Three decades of selective product formation via Griesbaum co-ozonolysis: insight and advances (1995–2025)

Abstract

The Griesbaum co-ozonolysis reaction was first described by Karl Griesbaum and his collaborators in the mid-1990s. Their pioneering work laid the foundation for synthesis of tetrasubstituted ozonides by the reaction of O-methyl oximes with carbonyl compounds in the presence of ozone. It has emerged as a powerful and selective alternative to conventional ozonolysis. This review highlights the unique advantages of the reaction, including its high selectivity, operational simplicity, mild conditions, and consistently good yields. Emphasis is placed on the distinctive features that make Griesbaum co-ozonolysis a preferred approach for constructing complex molecular frameworks, i.e., di-, tri- and tetrasubstituted ozonides, mainly spiro and dispiro-1,2,4-trioxolanes, triterpenoids and synthetic drug candidates. The reaction tolerates basic conditions, allowing for controlled functionalization, and has been applied in post-ozonolysis transformation and chemical sensor development. Importantly, it eliminates the need for tetrasubstituted alkenes, broadening accessibility to complex ozonides. A comprehensive survey of the literature from the past three decades is presented, focusing on the diverse range of products synthesized using this method. The review underscores the growing importance and synthetic utility of Griesbaum co-ozonolysis in modern synthetic, organic and medicinal chemistry.

---

1. Introduction

The direct ozonolysis of tetrasubstituted olefins seldom affords the corresponding tetrasubstituted ozonides. The challenge arises for several reasons: first, the synthesis of suitably substituted olefins is itself nontrivial as steric hindrance around the double bond often hinders their preparation. Second, even when accessible, tetrasubstituted olefins display reduced reactivity toward ozone due to steric shielding of the π-bond. Third, ozonolysis of hindered olefins frequently leads to an unstable or short-lived carbonyl oxide intermediate, which undergoes decomposition, side reactions, or rearrangement, rather than productive cycloaddition to stable ozonides. These combined limitations indicate that direct ozonolysis rarely furnishes isolable tetrasubstituted ozonides in useful yields. This fact provides a major impetus for the development of Griesbaum co-ozonolysis, which bypasses the need for tetrasubstituted olefins altogether.

Described by K. Griesbaum and colleagues in 1995, Griesbaum co-ozonolysis is an attractive conversion in organic chemistry that enables the formation of tetrasubstituted ozonides (particularly 1,2,4-trioxolanes) with enhanced stability compared to conventional ozonolysis products. The reaction employs O-methyl oximes and carbonyl compounds as substrates, using ozone as the oxidizing agent, to produce 1,2,4-trioxolanes (Fig. 1).

Fig. 1 General scheme of the Griesbaum co-ozonolysis reaction.

An O-methyl oxime reacts with ozone to form a carbonyl oxide, which is proposed to undergo 1,3-dipolar cycloaddition with a ketone to form a cyclic stable ozonide. One mechanistic hypothesis suggests that the reaction may proceed via a pathway analogous to the Story–Murray–Youssefyen mechanism,¹ rather than exclusively through the conventional carbonyl oxide pathway. In this proposal, the carbonyl compound 2 is inserted into the peroxide bond of intermediate 5, leading to the formation of adduct 6. This intermediate can then eliminate methyl nitrite, resulting in the formation of ozonide 3, as illustrated by the electron flow arrows in structure 6. Although this route has been reported to occur only to a negligible extent in the ozonolysis of olefin,² it may be operative under specific conditions, such as during the ozonolysis of ortho-methyl oximes (Fig. 2). Nevertheless, the precise mechanism of Griesbaum co-ozonolysis remains not fully understood, and further studies are required to clarify the operative pathways.

Fig. 2 Mechanism of the Griesbaum co-ozonolysis reaction.

In 1995, K. Griesbaum et al.^3,4 reported a distinctive type of ozonolysis that has since become known as Griesbaum co-ozonolysis. This process involves the ozonolysis of ortho-methyl oximes in the presence of substituted acyclic or cyclic carbonyl compounds to selectively form ozonides. Compared to conventional olefin ozonolysis, this method offers several significant advantages, as follows. With the elimination of the need for olefin precursors, it bypasses the synthesis of parent olefin entirely. With improved selectivity, the preparation of unsymmetrically substituted ozonides proceeds without the formation of undesired byproduct ozonides, which are typically encountered in the ozonolysis of unsymmetrical olefins. With the dual synthetic pathway (routes A and B), this flexibility enables optimization of the ozonide yields by selecting a pair of substrates wherein the more reactive carbonyl compound drives the reaction efficiently.⁵ With the access to otherwise unattainable ozonides, in certain cases, this co-ozonolysis method is the only viable route to obtain specific ozonides. For example, the ozonide of tetramethylethylene (3) could not be synthesized via conventional olefin ozonolysis in solution, but was successfully obtained through this approach (Fig. 2).

Three distinct classes of reactants have been identified in Griesbaum co-ozonolysis, based on the structural nature and branching of the starting materials. The type I reaction involves both O-methyl ketoximes and substituted ketones 2 that are acyclic in nature (Fig. 2). The type II reaction utilizes cyclic ketoximes 7 in combination with acyclic ketones 10 (Fig. 3). The type III reaction involves the use of cyclic ketoximes 7 with cyclic ketones 9 of varying ring sizes, leading to formation of ozonide products 12 and 11, respectively (Fig. 3).

Fig. 3 Cyclic O-methyl ketoxime treated with cyclic and acyclic ketones.

A key intermediate 6 is generated during the Griesbaum co-ozonolysis process, which is not observed in conventional olefin ozonolysis. This mechanistic distinction accounts for the structural difference between the product obtained via Griesbaum co-ozonolysis and those formed from standard olefin ozonolysis. This unique feature underscores the synthetic utility of the reaction. In particular, this methodology applied to the synthesis of spirocyclic trioxolanes via the reaction of cyclic ketones with O-methyl oximes, where the newly generated spiro-center can lead to four possible diastereomers.

These stereoisomers can be selectively obtained under stereo selectivity or regioselective reaction conditions.

2. N-Methoxy-1,2,4-dioxazolidines via ozonolysis reaction

Griesbaum et al.⁶ reported the formation of dioxazolidines as a continuation of his previous work. The synthesis carried out by the ozonolysis of ethyl vinyl ether in the presence of the O-methylated oxime of cyclohexanone resulted in the formation of N-methoxy-1,2-dioxa-4-azaspiro[4.5]decane. Additionally, the ozonolysis of O-methylated dioximes derived from 1,4- and 1,5-dicarbonyl compounds afforded N-methoxylated bicyclic 1,2,4-dioxazolidines 13.

This study extends the established concept of carbonyl oxide–imine cycloaddition commonly used to generate 2,3-dioxazolidines to include the reaction involving unsubstituted carbonyl oxide and O-methylated oxime, forming N-methoxy-substituted dioxazolidine. Notably, substituted carbonyl oxide did not undergo this transformation. The absence of such products in the ozonolysis reaction of O-methylated oximes with ketones, unlike the reaction with alkyl or aryl imines, suggests that O-methylated oximes are significantly less dipolarophilic compared to ketones. Importantly, this work represents the first report of a one-step synthesis for bicyclic N-methoxy-substituted 1,2,4-dioxazolidines, with compounds 13 offering the highest isolated yield. These findings introduce the known examples of these unique heterocyclic scaffolds (Fig. 4).

Fig. 4 N-Methoxylated bicyclic-1,2,4-dioxazolidines 13.

3. Post-ozonolysis transformations

With these ozonides in hand, attention was then directed to their post-ozonolysis functionalization. Although the Griesbaum co-ozonolysis is a powerful method, the synthetic utility of functionalized tetra-substituted ozonides remain underexplored.^7,8 Nevertheless, several successful transformations of di-, tri-, and tetra-substituted ozonides have been reported. For example, ozonide ketones and aldehydes⁹ have been reduced to their corresponding alcohols using lithium or sodium borohydride. Ozonide alcohol has been oxidized to aldehyde via Swern oxidation and ozonide aldehydes converted to dimethyl or 1,3-dioxolane acetals,^10,11 which serve as intermediates for further C–C bond forming reactions. These include allylation with allylzinc bromide allyl silanes or allylstannanes in the presence of TiCl₄ or SnCl₄ and Mukaiyama-type aldol condensations (Fig. 5).

Fig. 5 Decomposition of ozonide.

R-Oxo-ozonides, where the carbonyl is at the R-position, can be synthesized via Griesbaum co-ozonolysis¹² and further transformed into O-methyl oximes or epoxides using O-methylhydroxylamine or diazomethane, respectively.¹³ R-Chloro ozonides undergo substitution with alcohol to afford ozonide ether, either in the presence¹⁴ or absence of AgBF₄. R-Acetoxy ozonides similarly react to alcohols under NaHCO₃-mediated conditions.

