Thermodynamic stability and anticancer activity of 2,3-unsaturated and 2,3-dideoxy O³-betulin glycosides

Abstract

Derivatization of natural compounds with established cytotoxic activity is a common strategy in development of new anticancer agents. This approach is exemplified by betulin, a triterpenoid from birch bark that exhibits pronounced antiproliferative effects against various cancer cell lines. We used the Ferrier rearrangement to synthesize 2,3-unsaturated O³-betulin glycosides, which were converted into 2,3-dideoxy analogs via regioselective hydrogenation. The conformations of the 2,3-unsaturated O³-betulin glycosides are analyzed in relation to the stereoselectivity of the Ferrier rearrangement. Cytotoxicity studies of the synthesized O³-betulin glycosides were performed against breast cancer cells (MCF7 cell line), prostate cancer cells (PC3 cell line), and human keratinocytes (HaCaT cell line). These reveal that glycosylation of betulin with 2,3-unsaturated and 2,3-dideoxy sugars derived from D-xylose, L-arabinose, and L-fucose clearly enhances its activity against MCF7 cells. This improvement is accompanied by high selectivity of the active compounds. Structure-activity relationship analysis leads to the conclusion that enhanced activity is associated with the absence of a terminal hydroxymethyl group. Microscopic images of MCF7 cells treated with the synthesized glycosides illustrate vacuolization, membrane blebbing, and apoptotic disintegration. Presented studies show the anticancer potential of some of the O³-betulin glycosides and provide insight into the stereoselectivity of the Ferrier rearrangement.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-52341-w.

Introduction

In the search for new and effective pharmaceuticals, natural compounds with defined therapeutic profiles are often explored^1,2. Among these, our attention has been drawn to betulin ((3β)-lup-20(29)-en-3,28-diol) (BtOH), a triterpenoid occurring in the bark of Betula species, which has been used for centuries in traditional Chinese medicine. Owing to its diverse pharmacological properties, betulin has been the subject of several review articles^3–8. Betulin is known for its antioxidant⁹, neuroprotective¹⁰, gastroprotective, and hepatoprotective properties¹¹. Betulin-containing extracts from birch bark influence wound-healing processes under diabetic conditions¹², exhibit partial inhibition of α-amylase activity¹³, inhibit DNA topoisomerase activity¹⁴. Betulin isolated from P. incarnata exhibits anti-neuroinflammatory potential¹⁵. However, this trterpenoid has attracted significant interest among researchers mainly due to its anticancer properties. Betulin was reported to exhibit activity against the PANC-1 human pancreatic cancer cell line¹⁶, Hep-2 laryngeal cancer cells, THP-1 acute monocytic leukemia cells, and HeLa cervical cancer cells¹⁷, as well as against gastrointestinal cancers¹⁸. No wonder that numerous derivatives of this triterpenoid have been synthesized in the search for new and effective anticancer agents. Among them were betulin phosphate¹⁹, dicarboxylic acid²⁰, hydrazide-hydrazone²¹, thiazole²², and amino acid derivatives²³, as well as glycoconjugates of betulin with a succinic linker and a 1,2,3-triazole ring²⁴ tested in vitro for their cytotoxicity. Antineoplasic activities of betulin complex in γ-cyclodextrin derivatives were tested both in vitro and in vivo tumor models²⁵. Since glycosylated forms of betulin occur in nature, a number of its glycosides have been synthesized and their cytotoxicity has been determined²⁶. These studies included betulin O³-β-D-glucopyranoside, O³-α-D-mannopyranoside, and O³-α-L-arabinofuranoside²⁷, as well as O³-β-D-glucosaminopyranoside and O³-β-D-galactosaminopyranoside²⁸. Betulin glycosylated at both the O³ and O²⁸ positions (bidesmosidic) with β-D-glucopyranose, α-L-arabinopyranose, and/or α-L-rhamnopyranose has also been synthesized and tested against cancer cells^27,29.

This work presents a new class of O³-betulin glycosides in which the sugar ring has an atypical 2,3-unsaturated or 2,3-dideoxy structure. The effect of betulin glycosylation with sugars containing a 2,3-unsaturated ring on its biological properties has not been previously studied. However, this structural motif occurs in several biologically active compounds, such as the natural nucleoside antibiotic blasticidin S³⁰, the synthetic nucleoside Stavudine with anti-HIV activity³¹, and the 2,3-unsaturated glycosides of genistein, which were active towards cancer cells³². Additionally, glycosylation of betulin with a sugar ring containing a double bond between the C2 and C3 carbon atoms creates opportunities for further structural modification of such derivatives. These include various types of addition reactions, nucleophilic substitution at the allylic position, [3,3]-sigmatropic rearrangement, oxidative transformations and cycloaddition reactions³³. Double bond hydrogenation was employed by us to transform 2,3-unsaturated glycosides into 2,3-dideoxy glycosides in order to examine how the lack of the hydroxyl groups would affect the properties of betulin derivatives. For the glycosylation reactions we selected six monosaccharides differing in configuration and in their terminal groups. Our aim was to determine how these factors influence both the anticancer properties of betulin glycosides and the course of the reaction. For all synthesized O³-betulin glycosides, their effects on the viability of breast cancer (MCF7) and prostate cancer (PC3) cell lines, as well as normal human keratinocytes (HaCaT), were investigated. These studies demonstrate the potential of 2,3-unsaturated and 2,3-dideoxy betulin glycosides as candidates for anticancer agents. The influence of the structures of the synthesized compounds on their stability, the stereoselectivity of the reactions performed, and their anticancer activity is discussed.

Results and discussion

Synthesis

To obtain 2,3-unsaturated glycosides, the Ferrier rearrangement was employed³⁴. For this purpose, six acetylated glycals, namely D-glucal (1), D-galactal (2), L-rhamnal (3), L-fucal (4), D-xylal (5), and L-arabinal (6), were prepared³⁵. These structurally diverse glycals were subjected to reaction with O²⁸-acetylbetulin (AcO²⁸BtOH) in the presence of BF₃•OEt₂, followed by de-O-acetylation with MeONa in MeOH (Fig. 1). In this way, 2,3-unsaturated O³-betulin glycosides (7–12) were synthesized. Regioselective hydrogenation of compounds 7–12 using diazene generated in situ by the reaction of hydrazine with H₂O₂, afforded the corresponding 2,3-dideoxy O³-betulin glycosides (13–18).

Fig. 1.

Reactions carried out and the atom numbering system applied.

