared from 3b. Lastly 3q was prepared by chlorination of 3j. Note that the moderate to low yields reported for these compounds reflect the need for multiple recrystallizations needed to achieve ≥ 95% purity. We then prepared 3p, the 7-chloro analog of 3b by copper-catalyzed Finkelstein reaction (Fig. 3) 26. Lastly, the inactive compound 3m was converted to the 5-chloro derivative 3r by electrophilic chlorination, as described in Fig. 327.

Figure 2
figure 2

Synthesis of 4-chloro derivatives of 3a, 3b, 3j.

Figure 3
figure 3

Synthesis of additional chlorinated 2-naphthoic acid analogs.

As seen in Table 2, like 3b, all of the halogenated 2-naphthoic acids bearing a hydroxy at C3 or C1 (i.e. 3n3q) had improved potency relative to 1a (Table 1). Halogenated 2-napthoic acid 3r, which bears a hydroxy group at C6 rather than C3 or C1, is not potent.

Table 2 MICs (μg/mL) of synthesized halogenated 2-naphthoic acid analogs related to 3b.

To assess the importance of the carboxyl group to the antibacterial activity of 3b, the methyl (4b) and cyclic methylene ester (5b) were prepared, as were the 1° (6b) and methyl amides (7b) (Fig. 4). Further, since it has been shown that drug uptake by Gram-negative bacteria can sometimes be substantially improved by the addition of basic amine functionality28, 29, we synthesized basic amine-bearing amide 8b and Mannich base 9b from methyl ester 4b (Fig. 5).

Figure 4
figure 4

Synthesis of carboxyl derivatives of 3b.

Figure 5
figure 5

Synthesis of basic or zwitterionic derivatives of 3b.

As depicted in Table 3, the ester (4b, 5b) and amide (6b-8b) derivatives of 3b significantly lost potency, as did the zwitterionic Mannich base 9b.

Table 3 MICs (μg/mL) of synthesized analogs of 3a.

Exploration of 4,5-disubstituted salicylic acids

Knowing that naphthyl rings and naphthols are susceptible to CYP450 oxidation, we turned our attention to 4,5-disubstituted salicylic acids as potential isosteres for 3b. To jump-start this exploration, we purchased four 4,5-disubstituted salicylic acid derivatives 1i–l and screened them against the three N. gonorrhoeae isolates (Table 4).

Table 4 MICs (μg/mL) of commercial and synthesized 4,5-disubstituted salicylic acids.

Interestingly, compound 1i and 1j appeared to be nearly as potent as 3b with MIC = 8 µg/mL against N. gonorrhoeae CDC-166 and CDC-165 stains and MIC = 16 µg/mL against N. gonorrhoeae CDC-181 strain. It is noted that these compounds are significantly more potent than the mono-halogenated salicylic acids described in Table 1, including 5-bromosalicylic acid 1b and 5-chlorosalicylic acid 1c (MIC = 64 µg/mL). Thus, the presence of a second halogen (chlorine) at C4 increased potency by 4-fold. However, installation of a methyl group at position 4 in place of Cl negatively affected potency (Table 3, cf. 1k and 1i, 1l and 1j).

To further explore this scaffold, we synthesized two additional 3,4-dihalogenated salicylic acid derivatives from 5-fluorosalicylic acid 1f. Compounds 1m–n were prepared from 1f via electrophilic halogenation as described in Fig. 630.

Figure 6
figure 6

C5-halogenation of 4-fluorosalicylic acid 1f.

We concluded that similar to methyl substitution, fluorine substitution at position 4 seems to negatively affect potency of salicylic acid derivatives (Table 4, cf. 1m and 1i, 1n and 1j).

Anti-gonococcal activity of salicylic acid analogs

The antibacterial activity of the most promising salicylic acid derivatives (3b, 3o, 3p, 3q, 1i and 1j) was investigated against a panel of multidrug-resistant N. gonorrhoeae strains including ceftriaxone- and azithromycin-resistant strains (Table 5). The panel of strains tested contained 6 WHO reference strains with well-characterized genetic and phenotypic markers31. Salicylic acid analogs maintained consistent potency against the tested strains. Compounds 3b, 1i and 1j displayed the most potent activity, effectively inhibiting the tested strains at concentrations ranging from 2 to 8 µg/mL. Particularly intriguingly was their activity against azithromycin-resistant N. gonorrhoeae strains (WHO-P, WHO-U, WHO-V and WHO-Z), with MIC values ranging from 2 to 8 µg/mL. Moreover, the analogs demonstrated the comparable activity against ceftriaxone-resistant strains, including WHO-X, WHO-Y and WHO-Z, with MICs ranging from 2 to 32 µg/mL. To validate the MIC results, N. gonorrhoeae CDC-10328, a reference strain for antimicrobial susceptibility testing, was included among the tested strains. The MICs of the control antibiotics, azithromycin and ceftriaxone, against this strain matched the modal MICs previously reported32. These findings underscore the promising potential of the salicylic acid analogs as effective anti-gonococcal agents, even against multidrug-resistant strains (Table 5).

