Synthesis and Inhibitory Activity of Machaeridiol-Based Novel Anti-MRSA and Anti-VRE Compounds and Their Profiling for Cancer-Related Signaling Pathways

Three unique 5,6-seco-hexahydrodibenzopyrans (seco-HHDBP) machaeridiols A–C, reported previously from Machaerium Pers., have displayed potent activities against methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Enterococcus faecium, and E. faecalis (VRE). In order to enrich the pipeline of natural product-derived antimicrobial compounds, a series of novel machaeridiol-based analogs (1–17) were prepared by coupling stemofuran, pinosylvin, and resveratrol legends with monoterpene units R-(−)-α-phellandrene, (−)-p-mentha-2,8-diene-1-ol, and geraniol, and their inhibitory activities were profiled against MRSA ATCC 1708, VRE ATCC 700221, and cancer signaling pathways. Compounds 5 and 11 showed strong in vitro activities with MIC values of 2.5 μg/mL and 1.25 μg/mL against MRSA, respectively, and 2.50 μg/mL against VRE, while geranyl analog 14 was found to be moderately active (MIC 5 μg/mL). The reduction of the double bonds of the monoterpene unit of compound 5 resulted in 17, which had the same antibacterial potency (MIC 1.25 μg/mL and 2.50 μg/mL) as its parent, 5. Furthermore, a combination study between seco-HHDBP 17 and HHDBP machaeriol C displayed a synergistic effect with a fractional inhibitory concentrations (FIC) value of 0.5 against MRSA, showing a four-fold decrease in the MIC values of both 17 and machaeriol C, while no such effect was observed between vancomycin and 17. Compounds 11 and 17 were further tested in vivo against nosocomial MRSA at a single intranasal dose of 30 mg/kg in a murine model, and both compounds were not efficacious under these conditions. Finally, compounds 1–17 were profiled against a panel of luciferase genes that assessed the activity of complex cancer-related signaling pathways (i.e., transcription factors) using T98G glioblastoma multiforme cells. Among the compounds tested, the geranyl-substituted analog 14 exhibited strong inhibition against several signaling pathways, notably Smad, Myc, and Notch, with IC50 values of 2.17 μM, 1.86 μM, and 2.15 μM, respectively. In contrast, the anti-MRSA actives 5 and 17 were found to be inactive (IC50 > 20 μM) across the panel of these cancer-signaling pathways.


Introduction
There are currently considerable challenges with the treatment of infections caused by strains of clinically relevant bacteria that show multidrug-resistance (MDR), such as methicillin-resistant Staphylococcus aureus (MRSA) and Enterococci faecalis (VRE), as well In this paper, we report the synthesis of a series of new and diverse stemofuran-, pinosylvin-, and resveratol-type aralkyl compounds (1-11, 15-17) and geraniol-

Results
Seven phytocannabinoids with HHDBP and seco-HHDBP skeleta, machaeriols A-D and machaeridiols A-C, were previously isolated from Macherium Pers., of which the machaeriol chemotype possesses a unique pharmacophore with 10aS absolute configuration, an opposite configuration to 10aR of THC (tetrahydrocannabinol) / HHC (hexahydrocannabinol) [8]. In addition, the configuration of the machaeridiol chemotype at 1S, 3S, and 4S positions is opposite to those at 3R and 4R positions of CBD. Therefore, an unprecedented structural similarity for the HHDBP skeleton was observed between machaeriol and HHC, and for the 5,6-seco-HHDBP skeleton between machaeridiol and dihydro-CBD. Because machaeriols A-D are analogues to trans-HHC, they show an enantiomeric configuration at the ring junction, established by CD and enantioselective total syntheses [15][16][17]. Similarly, the macheridiol chemotype is similar to dihydro-CBD, with the n-alkyl mioety replaced by steryl and benzofuranyl forms. [18] The pseudo-enantiomeric configuration at C-3 and C-4, compared to CBD, was established by CD studies. Based on the uniqueness of the stereogenic ring junction of the machaeridiol chemotype, the synthesis of machaeridiol B was reported [19] to involve long steps; to repeat this method would be quite challenging.
The activities of compounds 5, 11, and 17 were found to be similar to those of natural products machaeridiol C (i.e., a stemofuran A type, 17) and machaeridiol A (i.e., a pinosylvin type, 11), with MIC values of 1.25 µg/mL to 2.5 µg/mL, and more potent than the positive drug control meropenem against MRSA and VRE.