Stereoselectivity of the Ferrier rearrangement

The reaction of glycals 1–6 with O²⁸-acetylbetulin in the presence of BF₃•OEt₂ was highly stereoselective, yielding 2,3-unsaturated glycosides with an α-anomeric configuration in the case of hexoses (7–10) and a predominant β-anomeric configuration (α:β ~ 1:5.7) in the case of pentoses (11 and 12) (Fig. 2). To determine the configuration of the anomeric carbon atom in 2,3-unsaturated pyranosides, it is necessary to assign the appropriate conformation of the sugar ring (⁰H₅ or ⁵H₀). This assignment was based on the measured J_4,5 coupling constant. A J_4,5 coupling constant in the range of 8–9 Hz (Fig. 2) indicates a pseudoaxial orientation of the H4 proton and an axial orientation of the H5 proton. Such an arrangement occurs in the ⁰H₅ conformation of 7α and 11α and in the ⁵H₀ conformation of 9α and 12α. In contrast, a J_4,5 value in the range of 0–2 Hz indicates a pseudoequatorial orientation of the H4 proton and an axial orientation of the H5 proton. The pseudoequatorially oriented H4 proton is nearly antiperiplanar to the endocyclic oxygen atom, resulting in a very small or undetectable coupling with axially oriented H5 proton³⁶. Such an arrangement occurs in the ⁰H₅ conformation of 8α and 12β and in the ⁵H₀ conformation of 10α and 11β. The J_3,4 coupling constants further support the conformational assignments of glycosides 7–12. This coupling constant (Fig. 2) is close to zero when the H4 proton adopts a pseudoaxial orientation (7α, 9α, 11α, 12α) and is 5.49–5.19 Hz when the H4 proton is pseudoequatorially oriented (8α, 10α, 11β, 12β). According to the findings of Ferrier and Sankey³⁷, the conformation adopted by 2,3-unsaturated pyranosides allows the configuration of the anomeric carbon atom to be inferred. This is because the C1‒O1 bond in such glycosides tends to adopt a pseudoaxial orientation due to the allylic effect. Thus, the ⁰H₅ conformation in 2,3-unsaturated pyranosides of the D series indicates an α configuration at the anomeric carbon atom, whereas in the L series it corresponds to a β configuration. Notice that the J_1,2 coupling constant in 2,3-unsaturated pyranosides, unlike in their saturated analogues, is not diagnostic of the configuration of the anomeric carbon atom. Both the α and β anomers of 2,3-unsaturated pyranosides exhibit a pseudoequatorial orientation of the anomeric proton. Therefore, the measured J_1,2 values for both anomers fall within the range of 0–3 Hz. In the present study, the configuration of the anomeric carbon atom in 2,3-unsaturated pyranosides (7–12) was additionally confirmed by analysis of their saturated analogues (13–18).

Fig. 2.

Structures of substrates (1–6) and products (7–18) with reaction yields given in brackets. The configuration and conformation of the sugar ring were established based on the indicated ³J_{H, H} coupling constants. Stability factors are highlighted using colors: red - equatorial orientation of the substituent, green - anomeric effect, blue - allylic effect.

As previously demonstrated³⁸, the stereoselectivity of the Ferrier rearrangement is governed by the thermodynamic stability of the resulting products. This stability is influenced by three factors, listed in order of importance: (i) the preference for an equatorial orientation of the substituent at the C5 carbon atom (red, Fig. 2), (ii) the anomeric effect, which in the case of 2,3-unsaturated glycosides means the tendency for a pseudoaxial orientation of the aglycone (green), and (iii) the allylic effect, defined as the preference for a pseudoaxial orientation of the OR substituent (here R = H) at the C4 carbon atom (blue). Only the α configuration of the anomeric carbon atom in 2,3-unsaturated hexopyranosides allows all three stability factors to act simultaneously, as in the case of 8α and 10α, or at least for the two most important factors to operate together, as in the case of 7α and 9α. The transition from the α to the β anomer in 2,3-unsaturated hexopyranoses requires forfeiting either the equatorial orientation of the C5 substituent or the anomeric effect. Since both effects are significant, the absence of either reduces the stabilization of the β anomer. Therefore, in the case of hexoses, the Ferrier rearrangement leads exclusively, or almost exclusively, to 2,3-unsaturated hexopyranosides with an α-anomeric configuration. Notice that these glycosides adopt a conformation that allows the stabilizing factors to operate simultaneously. This is the ⁰H₅ conformation in the case of D-series hexopyranosides (7α, 8α) and the ⁵H₀ conformation in the case of L-series hexopyranosides (9α, 10α). Pentopyranoses lack a substituent at the C5 carbon atom; therefore, the stability factor associated with this substituent (red, Fig. 2) does not apply to them. In their case, a β anomeric configuration, combined with an appropriate ring conformation, allows both the anomeric effect (green) and the allylic effect (blue) to operate. In the case of the α anomer, neither the ⁰H₅ nor the ⁵H₀ conformation allows both stability factors to act simultaneously. Ultimately, the α anomer adopts a conformation in which only the anomeric effect operates. This is the ⁰H₅ conformation in the case of D series pentopyranoside (11α) and the ⁵H₀ conformation in the case of L series pentopyranoside (12α). The fact that the α anomer is stabilized solely by the anomeric effect, whereas the β anomer benefits from both the anomeric and allylic effects, explains why the β anomer predominates among the Ferrier rearrangement products of pentopyranosides.

Course of hydrogenation

Diazene, generated in situ by the reaction of hydrazine with H₂O_2, was employed to hydrogenate the double bond in 2,3-unsaturated O³-betulin glycosides (7–12). Under these reaction conditions, no hydrogenolysis of the glycosidic bond occurs, which can take place when hydrogen is used in the presence of a catalyst^38,39. Glycosides 7–12 contain double bonds in both the sugar moiety and the aglycone. We found that performing the reaction at 5 °C enabled the exclusive formation of 2,3-dideoxy glycosides (13–18) while preserving the double bond in the betulin moiety. The conformation of glycosides 13–18 is determined by two stability factors: the equatorial orientation of the substituents (red, Fig. 2), which is particularly important for the terminal group, and the anomeric effect (green). 2,3-Dideoxy α-hexopyranosides adopt a conformation that allows the simultaneous equatorial orientation of the terminal substituent and axial orientation of the aglycone. This is the ⁴C₁ conformation in the case of D-series hexopyranosides (13α and 14α) and the ¹C₄ conformation in the case of L-series hexopyranosides (15α and 16α). When the pyranose ring lacks a terminal substituent, the anomeric effect governs the conformation of 2,3-dideoxy β-pentopyranosides. Accordingly, glycoside 17β adopts the ¹C₄ conformation, whereas glycoside 18β adopts the ⁴C₁ conformation.

In assigning the conformations adopted by glycosides 14α, 16α, 17β, and 18β, we were guided by observations reported in our previous paper³⁶. For these glycosides, in the respective chair conformation, the H4 proton is oriented antiperiplanarly to the ring oxygen atom, which results in the lack of coupling between the H4 and H5 protons.

