Improved Library of Pyrrolidine-2,3-dione Monomers as Antimicrobial and Antibiofilm Agents against Staphylococcus aureus

Abstract

The opportunistic nature of S. aureus, along with its tendency to develop resistance to antibiotics and ability to form biofilms readily, makes it a challenging pathogen to treat. This underscores the need for novel, evolving therapeutics. We report a new generation of pyrrolidine-2,3-dione monomers, accessed via a multicomponent reaction, with exciting antimicrobial and antibiofilm activities. The presence of a para-trifluoromethyl substituent on the scaffold’s N-aryl substituent improved the biological profile of these analogs relative to previous generations.

Staphylococcus aureus is a Gram-positive bacterium that is multidrug-resistant and has been deemed by the Centers for Disease Control and Prevention (CDC) as a serious pathogen that necessitates continuous innovation for effective treatment. , Due to its high gene plasticity, environmental adaptability, and production of a high number of virulence factors, S. aureus is particularly capable of developing antimicrobial resistance (AMR). − A case study conducted in a hospital in Vietnam from 2014 to 2021 highlighted the preponderance and rising trend of AMR in S. aureus, particularly against macrolide antibiotics such as erythromycin. Moreover, it also affects livestock and poultry and continues to be proficient in developing AMR. Methicillin, a β-lactam antibiotic, was discovered in 1960, and methicillin-resistant S. aureus (MRSA) emerged soon after, which posed significant challenges for treatment. In fact, the CDC report from 2019 recorded 10 000 deaths that could be attributed to MRSA, while the global estimation is greater than 100 000. ,

In addition to acquiring resistance through genetic mutations and the horizontal transfer of resistance genes, S. aureus can develop antibiotic tolerance by living in a biofilm. , Biofilms are surface-associated communities of bacteria encased in an extracellular matrix of polysaccharides, proteins, and DNA. Biofilms can form on biotic or abiotic surfaces, and the protection afforded to biofilm-associated bacteria due to the matrix and development of distinct phenotypic characteristics are important reasons why they are more challenging to treat compared to planktonic bacteria. , While the best method to treat biofilms is to prevent their formation in the first place, this is rarely feasible because S. aureus readily develops resistance. Other strategies to combat biofilms include physical removal via surgical methods, prevention of surface attachment using coatings, and treatment of the biofilm phenotype with small molecules or enzymes. ,

The pyrrolidine-2,3-dione scaffold was first observed in natural products in 1995 when phenopyrrozin (1, Figure ) was isolated from Penicillium sp. Similar natural products were discovered in the forthcoming years, but it was only in 2013 that any biological activity was reported for these compounds with low growth inhibition (IC₅₀ = 50–100 μg/mL) against a Gram-positive bacterium, Bacillus cereus. , In 2016, Dhavan and co-workers published the total synthesis of leopolic acid A (2, Figure ), and more recently our group published one. We learned that although leopolic acid A and its analogs show only moderate-to-low antimicrobial activity, modification of the 5-position of the scaffold to incorporate small-sized alkyl chains resulted in noteworthy antibiofilm activity.

1.

Representative examples of natural products containing the pyrrolidine-2,3-dione scaffold.

We have also done extensive work on unnatural pyrrolidine-2,3-diones − and determined that it is beneficial to have an unsubstituted enol moiety on the scaffold and a p-trifluoromethyl phenyl moiety at the 4-position. A common issue with pyrrolidine-2,3-dione monomers is their low aqueous solubility, which makes it challenging to evaluate their biological profile. While the pyrrolidine-2,3-dione dimers and monomers bearing N-polyheteroatom (N, O, S) linkers addressed this issue, it was observed that the dimeric forms were, in general, more active than their corresponding monomers. Thus, we desired to tune the simple monomers to make them amenable to biological evaluation and improve upon their activity as well. We turned our attention to the 5-position of the scaffold for further diversification; previous data sets published by Cusumano and co-workers explored substitution on the 5-position by incorporating small alkyl groups and some aromatic substituents, but none that systematically altered the benzylic position. The use of phenylacetaldehyde as one of the building blocks in the multicomponent reaction (vide infra) would enable access to pyrrolidine-2,3-diones containing a benzylic methylene coming off the 5-position of the scaffold, whereas using an aldehyde such as 3a arising from a chiral pool starting material, i.e., (R)-mandelic acid (4a), would incorporate a polar, chiral center at the benzylic position. This could potentially ameliorate the solubility issues that burdened the analogs tested by Nie and co-workers and additionally install a moiety that is known to possess a variety of beneficial biological activities. −

We leveraged a multicomponent reaction (Figure A), first developed by Shymanska and co-workers, to provide rapid access to the pyrrolidine-2,3-dione scaffold. The three components for this reaction (ester, aldehyde, and amine) were used from commercial sources or synthesized beforehand (Figure B). The stereocenter desired at the benzylic position could be controlled by using the enantiomer of the chiral pool starting material (see the Supporting Information for complete procedures).

2.

(A) Preparation of p-CF₃ phenylpyruvic ester 5 pursuant to use in a multicomponent reaction to access the pyrrolidine-2,3-dione scaffold. (B) Synthesis of aldehyde building block 3a originating from (R)-mandelic acid (4a).

