Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 May 31.
Published in final edited form as: ACS Comb Sci. 2020 Feb 17;22(3):150–155. doi: 10.1021/acscombsci.9b00186

Diversity-Oriented Library Synthesis from Steviol and Isosteviol-Derived Scaffolds

Trinh A D Holth 1, Michael A Walters 1, Oliver E Hutt 1, Gunda I Georg 1
PMCID: PMC13222091  NIHMSID: NIHMS2174770  PMID: 32065745

Abstract

The readily available natural product stevioside provides a unique diterpene core structure that can be explored for small molecule library development by diversity-oriented synthesis and functional group transformations. Validation arrays were prepared from steviol, isosteviol and related analogs, derived from stevioside, to produce over 90 compounds. These compounds were submitted to the NIH Molecular Libraries Small Molecule Repository for screening in the Molecular Libraries Screening Center Network. Micromolar hits were identified in multiple high throughput assays for several library members. A cheminformatics analysis of the compounds was performed that verified the expected diversity and complexity of this set of compounds. The screening results indicate that scaffolds derived natural products can provide screening hits against multiple target proteins.

Keywords: steviol, isosteviol, DOS, analogs, library synthesis

INTRODUCTION

Diversity-oriented synthesis (DOS) is an efficient approach to design and develop multiple small molecules that populate structurally diverse areas of chemical space. First introduced by Schreiber in early 2000,1 this method provides rapid access to structurally different and complex compound libraries for biological screening. In contrast to target-oriented synthesis (TOS), which focuses on the synthesis of closely related structural analogs (focused libraries), DOS explores a broad area of chemical space that features compounds with different skeletal and stereochemical moieties.2 Through DOS it is possible to create libraries of compounds from a single building block (scaffold) or collections of compounds from structurally similar scaffolds.

Natural products can serve as excellent scaffolds for chemical library development because of their inherent structural diversity and complexity, as compared to synthetic products.35 In addition, natural products have been shown to occupy areas of chemical space that are generally not well covered by synthetic compounds.67 Furthermore, natural products, their analogs, and natural product-inspired bioactive compounds have been sources of new drugs over many decades.8 Given the continuous research and effectiveness of exploring natural products for drug discovery and development, our group has investigated the library development of stevioside-derived scaffolds for biological evaluation. Our libraries were prepared under the Pilot Scales Libraries Initiative to synthesize and submit compounds to became part of the NIH Molecular Libraries Small Molecule Repository (MLSMR) and to be evaluated in the NIH-funded Molecular Libraries Screening Center Network centers.9

Stevioside is the primary natural product isolated from the Stevia rebaudiana Bertoni plant, indigenous to South America (Scheme 1).1011 Unlike many natural products, stevioside, which is used as a sweetener, is readily available in kilogram quantities and therefore ideally suited for the development of natural product-derived combinatorial libraries. Library synthesis is facilitated because stevioside chemistry has been explored extensively and chemical transformations leading to different types of scaffolds are well known.1214 Since stevioside, its analogs,1516 and related natural products such as gibberellic acid derivatives possess a variety of biological activities, we hypothesized that stevioside-derived libraries would provide promising opportunities to generate hits in screening campaigns.1720 The structure of stevioside contains an ent-kaurene core attached to three glucose molecules (Scheme 1). Upon enzymatic hydrolysis of the glycosidic bonds, steviol (1) is formed. Its structural analog isosteviol (2), which is stereochemically quite different, can be obtained through an acid-catalyzed Wagner-Meerwein rearrangement either from stevioside or steviol (Scheme 1).21 Both compounds provide stereochemically complex templates for library synthesis and can be converted to other diverse scaffolds in a few synthetic steps (Figure 1).12, 14, 22

Scheme 1.

Scheme 1.

Synthesis of steviol and isosteviol from stevioside

Figure 1.

Figure 1.

DOS approach to the ent-kaurene scaffold of steviol by D-ring modifications.