Unsaturated ozonides have also been derivatized¹⁵ by bromination of vinyl and vinylidene ozonides to yield dibromo derivatives. Diimide reduction of unsaturated bicyclic ozonides affords saturated analogs.¹⁶ Vinyl chloride ozonide undergoes methanolysis to form methyl ester ozonides, and the stepwise treatment with m-CPBA and AgBF₄ yields enone ozonides via epoxy intermediates.¹⁷

For post-ozonolysis transformations, reaction temperatures were maintained below 60 °C to prevent ozonide decomposition via Hock-type fragmentation into adamantane lactone and parent cyclohexanones (Fig. 6).¹⁸ Notably, these ozonides exhibit excellent stability under basic conditions and mild acidic treatments, including those for Boc deprotection. Additionally, reductive amination of ozonide aldehyde was feasible, with stability attributed to the absence of alpha hydrogen that otherwise promoted Kornblum–DeLaMare fragmentation.¹⁹

Fig. 6 Post-ozonolysis reduction of ozonide ester to ozonide alcohol.

The diastereoselectivity observed in the Griesbaum co-ozonolysis strongly suggests a preference for axial side attack of the carbonyl oxide,²⁰ governed primarily by the conformation rigidity of the cyclohexanone scaffold. While the electronic effect from the 4-substituent may contribute, the stereochemical outcomes appear largely controlled by kinetic factors, particularly transition state hyper-conjugation. This study also broadens the scope of ozonide-compatible transformations, including ester to alcohol reductions, reductive aminations, Mitsunobu reactions, nucleophilic substitution with mercaptides and azole anions, addition of organolithium and Grignard reagents, and phthalimide deprotections with hydrazine. While the reactivity of di-, tri-, and tetrasubstituted ozonides under these conditions has yet to be fully defined, the wide scope of post-ozonolysis transformations is reminiscent of those used in the derivatization of semisynthetic artemisinin analogs.^21,22 Trioxolanes 22, 23, 24, and 25 were synthesized via post-ozonolysis transformations of their corresponding trioxolane esters and phthalimide derivatives. The precursor ester and phthalimides were obtained through Griesbaum co-ozonolysis,²³ involving O-methyl oximes of 2-adamantanone 19 for compounds 22–24, and 4-tert-butyl cyclohexanone 20 for 25 with appropriate 4-substituted cyclohexanones 21 (Fig. 7).²⁴

Fig. 7 Synthesis of trioxolane by Griesbaum co-ozonolysis and post-ozonolysis transformation.

4. Anomalous ozonolysis product of 3β,28-di-O-acetyl-29-norlupan-20-one-O-methyloxime

In 2011,²⁵ a diacetylbetulin derivative 26 was employed to synthesize 3β,28-di-O-acetyl-29-norlupan-20-one-O-methyloxime 27 as a precursor to triterpene-1,2,4-trioxolane structures of interest for their potential anti-cancer and anti-malarial activity. Compound 26 was subjected to Griesbaum co-ozonolysis with various ketones (acetone, methyltrifluoro methyl acetone, or cyclohexanone) in a CH₂Cl₂: cyclohexane solvent mixture at 0 °C (Fig. 8).

Fig. 8 Synthesis of methyl-3β,28-di-O-acetyl-29,30-bisnorlupan-20-oate.

Unexpectedly, in all cases, the major product was methyl-3β,28-di-O-acetyl-29,30-bisnorlupan-20-oate (29), irrespective of the ketone used. The anticipated ozonolysis product was not detected. This anomalous outcome is attributed to a rearrangement of the secondary ozonide intermediate 28, involving cleavage of a C–Csp³ bond and migration of the C₃₀ methyl group to the electrophilic oxygen. This transformation is mechanistically analogous to the Baeyer–Villiger type rearrangement of ketones in the presence of peroxides.²⁶

A similar rearrangement was previously observed during ozonolysis of a 3-O-acetylbetulinic acid ester in EtOAc:MeOH at 70 °C, where the methyl ester of the 29,30-dinor-20-oic acid was isolated in 4% yield as a side product, alongside the expected 29-nor-20-ketone.²⁷ In contrast, compound 29 was obtained as the major product, an outcome likely influenced by the nature of the substituent on the trioxolane ring. Rearrangements of peroxide ozonolysis products are well documented,²⁸ particularly among substrates such as allyl alcohols, unsaturated ketones, indoles, and benzofurans.²⁹

The secondary ozonides at the C₂₀ position of botulin derivatives have been reported to be prone to decomposition. Notably, even the inclusion of a highly electrophilic trifluoromethyl ketone failed to yield the outcome of Griesbaum co-ozonolysis. The structure of 29 comprises six-membered rings A, B, C, and D in trans-chair conformations and five-membered ring E adopting a slightly distorted envelope geometry.

5. Synthesis of triterpenoid-based trioxolanes and dioxazolidines via ozonolysis of allobetulin derivatives

Another promising class of triterpenoid-based peroxide, the 1,2,4-dioxazolidine, was synthesized³⁰ via the Griesbaum method.³¹ The O-methylated dioxime 32, derived from the 1,5-dicarbonyl compound of A-secoallobetulin 31,³² was ozonized to completion, affording N-methoxy-substituted 1,2,4-dioxazolidines 33 and 34 in 55% and 38% yields, respectively, after chromatographic purification (Fig. 9). Notably, the reaction exhibited diastereoselectivity, yielding exclusively the (3S,5S)-configured peroxides as confirmed by X-ray crystallography and the established stereochemistry of allobetulin derivatives. These compounds represent rare examples of stable dioxazolidines, and the first example of N-methoxy-1,2,4-dioxazolidines derived from triterpenoids.

Fig. 9 Synthesis of triterpenoid-based 1,2,4-trioxolanes and 1,2,4-dioxazolidines by ozonolysis of allobetulin derivatives.

6. Stereochemistry of product

The use of symmetrical oxime ether eliminates the possibility of syn–anti isomerism in the resulting carbonyl oxide intermediates. This ensures that the stereochemical outcome of the subsequent cycloaddition is determined solely by the geometry of the starting ketones. In the case of achiral benzophenones, the resulting trioxolane products are themselves achiral. For 4-substituted cyclohexanones, the major trioxolane isomers consistently exhibit a cis-configuration, wherein the substituent occupies the equatorial position and the peroxo moiety is oriented axially.^33,34 Assuming the stereochemical pathway of the cycloaddition remains consistent across other 4-substituted cyclohexanones, the synthesized trioxolanes were assigned cis-configurations.

7. Diastereoselectivity in the Griesbaum co-ozonolysis reaction

The use of a symmetrical oxime ether prevents syn–anti isomerism in 2-adamantanone oxide formation, making the stereochemical outcomes dependent solely on the starting cyclohexanone. For 4-substituted cyclohexanones, only two achiral diastereomers cis and trans are possible (Fig. 10). Co-ozonolysis of compounds 35 and 36 was optimally performed in non-polar solvents at 0 °C, as recommended by Griesbaum.^35,36 For more polar substrates, mixed solvents with minimal dichloromethane content were employed to maintain favorable conditions for ozonide formation. The oxime ether-to-ketone ratio varied between 1.5 : 1 and 1 : 2 to afford 37 and 38 ranging from 20% to 70%, respectively. Product ratios were determined by integration of diagnostic signals in the ¹H NMR spectra of crude reaction mixtures. In most cases, only the major isomers 37 were isolated by flash chromatography or crystallization and fully characterized. Stereochemical cis or trans assignments are based on X-ray crystallography and the assumption of consistent stereochemical outcomes across all cycloaddition products. Stereochemistry assignments cannot be reliably inferred from the chemical shift, coupling constant, or chromatographic behavior. However, X-ray crystallography provides definitive configurations in representative cases. For example, acetoxy ozonide was assigned to a cis configuration based on the crystal structure of its hydrolysis product, revealing a chair conformation with the 4-hydroxyphenyl and peroxo groups in equatorial and axial positions, respectively. In group II, X-ray analysis confirmed the major isomer 37 as cis and the minor 38 as trans, with both the 4-phthalimidomethyl and peroxo groups equatorial in 37. Collectively, these findings support the conclusion that major cycloaddition products consistently adopt the cis configuration 4-substituted cyclohexanones.

Fig. 10 Diastereoselectivity of Griesbaum co-ozonolysis.

The results demonstrate that diastereoselectivity is influenced by the size and nature of the 4-substituent of the cyclohexanone dipolarophiles. Bulky substituents, such as tert-butyl, phenyl, and phthalimido, yielded exclusively one diastereomer, with cis/trans ratios exceeding 20 : 1. Substrates bearing functional groups linked via a single methylene unit exhibited moderate selectivity, with cis/trans ratios around 5 : 1, suggesting minimal influence from distal substituents. Cyclohexanone substituents with an ester or amide group at the 4-position gave cis/trans ratios ranging from 2.5 : 1 to 5 : 1.

8. Stereochemistry confirmation via analogue synthesis

Due to extensive overlap of the aliphatic signals in the ¹H NMR spectra of analogues, unambiguous stereochemical assignment proved difficult. To resolve this, several analogues were synthesized with the aim of obtaining diffraction-quality crystals. Finally, a single crystal of the 4-nitrobenzoate analogue was successfully grown via vapor diffusion of hexane into a saturated toluene/ethanol/methanol solution. High-resolution X-ray crystallographic analysis confirmed the anticipated stereochemistry: an axial orientation of the peroxide bridge and a trans relationship to the equatorially positioned 3-substituent.