After testing analogs 6 and 7 (Figure A) against two methicillin-susceptible and one methicillin-resistant strain of S. aureus, it was interesting to note that the presence of the chiral hydroxyl group seemed to turn on antimicrobial activity (Figure B), with 6 showing MIC values of 4–8 μg/mL across the three strains. This pattern persisted for the halogens (8–13). We tested the para-bromo analog bearing inverse stereochemistry at the benzylic center (14) and observed reduced activity. We also observed that it was necessary for the hydroxyl group to be benzylic, as 15 was inactive.

3.

(A) Pyrrolidine-2,3-dione analogs synthesized and tested for biological activity; major diastereomer shown. (B) Results of antimicrobial and antibiofilm evaluation (tested as diastereomeric mixtures). Vancomycin was used as a positive control.

At this point, we turned to the Topliss Tree for guidelines on aromatic substitutions in pursuit of improving antimicrobial activity; while the meta-bromo substituted analog (16) was unremarkable, disubstituted 17 presented enhanced activity, and the best activity was displayed by 18 with MIC values of 2 μg/mL across all the S. aureus strains in our scope. These values represent the apogee of antimicrobial activity among standalone pyrrolidine-2,3-diones tested in our lab and are comparable to those of FDA-approved antibiotics such as linezolid. Curiously, the p-CF₃ substituent seemed to defy the trend displayed by the halogens, as 19 shows striking activity, albeit slightly reduced as compared to its benzylic hydroxyl counterpart 18. It is unknown why this deviation from the expected trend is observed with the p-CF₃ substituent. As expected, the analog possessing the inverse hydroxyl stereocenter (20) followed the trend of reduced activity, as did the analog when the alcohol was not benzylic (21). Fathalla and co-workers observed modest antimicrobial activity with a methylene linker between N1 and an aryl group, as was the case for us with analog 22. It did not, however, usurp 18 as our lead analog.

To further understand trends associated with modifications at the 5-position of the pyrrolidine-2,3-dione scaffold while keeping the moiety originating from 3 sacrosanct, we used various amines as one of the building blocks of the multicomponent reaction. From the analogs obtained from these reactions, we observed that an electronically neutral aromatic ring (23) is active compared to an aliphatic chain (24), however, if the substituent at the para position is electronically activating (25), it renders the compound inactive as well. For monosubstitutions, the more electronically deactivating the substituent, the better. Efforts to make ortho-substituted N-aryl substituents were unsuccessful, likely due to the role of steric hindrance in the multicomponent reaction. It is important to note that many of the compounds have been tested as mixtures of diastereomers (ranging from 1.7:1 to >10:1), with the major diastereomer shown in Figure A.

Although 18 was the most active analog in terms of antimicrobial activity, its antibiofilm activity, as reflected in its MBEC value of 64 μg/mL, was not the best-in-class. Analogs 19 and 21, which contained a second trifluoromethyl group, displayed MBEC values of 32 μg/mL, as did 10, albeit sans the second trifluoromethyl substituent. The latter two also showed a striking MBEC/MIC ratio of 4. While discernible trends are observed for the pyrrolidine-2,3-dione analogs mentioned herein with respect to their MIC values, the trend is unclear for their MBEC values. It is worth noting that FDA-approved antimicrobial therapeutics lack significant biofilm-eradication capability, resulting in high MBEC/MIC ratios. They would need to be used in combination with other antibiotics, else the working concentration for standalone antibiotics to eradicate biofilms would cause harmful side effects to the patient. This class of monomeric pyrrolidine-2,3-diones represents the first of its kind to possess both significant antimicrobial and antibiofilm properties; however, we have not evaluated these new scaffolds against Gram-negative pathogens or conducted studies to evaluate their cytotoxicity and selectivity index. Those studies will be part of future efforts on lead compounds from this work.

In summary, we have expanded the library of pyrrolidine-2,3-dione monomers, which now includes analogs with noteworthy antimicrobial activity. The most potent analog, compound 18, has remarkable MIC values of 2 μg/mL against the 3 S. aureus strains tested. For N-aryl substitutions on the pyrrolidine-2,3-dione, the more electronically deactivating the substituent at the para position, the better the antimicrobial activity. For halogens, the presence of a chiral hydroxyl group at the benzylic position of the 5-substitution turned on activity. The antibiofilm activity is less predictable, however, but compounds 19, 21, and 10 all have MBEC values of 32 μg/mL with 21 and 10 showing MBEC/MIC ratios of 4, which is uncommon in most FDA-approved antimicrobials.

Glossary

Abbreviations

AMR

antimicrobial resistance

CDC

Centers for Disease Control and Prevention

DNA

deoxyribonucleic acid

FDA

United States Food and Drug Administration

IC₅₀

half maximal inhibitory concentration

MBEC

minimum biofilm eradication concentration

MIC

minimum inhibitory concentration

MRSA

methicillin-resistant S. aureus

MSSA

methicillin-susceptible S. aureus

S. aureus

Staphylococcus aureus

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmedchemlett.6c00283.

• Detailed experimental procedures, spectroscopic data, and ¹H and ¹³C NMR spectra (PDF)

The manuscript was written through the contributions of all authors.

The authors declare no competing financial interest.