Cytotoxic activity

The MTT assay⁴⁰ was performed to evaluate the effects of betulin (BtOH) and its 2,3-unsaturated (7–12) and 2,3-dideoxy (13–18) O³-glycosides on the viability of human keratinocytes (HaCaT line), breast cancer cells (MCF7 line), and prostate cancer cells (PC3 line). The cell lines were exposed for 48 h to a medium containing the tested compound over a wide range of concentrations. The viability of HaCaT (A), MCF7 (B), and PC3 (C) cells is presented in Fig. S1. The IC₅₀ values (the concentration of compound causing a 50% reduction of cell viability) and the selectivity indexes (SI), defined as the ratio of the IC₅₀ for the normal cell line (HaCaT) to that for the cancer cell line (PC3 or MCF7), are summarized in Table 1.

Table 1.

IC₅₀ values (µM) of betulin (BtOH) and its O³-glycosides (7α-18β) towards normal (HaCaT) and cancer (PC3 and MCF7) cells, together with selectivity indexes (SI).

Comp.	IC50 [µM]			SI
	HaCaT	MCF7	PC3	MCF7	PC3
BtOH	2.72	2.10	6.16	1.29	0.44
7α	16.13	17.31	20.42	0.93	0.79
8α	15.65	14.83	16.25	1.06	0.96
9α	14.50	23.76	20.10	0.61	0.72
10α	2.28	1.18	10.05	1.93	0.23
11α	16.54	13.86	22.20	1.19	0.75
11β	1.64	0.56	9.11	2.91	0.18
12β	2.12	0.60	7.34	3.56	0.29
13α	19.43	19.56	18.51	0.99	1.05
14α	16.89	12.50	16.65	1.35	1.01
15α	> 50.00	> 50.00	39.92	—	—
16α	7.13	8.53	11.38	0.84	0.63
17β	1.77	1.20	8.51	1.47	0.21
18β	2.18	1.91	12.41	1.14	0.18

General considerations

As shown in Table 1, 2,3-unsaturated O³-betulin glycosides such as 10α, 11β and 12β as well as 2,3-dideoxy O³-betulin glycosides such as 16α, 17β and 18β are effectively active against prostate cancer cells (PC3 line) at concentrations ranging from 7.34 to 12.41 µM. The remaining glycosides, with the exception of 15α, also show activity against PC3 cancer cells, but at higher concentrations of 16.25–22.20 µM. Compared with betulin itself (IC₅₀ 6.16 µM), it is evident that the glycosylation did not enhance the cytotoxicity of betulin against PC3 cancer cells. Instead, glycosylation slightly increased the cytotoxicity of betulin toward human keratinocytes (HaCaT). In their case, the IC₅₀ values ​​are 2.72 µM for betulin (BtOH), 2.28 µM for 10α, 1.64 µM for 11β, 2.12 µM for 12β, 1.77 µM for 17β, and 2.18 µM for 18β. This is reflected in the SI indexes for PC3 cells, which are 0.44 for betulin (BtOH), 0.23 for 10α, 0.18 for 11β, 0.29 for 12β, 0.21 for 17β and 0.18 for 18β.

Presented glycosylation that did not enhance the cytotoxicity of betulin against PC3 prostate cancer cells clearly improved the activity of betulin against breast cancer cells (MCF7 line). In the case of MCF7 cells, the IC₅₀ value is 2.10 µM for betulin (BtOH), 1.18 µM for 10α, 0.56 µM for 11β, 0.60 µM for 12β, 1.20 µM for 17β and 1.91 µM for 18β. Importantly, the SI indexes of these glycosides related to the MCF7 cells are in majority more favorable than that of betulin itself being 1.29 (BtOH), 1.93 (10α), 2.91 (11β), 3.56 (12β), 1.47 (17β), and 1.14 (18β), respectively. Thus, the listed O³-betulin glycosides, particularly the 2,3-unsaturated 11β and 12β, are not only highly cytotoxic but also highly selective, preferentially targeting MCF7 breast cancer cells at low concentrations while exerting reduced effects on human keratinocytes (HaCaT line) and prostate cancer cells (PC3 line).

Discussion on SARs

Considering the structure-activity relationships (SARs), it is evident that the incorporation of a 2,3-unsaturated 6-deoxyhexopyranose derived from L-fucose (10α) as well as 2,3-unsaturated pentopyranoses, derived from D-xylose (11β) and L-arabinose (12β), led to a significant enhancement of betulin’s anticancer activity against MCF7 breast cancer cells. Among the 2,3-dideoxy glycosides, derivatives of pentopyranoses (17β and 18β) again exhibited the highest activity against MCF7 cancer cells. It is noteworthy that the sugar moieties of 11β and 12β, as well as those of 17β and 18β, respectively, constitute pairs of enantiomers. In contrast, the incorporation of both 2,3-unsaturated and 2,3-dideoxy hexopyranoses derived from D-glucose (7α and 13α) and D-galactose (8α and 14α) had a detrimental effect on the anticancer activity of betulin. These poorly active derivatives differ from the more effective D-xylose and L-arabinose derivatives by the presence of a terminal hydroxymethyl group. Taking into account that the D-xylose and L-arabinose derivatives also exhibit favorable SI values, it can be concluded that the absence of a terminal hydroxymethyl group enhances both the anticancer activity and selectivity of the presented compounds. A similar conclusion was previously drawn from studies of analogous diosgenin glycosides³⁸.

The 2,3-unsaturated ring derived from L-fucose in 10α also exerts a beneficial effect on the activity of betulin against MCF7 cancer cells. L-Fucose is a 6-deoxyhexose commonly incorporated into human glycoproteins and glycolipids in the form of an α glycoside. It is typically found at the terminal positions of glycoconjugates, where it functions as a sugar determinant involved in cell recognition and adhesion-signaling pathways⁴¹. These properties of L-fucose may explain the high activity of glycoside 10α, particularly that it possesses the α configuration of the anomeric carbon atom, which is characteristic of biologically active L-fucose residues.

The enhancement of the anticancer properties of betulin (BtOH) upon its conversion into 2,3-unsaturated (10α, 11β, 12β) and 2,3-dideoxy glycosides (17β, 18β) cannot be attributed to the increased hydrophilicity typically associated with glycosylation, since only one hydroxyl group from the sugar moiety replaces a hydroxyl group originally present in betulin. It appears that this improved activity against MCF7 cancer cells is rather attributed to the introduction of an additional rigid ring lacking a terminal hydroxymethyl group into the betulin structure. Both the 2,3-unsaturated and 2,3-dideoxy sugar rings attached to betulin are relatively rigid, but they differ structurally. Comparing pairs of glycosides with the same sugar ring configuration, e.g., 10α and 16α (α-L-threo), 11β and 17β (β-D-glycero), 12β and 18β (β-L-glycero), it can be concluded that the 2,3-unsaturated ring exerts a stronger positive influence on the anticancer properties of betulin glycosides than the 2,3-dideoxy ring (Table 1). What differentiates these two groups of glycosides with the same configuration is the conformation of the sugar ring. In the more active 2,3-unsaturated glycosides, the sugar ring adopts a half-chair conformation, ⁵H₀ in the case of 10α and 11β and ⁰H₅ in the case of 12β. In turn, in the 2,3-dideoxy glycosides, the sugar ring adopts a chair conformation, ¹C₄ in the case of 17β and ⁴C₁ in the case of 18β. The conformations adopted by the sugar rings in 11β and 12β, as well as in 17β and 18β, are mirror images, since the sugar moieties in each pair are enantiomers. As can be seen, belonging to the D series (11β and 17β) or L series (12β and 18β) does not differentiate the activity of either the 2,3-unsaturated (11β and 12β) or the 2,3-dideoxy (17β and 18β) O³-betulin glycosides against MCF7 breast cancer cells.