9. Diastereoselective synthesis of triterpenoid trioxolanes via Griesbaum co-ozonolysis

Three principal strategies are commonly employed for the synthesis of ozonides in organic chemistry: (i) low-temperature ozonolysis of olefins in aprotic solvents,³⁷ (ii) co-ozonolysis of O-methyl oximes and carbonyl compounds,³⁸ and (iii) ozone-free methods involving the reaction of 1,5-dicarbonyl compounds and hydrogen peroxide.³⁹ Among these, the Griesbaum co-ozonolysis method⁵ is the most widely adopted, offering high efficiency and structural tenability in the synthesis of peroxide. Using this approach, Vennerstrom⁴⁰ synthesized over 200 spiro-1,2,4-trioxolanes incorporating adamantanone scaffolds, which were evaluated for anti-malarial, anti-tumor, and anti-schistosomal activities. The most promising candidates included OZ277 (Fig. 11), but exhibited potent antimalarial efficacy. As part of ongoing efforts to expand the diversity of biologically active peroxides, the present work⁴¹ describes the synthesis of CF₃-methylated 1,2,4-trioxolanes derived from oleanane-type triterpenoids.

Fig. 11 Peroxides example and synthesis of the CF₃-methylated 1,2,4-trioxolanes derived from oleanane-type triterpenoids.

10. Identification of synthetic trioxolane candidate drug development

As reported in ref. 42, in comparison to semi-synthetic artemisinins, the selected 1,2,4-trioxolane drug development candidate trioxolane 53, also known as OZ277 or RBx-11160 (Fig. 12) offers significant advantages, including structural simplicity, and a cost-effective, scalable economically feasible and scalable synthesis. To avoid the synthesis of chiral analogues of the trioxolane prototype, the author employed 4-substituted cyclohexanones, which when reacted with symmetrical O-methyl-2-adamantanone oxime via Griesbaum co-ozonolysis, yield only two achiral diastereomers (cis and trans). Two key trioxolane intermediates were efficiently synthesized using the Griesbaum co-ozonolysis reaction on a 100 mmol scale, achieving yields of 68% to 78%. Notably, without the need for chromatography purification, X-ray crystallographic analyses of trioxolane 53 and the phthalimide precursor of trioxolane 54 confirmed that the peroxide linkage adopts an axial orientation, while the substituent is equatorially positioned, thereby confirming cis-configuration. As exemplified by trioxolanes 53–55 (Fig. 12), a series of relatively polar trioxolane derivatives were synthesized through a variety of post-ozonolysis transformations, demonstrating the versatility and modularity of this synthetic platform.

Fig. 12 Synthesis of trioxolanes 53–55.

11. Efficient, stereocontrolled synthesis of trioxolanes for ferrous iron-dependent drug delivery

A stereocontrolled, three-step synthesis of 1,2,4-trioxolanes bearing the critical 3-substitution pattern required for ferrous iron-dependent drug activation has been developed,⁴³ affording the product in 67–71% overall yield. The key transformation is a diastereoselective Griesbaum co-ozonolysis, which predominantly yields trioxolanes in a trans relationship between the 3-substituent and the peroxide bridge. This stereochemistry was confirmed by X-ray crystallographic analysis of a representative 3-substituted 4-nitrobenzoate analogue.

An effective strategy for relative stereochemistry in the formation of the trioxolane ring involves Griesbaum co-ozonolysis, which culminates in a stereochemistry-defining [3 + 2] cycloaddition between a carbonyl oxide and ketone. Vennerstrom and co-workers demonstrated that in 4-substituted cyclohexanones, this cycloaddition occurs preferentially via axial attack of the carbonyl oxide, yielding a product comprising equatorial 4-substituents and a cis relationship to the axial peroxide.

By analogy, an axial attack on a 3-substituted cyclohexanone is expected to furnish trioxolanes with equatorial 3′′ substituents in a trans relationship with the peroxide bridge (Fig. 13). To investigate this, tert-butyldiphenylsilyl ether 57 was synthesized and subjected to Griesbaum co-ozonolysis with 2.5 equivalents of oxime 11 in CCl₄, affording the trioxolanes 58 in high yield. Subsequently, treatment with TBAF in THF provided alcohol 60, completing a three-step sequence from precursor 56 in 67–71% overall yield. The ¹H NMR analysis of 58 and 60 suggested high diasteroselectivity, which was further confirmed through conversion to methyl ether 61. Integration of the methoxy signal in the proton NMR spectrum of 61 indicated a diastereomeric ratio of 90 : 10 d.r., confirming the stereochemical outcomes of the Griesbaum cycloaddition.

Fig. 13 Stereocontrolled synthesis of the 3-substituted cyclohexane derivative.

11.1 Synthetic strategy towards anti-malarial ozonide via Griesbaum co-ozonolysis

Using a modified procedure, keto phenol 62 was transformed into ketoepoxide 63 with an 87% yield.⁴⁴ Griesbaum co-ozonolysis of 63 with O-methyl-2-adamantanone oxime 53 afforded epoxy ozonide 64 (54%) yield. Subsequently, treatment of 64 with morpholine followed by mesylate furnished compound 65 in 60% yield (Fig. 14).

Fig. 14 Stereoselective ozonide derivative synthesis.

Griesbaum co-ozonolysis of keto acetate 66 with fluorinated O-methyl-2-adamantanone oximes 67 and 68 (ref. 45) afforded the corresponding ozonide esters 69 (54%) and 70 (44%), respectively (Fig. 15). These intermediates were then subjected to a one-pot acetate hydrolysis and alkylation with N-(2-chloroethyl)morpholine, followed by mesylate salt formation, yielding fluorinated ozonides 71 and 71 in overall yields of 77% and 72%, respectively.

Fig. 15 Synthesis of fluorinated ozonide.

Ozonide 71 was obtained as a 1 : 1 mixture of diastereomers. Chlorosulfonylation of ketone yielded crude 73 (10%) yield, which was subsequently converted into the corresponding pentafluorosulfonate ester 75 (70%) (Fig. 16).

Fig. 16 Synthesis of p-sulfonyl phenyl ozonide derivatives.

Griesbaum co-ozonolysis of pentafluorosulfonate ester 75 prepared from ketone 73 chlorosulfonylation (10%) and subsequent fluorination (70%) (Fig. 16), with O-methyl-2-adamantanone oxime 35, afforded the corresponding per-fluorosulfonyl substituted ozonide.

11.2 Desymmetrized anti-malarial trioxolanes: synthesis and in vivo profiling

To expand upon the earlier investigation of trans-3′′ carbamates-containing trioxolanes,⁴⁶ both enantiomeric forms of the newly designed analogues were synthesized and evaluated. This included the preparation of the (S,S) diastereomers of compounds that had previously been accessed only as (R,R) forms. To this end, the (S,S) trans-3′′ analogues were synthesized via an enantioselective route, beginning with a (R)-taniaphos-mediated asymmetric borylation of cyclohexanone, which provided the chiral boronic ester intermediate bearing the desired S-configuration. This stereochemically defined intermediate laid the foundation for subsequent stereocontrolled elaboration into the target (S,S)-configured trioxolane carbamates. Both enantiomers of the new analogues were studied, and the (S,S) forms were prepared for analogues earlier available only as (R,R). Thus, (S,S) trans-3′′ analogues were synthesized in an enantiocontrolled fashion, starting with (R)-taniaphos-mediated asymmetric borylation of cyclohexenone to prepare intermediate 80.

Oxidation of the C–B bond in intermediate 80, followed by protection of the resulting alcohol as a TBDPS ether, furnished ketone 82. This intermediate underwent a diastereocontrolled Griesbaum co-ozonolysis with adamantan-2-one O-methyloxime, affording the trans-trioxolane intermediate 83 in a 12 : 1 diastereomeric ratio (d.r.), as determined by ¹H-NMR analysis. Subsequent TBDPS deprotection of 83, followed by conversion to the corresponding p-nitrophenyl carbonate 85, enable late-stage diversification via carbamate formation to yield the desired (S,S)-trans-3 carbamate analogues 86a–d. An analogous enantioselective sequence, employing the enantiomeric taniaphos ligand, was used to access the complementary (R,R)-trans-3-carbamates analogues 86a–d (Fig. 17).

Fig. 17 Representative enantiocontrolled synthesis of (R,R)-trans-3′′ analogues 86a–d.

12. Effect of functional group polarity on the anti-malarial activity of spiro and dispiro trioxolanes

In this study,⁴⁷ physicochemical characterization and antimalarial evaluation of structurally diverse derivatives were reported based on early trioxolane scaffolds, encompassing a broad range of polarity. The polarity objective was to determine which functional groups contributed to favorable antimalarial activity, and to identify lead compounds exhibiting an optimal balance between lipophilicity (log P/log D) and aqueous solubility. A key emphasis was placed on the carboxylic acid moiety due to its twofold benefits: its capacity to generate water-soluble salts and its therapeutic relevance demonstrated by semisynthetic artemisinin derivatives such as artesunate and artelinate.⁴⁸

Trioxolanes 93 (50%), 94 (18%), 95 (25%), 96 (45%), and 99 (49%) were synthesized directly via Griesbaum co-ozonolysis involving the reactions of O-methyl-2-adamantanone oxime 87 with either 4-substituted cyclohexanones 88 or benzophenones 89. The corresponding trioxolane carboxylic acids 91 (58%), 95 (84%), 22 (90%), diol 90 (46%), alcohol 92 (87%), and diamine 99 (29% overall) were obtained through aqueous KOH hydrolysis of their respective trioxolane ester precursor. These ester precursors were themselves synthesized directly by co-ozonolysis, yielding between 25–63%. The synthesis of compound 92 required three additional post-ozonolysis transformations: reduction using lithium borohydride or lithium triethyl borohydride for 92, phthalimide formation via the Mitsunobu reaction, and hydrazinolysis for phthalimide deprotection (Fig. 18).