Antimicrobial Combination Studies
Based on our previous investigation on the synergistic activity between machaeridiol B and machaeriol C against MRSA [9], analogs 5 and 17 (i.e., machaeridiol type analogs) were subjected to combination studies with machaeriol C. Because both compounds 17 and machaeriol C showed potent activity against MRSA (Tables 1 and 2), they were subjected to a broth microdilution checkerboard assay [9], using an MRSA strain. The presence of machaeriol C improved the anti-MRSA activity of 17; for example, 17 alone showed an MIC of 0.625 µg/mL, but in combination with machaeriol C, the MIC of 17 was decreased to 0.156 µg/mL. This combination study of 17 and machaeriol C displayed a synergistic effect with a fractional inhibitory concentration (FIC) value of 0.5 against MRSA, showing a four-fold decrease in MIC of 17 to 0.312 µg/mL ( Table 2). On the other hand, compound 5 + machaeriol C displayed an additive effect with an FIC value of 0.75, showing a four-fold reduction of MIC for 5 from 1.25 µg/mL to 0.312 µg/mL and a two-fold reduction of MIC of macheriol C from 1.25 µg/mL to 0.625 µg/mL. However, a combination of either 5 or 17 with antibiotic vancomycin did not show any such effect against MRSA ( Table 2). All checkerboard assays were run in triplicate.

In Vivo Antimicrobial Activity Studies agaisnt Nosocomial MRSA
Compounds 11 and 17 were evaluated in vivo against MRSA strain USA300, using a nosocomial assay protocol in a murine model. Both compounds were found to be unable to reduce bacterial loads in the nasopharynx and lung at a dose of 30 mg/kg, using ciprofloxacin as a reference standard. A single intranasal dose of compound, vehicle, or ciprofloxacin was administered 12 h following infection, and this regimen produced significant but modest results for ciprofloxacin (Table 3). Alternate concentrations, additional treatments, parental versus intranasal routes, and/or alternate treatment schedules need to be investigated to determine the optimal results for ciprofloxacin control, to better compare the efficacy of compounds 11 and 17.

Activities against Cancer-Related Signaling Pathways
In order to evaluate the anticancer activities and understand the mechanism of action, compounds 1-17 were profiled against a panel of luciferase reporter genes that assessed the activity of complex cancer-related signaling transduction pathways, using T98G glioblastoma multiforme cells (i.e., transcription factors; TF; Scheme 5). The compounds were considered inactive if their inhibition was less than 40% of the induction at the highest concentration tested (20 µM). These inactive compounds were not further advanced for a dose response assay for IC 50 determination. Among the compounds (Table 4), the geran-iol derivative of pinosylvin legend 14 exhibited strong activity against many signaling pathways (TFs), notably with IC 50 values between 1.86 µM to 4.45 µM against the Stat, Smad, Myc, Ap-1, NF-kB, E2F, ETS, and Notch pathways. However, the MRSA active stemofurans 5 and 17 and the pinosylvin 11 analogs were found to be inactive and weakly active, respectively, against the panel of these signaling pathways.
In order to evaluate the anticancer activities and understand the mechanism of action, compounds 1-17 were profiled against a panel of luciferase reporter genes that assessed the activity of complex cancer-related signaling transduction pathways, using T98G glioblastoma multiforme cells (i.e., transcription factors; TF; Scheme 5). The compounds were considered inactive if their inhibition was less than 40% of the induction at the highest concentration tested (20 µ M). These inactive compounds were not further advanced for a dose response assay for IC50 determination. Among the compounds (Table 4), the geraniol derivative of pinosylvin legend 14 exhibited strong activity against many signaling pathways (TFs), notably with IC50 values between 1.86 µM to 4.45 µ M against the Stat, Smad, Myc, Ap-1, NF-kB, E2F, ETS, and Notch pathways. However, the MRSA active stemofurans 5 and 17 and the pinosylvin 11 analogs were found to be inactive and weakly active, respectively, against the panel of these signaling pathways.

Scheme 5.
Schematic assay protocol of cancer-related signaling transduction pathways using T98G glioblastoma multiforme cells.  * Values are IC 50 in µM that inhibited luciferase induction by 50%. Test agents were added to cells 30 min before the addition of the indicated inducer and were harvested 4 or 6 h later for the luciferase assay (Notch, FoxO, Wnt, Hedgehog, miR-21). No inducer was added to cells transfected with the FoxO, miR-21, or pTK control vector. # This unique assay panel had a special feature of evaluating which pathways were sensitive (modulated) to a test agent. Each test agent was run at various concentrations and each concentration was tested in duplicate. As this multiplex assay offered many advantages for the simultaneous measurement and analysis of multiple pathways for any test agent, the intra-and inter-assay precision, accuracy, and reproducibility were high with this method. The results were normalized to the control vector, which was pTK. This empty vector control allowed us to see any non-specific cytotoxic effects of the test compounds on the cells. -: No activity detected.