Table 5 MICs (µg/mL) of select salicylic acid analogs against a panel of N. gonorrhoeae strains.

Killing kinetics of salicylic acid analogs

Following the confirmation of potent antibacterial activity of the analogs against multidrug-resistant N. gonorrhoeae strains, we assessed their killing kinetics against N. gonorrhoeae CDC-166 (Fig. 7). Compounds 3q, 1i and 1j displayed a rapid bactericidal activity against N. gonorrhoeae, completely eradicating the high bacterial burden below the detection limit within 8 h. This was superior to the activity of azithromycin which exerted its bactericidal activity after 10 h and required 12 h to completely eradicate the bacterial burden. Compounds 3b, 3o and 3p were as effective as azithromycin in their bactericidal activity where they reduced the bacterial count below the detection limit within 10 h (Fig. 7). The rapid bactericidal activity observed in these salicylic acid analogs is a highly desirable trait for anti-gonococcal agents. It not only limits the spread of infection but also plays a pivotal role in reducing the development of bacterial resistance and preventing disease progression33,34.

Figure 7
figure 7

Time-kill kinetics of salicylic acid analogs against N. gonorrhoeae CDC-166. A log-phase bacterial culture was exposed to either compounds 3b, 3o, 3p, 3q, 1i, 1j or azithromycin (at 5 × MIC). DMSO (solvent) served as a negative control. The error bars represent standard deviation values for each test agent.

Effects on commensal bacteria

One of the disadvantages of the anti-gonococcal standard of care therapeutics is that they inhibit growth of both gonococci and commensal bacteria35. The vaginal microbiota are the primary defense line against N. gonorrhoeae infection. They compete with N. gonorrhoeae for adhesion to the genitourinary tract in addition to creating an acidic environment that prevents gonococcal colonization. Hence, disruption of the healthy microbiota present in the genitourinary tract could enhance the gonococcal infection36. The microbiome of the genitourinary tract is dominated by Lactobacillus. It was reported that Lactobacillus could significantly reduce N. gonorrhoeae viability by creating acidic environment, producing bacteriocins, releasing biosurfactants, co-aggregating with gonococci, and reducing the gonococcal adhering to epithelial cells37. We evaluated the antibacterial activity of the prioritized salicylic acid derivatives (3b, 3o, 3p, 3q, 1i and 1j) alongside azithromycin, against representative members of the vaginal microbiota of the Lactobacillus species. As reported previously, azithromycin inhibited growth of Lactobacillus strains with MIC values of ≤ 1 µg/ml38,39. To our delight, the salicylic acid analogs, 3b, 3o, 3p, 3q, 1i and 1j did not inhibit growth of Lactobacillus strains at concentrations as high as 128–256 µg/mL (Table 6), which indicates the high selectivity of the salicylic acid analogs against N. gonorrhoeae over the vaginal microbiota.

Table 6 MICs (µg/mL) of AZM and select salicylic acid analogs against N. gonorrhoeae and different Lactobacillus species.

Cytotoxicity of salicylic acid analogs.

An essential attribute in the development of new drugs is their lack of toxicity. To assess this aspect, we evaluated the prioritized salicylic acid analogs (3b, 3o, 3p, 3q, 1i and 1j) for their cytotoxicity against kidney epithelial (Vero) cells (Fig. 8). We aimed to identify any potential cytotoxicity effects on mammalian cells. Encouragingly, the compounds exhibited an excellent safety profile and demonstrated high tolerability to Vero cells. The CC50, representing the concentration required to reduce cell viability by 50%, were found to be higher than 128 μg/mL for all analogs, with the exception of 3q, which exhibited reduced viability at 128 μg/mL. Notably, all the cells were viable at 64 μg/mL (Fig. 8). These results highlight the favorable cytotoxicity profile of the salicylic acid analogs, signifying their potential as safe and non-toxic candidates for further drug development.

Figure 8
figure 8

In vitro cytotoxicity assessment of salicylic acid analogs against kidney epithelial cells (Vero). Compounds were incubated with Vero cells for 24 h. Then, cells were incubated with MTS reagent before measuring absorbance values. Results are presented as percent viable cells relative to DMSO (negative control). The absorbance values represent an average of four samples analyzed for each compound. Error bars represent standard deviation values.