Fig. 18 Griesbaum co-ozonolysis reaction to prepare spiro and dispiro-1,2,4-trioxolanes.

13. Characterization of major CYP450 metabolites of ozonide OZ277

In 2008, L. Zhou⁴⁹ reported the synthesis of the minor metabolite G via a four-step sequence (Fig. 19), starting with a Griesbaum co-ozonolysis reaction between 5-acetoxy-2-adamantanone O-methyl oxime 101 and keto ester 102. Oxime ether 101 was prepared in two steps: initially by the reaction of 5-hydroxy-2-adamantanone with methoxylamine HCl and pyridine (yield: 97%), followed by acetylation with acetic anhydride, pyridine, and catalytic DMAP (yield: 91%). The ozonolysis of unsymmetrical oxime ethers such as 101 generates enantiomeric carbonyl oxide intermediates, which can undergo cycloaddition reactions with 4-substituted cyclohexanones 102 to potentially yield four isomeric ozonides.⁵⁰ Based on previously established data, the formation of two major ozonide isomers was anticipated, with substituent and peroxo groups adopting equatorial and axial orientations, respectively, in the cyclohexane ring. Indeed, analysis of the crude reaction mixture (30%, silica gel chromatography, 10% ether in hexanes) using ¹H and ¹³C NMR revealed two predominant ozonide diester isomers, from which the major isomer 103 was isolated by crystallization in 17% yield. The trans, cis configuration was assigned based on the conversion of the trans, cis ozonide diester 103 to 105 (86%).

Fig. 19 Synthesis of ozonide diester diastereomers.

Similarly, the major metabolite 106 F was synthesized via a parallel four step sequence (Fig. 19), starting from a Griesbaum co-ozonolysis between 5-(4-methylbenz oxy)-2-adamantanone O-methyloxime and keto ester 102, which upon chromatography purification afforded a mixture of four ozonide diester diastereomers (103 and 104). The oxime ether precursor was synthesized by benzoylation of 5-hydroxy-2-adamantanone using 4-methylbenzoyl chloride in pyridine (90%), followed by conversion to the O-methyloxime using methoxylamine hydrochloride pyridine (94%).

The 4-methylbenzoate group was strategically chosen over acetate to increase the molecular weight, facilitating fractional crystallization of the minor ozonide diester isomers, and to provide distinct benzylic singlet proton NMR signals for effective diastereomer discrimination. Repeated chromatography yielded four fractions of isomers. Crystallization of the isomer from acetone afforded 103 as colorless crystals, whose trans cis diastereomeric structure was confirmed by X-ray crystallographic analysis.

14. SAR of an ozonide carboxylic acid (OZ78) against Fasciola hepatica

Following the method of Tsandi,⁵¹ acyl sulfonamides were synthesized by coupling with methanesulfonamide in the presence of DMAP and DCC. Corresponding hydrazides were obtained via direct reaction of the methyl ester with hydrazine. Glycine and taurine conjugates of ozonides were prepared by coupling the HOBt activated ester⁵² with glycine ethyl ester and taurine, respectively.

Ozonide dicarboxylic acid 109 was synthesized through five-step sequence starting from compound 107, the Knoevenagel condensation product of 1,4-cyclohexanedione monoethylene ketal and isopropylidene malonate (Fig. 20).⁵³ Subsequently, reduction (99%), deketalization (27%), and esterification (37%) yielded diester ketone 108, which underwent Griesbaum co-ozonolysis with oxime ether 35 (ref. 54) to give diester ozonide 109 in 75% yield. Hydrolysis of 109 furnished the target dicarboxylic acid 50 in high yield.

Fig. 20 Five-step synthesis of diester ozonide and diacid ozonide.

15. Machines vs. malaria: flow synthesis of anti-malarial candidate OZ439

Processing conditions for the Griesbaum co-ozonolysis were optimized using commercially available 4-phenylcyclohexanone 66 as a model substrate.^55,56 A solution oxime 11 and ketone 66 (X = H) in EtOAc was exposed to an ozone stream, with both liquid and gas flow rates adjusted to maximize the product yield.⁵⁷ Using an excess of ketone gave moderate yields, whereas using an excess of oxime significantly improved the outcome, affording compound 68 in 76% isolated yield after purification (Fig. 21). The continuous flow reaction ran smoothly for 45 min; however, extended reaction times led to blockages due to slight solvent evaporation.

Fig. 21 Flow setup for ozonide formation.

This issue was resolved by reducing the concentration of oxime 35 (0.4 to 0.2 M) and ketone 110 (0.2 to 0.1 M), while increasing the flow rates of both liquid (1 mL min⁻¹) and gas (1 L min⁻¹), to maintain system throughput. Under these diluted conditions, the reaction proceeded continuously for over 3 hours without clogging, resulting in 78% yield and a throughput of 1.6 g h⁻¹ (38.4 g d⁻¹).

The same reactor setup was successfully applied to the co-ozonolysis of 35 and 110, delivering the 1,2,4-trioxolane 111 in 70% yield with 9 : 1 cis/trans selectivity. This corresponded to a production rate of 1.9 g h⁻¹ (45.6 g d⁻¹). Notably, this continuous flow protocol offers advantages over batch ozonolysis by enhancing safety and efficiency, and eliminating the use of flammable solvents such as pentane through substitution with ethyl acetate.

A robust and efficient machine-assisted protocol was developed for the synthesis of the anti-malarial drug candidate OZ439,⁵⁸ leveraging flow chemistry to overcome limitation batch processing. Three key transformations (selective partial hydrogenation, phenolic acetylation, and Griesbaum co-ozonolysis) were successfully adapted to continuous flow conditions, enabling improved process control, safety, and scalability. Notably, this new synthetic route avoided the use of the genotoxic reagent 4-(2-chloroethyl)morpholine, offering a safer and more sustainable alternative for large scale production.

16. Targeted drug delivery to malaria parasite using an arterolane-like scaffold

Gratifyingly, the partially protected diketone 112 underwent efficient Griesbaum co-ozonolysis with O-methyl-2-adamantanone oxime 11, yielding the desired trioxolane 113 in nearly quantitative yields (Fig. 22). The use of two or more equivalents of oxime was critical for achieving high conversion. Subsequently, deprotection of ketal moiety 113 using ferric chloride hexahydrate in acetone/dichloromethane proceeded smoothly, affording ketone 115 in excellent yield. A modified reduction protocol, based on conditions previously reported for 4-keto trioxolanes, was successfully applied to reduce 3-keto trioxolane 115, yielding alcohol 114 in 67%. The reduction showed minimal diastereofacial selectivity, producing a roughly 1 : 1 mixture of cis and trans racemic diastereomers.

Fig. 22 Synthesis of the 3-keto trioxolane alcohol derivative.

This work⁵⁹ presents an improved chemical scaffold for trioxolane-mediated drug delivery supported by the efficient, concise, and scalable synthesis of key intermediate 114, which will facilitate further exploration of this class of compounds. Additionally, the amino nucleoside was employed as a chemical biological probe to monitor drug release in P. falciparum, demonstrating the utility of this approach. The use of puromycin as a drug surrogate is anticipated to enable future studies on trioxolane-mediated drug delivery across malaria and other disease models.

17. Endoperoxide-based FRET probe for ratiometric imaging of labile iron in living cells

The design, synthesis, and biological evaluation of FRET Iron Probe 1 (FIP⁻¹) was reported in ref. 60, as a reactivity-based probe for ratiometric fluorescence imaging of labile Fe(II) pools in living systems. Inspired by antimalarial natural products and their mechanisms of iron-dependent activation, FIP⁻¹ incorporates two fluorophores, fluorescein (donor) and CY3 (acceptor) linked via an Fe(II)-cleavable endoperoxide bridge that is disrupted upon activation, separating the fluorophores and reducing FRET. FIP⁻¹ exhibits high selectivity for Fe(II) over other metal ions in aqueous buffer and enables dynamic visualization of intracellular iron fluctuation under iron supplementation or chelation conditions.^61–64

The synthesis of FIP⁻¹ involves the five-step preparation of a dual-functionalized adamantyl-endoperoxide linker from commercially available starting materials (Fig. 23). Starting with 2-adamantanone-5-carboxylic acid 116, Fischer esterification yields the methyl ester 75, which was subsequently converted to oxime ether 118 via reaction with hydroxylamine hydrochloride.^65,66 Compound 118 underwent Griesbaum co-ozonolysis with 1,4-cyclohexanedione and ozone to afford endoperoxide 119. Reductive amination of 119, followed by saponification, provided the carboxylic acid-functionalized endoperoxide linker 121. Separately, Cy3-NHS-ester and 5-AMF were synthesized according to a previously published procedure.⁴³ These fluorophores were then subsequently coupled to linker 121 to furnish the final ratiometric probe FIP⁻¹ 123.⁶⁷

Fig. 23 Synthetic route of FIP⁻¹.