Discussion of morphological changes in cancer cells

According to the results presented in Fig. S1, the drop of viability of MCF7 cancer cells treated with betulin (BtOH) and its O³-glycosides (7α–18β) at concentrations of 25 µM and 50 µM, respectively, is high. This is further illustrated by the pronounced morphological changes observed in MCF7 cells following exposure to betulin (BtOH) and its O³-glycosides (7α–18β). Selected images showing signs of cell stress, such as vacuolization, membrane blebbing and/or shrinkage, and cell death are presented in Fig. 3. Comprehensive images depicting the morphological changes after 24 h and 48 h exposure to all tested compounds at concentrations of 25 µM and 50 µM, respectively, are provided in Table S1. A list of the observed morphological alterations in MCF7 cancer cells is given in Table S2. These results indicate that each of the tested compounds, at both concentrations and after both exposure times, induces substantial cellular stress and hallmarks consistent with apoptosis-like cell death in MCF7 cancer cells. Cell death, particularly after 48 h exposure to betulin or its O³-glycosides, is in most cases accompanied by cell shrinkage and/or membrane blebbing, a process in which bulges form on the cell membrane due to its breakdown. Blebbing is known to be associated with dynamic cell reorganization and, in the case of cancer cells, can promote their invasion into the bloodstream without necessarily leading to apoptosis⁴². However, when apoptosis does occur, blebbing is an invariable hallmark of the process⁴³.

Fig. 3.

Microscopic images of MCF7 cells after 24 h exposure to O³-betulin glycosides showing stress-associated morphological changes: vacuolisation induced by 9α (25 µM), cell blebbing and/or shrinkage induced by 11β (25 µM), and cell death induced by 10α (50 µM).

In some cases, in addition to MCF7 cancer cell death, vacuolization is observed either instead of or alongside blebbing and/or shrinkage. This phenomenon is primarily seen after 24 h exposure of the cancer cells to 8α, 9α, 11α, 14α–16α (Table S2). Cytoplasmic vacuolization is a morphological alteration involving the formation of vacuole-like structures. This process is a cellular response to stress caused by exposure to toxins, therapeutic agents⁴⁴, heat⁴⁵ or high salt concentrations⁴⁶. Vacuolization can lead to cell damage or death; however, its precise role in these processes remains unclear.⁴⁷ Typically, the vacuolization observed here (Fig. 3, Table S2) occurred after short-term (24 h) treatment with glycosides characterized by relatively low cytotoxicity. After prolonged exposure (48 h) to betulin and its O³-glycosides in MCF7 cancer cells, vacuolization was no longer observed, with only blebbing and cell death remaining, regardless of the concentration used. Glycoside 15α is an exception, as it exhibits the lowest cytotoxicity among the compounds tested.

Summary

A series of 2,3-unsaturated O³-betulin glycosides was synthesized via the Ferrier rearrangement. The stereoselectivity of the reaction was shown to depend closely on the conformational stability of the products. In the case of hexopyranoses, α glycosides were formed exclusively, whereas in the case of pentopyranoses, β anomers dominated among the reaction products. Regioselective hydrogenation of the 2,3-unsaturated glycosides afforded a series of 2,3-dideoxy O³-betulin glycosides.

Cytotoxicity studies of synthesized compounds revealed that glycosylation did not enhance the activity of betulin against prostate cancer cells (PC3 line). The situation is different for breast cancer cells (MCF7 line). Some of the glycosides tested exhibited activity at much lower concentrations than betulin itself. In particular, the 2,3-unsaturated glycosides derived from D-xylose (11β) and L-arabinose (12β) were highly effective, with comparable IC₅₀ values of 0.56 µM and 0.60 µM, respectively. Additionally, the 2,3-unsaturated glycoside derived from L-fucose (10α) and 2,3-dideoxy glycosides derived from D-xylose (17β) and L-arabinose (18β) also demonstrated notable activity against MCF7 cancer cells. Examination of the active O³-betulin glycosides suggests that the absence of a terminal hydroxymethyl group is beneficial for their activity.

Microscopic images showing morphological changes of MCF7 cancer cells after exposure to betulin and its O³-glycosides indicate that all tested compounds induce substantial cellular stress, often manifested by membrane blebbing and hallmarks of cell death. Cytoplasmic vacuolization is observed primarily after short-term (24 h) treatment with the less cytotoxic compounds.

In vivo animal model should be performed to accurately evaluate the obtained results.

Experimental

Synthetic procedures and results of analysis

General methods

Solvents and chemical reagents were purchased and used without further purification. The ¹H and ¹³C NMR spectra were recorded on the Bruker Avance III 500 MHz (500.13/125.76 MHz), using CDCl₃ as the solvent with Me₄Si as an internal standard. Structural assignments were based on the COSY and HSQC techniques. The digital resolution of the ¹H NMR spectra was 0.305 Hz/point. If two different values of the same coupling constant were recorded, both are presented and separated by a slash. All ¹H and ¹³C spectra were processed and prepared using MestreNova software and are presented in the Supporting Information. Mass spectra were recorded using an AB Sciex TripleTOF 5600 + mass spectrometer equipped with a duoelectrospray interface and operated in the negative ionization mode. Signals were recorded in the form of formic acid adducts, however, adducts with isopropyl alcohol or methanol (solvents) were also detected. Thin-layer chromatography (TLC) was performed on aluminium plates coated with E. Merck Kieselgel 60 F254, using the following eluent systems (v/v): A, 4:1 toluene : AcOEt; B, 6:1 toluene : AcOEt; C, 1:3 toluene : AcOEt; D, 1:1 toluene : AcOEt; E, 2:1 toluene : AcOEt. For the compounds detection, the dry plates were sprayed with 1% H₂SO₄ in MeOH and heated at ca. 200 °C in the stream of the hot air. Column chromatography was performed on SiliaFlash Irregular Silica Gels, G60, 60–200 μm (70–230 mesh), 60 Å, using one of the above-mentioned eluent systems.

Per-O-acetylated glycals (1–6)

These were obtained according to the known procedures in which the bromoacetate of the respective sugar was subjected to a reductive elimination reaction (Zn, AcOH, CuSO₄)³⁵.