RESULTS AND DISCUSSION

Figure 1 illustrates several approaches that can be taken. For path II, the Wagner-Meerwein rearrangement, as discussed above, is achieved with hydrochloric acid to yield isosteviol. From isosteviol (2), several different scaffolds can be generated after one or two synthetic steps. Through route III, a Baeyer-Villiger oxidation of isosteviol (2) produces lactone 4.23 Formation of an oxime and subsequent Beckmann rearrangement (path IV) yields lactam 5,12 and photo-Beckman rearrangement leads to the corresponding amide regioisomer 6 (path V).12 As shown for path VI, the Beckman fragmentation product 7 can also be obtained depending on reaction conditions.12 Optimized protocols for isosteviol oxime formation, Beckman rearrangement and Beckmann fragmentation have been reported by us.12 In addition to forming novel scaffolds, the functional groups in the molecules can be easily modified for library synthesis.14 Path I involves an allylic oxidation of the C17 methylene of steviol to furnish diol 3a, which can be further oxidized to the C15 ketone analog 3b.24 Reductive amination of isosteviol (2) via route VII yields amine 8.2425 Other obvious modifications involve the carboxylic acid moiety, which can be reduced to the corresponding alcohol or a methyl group.2627 Curtius rearrangement of the C4 acyl azide leads to the formation of a 4-amine analog.16 The double bond of steviol can be transformed to an epoxide,28 which then can be transformed to a diol,28 and ozonolysis of the double bond provides the corresponding ketone.14

In this report, we focus on steviol (1) and D ring modified compounds 2, 5, 7, and 8 as the scaffolds for library development. We describe the synthesis of several pilot scale libraries, their evaluation in multiple high throughput screens and an in silico analysis of the chemical and drug-like properties of the compounds prepared in this study.

RESULTS AND DISCUSSION

Hydrolysis of stevioside under well-known reaction conditions provided steviol (1) and isosteviol (2) in good yields (Scheme 1).2930 These two molecules served as scaffolds for library development with reactions aimed at transforming the carboxylic acid moiety on ring A in both compounds to amides. We set up validation arrays for the steviol and isosteviol scaffolds for amide bond formation utilizing both solution and solid phase chemistry (only the isosteviol validation array is shown in Scheme 2).

Scheme 2.

Scheme 2.

Validation array for amide formation from isosteviol 2 by solution phase (A) and solid phase synthesis (B)

These reactions were carried out to compare product yields, ease of work-up, and purity of products. For the solution phase chemistry, isosteviol was converted to the corresponding acid chloride by reaction with oxalyl chloride and then reacted with amines 9{1-4} to furnish amides 10{1-4}. For the second array, we used polymer-bound 1-hydroxybenzotriazole (HOBt), a reagent that reacts with the acid moiety to form a reactive ester intermediate. Isosteviol was loaded onto the polymer using the coupling reagent EDCI in the presence of DMAP. The polymer-bound activated ester was purified by several washes and then reacted with amines 9{1-4} to form the corresponding amides 10{1-4}. Use of the amine as the limiting reagent in this reaction assures high purity of the resulting amide reaction products. After careful analysis and comparison of both the solution and solid phase approaches, we decided to carry out polymer-supported parallel synthesis to construct the library. The resin-bound synthesis approach allowed for a simple filtration to obtain the final products in high purity. Using parallel solid phase synthesis, we then prepared a small library of approximately 60 amides from steviol, isosteviol, lactam acid 12, nitrile acid 14, and amine acid 16 with amines 9 (Scheme 3). A selection of diverse primary amines was used in the library production.

Scheme 3.

Scheme 3.