18. Expanded scope of Griesbaum co-ozonolysis for structurally diverse ferrous iron sensors

Sterically shielded 1,2,4-trioxolanes, synthesized via Griesbaum co-ozonolysis, have emerged as key scaffolds in the development of chemical probes for detecting labile ferrous iron. Another report⁶⁸ optimized co-ozonolysis conditions that proceeded with high efficiency diastereoselectivity across a broader substrate scope, enabling the design of next-generation iron ferrous sensors with tailored reactivity and physicochemical properties.

When cyclic ketone and oxime are employed as co-reactants, the resulting trioxolane adducts exhibit notable stability, attributed to steric shielding of the endoperoxide bond by an adjacent axial C–H bond rigid carboxylic framework. Vennerstrom and co-workers exploited this effect in the development of the antimalarial agents arterolane⁶⁹ and artefenomel,⁷⁰ where the adamantane-derived trioxolane core stabilized the peroxide moiety, enhancing its selective reactivity with labile Fe(II) in Plasmodium parasites. This Fenton-type activation mirrors the mechanism of artemisinin derivatives, underscoring the therapeutic relevance of peroxide stability and iron-triggered activation. Growing recognition of labile iron as the bioavailable intracellular iron pool has driven the development of oxidation-state specific probes, primarily via reactivity-based strategies.⁷¹ Ferrous iron-triggered reduction of N–O or O–O bonds enables the activation of fluorophores (e.g. TRX-FIP-1 (ref. 72)), modulation of FRET pairs (TRX-FRET), release of tethered reporters (e.g. TRX-PURO,⁷³ ICL-1,⁷⁴ HNG⁷⁵), or covalent capture of PET radiolabels (18F-TRX^76,77). Trioxane and trioxolane-based reagents have also found utility in chemoproteomic profiling of malarial parasite and mammalian cancer cells (e.g. FIPC-1 (ref. 78)), further demonstrating their versatility in bioanalytical applications (Fig. 24).

Fig. 24 Conversion of oxime to carbonyl oxide, followed by diastereoselective reaction with the ketone co-reactant, with axial addition favored.

While arterolane-like pharmacophores have played a central role in the design of first-generation ferrous iron probes, further optimization will likely require access to trioxolane scaffolds with expanded iron reactivity profiling and improved physiochemical properties that are suitable for in vivo applications. The Griesbaum co-ozonolysis was revisited as a strategy to access new structural frameworks for ferrous iron-reactive therapeutics and chemical probes. Mechanistically, the Griesbaum co-ozonolysis proceeds via [3 + 2]/retro-[3 + 2] sequence, wherein O-methyloxime undergoes ozonolysis to generate a carbonyl oxide intermediate. This transient then engages in stereochemistry determining [3 + 2] cycloaddition with a ketone to form the 1,2,4-trioxolane adducts. The reactions between adamantane-derived oximes and substituted cyclohexanones are known to favor axial attack of the carbonyl oxide on the ketone, typically furnishing cis-40 or trans-30 configured trioxolanes with useful diastereoselectivity (d.r. 9 : 1) (Fig. 22).⁵⁴

Initial reports by Griesbaum primarily focused on simple alkyl and cycloalkyl substrates, while subsequent studies have emphasized adamantane-derived oximes, which yield pharmacologically active trioxolanes.⁷⁹ Other groups have extended this chemistry to non-adamantane substrates.^80,81 An optimized Griesbaum co-ozonolysis protocol was previously reported, employing adamantanone oxime as the access reagent and using the ketone as the limiting component, which afforded hydroxy trioxolanes (70% yield) at 0 °C in CCl₄. However, this method was found to be inadequate for substituted adamantanes, and particularly non-adamantyl ketones, where the yield decreased to 5–23%, or reactions failed altogether. These poor outcomes were attributed to possible side-reactions involving ozone or the carbonyl oxide intermediate (Fig. 25). To address these limitations, subsequent studies investigated the co-ozonolysis of enantiopure ketone 128 ⁸² with various substituted adamantanone oxime 127 under low temperature conditions (−78 °C in hexanes, O₃, flow 6 g h⁻¹ or 1.1 L per min O₂). These revised conditions (condition B) yielded substantial improvement: compound 130a was isolated in 77–94% yield (vs. 48–91% under previous conditions), and substituted trioxolanes 130b–e were obtained up to 94% yield. Diasteroselectivity was also improved, with exclusive formation of the trans isomers for 130a–e under condition B compared to 90 : 10 d.r. under the original conditions. Notably, no diastereofacial selectivity was observed when unsymmetrical carbonyl oxides were employed.

Fig. 25 Scope of Griesbaum co-ozonolysis involving structurally diverse oxime 127 and ketone 128.

Given the role of axial C–H shielding in modulating the Fe(II)-specific Fenton reactivity of endoperoxide probes (e.g. TRX-PURO, ICL-1),⁷² other bicyclic aliphatic ring systems have been evaluated computationally for potential steric shielding effects. The ring system including bicycle[2.2.1]heptane, bicycle[2.2.2]heptane, and bicycle[3.3.1]nonane were identified as promising candidates. Applying condition B, co-ozonolysis of bicycle[2.2.1]heptan-2-one methyl oxime with 128 afforded troxolan 130f in 81% as a mixture of stereoisomers (13 : 8 : 1 by ¹H NMR). Notably, this transformation failed under the original condition A. Building on these results, more oxime substrates were tested at low temperature (Fig. 26), which expanded the range of trioxolane adducts that could be obtained.

Fig. 26 Structures of adducts 130a–f with their respective yields in conditions A and B.

Under the optimized low temperature conditions, the reaction of ketone 128 with bicyclo[2.2.2]octan-2-one methyl oxime and bicyclo[3.3.1]nonan-9-one methyl oxime yielded trioxolane adduct 131g and 131h in modest (26–44%) and excellent (80–87%) yields, respectively. In contrast, camphor methyl oxime failed to afford adduct 131i, likely due to steric hindrance around the oxime moiety. Further substrate exploration included substituted cyclohexanone, cyclopentanone, and acyclic oximes. Cyclohexanone-derived oximes with 4-substitution provided adducts 131j and 131l in 71% and 53% yields, respectively, while 2-bromocyclohexan-1-one methyl oxime afforded 131k in 53% yield. Unexpectedly, cyclopentanone oxime gave 131k in 53% yield. Notably, cyclopentanone oximes failed to yield quantities of the expected trioxolane adducts (131m–o), likely due to unfavorable ring strain or instability of the corresponding intermediates under the reaction conditions (Fig. 27).

Fig. 27 Structures of trioxolane adducts 131f–q and their yields.

The failure of certain Griesbaum co-ozonolysis reactions may be attributed to intrinsic instability of the resulting trioxolane adducts, or competitive side reactions of the carbonyl oxide intermediate, such as dimerization. For example, attempts to form trioxolanes 131p and 131q from methyl oximes of acetophenones and 4-methoxyphenylacetone, respectively, yielded no isolable products (Fig. 27).

To assess the Fe²⁺-dependent reactivity of selected novel trioxolanes like 131f, 131g, and 131h, their incorporation into payload-bearing trioxolane conjugates has been investigated. As a surrogate for physiological Fe²⁺ reactivity, these studies evaluated their ability to undergo iron(II)-triggered payload.^83,84 Conjugates bearing mefloquine as the payload exhibited potent in vitro anti-plasmodial activity, with IC₅₀ values of 74 nM and 24 nM, comparable to the positive control (IC₅₀ = 17 nM). These findings support a canonical Fenton-type activation mechanism,⁸⁵ wherein Fe(II)-mediated cleavage of the trioxolane ring triggers payload release, thereby accounting for the observed anti-malarial potency of the conjugates.

19. Targeting ferrous iron mobilization in Pseudomonas aeruginosa via Iron(II)-caged LpxC inhibitor

To realize the proposed disconnection, synthetic access to both the LpxC inhibitor (LpxCi) carboxylate and an aminooxy-functionalized trioxolane (TRX) intermediate was required (Fig. 28).⁸⁶ The LpxCi carboxylate corresponded to a late-stage intermediate in the synthesis of hydroxamate-based inhibitors, while the TRX component was designed to enable site-specific conjugation via the hydroxamate oxygen. Following extensive optimization, a robust and scalable synthetic route was established that enabled multigram scale production of the trioxolane hydroxamate conjugates. This strategy provides a versatile platform for constructing iron(II) response drug conjugates, expanding the utility of trioxolane scaffolds for targeted delivery applications.

Fig. 28 Stereo-controlled synthesis of key trioxol intermediate, 138, bearing a 3-aminooxy function.

The synthesis of the aminooxy-functionalized trioxolane intermediate (±)-138 was achieved via a streamlined route (Fig. 28). Monobenzylation of cyclohexane-1,3-diol yielded ether 133, which underwent a Mitsunobu reaction with N-hydroxyphthalimide to afford intermediate 134. Subsequently benzyl deprotection of BCl₃·SMe₂ furnished alcohol 135 in 89% yield. Oxidation of 135 to ketone 136 was initially hindered by its tendency to undergo beta-elimination; however, this step was successfully optimized using Dess–Martin periodinane or Swern oxidation under carefully controlled conditions. The resulting ketone 136 underwent Griesbaum co-ozonolysis with the O-methyl oxime of adamantan-2-one to afford trioxolane 137 in 92% yield, with diastereoselectivity (13 : 1 d.r) favoring the trans isomer, consistent with prior observations.⁸⁷

To assess the in vivo stability and pharmacokinetics, compound 140 was administered via intraperitoneal (IP) injection to female NSG mice. Plasma samples were analyzed for both intact conjugate 140 and liberated LpxC inhibitor 140 to evaluate potential Fe²⁺-dependent or-independent release mechanisms. Compound 140 displayed a rapid elimination profile with a half-life (T_1/2) of approximately 0.7 hours, closely resembling the pharmacokinetics of PF-508109037 (Fig. 29 and 30).