2,3-unsaturated betulin-3-yl glycosides (7–12)

General procedure

O²⁸-Acetylbetulin (1.0-1.1 mmol), prepared from betulin according to the literature procedure⁴⁸, and respective per-O-acetylated glycal (1–6) (1.0 mmol) were dissolved in a mixture of anhydrous Et₂O and DCM (2:1). The solution was cooled in an ice-water bath. Next, BF₃•OEt₂ was added (0.05–0.1 mmol). The mixture was stirred, and after melting the ice in a bath, it was allowed to reach rt. The reaction was continued for an additional 12 h and monitored with TLC (eluent A). Then, the reaction mixture was diluted with DCM and washed three times with water. The organic layer was dried with MgSO₄, filtered, and evaporated. The residue was treated with anhydrous MeOH (20 mL) and 0.3 M MeONa in MeOH (3–5 ml) and stirred at rt for 16–18 h. De-O-acetylation was monitored with TLC (eluent A). Finally, the solvents were evaporated, and the residue was purified using column chromatography (eluent A or B).

Betulin-3-yl 2,3-dideoxy-α-D-erythro- (7α) and β-D-erythro-hex-2-enopyranoside (7β)

Reaction of O²⁸-acetylbetulin (400 mg, 0.83 mmol) with 1 (200 mg, 0.73 mmol) followed by de-O-acetylation and column chromatography (eluent A) gave first 7α (265 mg, 64%, a white amorphous solid); R_f = 0.56 (eluent C); ¹H NMR (CDCl₃, 500 MHz): δ 5.94 (d, 1H, J = 10.07 Hz, H3’), 5.68 (dt, 1H, J = 10.07 Hz, J = 2.14 Hz H2’), 5.09 (s, 1H, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 4.23 (d, 1H, J = 8.54 Hz, H4’), 3.85 (d, 2 H, J = 3.35 Hz, 2 × H6’), 3.80 (d, 1H, J = 10.99 Hz, H28_a), 3.75 (dt, 1H, J = 9.15/8.54 Hz, J = 4.58/3.97, H5’), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.25 (dd, 1H, J = 11.59 Hz, J = 4.27 Hz, H3), 2.39 (td, 1H, J = 10.99/10.68 Hz, J = 5.80 Hz, H19), 1.97 (m, 1H, H21_a), 1.91 (m, 1H, H16_a), 1.86 (dd, 1H, J = 12.21 Hz, J = 8.24 Hz, H22_a), 1.71 (m, 1H, H12_a), 1.69 (m, 1H, H1_a), 1.69 (s, 3 H, H30), 1.65 (m, 3 H, H2_b, H13, H15_a), 1.57 (m, 1H, H18), 1.53 (m, 1H, 6_a), 1.49 (m, 1H, H2_b), 1.42 (m, 1H, H11_a), 1.41 (m, 2 H, H7_a, H7_b), 1.39 (m, 2 H, H6_b, H21_b), 1.27 (bd, 1H, H9), 1.24 (m, 1H, H11_b), 1.22 (td, 1H, J = 13.13/12.51 Hz, J = 3.96 Hz, H16_b), 1.06 (m, 1H, H12_b), 1.05 (m, 1H, H15_b), 1.04 (m, 1H, H22_b), 1.02 (s, 3 H, H26), 0.97 (2 x s, 6 H, H23, H27), 0.86 (m, 1H, H1_b), 0.82 (s, 3 H, H24), 0.77 (s, 3 H, H25), 0.72 (d, 1H, J = 9.76 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 132.6 (C3’), 127.4 (C2’), 109.7 (C29), 91.1 (C1’), 83.6 (C3), 71.4 (C5’), 64.5 (C4’), 62.9 (C6’), 60.6 (C28), 55.8 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 38.6 (C4), 38.4 (C1), 37.3 (C10), 37.1 (C13), 34.3 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.2 (C23), 27.0 (C12), 25.2 (C15), 22.9 (C2), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.3 (C25), 16.2 (C24), 16.0 (C26), 14.7 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₇H₅₉O₇, 615.4266; found, 615.5101, [M + C₃H₈O] calcd for C₃₉H₆₆O₆, 630.4859; found, 630.5488. Eluted second was the 10:1 mixture of 7α and 7β (30 mg, 7%); R_f = 0.49 (eluent C).

Betulin-3-yl 2,3-dideoxy-α-D-threo-hex-2-enopyranoside (8α)

Reaction of O²⁸-acetylbetulin (400 mg, 0.83 mmol) with 2 (200 mg, 0.73 mmol) followed by de-O-acetylation and column chromatography (eluent A) provided 8α (174 mg, 42%, a white amorphous solid); R_f = 0.42 (eluent C); ¹H NMR (CDCl₃, 500 MHz): δ 6.13 (dd, 1H, J = 10.08 Hz, J = 5.49 Hz, H3’), 5.85 (dd, 1H, J = 10.07 Hz, J = 3.05 Hz, H2’), 5.17 (d, 1H, J = 2.75 Hz, H1’), 4.68 (d, 1H, J = 1.83 Hz, H29_a), 4.58 (s, 1H, H29_b), 4.12 (td, 1H, J = 5.19 Hz, J = 1.83 Hz, H5’), 3.95 (dd, 1H, J = 11.90 Hz, J = 5.49 Hz, H6_a’), 3.93 (m, 1H, H4’), 3.87 (dd, 1H, J = 11.90 Hz, J = 4.88 Hz, H6_b’), 3.80 (d, 1H, J = 10.99 Hz, H28_a), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.31 (dd, 1H, J = 12.21/11.60 Hz, J = 3.97/3.36 Hz, H3), 2.39 (td, 1H, J = 10.99 Hz, J = 5.80 Hz, H19), 1.96 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.85 (dd, 1H, J = 12.20 Hz, J = 8.24 Hz, H22_a), 1.70 (m, 2 H, H1_a, H12_a), 1.69 (m, 1H, H2_a), 1.69 (s, 3 H, H30), 1.65 (m, 1H, H13), 1.64 (m, 1H, H15_a), 1.61 (m, 1H, H18), 1.52 (m, 1H, H6_a), 1.50 (m, 1H, H2_b), 1.41 (m, 4 H, H7_a, H7_b, H11_a, H21_b), 1.39 (m, 1H, H6_b), 1.28 (dd, 1H, J = 12.51 Hz, J = 2.14 Hz, H9), 1.24 (m, 1H, H11_b), 1.22 (m, 1H, H16_b), 1.06 (m, 1H, H12_b), 1.05 (m, 1H, H15_b), 1.04 (m, 1H, H22_b), 1.02 (s, 3 H, H26), 0.98 (s, 3 H, H23), 0.97 (s, 3 H, H27), 0.87 (m, 1H, H1_b), 0.84 (s, 3 H, H24), 0.76 (s, 3 H, H25), 0.71 (d, 1H, J = 10.69 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 129.8 (C2’), 128.7 (C3’), 109.7 (C29), 90.9 (C1’), 83.0 (C3), 70.1 (C5’) 63.1 (C4’, C6’), 60.6 (C28), 55.8 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 38.5 (C4), 38.4 (C1), 37.3 (C10), 37.1 (C13), 34.2 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.2 (C23), 27.0 (C12), 25.2 (C15), 22.7 (C2), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.3 (C25), 16.2 (C24), 16.0 (C26), 14.7 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₇H₅₉O₇, 615.4266; found, 615.5098, [M + C₃H₈O] calcd for C₃₉H₆₆O₆, 630.4859; found, 630.5312.