Amide library formation by combination of 2, 1, 12, 14, and 16 with amines 9

After modifying the carboxylic acid moiety on the A ring, modifications of the C and D rings were explored through alkylation of the hydroxyl group of steviol, reduction of the ketone of isosteviol, and alkylation of the D-ring lactam of scaffold 5 (Scheme 4). We converted the acid functionalities of scaffolds 1, 2, 5, and 8 to the corresponding methyl esters. Reacting the hydroxyl group of the steviol methyl ester with alkylating agents 18{1-9} provided ethers 19{1-9}. The C16 ketone group of the isosteviol methyl ester was reduced with sodium borohydride to yield the corresponding secondary alcohol, which then was alkylated with 13{1-6} to provide ethers 20{1-6}.24 N-Alkylated lactams 21{1-9} were obtained by reaction of lactam 5 with 13{1-9}, while N-alkylated amines 22{1-5} were generated from reacting amine 8 with 13{1-5}. The alkylations were carried out with sodium hydride, because a strong base was necessary to facilitate the removal of the hydroxyl proton for ether formation and the amide proton for lactam alkylation. These routes created a 33-membered library of O- and N-alkylated reaction products.

Scheme 4.

Scheme 4.

Synthesis of an O- and N-alkyl library

Bifunctional libraries can be prepared using alkylated products 19-22 to synthesize amides. Scheme 5 shows a representative example of one such library. Reaction of N-methyl amide analog 21{1} with amines 9{1-4} by first hydrolyzing the ester and then standard amide formation from the corresponding acid chlorides yielded amides 23{1-4}.

Scheme 5.

Scheme 5.

Synthesis of a bifunctional library from amide 21{1}

All compounds were submitted to the NIH’s Molecular Libraries Small Molecule Repository (MLSMR).9 All submitted compounds were analyzed for identity and purity (≥90%) by the MLSMR. Supplemental Information Table S1 provides a listing of the 22 compounds that were selected for screening by the NIH MLSMR. These compounds were evaluated in 56 assays by the Molecular Libraries Screening Center Network. Several library members, including previously prepared compounds 21{1} and 24-27 (Table 1),12 displayed single or double digit micromolar inhibition of target proteins and are shown in Table 1, which shows a representative sample of current PubChem data on our submitted compounds that have been tested by the MLSMR.3132 A table of all currently-tested compounds is available in the supporting information section that only shows results from active primary screens or confirmatory assays; inconclusive or inactive results, which make up a majority of the data, are not listed (Table S1). Some assays in which steviol and isosteviol and their corresponding libraries showed activity include target proteins neuropilin-1 (NRP-1); tyrosyl-DNA phosphodiesterase 1 (TDP1); NACHT, leucine-rich repeat, Pyrin domain-containing-3 (NALP3); hepatitis C virus (HCV); glucagon-like peptide-1 (GLP1); and lipid storage modulators. NRP-1 is a protein receptor for vascular endothelial growth factor, thus playing a role in angiogenesis and is being studied for its role in tumor growth progression for anticancer research.3334 The TDP1 enzyme is also being investigated for anticancer development due to its mechanism of DNA repair and treatment for spinocerebellar ataxia, which is caused by a TDP1 mutation.35 Furthermore, assays also identified NALP3, a protein involved in the inflammatory process of the human innate immune system.36 Primary screens for inhibitors of the NALP3 protein can provide potential starting points for anti-inflammatory therapeutic development.37 Compound 21{3} exhibited low micromolar activity in a confirmatory inhibitor assay for HCV (Table 1). Recently, Lin et al. created 33 ureido- and amide-substituted steviol analogs, similar to amide product 11 of steviol, to study their inhibitory effects against the Hepatitis B Virus (HBV), which similar to HCV infections can lead to liver cirrhosis and cancer.16 Their work demonstrates the importance of using natural products and their analogs as sources for therapeutic research as well as implies the potential for isosteviol analogs to be further investigated in HBV drug development. Finally, both GLP1 and lipid storage modulators have been sought in diabetes and anti-obesity research.3839

Table 1.