Discussion
The synthesis of the compounds 1-17, using stemofuran A, pinosylvin, and resveratrol legends, was inspired by the potent anti-MRSA and anti-VRE activities of machaeridiol A and machaeridiol C [9], CBD [21], and CBG [22] to develop a new pipeline of anti-MRSA active compounds. This appears to be the first report of compounds 1, 2, 4-12, 14, 15, and 17 from either a natural or synthetic source. Compounds 3 and 16 were previously synthesized by coupling reactions similar to those described in the current study, but using a different catalyst (BF 3 ·OEt 2 ) [16] and a different solvent (DCM) [23], respectively. Compound 13 (amorphastibol) was reported from three Amorpha species [24]. Compounds 9 and 10 were isolated as a 5:1 mixture and very likely formed by an acid-catalyzed intermolecular cyclization of compound 11, as previously reported for CBD under similar conditions [25]. Compound analogs to 6-8 and 11, but without an exocyclic double bond in the monoterpene moiety, were reported from Lindera reflexa [26,27].
The HHDBP and 5,6-seco-HHDBPs, machaeriols and machaeridiols, were only reported from Machaerium Pers. (Rimachi 12161), despite our search for other Machaerium species available at the the repository of the National Center for Natural Products Research at the University of Mississippi. However, the stereo-specific total synthesis of machaeriols A-D and mechaeridiol B (Figure 1) has been reported [15,19,28]. In addition, several HHC-related machaeriol analogs were synthesized, showing a strong inhibition of tumor (breast cancer) growth by targeting VEGF-mediated angiogenesis signaling in endothelial cells and suppressing VEGF production and cancer cell growth [29]; however, to best of our knowledge, machaeridiol analogs have not been synthesized and reported for anti-MRSA or anticancer activities.
The antibacterial activities against MRSA and VRE, as well as anticancer activities toward signal transduction pathways, were assessed; compounds 5, 11, and 17 showed potent activities against MRSA and VRE, with MIC values in the range of 1.25 µg/mL to 2.50 µg/mL, while compound 14 exhibited moderate activity (MIC 5.0 µg/mL). It is intriguing to note that the combination of seco-HHDBP 17 with HHDBP machaeriol C displayed a synergistic effect against MRSA, showing a four-fold decrease in the MIC of both 17 and machaeriol C, while compound 5 showed an additive effect with machaeriol C and a four-fold decrease in the MIC of 5 ( Table 2). The strong in vitro efficacy of compounds 11 and 17 against MRSA prompted investigation of their efficacy in vivo. A preliminary experiment employing a single intranasal dose of each compound, vehicle, and ciprofloxacin yielded modest efficacy results for ciprofloxacin only. Different parameters in the treatment regimen must be tested to fully determine whether compounds 11 and 17 are able to reduce bacterial loads in vivo. Because the results were modest for ciprofloxacin, it is possible that repeated treatment or systemic administration may be necessary to achieve efficacy. Although compound 17 was found to be inactive against in vivo nosocomial MRSA at a low dose using a murine model, our study facilitated a way forward for further mechanistic investigation on these new leads.
Finally, three geranyl-substituted structural analogs (12)(13)(14) of cannabigerol (CBG) were synthesized. Among these, compound 14 was found to be strongly active against the cancer signaling pathways, notably Smad, Myc, and Notch, with IC 50 values between 1.2 µM and 4.8 µM in T98G glioblastoma multiforme cells. On the other hand, the anti-MRSA compounds 5 and 17 were found to be inactive against the panel of these transporter genes. A recent report by Morch et al. in 2021 [30] showed that Smad/TGF-β is a key player in probiotic protection against MRSA in C. elegans. In addition, in 2013, Choi et al. [31] reported that Stat3 induction helps host defense against MRSA pneumonia. Interestingly, the Notch pathway is activated by S. aureus toxins, in both in vivo and in vitro conditions [32], which suggests that compound 14 could have strong therapeutic potential against MRSA, due to its strong selective inhibition against the Notch pathway in this current study. Several reports suggested as association of MRSA with a significant increase in cancer mortality [5,6]. Thus, the inhibition of compound 14 against these cancer-signaling pathways could possibly decrease the morbidity of cancer conditions. Earlier, CBG, a close structural analog of 13, was not only reported for anti-MRSA activity [22], but also for its strong activity in colon, breast, and oral carcinomas [33][34][35]. In addition, CBG has been regarded as a potential therapeutic agent for glioblastoma [36]. This combination of earlier reports and our observations from the current study strongly suggest that compound 14 should be explored further for several cancers, particularly glioblastomas (brain tumors), and in combination therapies against MRSA. The co-occurrence of alterations in the multiple pathways suggests that compound 14 could be a potential lead molecule for targeted and combination therapies. Certain pathways, such as the RAS signaling pathway, are altered across many different tumor types [37], and the inhibition of compound 14 against RAS (moderate activity with IC 50 value of 6.5 µM) indicates that it could be explored further for several cancers. The cross-talk between the other pathways in response to compound 14 treatment (Stat 3, AP-1, NFkB, E2F, and ETS with IC 50 in the range of 4 µm to 5 µm) reflect functional interactions and dependencies. Overall, as Yu et al. [38] reported in 2022, molecular targeted therapies play a key role in the treatment of various cancers, and our results could be the starting point for the further development of compound 14.