Betulin-3-yl 2,3,6-trideoxy-α-L-erythro-hex-2-enopyranoside (9α)

Reaction of O²⁸-acetylbetulin (700 mg, 1.44 mmol) with 3 (300 mg, 1.40 mmol) followed by de-O-acetylation and by column chromatography (eluent B) provided 9α (567 mg, 73%, a white amorphous solid); R_f = 0.61 (eluent D); ¹H NMR (CDCl₃, 500 MHz): δ 5.89 (d, 1H, J = 10.38 Hz, H3’), 5.77 (br d, 1H, J = 10.07 Hz, H2’), 4.98 (s, 1H, H1’), 4.68 (s, 1H, H29_a), 4.59 (s, 1H, H29_b), 3.83 (d, 1H, J = 8.24 Hz, H4’), 3.80 (d, 1H, J = 11.30 Hz, H28_a), 3.74 (dq, 1H, J = 8.54 Hz, J = 6.11 Hz, H5’), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.14 (dd, 1H, J = 10.07 Hz, J = 6.10 Hz, H3), 2.38 (td, 1H, J = 10.99 Hz, J = 5.79 Hz, H19), 1.96 (m, 1H, H21_a), 1.92 (m, 1H, H16_a), 1.85 (dd, 1H, J = 12.21 Hz, J = 8.54 Hz, H22_a), 1.71 (m, 3 H, H2_a, H2_b, H12_a), 1.69 (s, 3 H, H30), 1.67 (m, 1H, H1_a), 1.64 (m, 2 H, H13, H15_a), 1.59 (m, 1H, H18), 1.52 (m, 1H, H6_a), 1.42 (m, 1H, H11_a), 1.41 (m, 2 H, H6_b, H21_b), 1.39 (m, 2 H, H7_a, H7_b), 1.30 (d, 3 H, J = 6.41 Hz, H6’), 1.26 (m, 1H, H9), 1.21 (m, 2 H, H11_b, H16_b), 1.06 (m, 1H, H12_b), 1.05 (m, 1H, H15_b), 1.03 (m, 1H, H22_b), 1.02 (s, 3 H, H26), 0.97 (s, 3 H, H27), 0.94 (m, 1H, H1_b), 0.91 (s, 3 H, H23), 0.83 (s, 3 H, H24), 0.77 (s, 3 H, H25), 0.71 (d, 1H, J = 10.68 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 132.9 (C3’), 127.0 (C2’), 109.7 (C29), 96.7 (C1’), 87.9 (C3), 69.9 (C4’), 67.7 (C5’), 60.6 (C28), 55.5 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 39.2 (C4), 38.9 (C1), 37.3 (C13), 36.9 (C10), 34.2 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.0 (C23), 27.1 (C12), 26.4 (C2), 25.2 (C15), 20.9 (C11), 19.1 (C30), 18.3 (C6), 18.0 (C6’), 16.4 (C25), 16.2 (C24), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₇H₅₉O₆, 599.4317; found, 599.5108, [M + C₃H₉O]⁺ calcd for C₃₉H₆₇O₅, 615.4988; found, 615.5086.

Betulin-3-yl 2,3,6-trideoxy-α-L-threo-hex-2-enopyranoside (10α)

Reaction of O²⁸-acetylbetulin (500 mg, 1,03 mmol) with 4 (200 mg, 0,93 mmol) followed by de-O-acetylation and by column chromatography (eluent B) led to 10α (195 mg, 38%, a white amorphous solid); R_f 0.56 (eluent D); ¹H NMR (CDCl₃, 500 MHz): δ 6.15 (dd, 1H, J = 10.07/9.76 Hz, J = 5.80/5.49 Hz, H3’), 5.90 (dd, 1H, J = 9.77 Hz, J = 2.44 Hz, H2’), 5.00 (s, 1H, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 4.16 (q, 1H, J = 6.72/6.41 Hz, H5’), 3.80 (d, 1H, J = 10.68 Hz, H28_a), 3.57 (br, 1H, H4’), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.14 (dd, 1H, J = 11.29/10.68 Hz, J = 5.18/4.57 Hz, H3), 2.39 (td, 1H, J = 10.68 Hz, J = 6.10 Hz, H19), 1.97 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.85 (dd, 1H, J = 12.21 Hz, J = 8.24 Hz, H22_a), 1.71 (m, 3 H, H2_a, H2_b, H12_a), 1.69 (s, 3 H, H30), 1.67 (m, 1H, H1_a), 1.65 (m, 1H, H15_a), 1.64 (m, 1H, H13), 1.60 (m, 1H, H18), 1.51 (m, 1H, H6_a), 1.43 (m, 1H, H11_a), 1.42 (m, 1H, H21_b), 1.40 (m, 3 H, H6_b, H7_a, H7_b), 1.28 (m, 1H, H9), 1.27 (d, 1H, J = 6.71 Hz, H6’), 1.21 (m, 1H, H16_b), 1.20 (m, 1H, H11_b), 1.06 (m, 1H, H12_b), 1.05 (m, 1H, H15_b), 1.04 (m, 1H, H22_b), 1.02 (s, 3 H, H26), 0.91 (s, 3 H, H23), 0.98 (s, 3 H, H27), 0.95 (m, 1H, H1_b), 0.83 (s, 3 H, H24), 0.76 (s, 3 H, H25), 0.71 (d, 1H, J = 10.07 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 129.9 (C3’), 128.4 (C2’), 109.7 (C29), 96.9 (C1’), 87.5 (C3), 66.0 (C5’), 64.1 (C4’), 60.6 (C28), 55.5 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 39.2 (C4), 38.9 (C1), 37.3 (C13), 36.9 (C10), 34.2 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.0 (C23), 27.0 (C12), 26.2 (C2), 25.2 (C15), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.3 (C25), 16.2 (C24), 16.1 (C6’), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₇H₅₉O₆, 599.4317; found, 599.5098, [M + C₃H₉O]⁺ calcd for C₃₉H₆₇O₅, 615.4988; found, 615.5070.