Activity of Steviol and Isosteviol Analogs

graphic file with name nihms-2174770-t0003.jpg
graphic file with name nihms-2174770-t0004.jpg
a

IC50 = concentration of an inhibitor required for 50% inhibition of maximum control response;

b

AbsAC = Absolute active concentration with compounds below 10 μM to be considered active hits;

c

AC50 = concentration required to elicit a 50% response in an in vitro assay;

d

EC50 = concentration of an agonist required to produce 50% maximum (effective) response

A PMI (principal moment of inertia) plot shows that this library of compounds is both drug-like and diverse (Figure 2). This type of plot places rod shaped molecules (e.g., acetylene) at the upper left corner (A), pancake shaped molecules (e.g. benzene) at the lower apex (B), and spherical molecules (e.g. adamantane) in the upper right corner (C). As described previously calculated PMI for most drugs tend to congregate along the A-B axis from rod to pancake shape.40 This plot demonstrates that our library lies in relevant three-dimensional space near this axis and that the library features shape diversity relative to commercially available lead-finding libraries (e.g. Maybridge HitFinder Collection; 14,400 compounds selected to represent the overall diversity of the screening collection, maybridge.com).

Figure 2.

Figure 2.

PMI plot of stevioside library overlayed with a commercial “diversity” library demonstrating that the stevioside library is both drug-like and diverse.

More detailed comparisons of these stevioside-derived compounds with commercial libraries (cLogP, Fsp3, number of stereocenters, and pairwise similarity) confirm the uniqueness of this library (see Supporting Information).

CONCLUSION

We have developed a small library of compounds using a diversity-oriented synthesis approach based on the steviol and isosteviol scaffolds. Testing of the libraries provided screening hits in several assays against diverse protein targets. We analyzed the compounds using an in silico technique to determine their complexity and structural diversity relative to two commercial libraries. This analysis supported our hypothesis that the structures of the compounds we prepared are complex and diverse. Further analysis showed that our libraries are drug-like.

Supplementary Material

SI final

The Supporting Information is available free of charge on the ACS Publications website at DOI:

Calculations of drug like properties of compounds, experimental procedures and compound characterization data of 20 library members that were tested, including full 1H and 13C NMR spectra. This material is available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGMENTS

We gratefully acknowledge financial support from the National Institutes of Health (GM081267) and the University of Minnesota through the Vince and McKnight Endowed Chairs.