Fig. 29 Synthesis of hydroxamate and amid-linked TRX conjugates from known LpxC inhibitors and reaction of 9 with ferrous ammonium sulfate.

Fig. 30 Plasma exposure profile of prodrug 140 (red) and systematically released 142 (blue).

20. Artefenomel regioisomer RLA-3107: a promising lead for the next-generation endoperoxide anti-malarial

The Griesbaum co-ozonolysis reaction, widely employed for the synthesis of 1,2,4-trioxolanes, is intrinsically diastereoselective, typically yielding cis or trans diastereomers with a 9 : 1 diastereomeric ratio when using C4- or C3-substituted cyclohexanones, respectively. Full stereocontrol can thus be achieved using enantiopure cyclohexanone substrates, as demonstrated in the enantioselective synthesis of arterolane-like trans-R4 amide and carbamate, as well in O'Neill's desymmetrized tetraoxanes synthesis.⁸⁸ In the present work,⁸⁹ racemic cyclohexanone was employed to afford the (±)-2 trans diastereomer as the major product.

A stereocontrolled synthesis of RLA-3107 148, a regioisomer of the clinical studied drug candidate artefenomel, was accomplished starting with a palladium(II) catalyzed 1,4-conjugate addition. A dicationic palladium(II) complex, formed in situ from Pd(acac) and Cu(BF₄)₂·H₂O, effectively catalyzed the 1,4-addition of 4-(acetoxy)phenyl boronic acid to cyclohexen-1-one 144, affording ketone 145 in 91% yield. Subsequently, Griesbaum co-ozonolysis of 145 with adamantan-2-one O-methyl oxime 35 in the presence of ozone at 0 °C yielded 3-aryl substituted 1,2,4-trioxolane intermediate 146 in nearly quantitative yield with a diastereomeric ratio (d.r.) of 8 : 1. Hydrolysis of the acetate protecting group under basic condition (KOH in MeOH/THF at 50 °C) afforded phenol 147 in 95% yield. This intermediate was then alkylated with 4-(2-chloroethyl)morpholine hydrochloride using NaOH and [(Bu)₄NHSO₄] in acetonitrile at 55 °C, furnishing (±)-148 in 62% yield or (50% overall yield for four steps). Final purification of both 147 and 148 via column chromatography removed any residual minor diastereomer, yielding pure trans (±)-148, as confirmed by ¹H NMR analysis. Since the precursor ketone has been previously synthesized in the enantiopure form,⁸⁰ this route is fully amenable to enantiomerically synthesizing 116 (Fig. 31).

Fig. 31 Stereocontrolled synthesis of artefenomel regioisomers trans-(±)-148.

This work validates the conformational prediction that trans 3′′-aryl substituted trioxolanes exhibit comparable Fe²⁺ reactivity and anti-plasmodial activity to their C4-aryl regioisomers counterparts. The in vitro and in vivo studies presented here demonstrate the 3′′-aryl trioxolane scaffold as a promising platform for the design of next-generation peroxide antimalarial drugs including RLA-3107. Building on this, a diverse series of desymmetrized anti-malarial trioxolanes bearing carbomate side chains was synthesized, targeting both (R,R) and (S,S) trans isomers. A stereocontrolled synthesis of RLA-3107 (148), a regioisomer of artefenomel, was carried out, and conformational analysis confirmed the expected trans-3′′ stereoisomers. In particular, (S,S) analogues were accessed via enantioselective borylation of cyclohexanone, mediated by (R)-taniaphos, to yield the intermediate with a defined S configuration. These analogues are currently undergoing biological profiling.

21. Exploration of next-generation trioxolanes among artemisinin analogues

To identify next-generation trioxolane agents with differentiated pharmacological properties, a series of novel analogs bearing the trans-4′′ substitution inspired by conformational analysis has recently been reported,⁹⁰ which predicted stabilization of the endoperoxide bond comparable to that observed with conventional cis-4′′ substitution. Indeed, congeneric trans-4′′ analogs of arterolane 150 demonstrated promising anti-plasmodial activity in both in vitro and in vivo models, validating the hypothesis that trans-4′′ substitution can modulate ferrous iron reactivity in a pharmacologically favorable manner. Continued investigations have led to the identification of compound 152 (Fig. 32), an advanced lead with efficacy against artemisinin-resistant (ART-R) Plasmodium falciparum and potential as a trioxolane-based development candidate.

Fig. 32 Structures of dihydroartemisinin, arterolane (150), and trans-trioxolane analogs (±)-152 and (S,S)-152, with the stereochemistry indicated.

The transition to trans-3′′ substitution in the 1,2,4-trioxolane pharmacophore introduces two stereogenic centres, giving rise to four possible diastereomers. However, the Griesbaum co-ozonolysis employed for trioxolane synthesis is intrinsically stereoselective, favoring cis-4 substitution and trans-3′′ substitution consistent with prior reports. Leveraging this inherent selectivity, a concise six-step synthetic route was developed to access racemic (±)-152 and related analogs (Fig. 32 and 33), closely aligned with the four-step synthesis previously reported for the artefenomel regioisomer. Access to enantiopure forms, namely (R,R)-152 and (S,S)-152, was achieved by simply utilizing enantiomerically pure cyclohexanone intermediates in the Griesbaum reaction.

Fig. 33 Synthesis of racemic (±)-151 (RLA-4735) comprising both (R,R) and (S,S).

For the preparation of the (R,R)-152 enantiomer, the following conditions were employed: (a) 4-(benzyloxy)phenylboronic acid, 4 mol% acetylacetonatobis(ethylene)rhodium(I), 10 mol% R-BINAP, aqueous KOH, dioxane, 100 °C, 16 hours, 54%. For the corresponding (S,S)-152 enantiomer, the reaction was performed using S-BINAP under otherwise identical conditions, affording the desired product in 33% yield.

These finding underscore the therapeutic potential of (S,S)-151 and structurally related, further optimized analogs for the treatment of both artemisinin-sensitive and artemisinin-resistant (P. falciparum) malarial drugs. Ideally, next-generation trioxolane derivatives will demonstrate curative efficacy in infected patients irrespective of the resistance status, while also offering reduced dosing frequencies and favorable pharmacokinetic properties. Moreover, such compounds should be amenable to co-formulation with a suitable partner drug, facilitating the development of robust combination therapies for global malaria control and eradication efforts (Fig. 34).

Fig. 34 Iron reactivity of the trioxolane antimalarial is influenced by cyclohexane conformation cis-4′′ (155) and trans-3′′ (156).

These stabilizes the endoperoxide bridge (highlighted in green and red) within the optimal pharmacological range. Considering these findings, the artefenomel-adjacent chemo-type described herein exemplified by (S,S)-151 represents a significant advancement toward the development of next-generation antimalarial combination therapies capable of delivering reliable efficacy in the context of ART-R malaria.

22. Conclusions

Over the past three decades, Griesbaum co-ozonolysis has proven to be a versatile and highly selective method for constructing 1,2,4-trioxolane scaffolds with significant biomedical relevance. Its mild conditions, broad substrate scope, and predictable diastereoselectivity have enabled applications, ranging from antimalarial and anticancer agents to diagnostic probes. Recent advances, including continuous flow ozonolysis, stabilization strategies, and rational substituent design have further enhanced its utility and scalability.

Despite these achievements, challenges remain in expanding the substrate diversity, improving the regio and stereo control in complex systems, and deepening our mechanistic understanding. Further work in these areas is expected to unlock new opportunities in drug discovery and chemical biology, solidifying Griesbaum co-ozonolysis as a cornerstone transformation in modern synthetic chemistry.

23. Biography of Karl Griesbaum

Karl Griesbaum (1932–2019) was a German organic chemist and professor, born in Ettenheimmünster. He studied chemistry at the University of Karlsruhe and earned his PhD in 1960 under Rudolf Criegee, focusing on the mechanism of olefin ozonolysis. After a postdoctoral stint at Ohio State University (1960–1961), he worked as an industrial chemist at Esso Research (1961–1968) and then completed his habilitation in organic chemistry at Karlsruhe in 1968. He was appointed to the newly created Chair of Petrochemistry and Organic Technology at the University of Karlsruhe in 1972. Griesbaum's research spanned acetylenic chemistry, oxidation reactions, and petrochemical intermediates, with a notable focus on ozonolysis. He is best remembered for Griesbaum co-ozonolysis, first reported in 1995 as a selective method for synthesizing tetrasubstituted 1,2,4-trioxolanes via O-methyl ketoximes and carbonyl compounds in the presence of ozone.

Consent for publication

All the authors have given consent for the publication of the manuscript.