Betulin-3-yl 2,3-dideoxy-α-D-glycero- (11α) and -β-D-glycero-hex-2-enopyranoside (11β)

Reaction of O²⁸-acetylbetulin (750 mg, 1.55 mmol) with 5 (300 mg, 1.50 mmol) followed by de-O-acetylation and column chromatography (eluent B) provided first 11α (52 mg, 6%, a white amorphous solid); R_f 0.59 (eluent D); ¹H NMR (CDCl₃, 500 MHz): δ 5.98 (d, 1H, J = 10.38 Hz, H3’), 5.70 (d, 1H, J = 10.38 Hz, H2’), 5.02 (s, 1H, H1’), 4.69 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 4.24 (br, 1H, H4), 3.79 (d, 1H, J = 10,70 Hz, H28_a), 3.75 (dd, 1H, J = 10.68 Hz, J = 5.19 Hz, H5_a’), 3.71 (dd, 1H, J = 10.68 Hz, J = 8.55 Hz, H5_b’), 3.33 (d, 1H, J = 10.70 Hz, H28_b), 3.24 (dd, 1H, J = 11.91 Hz, J = 3.96 Hz, H3), 2.38 (td, 1H, J = 10.99 Hz, J = 5.80 Hz, H19), 1.96 (m, 1H, H21_a), 1.92 (m, 1H, H16_a), 1.86 (dd, 1H, J = 12.20 Hz, J = 8.85 Hz, H22_a), 1.71 (m, 1H, H12_a), 1.68 (m, 2 H, H1_a), 1.68 (s, 3 H, H30), 1.67 (m, 1H, H2_a), 1.64 (m, 1H, H15_a), 1.63 (m, 1H, H13), 1.60 (m, 1H, H18), 1.55 (m, 1H, H6_a), 1.51 (m, 1H, H2_b), 1.42 (m, 1H, H11_a), 1.41 (m, 2 H, H6_b, H21_b), 1.40 (m, 2 H, H7_a, H7_b), 1.27 (br d, 1H, J = 12.51 Hz, H9), 1.24 (m, 1H, H11_b), 1.21 (td, 1H, J = 13.13/12.51 Hz, J = 4.28/3.66 Hz, H16_b), 1.06 (m, 1H, H12_b), 1.04 (m, 2 H, H15_b, H22_b), 1.02 (s, 3 H, H26), 1.00 (s, 3 H, H23), 0.97 (s, 3 H, H27), 0.85 (m, 1H, H1_b), 0.84 (s, 3 H, H24), 0.78 (s, 3 H, H25), 0.72 (d, 1H, J = 9.46 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 132.7 (C3’), 128.6 (C2’), 109.7 (C29), 91.2 (C1’), 83.5 (C3), 63.7 (C5’), 63.3 (C4’), 60.6 (C28), 55.8 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 38.6 (C4), 38.3 (C1), 37.3 (C10), 37.1 (C13), 34.3 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.1 (C23), 27.0 (C12), 25.2 (C15), 23.0 (C2), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.4 (C25), 16.1 (C24), 16.0 (C26), 14.7 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₆H₅₇O₆, 585.4161; found, 585.4890, [M + C₃H₉O]⁺ calcd for C₃₈H₆₅O₅, 601.4832; found, 601.5328. Eluted second was 11β (295 mg, 36%); R_f 0.53 (eluent D); ¹H NMR (CDCl₃, 500 MHz): δ 6.10 (dd, 1H, J = 9.77 Hz, J = 5.19 Hz, H3’), 5.90 (dd, 1H, J = 9.77 Hz, J = 3.05 Hz, H2’), 4.98 (d, 1H, J = 3.05 Hz, H1’), 4.68 (s, 1H, J = 1.83 Hz, H29_a), 4.58 (s, 1H, H29_b), 4.14 (dd, 1H, J = 12.21 Hz, J = 2.44 Hz, H5_a’), 3.80 (m, 2 H, H4’, H28_a), 3.74 (d, 1H, J = 12.52 Hz, H5_b’), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.14 (dd, 1H, J = 11.60 Hz, J = 4.58 Hz, H3), 2.39 (td, 1H, J = 10.99/10.68 Hz, J = 5.80 Hz, H19), 1.97 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.85 (dd, 1H, J = 12.21 Hz, J = 8.85 Hz, H22_a), 1.74 (m, 1H, H2_a), 1.71 (m, 1H, H12_a), 1.69 (s, 3 H, H30), 1.66 (m, 1H, H1_a), 1.67 (m, 1H, H2_b), 1.66 (m, 1H, H13), 1.65 (m, 1H, H15_a), 1.60 (m, 1H, H18), 1.50 (m, 1H, H6_a), 1.42 (m, 1H, H21_b), 1.41 (m, 1H, H11_a), 1.40 (m, 2 H, H7_a, H7_b), 1.37 (m, 1H, H6_b), 1.27 (dd, 1H, J = 12.82 Hz, J = 2.14 Hz, H9), 1.22 (m, 1H, H16_b), 1.18 (m, 1H, H11_b), 1.06 (m, 1H, H12_b), 1.05 (m, 2 H, H15_b, H22_b), 1.02 (s, 3 H, H26), 0.98 (s, 3 H, H27), 0.93 (m, 1H, H1_b), 0.92 (s, 3 H, H23), 0.83 (s, 3 H, H24), 0.76 (s, 3 H, H25), 0.71 (d, 1H, J = 10.68 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 128.7 (C2’, C3’), 109.7 (C29), 95.7 (C1’), 87.9 (C3), 64.0 (C5’), 61.8 (C4’), 60.6 (C28), 55.5 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 39.2 (C1), 38.9 (C4), 37.3 (C13), 37.0 (C10), 34.2 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.0 (C23), 27.1 (C12), 26.1 (C2), 25.2 (C15), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.3 (C25), 16.2 (C24), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₆H₅₇O₆, 585.4161; found, 585.5065, [M + C₃H₉O]⁺ calcd for C₃₈H₆₅O₅, 601.4832; found, 601.5332.

Betulin-3-yl 2,3-dideoxy-α-L-glycero- (12α) and -β-L-glycero-hex-2-enopyranoside (12β)