Footnotes

The authors declare no competing financial interests

REFERENCES

  • 1.Schreiber SL Target-oriented and diversity-oriented organic synthesis in drug discovery. Science 2000, 287, 1964–1969. [DOI] [PubMed] [Google Scholar]
  • 2.Burke MD; Schreiber SL A planning strategy for diversity-oriented synthesis. Angew. Chem. Int. Ed. Engl 2004, 43, 46–58. [DOI] [PubMed] [Google Scholar]
  • 3.Tan DS Diversity-oriented synthesis: exploring the intersections between chemistry and biology. Nat. Chem. Biol 2005, 1, 74–84. [DOI] [PubMed] [Google Scholar]
  • 4.Yi S; Varun BV; Choi Y; Park SB A brief overview of two major strategies in diversity-oriented synthesis: Build/couple/pair and ring-distortion. Front. Chem 2018, 6, 507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Gu Y; Tan C; Gong J; Yang Z Diversity-oriented synthesis of natural products via gold-catalyzed cascade reactions. Synlett 2018, 29, 1552–1571. [Google Scholar]
  • 6.Lachance H; Wetzel S; Kumar K; Waldmann H Charting, navigating, and populating natural product chemical space for drug discovery. J. Med. Chem 2012, 55, 5989–6001. [DOI] [PubMed] [Google Scholar]
  • 7.Rosen J; Gottfries J; Muresan S; Backlund A; Oprea TI Novel chemical space exploration via natural products. J. Med. Chem 2009, 52, 1953–1962. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Newman DJ; Cragg GM Natural products as sources of new drugs from 1981 to 2014. J. Nat. Prod 2016, 79, 629–661. [DOI] [PubMed] [Google Scholar]
  • 9.Austin CP; Brady LS; Insel TR; Collins FS NIH Molecular Libraries Initiative. Science 2004, 306, 1138–1139. [DOI] [PubMed] [Google Scholar]
  • 10.Shibata H; Sawa Y; Oka T; Sonoke S; Kim KK; Yoshioka M Steviol and steviol-glycoside: glucosyltransferase activities in Stevia rebaudiana Bertoni--purification and partial characterization. Arch. Biochem. Biophys 1995, 321, 390–396. [DOI] [PubMed] [Google Scholar]
  • 11.Geuns JM Stevioside. Phytochemistry 2003, 64, 913–921. [DOI] [PubMed] [Google Scholar]
  • 12.Hutt OE; Doan TL; Georg GI Synthesis of skeletally diverse and stereochemically complex library templates derived from isosteviol and steviol. Org. Lett 2013, 15, 1602–1605. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Wu Y; Yang JH; Dai GF; Liu CJ; Tian GQ; Ma WY; Tao JC Stereoselective synthesis of bioactive isosteviol derivatives as alpha-glucosidase inhibitors. Bioorg. Med. Chem 2009, 17, 1464–1473. [DOI] [PubMed] [Google Scholar]
  • 14.Moons N; De Borggraeve W; Dehaen W Stevioside and steviol as starting materials in organic synthesis. Curr. Org. Chem 2012, 16, 1986–1995. [Google Scholar]
  • 15.Lin Z; Guo Y; Gao Y; Wang S; Wang X; Xie Z; Niu H; Chang W; Liu L; Yuan H; Lou H ent-Kaurane diterpenoids from chinese liverworts and their antitumor activities through michael addition as detected in situ by a fluorescence probe. J. Med. Chem 2015, 58, 3944–56. [DOI] [PubMed] [Google Scholar]
  • 16.Lin SJ; Su TC; Chu CN; Chang YC; Yang LM; Kuo YC; Huang TJ Synthesis of C-4-substituted steviol derivatives and their inhibitory effects against hepatitis B virus. J. Nat. Prod 2016, 79, 3057–3064. [DOI] [PubMed] [Google Scholar]
  • 17.Brahmachari G; Mandal LC; Roy R; Mondal S; Brahmachari AK Stevioside and related compounds - molecules of pharmaceutical promise: a critical overview. Arch. Pharm. Chem. Life Sci 2011, 344, 5–19. [DOI] [PubMed] [Google Scholar]
  • 18.Zhang Y; Zhang H; Chen J; Zhao H; Zeng X; Zhang H; Qing C Antitumor and antiangiogenic effects of GA-13315, a gibberellin derivative. Invest. New Drugs 2012, 30, 8–16. [DOI] [PubMed] [Google Scholar]
  • 19.Zhou G-B; Kang H; Wang L; Gao L; Liu P; Xie J; Zhang F-X; Weng X-Q; Shen Z-X; Chen J; Gu L-J; Yan M; Zhang D-E; Chen S-J; Wang Z-Y; Chen Z Oridonin, a diterpenoid extracted from medicinal herbs, targets AML1-ETO fusion protein and shows potent antitumor activity with low adverse effects on t(8;21) leukemia in vitro and in vivo. Blood 2007, 109, 3441–3450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Mokarram P; Mohammadi Z; Khazayel S; Dayong Z Induction of epigenetic alteration by CPUK02, an ent- kaurenoid derivative of stevioside. Avicenna J. Med. Biotechnol 2017, 9, 13–18. [PMC free article] [PubMed] [Google Scholar]