Conclusion

In this report, we demonstrated that the significant anti-gonococcal activity of salicylic acid 1a against AZM-resistant N. gonorrhoeae (MIC = 16 µg/mL) is not a simple characteristic of its o-hydroxy benzoic acid moiety. Inopportune substitution of the benzo ring (cf. Table 1) greatly reduced anti-gonococcal potency. In contrast, we found that appropriately substituted 3-hydroxy-2-naphthoic acids (3b, 3o, 3p) and 1-hydroxy-2-naphthoic acid (3q) have twice the anti-gonococcal potency of 1a. In addition, these compounds, and less-potent 4,5-dihalogenated salicylic acids (1i, 1j) are significantly less toxic to commensal vaginal bacteria than AZM. Moreover, these compounds exhibited rapid bactericidal activity against N. gonorrhoeae, were tolerable to Vero cells, and retain activity against ceftriaxone- and azithromycin-resistant strains. Thus, although greater anti-gonococcal potency is required for a new therapeutic, the other favorable properties exhibited by the simple molecular scaffolds displayed in Table 6 suggest they warrant further development.

Materials and methods

Procurement and synthesis of tested compounds

Compounds 1ah and 3am were purchased from a variety of suppliers. In each case, 1H NMR was used to confirm the identity and purity of the compound. Compounds 3nr, 49b, and 1mn were prepared as described above. Full synthetic procedures and analytical characterization data (1H, 13C NMR, HRMS) are provided in the Supplementary Information.

Bacterial strains, media, reagents and antibacterial assay procedures

N. gonorrhoeae strains (Table 7) used in the study were clinical isolates obtained from the CDC, the WHO and the American Type Culture Collection (ATCC). Lactobacillus isolates were obtained from the Biodefense and Emerging Infections Research Resources Repository (BEI Resources). Azithromycin (AZM) and ceftriaxone (CEF) were purchased from TCI America, (Portland, OR, USA). Media and reagents were purchased commercially: brucella broth, IsoVitaleX, chocolate II agar plates and MRS broth (Becton, Dickinson and Company, Cockeysville, MD, USA), yeast extract and dextrose (Fisher Bioreagents, Fairlawn, NJ, USA), protease peptone (Oxoid, Lenexa, KS, USA), hematin, pyridoxal, and nicotinamide adenine dinucleotide (NAD) (Chem-Impex International, Wood Dale, IL, USA), Eagle's Minimum Essential Medium and fetal bovine serum (Corning, Manassas, VA, USA) and phosphate-buffered saline (PBS) (Corning, Manassas, VA, USA). Compounds were prepared as stock solutions in DMSO, diluted in media, to give a final DMSO concentration of less than 2%.

Table 7 Bacterial isolates used in the study.

Antibacterial activity of salicylic acid analogs against N. gonorrhoeae strains

The determination of MICs for compounds was carried out as described previously11,13,40,41,42. Briefly, N. gonorrhoeae strains were grown overnight on chocolate agar plates. A bacterial solution equivalent to 1.0 McFarland standard was prepared and diluted in brucella broth supplemented with yeast extract, dextrose, protease-peptone, NAD, pyridoxal, hematin and IsoVitaleX to reach about 1 × 106 CFU/mL. Serial dilutions of test agents were incubated with bacteria at 37 °C in presence of 5% CO2 for 24 h before recording the MICs as observed visually. MICs reported are the lowest concentrations of each test agent that could completely inhibit the visual bacterial growth.

Killing kinetics analysis of salicylic acid analogs against N. gonorrhoeae

In order to determine if salicylic acid analogs exhibit bacteriostatic or bactericidal activity against N. gonorrhoeae, a standard time-kill assay was performed against N. gonorrhoeae CDC-166, as described previously11,38,39. Briefly, a log-phase culture of N. gonorrhoeae was diluted to  106 CFU/mL in the supplemented brucella broth. Test agents were then added (at 5 × MIC in triplicates) and incubated with bacteria at 37 °C in presence of 5% CO2. An aliquot from each treatment was collected after the corresponding times of incubation and subsequently serially diluted and plated onto chocolate II agar plates. Plates were incubated for 24 h at 37 °C before viable CFU/mL was determined.

In vitro cytotoxicity evaluation of the salicylic acid analogs

The in vitro cytotoxicity assessment for salicylic acid analogs was carried out against kidney fibroblast (Vero) cells as described elsewhere43,44,45,46,47,48. Briefly, compounds were incubated with Vero cells for 24 h and DMSO served as a negative control. Then, cells were incubated with MTS reagent for 3 h before measuring absorbance values (OD490).

Antibacterial activity of salicylic acid analogs against genitourinary tract normal microbiota strains

The MICs of salicylic acid analogs against representative commensal members of the genitourinary tract were determined, as described elsewhere16,45,49,50,51,52. Lactobacilli were grown onto MRS agar for 48 h at 37 °C in presence of 5% CO2. A bacterial solution equivalent to 0.5 McFarland standard was diluted in MRS broth to achieve a bacterial concentration of ~ 5 × 105 CFU/mL and incubated with serial dilutions of the test agents as described, before recording the MIC values by visual inspection of growth.