Author contributions

Muhammad Zain-Ul Abideen: literature survey, compilation, writing, structure drawing, Aamer Saeed: conceptualization, supervision, writing – review & editing final version; Mian Bilal Haider: editing, compilation, writing the final version, Ghulam Shabir: review and editing; Hesham R. El-Seedi: review & editing.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

References

1. P. R. Story, R. K. Murray and H. Youssefyen, Liebigs Ann., 1998, 90(9), 2194–2195.
2. P. S. Bailey, Ozonation in Organic Chemistry, Academic Press, New York, 1982, vol. II, p. 55.
3. (a) K. Griesbaum, B. Övez and T. S. Huh, Liebigs Ann., 1995, 1571–1574; (b) K. Griesbaum, P. Krieger-Beck and J. Beck, Chem. Ber., 1991, 124, 391–396.
4. Y. Tang, Y. Dong, J. M. Karle, C. A. DiTusa and J. L. Vennerstrom, J. Org. Chem., 2004, 69(19), 6470–6473.
5. K. Griesbaum, X. Liu, A. Kassiaris and M. Scherer, Liebigs Ann., 1997,(7), 1381–1390.
6. (a) K. Griesbaum, X. Liu and H. Henke, J. Org. Chem., 1998, 63(4), 1086–1089; (b) K. J. McCullough and M. Nojima, Peroxides from Ozonization, in Organic Peroxides, ed. W. Ando, Wiley, Chichester, 1992, pp. 661–728.
7. K. Griesbaum, Trends Org. Chem., 1997, 6, 145.
8. K. J. McCullough and M. Nojima, Peroxides from Ozonization, in Organic Peroxides, ed. W. Ando, Wiley, Chichester, 1992, pp. 661–728.
9. Y. S. Hon and S. L. Yan, Tetrahedron Lett., 1993, 34, 6591–6594.
10. Y. S. Hon and J.-L. Yan, Tetrahedron, 1997, 53(14), 5217–5232.
11. Y. S. Hon and J.-L. Yan, Tetrahedron Lett., 1994, 35(11), 1743–1746.
12. Y. Dong, K. Griesbaum and K. J. Mc Cullough, J. Chem. Soc., 1997, 1601–1604.
13. K. Griesbaum, Y. Dong and J. McCullough, J. Org. Chem., 1997, 62(18), 6129–6136.
14. K. Griesbaum and K. Schlindwein, J. Org. Chem., 1995, 60(24), 8062–8066.
15. Y. Tang, Y. Dong, J. M. Karle, C. A. DiTusa and J. L. Vennerstrom, J. Org. Chem., 2004, 69(19), 6470–6473.
16. W. Adam, H. J. Eggelte and A. Rodriguez, Synthesis, 1979, 5, 383–384.
17. J. L. Vennerstrom, Y. Dong and S. A. Charman, Acc. Chem. Res., 2010, 43(1), 145–154.
18. (a) K. Griesbaum, B. Övez and T. S. Huh, Liebigs Ann., 1995, 1571–1574; (b) K. Griesbaum, P. Krieger-Beck and J. Beck, Chem. Ber., 1991, 124, 391–396; (c) W. H. Bunnelle, Chem. Rev., 1991, 91, 335–362.
19. M. C. Kimber and D. S. Lee, Nat. Prod. Rep., 2024, 41(5), 813–833.
20. B. W. Gung, Chem. Rev., 1999, 99(5), 1377–1386.
21. J. A. Vroman, M. Alvim-Gaston and M. A. Avery, Curr. Pharm. Des., 1999, 5(2), 101–138.
22. G. Bez, B. Kalita, P. Sarmah, N. C. Barua and D. K. Dutta, Curr. Org. Chem., 2003, 7(12), 1231–1255.
23. W. H. Bunnelle, L. A. Meyer and E. O. Schlemper, J. Am. Chem. Soc., 1989, 111(19), 7612–7613.
24. Y. Tang, Y. Dong, S. Wittlin, S. A. Charman, J. Chollet, F. C. Chiu and J. L. Vennerstrom, Bioorg. Med. Chem. Lett., 2007, 17(5), 1260–1265.
25. O. B. Kazakova, E. Y. Yamansarov, O. S. Kukovinets, N. I. Medvedeva, D. V. Kazakov, O. K. Kornilov and K. Y. Suponitskii, Chem. Nat. Compd., 2011, 47, 738–740.
26. G. J. Ten Brink, I. W. C. E. Arends and R. A. Sheldon, Chem. Rev., 2004, 104(9), 4105.
27. G. G. Holmes and J. M. Smith, Org. Biomol. Chem., 2018, 16(22), 4530–4542.
28. Y. Tang, Y. Dong, J. M. Karle, C. A. DiTusa and J. L. Vennerstrom, J. Org. Chem., 2004, 69(19), 6470–6473.
29. F. W. Semmler, Ber. Dtsch. Chem. Ges., 1909, 42, 246.
30. O. B. Kazakova, D. V. Kazakov, E. Y. Yamansarov, N. Y. I. Medvedeva, G. A. Tolstikov, K. Y. Suponitsky and D. E. Arkhipov, Tetrahedron Lett., 2011, 52(9), 976–979.
31. K. Griesbaum, X. Liu and H. Henke, J. Org. Chem., 1998, 63(4), 1086–1089.
32. (a) Y. Dong, X. Wang, S. Kamaraj, V. J. Bulbule, F. C. Chiu, J. Chollet and J. L. Vennerstron, J. Med. Chem., 2017, 60(7), 2654–2668; (b) N. I. Medvedeva, O. B. Flekhter, L. A. Baltina, F. Z. Galin and G. A. Tolstikov, Chem. Nat. Compd., 2004, 40, 247–249.
33. Y. Tang, Y. Dong, J. M. Karle, C. A. DiTusa and J. L. Vennerstrom, J. Org. Chem., 2004, 69(19), 6470–6473.
34. Y. Dong, J. Chollet, H. Matile, S. A. Charman, F. C. K. Chiu, W. N. Charman, B. Scorneaux, H. Urwyler, J. Santo Tomas, C. Scheurer, C. Snyder, A. Dorn, X. Wang, J. M. Karle, Y. Tang, S. Wittlin, R. Brun and J. L. Vennerstrom, J. Med. Chem., 2005, 48(15), 4953–4961.
35. K. Griesbaum, B. Övez and T. S. Huh, Liebigs Ann., 1995, 1571–1574.
36. K. Griesbaum, X. Liu, A. Kassiaris and M. Scherer, Liebigs Ann., 1997,(7), 1381–1390.
37. (a) T. J. Fisher and P. H. Dussault, Tetrahedron, 2017, 73(30), 4233–4258; (b) Y. Dong, X. Wang, S. Kamaraj, V. J. Bulbule, F. C. Chiu, J. Chollet and J. L. Vennerstrom, J. Med. Chem., 2017, 60(7), 2654–2668.
38. A. O. Terent’ev, D. A. Borisov, V. A. Vil and V. M. Dembitsky, Beilstein J. Org. Chem., 2014, 10(1), 34–114.
39. G. Gabriel dos Passos Gomes, I. A. Yaremenko, P. S. Radulov, R. A. Novikov, V. V. Chernyshev, A. A. Korlyukov, G. I. Nikishin, I. V. Alabugin and A. O. Terent’ev, Angew. Chem., Int. Ed., 2017, 56, 4955–4959.
40. J. L. Vennerstrom, S. Arbe-Barnes, R. Brun, S. A. Charman, F. C. Chiu, J. Chollet and W. N. Charman, Nature, 2004, 430(7002), 900–904.
41. O. B. Kazakova, E. F. Khusnutdinova, G. A. Tolstikov and K. Y. Suponitsky, Russ. J. Bioorg. Chem., 2010, 36(4), 512–515.
42. J. L. Vennerstrom, S. Arbe-Barnes, R. Brun, S. A. Charman, F. C. Chiu, J. Chollet and W. N. Charman, Nature, 2004, 430(7002), 900–904.
43. S. D. Fontaine, A. G. DiPasquale and A. R. Renslo, Org. Lett., 2014, 16(21), 5776–5779.
44. Y. Dong, X. Wang, S. Kamaraj, V. J. Bulbule, F. C. Chiu, J. Chollet and J. L. Vennerstrom, J. Med. Chem., 2017, 60(7), 2654–2668.
45. Y. Dong, S. Wittlin, K. Sriraghavan, J. Chollet, S. A. Charman, W. N. Charman, C. Scheurer, H. Urwyler, J. Santo Tomas, C. Snyder, D. J. Creek, J. Morizzi, M. Koltun, H. Matile, X. Wang, M. Padmanilayam, Y. Tang, A. Dorn, R. Brun and J. L. Vennerstrom, J. Med. Chem., 2010, 53(1), 481–491.
46. M. T. Klope, J. A. Tapia Cardona, J. Chen, R. L. Gonciarz, K. Cheng, P. Jaishankar and A. R. Renslo, ACS Med. Chem. Lett., 2024, 15(10), 1764–1770.
47. Y. Dong, Y. Tang, J. Chollet, H. Matile, S. Wittlin, S. A. Charman and J. L. Vennerstrom, Bioorg. Med. Chem., 2006, 14(18), 6368–6382.
48. A.-J. Lin, D. L. Klayman and W. K. Milhous, J. Med. Chem., 1987, 30(11), 2147–2150.