General Experimental Procedures
Optical rotation was determined by an AUTOPOLVR IV polarimeter. The 1H-and 13C NMR spectra were recorded on a Bruker Avance 400 or 500 MHz spectrometers. HMBC, HSQC, and ROESY were measured on an Agilent DD2-500 NMR spectrometer.
The ESI-HRMS data was obtained by utilizing Agilent 6545 LC/Q-ToF and Model 6230 ToF (controlled by Agilent MassHunter Work Station, A.08.00) systems. All acquisitions were performed under a positive and negative ionization mode with a capillary voltage of 3000 V. Nitrogen was used as the nebulizer gas (25 psig) as well as the drying gas at 7 L/min at a drying gas temperature of 325 • C. Other parameters included sheath gas temperature, 300 • C; sheath gas flow, 7 L/min; skimmer, 65 V; Oct RF V, 750 V; and fragmentor voltage, 150 V. Ten microliters of sample were injected. Full scan mass spectra were acquired from m/z 100-1100. Data acquisition and processing was done using the MassHunter Workstation software (Qualitative Analysis Version B.10.00, Santa Clara, CA, USA). Accurate mass measurements of each spectrum from the data collected were obtained by means of reference ion correction using reference masses at m/z 121.0509 (protonated purine) and 922.0098 [protonated hexakis (1H, 1H, 3H-tetrafluoropropoxy) phosphazine or HP-921] in positive ion mode and at m/z 112.9856 (deprotonated trifluoroacetic acid-TFA) and 1033.9881 (TFA adducted HP-921) in negative ion mode. LC/MS data was measured using an Agilent 1290 Infinity series UHPLC instrument, coupled to an Agilent 6120 quadrupole mass spectrometer with a dual ESI and APCI interface. HPLC analysis was conducted on a Agilent technologies 1100 series with a diode array detector with semi preparative RP-HPLC (column: Luna C18(2) 10 µ, 250 × 10 mm; detector; UV 254 nm), using MeCN-H 2 O as the solvent. TLC analysis was carried out using analytical silica gel 60 PF254 and pre-coated alumina plates (Merck, Rahway, NJ, USA; 0.25 mm thick). The plates were visualized under a UV lamp (254 nm) and sprayed with an anisaldehyde reagent, followed by heating.

Synthesis of Compound 17
Compound 5 (45 mg, 0.12 mmoL) in ethyl acetate was hydrogenated with Adam's catalyst, PtO 2 (10%, 4.5 mg) at 40 psi at RT for 24 h, as previously published [40]. The catalyst was removed by filtration and the filtrate was evaporated to obtain a complete hydrogenated product (42 mg).

Antimicrobial Assay
Antimicrobial assays were carried out using a published method [10]. All organisms were obtained from the American Type Culture Collection (Manassas, VA, USA). These organisms included Candida albicans ATCC 90028, Cryptococcus neoformans ATCC 90113, Aspergillus fumigatus ATCC 204305, methicillin-resistant Staphylococcus aureus ATCC 1708 (MRSA), Escherichia coli ATCC 2452, Pseudomonas aeruginosa ATCC BAA-2018, Klebsiella pneumonia ATCC 2146, and vancomycin resistant Enterococcus faecium (VRE) ATCC 700221. Briefly, the antimicrobial activity was determined through a high throughput screening assay performed in a 384-well plate. Appropriate drug controls for bacteria and fungi were included in each assay. The concentration of compound/fraction responsible for 50% growth inhibition (IC 50 ) was calculated using XLfit 4.2 software (IDBS, Alameda, CA), with fit model 201. Minimum inhibitory concentration (MIC) was defined as the lowest test concentration that afforded no visual growth. Susceptibility testing was performed using a modified version of the CLSI (formerly NCCLS) method [41,42].