  • 21.Holth TA; Hutt OE; Georg GI, Beckmann rearrangements and fragmentations in organic synthesis. In Molecular Rearrangements in Organic Synthesis, Rojas CM, Ed. Wiley: 2015; pp 111–150. [Google Scholar]
  • 22.Roy A; Roberts FG; Wilderman PR; Zhou K; Peters RJ; Coates RM 16-Aza-ent-beyerane and 16-Aza-ent-trachylobane: potent mechanism-based inhibitors of recombinant ent-kaurene synthase from Arabidopsis thaliana. J. Am. Chem. Soc 2007, 129, 12453–12460. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Coates RM; Bertram EF Structural modifications of isosteviol. Partial synthesis of atiserene and isoatiserene. J. Org. Chem 1971, 36, 2625–2631. [Google Scholar]
  • 24.Wu Y; Dai GF; Yang JH; Zhang YX; Zhu Y; Tao JC Stereoselective synthesis of 15- and 16-substituted isosteviol derivatives and their cytotoxic activities. Bioorg. Med. Chem. Lett 2009, 19, 1818–1821. [DOI] [PubMed] [Google Scholar]
  • 25.Ma C; Lazo JS; Xie XQ Compound acquisition and prioritization algorithm for constructing structurally diverse compound libraries. ACS Comb. Sci 2011, 13, 223–231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Coates RM; Kang HY Synthesis and evaluation of cyclobutylcarbinyl derivatives as potential intermediates in diterpene biosynthesis. The Journal of Organic Chemistry 1987, 52, 2065–2074. [Google Scholar]
  • 27.Paradine SM; Griffin JR; Zhao J; Petronico AL; Miller SM; White CM A manganese catalyst for highly reactive yet chemoselective intramolecular C(sp3)–H amination. Nat. Chem 2015, 7, 987–994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Khaibullin RNS, Yu I.; Kataev VE; Lodochnikova OA; Gubaidullin AT; Balandina AA; Latypov SK Wagner-Meerwein rearrangement of steviol 16α,17- and 15α,16-epoxides. Russ. J. Org. Chem 2010, 46, 1006–1012. [Google Scholar]
  • 29.Pezzuto JM; Compadre CM; Swanson SM; Nanayakkara D; Kinghorn AD Metabolically activated steviol, the aglycone of stevioside, is mutagenic. Proc. Natl. Acad. Sci .U S A 1985, 82, 2478–2482. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Avent AG; Hanson JR; De Oliviera BH Hydrolysis of the diterpenoid glycoside, stevioside. Phytochemistry 1990, 29, 2712–2715. [Google Scholar]
  • 31.Kim S; Thiessen PA; Bolton EE; Chen J; Fu G; Gindulyte A; Han L; He J; He S; Shoemaker BA; Wang J; Yu B; Zhang J; Bryant SH PubChem substance and compound databases. Nucleic Acids Res. 2016, 44, D1202–1213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Wang Y; Suzek T; Zhang J; Wang J; He S; Cheng T; Shoemaker BA; Gindulyte A; Bryant SH PubChem BioAssay: 2014 update. Nucleic Acids Res. 2014, 42, D1075–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Chaudhary B; Khaled YS; Ammori BJ; Elkord E Neuropilin 1: function and therapeutic potential in cancer. Cancer Immunol. Immunother 2014, 63, 81–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Ellis LM The role of neuropilins in cancer. Mol. Cancer Ther 2006, 5, 1099–1107. [DOI] [PubMed] [Google Scholar]
  • 35.Comeaux EQ; van Waardenburg RC Tyrosyl-DNA phosphodiesterase I resolves both naturally and chemically induced DNA adducts and its potential as a therapeutic target. Drug Metab. Rev 2014, 46, 494–507. [DOI] [PubMed] [Google Scholar]
  • 36.Campbell L; Raheem I; Malemud CJ; Askari AD The Relationship between NALP3 and Autoinflammatory Syndromes. Int. J. Mol. Sci 2016, 17, 1–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Leung YY; Yao Hui LL; Kraus VB Colchicine--Update on mechanisms of action and therapeutic uses. Semin. Arthritis Rheum 2015, 45, 341–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Morris A Obesity: 5-HT2A in GLP1-mediated weight loss. Nat. Rev. Endocrinol 2017, 13, 127. [DOI] [PubMed] [Google Scholar]
  • 39.Beller M; Thomas C; Shen M; Auld D, Identification of lipid storage modulators. In Probe Reports from the NIH Molecular Libraries Program, Bethesda (MD), 2010. [Google Scholar]
  • 40.Akritopoulou-Zanze I; Metz JT; Djuric SW Topography-biased compound library design: the shape of things to come? Drug Discov. Today 2007, 12, 948–952. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SI final

RESOURCES