49. L. Zhou, A. Alker, A. Ruf, X. Wang, F. C. Chiu, J. Morizzi and J. L. Vennerstrom, Bioorg. Med. Chem. Lett., 2008, 18(5), 1555–1558.
50. (a) W. Adcock and N. A. Trout, Chem. Rev., 1999, 99, 1415; (b) B. W. Gung, Chem. Rev., 1999, 99, 1377.
51. E. Tsandi, C. G. Kokotos, S. Kousidou, V. Ragoussis and G. Kokotos, Tetrahedron, 2009, 65(7), 1444–1449.
52. Y. Dong, S. Wittlin, K. Sriraghavan, J. Chollet, S. A. Charman, W. N. Charman, C. Scheurer, H. Urwyler, J. Santo Tomas, C. Snyder, D. J. Creek, J. Morizzi, M. Koltun, H. Matile, X. Wang, M. Padmanilayam, Y. Tang, A. Dorn, R. Brun and J. L. Vennerstrom, J. Med. Chem., 2010, 53(1), 481–491.
53. T. Itoh, T. Wanibe and S. Iwatsuki, J. Polym. Sci., 1996, 34(6), 963–969.
54. Y. Dong and J. L. Vennerstrom, J. Org. Chem., 1998, 63(23), 8582–8585.
55. (a) C. Battilocchio, I. R. Baxendale, M. Biava, M. O. Kitching and S. V. Ley, Org. Process Res. Dev., 2012, 16(5), 798–810; (b) M. O'Brien, I. R. Baxendale and S. V. Ley, Org. Lett., 2010, 12(7), 1596–1598.
56. T. Ouchi, C. Battilocchio, J. M. Hawkins and S. V. Ley, Org. Process Res. Dev., 2014, 18(11), 1560–1566.
57. (a) K. Borstnik, I. Paik, T. Shapiro and G. H. Posner, Int. J. Parasitol., 2002, 32(13), 1661; (b) S. H. Lau, A. Galván, R. R. Merchant, C. Battilocchio, J. A. Souto, M. B. Berry and S. V. Ley, Org. Lett., 2015, 17(13), 3218–3221.
58. S. H. Lau, A. Galván, R. R. Merchant, C. Battilocchio, J. A. Souto, M. B. Berry and S. V. Ley, Org. Lett., 2015, 17(13), 3218–3221.
59. S. D. Fontaine, B. Spangler, J. Gut, E. M. Lauterwasser, P. J. Rosenthal and A. R. Renslo, ChemMedChem, 2015, 10(1), 47–51.
60. A. T. Aron, M. O. Loehr, J. Bogena and C. J. Chang, J. Am. Chem. Soc., 2016, 138(43), 14338–14346.
61. S. V. Torti and F. M. Torti, Nat. Rev. Cancer, 2013, 13(5), 342–355.
62. S. V. Torti and M. F. Torti, Cancer Res., 2011, 71(95), 1511–1514.
63. B. Spangler, C. W. Morgan, S. D. Fontaine, M. N. Vander Wal, C. J. Chang, J. A. Wells and A. R. Renslo, Nat. Chem. Biol., 2016, 12(9), 680–685.
64. A. R. Bogdan, M. Miyazawa, K. Hashimoto and Y. Tsuji, Trends Biochem. Sci., 2016, 41(3), 274–286.
65. S. J. Dixon and B. R. Stockwell, Nat. Chem. Biol., 2014, 9–17.
66. C. Zhang and F. Zhang, Protein Cell, 2015, 6(2), 88–100.
67. W. S. Yang and B. R. Stockwell, Chem. Biol., 2008, 15(3), 234–245.
68. J. Chen, R. L. Gonciarz and A. R. Renslo, RSC Adv., 2021, 11(54), 34338–34342.
69. J. Matsumoto, Y. Dong and J. L. Vennerstrom, Org. Biomol. Chem., 2018, 16(12), 2187–2195.
70. S. A. Charman, S. Arbe-Barnes, I. C. Bathurst, R. Brun, M. Campbell, W. N. Charman and J. L. Vennerstrom, Proc. Natl. Acad. Sci. U. S. A., 2011, 108(11), 4400–4405.
71. (a) A. T. Aron, A. G. Reeves and C. J. Chang, Curr. Opin. Chem. Biol., 2018, 43, 113; (b) C. J. Chang, Nat. Chem. Biol., 2015, 11(10), 744; (c) K. J. Bruemmer, S. W. M. Crossley and C. J. Chang, Angew. Chem., Int. Ed., 2020, 59(33), 13734; (d) T. Hirayama, Free Radical Biol. Med., 2019, 133, 38–45.
72. B. Spangler, C. W. Morgan, S. D. Fontaine, M. N. VanderWal, C. J. Chang, J. A. Wells and A. R. Renslo, Nat. Chem. Biol., 2016, 12(9), 680–685.
73. A. T. Aron, M. C. Heffern, Z. R. Lonergan, M. N. Vander Wal, B. R. Blank, B. Spangler, Y. Zhang, H. M. Park, A. Stahl, A. R. Renslo, E. P. Skaar and C. J. Chang, Proc. Natl. Acad. Sci. U. S. A., 2017, 114(48), 12669–12674.
74. S. Xu, H. W. Liu, L. Chen, J. Yuan, Y. Liu, L. Teng, S. Y. Huan, L. Yuan, X. B. Zhang and W. Tan, J. Am. Chem. Soc., 2020, 142(5), 2129–2133.
75. N. Zhao, Y. Huang, Y. H. Wang, R. K. Muir, Y. C. Chen, N. Hooshdaran, J. Wei, P. Viswanath, Y. Seo, D. Ruggero, A. R. Renslo and M. J. Evans, J. Nucl. Med., 2021, 62(7), 949–955.
76. R. K. Muir, N. Zhao, J. Wei, Y. H. Wang, A. Moroz, Y. Huang, Y. C. Chen, R. Sriram, J. Kurhanewicz, D. Ruggero, A. R. Renslo and M. J. Evans, ACS Cent. Sci., 2019, 5(4), 727–736.
77. Y. C. Chen, J. A. Oses-Prieto, L. E. Pope, A. L. Burlingame, S. J. Dixon and A. R. Renslo, J. Am. Chem. Soc., 2020, 142(45), 19085–19093.
78. S. D. Fontaine, A. G. DiPasquale and A. R. Renslo, Org. Lett., 2014, 16(21), 5776–5779.
79. J. Wu, X. Wang, F. C. K. Chiu, C. Haberli, D. M. Shackleford, E. Ryan, S. Kamaraj, V. J. Bulbule, A. I. Wallick, Y. Dong, K. L. White, P. H. Davis, S. A. Charman, J. Keiser and J. L. Vennerstrom, J. Med. Chem., 2020, 63(7), 3723–3736.
80. E. Y. Yamansarov, O. B. Kazakova, N. I. Medvedeva, D. V. Kazakov, O. S. Kukovinets and G. A. Tolstikov, Russ. J. Org. Chem., 2014, 50(7), 1043–1047.
81. O. B. Kazakova, E. F. Khusnutdinova, A. V. Petrova, E. Y. Yamansarov, A. N. Lobov, A. A. Fedorova and K. Y. Suponitsky, J. Nat. Prod., 2019, 82(9), 2550–2558.
82. B. R. Blank, R. L. Gonciarz, P. Talukder, J. Gut, J. Legac, P. J. Rosenthal and A. R. Renslo, ACS Infect. Dis., 2020, 6(7), 1827–1835.
83. E. M. W. Lauterwasser, S. D. Fontaine, H. Li, J. Gut, K. Katneni, S. A. Charman, P. J. Rosenthal, M. Bogyo and A. R. Renslo, ACS Med. Chem. Lett., 2015, 6(11), 1145–1149.
84. P. S. Sijwali and P. J. Rosenthal, Proc. Natl. Acad. Sci. U. S. A., 2004, 101(13), 4384–4389.
85. S. D. Fontaine, B. Spangler, J. Gut, E. M. Lauterwasser, P. J. Rosenthal and A. R. Renslo, ChemMedChem, 2015, 10(1), 47–51.
86. (a) B. R. Blank, P. Talukder, R. K. Muir, E. R. Green, E. P. Skaar and A. R. Renslo, ACS Infect. Dis., 2019, 5(8), 1366–1375; (b) B. R. Blank, J. Gut, P. J. Rosenthal and A. R. Renslo, ACS Med. Chem. Lett., 2023, 14(4), 493–498.
87. S. S. Mahajan, E. Deu, E. M. W. Lauterwasser, M. J. Leyva, J. A. Ellman, M. Bogyo and A. R. Renslo, ChemMedChem, 2011, 6(3), 415–419.
88. C. M. Woodley, G. L. Nixon, N. Basilico, S. Parapini, W. D. Hong, S. A. Ward, G. A. Biagini, S. C. Leung, D. Taramelli, K. Onuma, T. Hasebe and P. M. O'Neill, ACS Med. Chem. Lett., 2021, 12(7), 1077–1085.
89. B. R. Blank, J. Gut, P. J. Rosenthal and A. R. Renslo, ACS Med. Chem. Lett., 2023, 14(4), 493–498.
90. (a) M. T. Klope, J. A. Tapia Cardona, J. Chen, R. L. Gonciarz, K. Cheng, P. Jaishankar and A. R. Renslo, ACS Med. Chem. Lett., 2024, 15(10), 1764–1770; (b) M. T. Klope, P. Talukder, B. R. Blank, S. Chelebieva, J. Chen, S. D. Fontaine and A. R. Renslo, bioRxiv, 2025, preprint,  DOI:10.1101/2025.01.02.631109.

---