Antimicrobial Combination Study by Checkerboard Method
The combination study of the compounds was carried out in MRSA using the standard checkerboard method by Norden et al. 1979 [43]. Test samples were dissolved in 100% DMSO to the desired concentrations, and serially diluted (1:2) with 20% DMSO/saline. For checkerboard, compound 5, 17 and machaeriol C (i.e., previously isolated from Machaerium sp. [8,9]) were tested in the range of 2.5 µg/mL to 0.039 µg/mL. Inocula was prepared in cation-adjusted Mueller-Hinton broth to afford 5 × 10 5 colony forming units per mL. Samples were transferred to 96-well assay plates (10 µL) in a checkerboard layout followed by inocula (180 µL

Bacterial Cultures for In Vivo Experiment
MRSA USA300 (kindly provided by Jorge Vidal, University of Mississippi Medical Center, Jackson, MS, USA) was isolated on tryptic soy agar (TSA), then grown in culture for 16 h at 37 • C with shaking in tryptic soy broth (TSB). The bacteria were then diluted 100-fold in sterile TSB and grown to mid-logarithmic phase in tryptic soy broth (TSB). Serial dilutions were prepared in sterile PBS to achieve a target inoculum of 10 7 colony-forming units (CFU) per 0.03 mL.

In Vivo Anti-MRSA Nosocomial Assay in Murine Model
Six-to eight-week-old female C57BL/6J mice (the Jackson Laboratory, Bar Harbor, Maine) were anesthetized with a mixture of ketamine and xylazine and weighed. Each anesthetized mouse received a target inoculum of 10 7 CFU, as performed by Achouiti et al. [46]. The serial dilution and plating of the inoculum were carried out to verify the purity and accuracy of the dose. The actual dose was determined to be 2 × 10 7 CFU. Twelve hours after infection, the mice were anesthetized and administered 0.048 mL of vehicle ciprofloxacin (30 mg/kg), compound 5 (30 mg/kg), or compound 17 (30 mg/kg). The vehicle composition, which was also used for suspension of the compounds and dilution of ciprofloxacin, was ethanol:DMSO:Cremophor EL:PBS (5:5:10:80 v/v).
Twenty-four hours after infection, the mice were anesthetized, then euthanized with an overdose of sodium pentobarbital. Using a midline incision, each trachea was exposed and another incision was made partially through the middle of the trachea, being careful not to completely sever the trachea all the way through. Nasopharyngeal lavage was performed by placing a pipette into the tracheal incision, pushing 0.05 mL of sterile PBS through the trachea 2 times and into a 1.5 mL tube placed under the nose. The lavage fluid was then diluted and plated on TSA for bacterial load quantitation. Next, the whole lung was removed and homogenized before being plated and counted to determine bacterial loads.

Transcriptional Reporter Assays
T98G glioblastoma multiforme cells from ATCC were plated in white opaque 384well plates at a density of 4300 cells/well in 30 µL of growth medium (DMEM with 10% FBS and 1% Pen/Strep). On the next day, the medium was aspirated and replaced with DMEM containing 10% FBS. The cells were transfected with respective plasmids using X-tremeGENE HP transfection reagent (Roche). The luciferase vectors used in this assay are summarized in (Table S1, Supplementary Materials). After 24 h of transfection, the test agents were added to the transfected cells, followed 30 min later by an inducing agent (IL-6 for Stat 3, TGF-β for Smad, m-wnt3a for Wnt and PMA for AP-1, NFkB, E2F, Myc, ETS and Hedgehog). No inducer was added for FoxO, miR-21, Ras, AhR and pTK vector control. After 4 h or 6 h of induction, the cells were lysed by the addition of a One-Glo luciferase assay system (Promega, Madison, WI, USA). The light output was detected in a Glomax Multi+ detection system with Instinct Software (Promega). This luciferase assay determined whether the test agent was able to inhibit the activation of the cancer-related signaling pathways. In the case of FoxO, mi-R21-, Ras-, and AhR-enhanced luciferase activity by the test agents was assessed [47].