Reaction of 28-O-acetylbetulin (1000 mg, 2.06 mmol) with 6 (400 mg, 2.00 mmol) followed by de-O-acetylation and column chromatography (eluent B) provided first 12α (69 mg, 6%); R_f 0.59 (eluent D). Eluted second was 12β (390 mg, 36%); R_f 0.54 (eluent D); R_f 0.52; ¹H NMR (CDCl₃, 500 MHz): δ 6.11 (dd, 1H, J = 10.07/9.77 Hz, J = 5.50/5.19 Hz, H3’), 5.81 (dd, 1H, J = 10.07/9.76 Hz, J = 3.05/2.72 Hz, H2’), 5.05 (d, 1H, J = 2.75 Hz, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 4.16 (dd, 1H, J = 12.21/11.90 Hz, J = 1.84/1.53 Hz, H5_a’), 3.81 (br s, 1H, H4’), 3.80 (d, 1H, J = 10.37 Hz, H28_a), 3.76 (d, 1H, J = 12.51 Hz, H5_b’), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.27 (dd, 1H, J = 12.21/11.90 Hz, J = 3.67/3.36 Hz, H3), 2.39 (td, 1H, J = 10.99/10.68 Hz, J = 6.11/5.80 Hz, H19), 1.97 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.85 (dd, 1H, J = 12.21 Hz, J = 8.54 Hz, H22_a), 1.69 (m, 1H, H2_a), 1.71 (m, 1H, H12_a), 1.69 (m, 2 H, H1_a, H2_a), 1.68 (s, 3 H, H30), 1.64 (m, 2 H, H13, H15_a), 1.60 (m, 1H, H18), 1.53 (m, 1H, H6_a), 1.48 (m, 1H, H2_b), 1.42 (m, 1H, H11_a), 1.41 (m, 1H, H21_b), 1.40 (m, 3 H, H6_b, H7_a, H7_b), 1.27 (m, 1H, H9), 1.22 (td, 1H, J = 13.42/13.13 Hz, J = 3.96/3.67 Hz, H16_b), 1.21 (m, 1H, H11_b), 1.06 (m, 1H, H12_b), 1.04 (m, 2 H, H15_b, H22_b), 1.02 (s, 3 H, H26), 0.98 (s, 3 H, H27), 0.86 (m, 1H, H1_b), 0.97 (s, 3 H, H23), 0.84 (s, 3 H, H24), 0.77 (s, 3 H, H25), 0.73 (d, 1H, J = 10.38 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 129.7 (C2’), 128.6 (C3’), 109.7 (C29), 89.9 (C1’), 83.1 (C3), 64.5 (C5’), 61.8 (C4’), 60.7 (C28), 55.8 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 38.5 (C1), 38.4 (C4), 37.3 (C13), 37.1 (C10), 34.3 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.1 (C23), 27.0 (C12), 22.8 (C2), 25.2 (C15), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.4 (C25), 16.1 (C24), 16.0 (C26), 14.7 (C27). MS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₆H₅₇O₆, 585.4161; found, 585.4882, [M + C₃H₉O]⁺ calcd for C₃₈H₆₅O₅, 601.4832; found, 601.5326.

2,3-Dideoxy betulin-3-yl glycosides (13–18)

General procedure

Respective 2,3-unsaturated betulin glycoside (7–12) (0.05–0.10 mmol) was dissolved in EtOH (20 ml). The mixture was stirred at 5 °C and hydrazine hydrate (1 ml) and 30% water solution of H₂O₂ (2 ml) were added. The reaction was continued for the next 4–28 days and monitored with TLC (eluent C). If necessary, another portion of hydrazine hydrate and H₂O₂ were added every few days (up to three additional portions). Next, EtOH was evaporated, and DCM was added to the residue. The resulting mixture was washed with water (3×). The organic layer was dried with MgSO₄, filtered, and evaporated to give the product.

Betulin-3-yl 2,3-dideoxy-α-D-erythro-hexopyranoside (13α)

Hydrogenation of 7α (51 mg, 0.089 mmol) followed by column chromatography (eluent E) led to 13α (35 mg, 69%); R_f 0.44 (eluent C); ¹H NMR (CDCl₃, 500 MHz): δ 4.97 (s, 1H, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 3.80 (m, 3 H, 2 × H6’, H28_a), 3.69 (dt, 1H, J = 9.47/8.85 Hz, J = 4.58/3.96 Hz, H5’), 3.62 (td, 1H, J = 10.07/9.46 Hz, J = 5.19/4.58 Hz, H4’), 3.33 (d, 1H, J = 10.99 Hz, H28_b), 3.16 (dd, 1H, J = 11.60/11.29 Hz, J = 3.97/3.66 Hz, H3), 2.39 (td, 1H, J = 10.99 Hz, J = 5.80 Hz, H19), 1.96 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.87 (m, 1H, H3_a’), 1.86 (m, 1H, H22_a), 1.76 (m, 3 H, H3_b’, H2_a’, H2_b’), 1.70 (m, 1H, H12_a), 1.72 (m, 1H, H2_a), 1.68 (m, 1H, H1_a), 1.68 (s, 3 H, H30), 1.64 (m, 2 H, H13, H15_a), 1.58 (m, 1H, H18), 1.54 (m, 1H, H6_a), 1.41 (m, 2 H, H11_a, H21_b), 1.40 (m, 4 H, H2_b, H6_b, H7_a, H7_b), 1.27 (m, 1H, H9), 1.21 (m, 2 H, H11_b, H16_b), 1.06 (m, 1H, H12_b), 1.05 (m, 1H, H15_b), 1.04 (m, 1H, H22_b), 1.02 (s, 3 H, H26), 1.01 (s, 3 H, H23), 0.98 (s, 3 H, H27), 0.84 (m, 1H, H1_b), 0.83 (s, 3 H, H24), 0.79 (s, 3 H, H25), 0.72 (d, 1H, J = 9.77 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 109.7 (C29), 91.8 (C1’), 81.5 (C3), 73.3 (C5’), 67.6 (C4’), 63.4 (C6’), 60.6 (C28), 55.6 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 38.6 (C4), 38.4 (C1), 37.3 (C10), 37.1 (C13), 34.3 (C7), 34.0 (C22), 30.0 (C2’), 29.8 (C21), 29.2 (C16), 28.7 (C23), 27.4 (C3’), 27.0 (C12), 25.2 (C15), 22.0 (C2), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.4 (C25), 16.2 (C24), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₇H₆₁O₇, 617.4423; found, 617.5197, [M + C₃H₉O]⁺ calcd for C₃₉H₆₉O₆, 633.5094; found, 633.5697.

Betulin-3-yl 2,3-dideoxy-α-D-threo-hexopyranoside (14α)

Hydrogenation of 8α (50 mg, 0.088 mmol) gave 14α (38 mg, 75%); R_f 0.38 (eluent C); ¹H NMR (CDCl₃, 500 MHz): δ 5.08 (br s, 1H, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 3.93 (br s, 1H, H4’), 3.90 (t, 1H, J = 3.97 Hz, H5’), 3.86 (dd, 1H, J = 11.90 Hz, J = 3.97 Hz, H6_a’), 3.81 (dd, 1H, J = 11.90 Hz, J = 3.97 Hz, H6_b’), 3.80 (d, 1H J = 10,08 Hz, H28_a), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.17 (dd, 1H, J = 11.90/11.60 Hz, J = 4.27/3.97 Hz, H3), 2.39 (td, 1H, J = 10.99/10.68 Hz, J = 6.11/5.80 Hz, H19), 2.10 (tt, 1H, J = 13.12 Hz, J = 3.66 Hz, H2_a’), 2.01 (m, 1H, H3_a’), 1.96 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.85 (m, 1H, H22_a), 1.74 (m, 1H, H2_a), 1.72 (m, 1H, H3_b’), 1.70 (m, 1H, H12_a), 1.68 (s, 3 H, H30), 1.68 (m, 1H, H1_a), 1.64 (m, 2 H, H13, H15_a), 1.58 (m, 1H, H18), 1.53 (m, 1H, H6_a), 1.47 (m, 1H, H2_b’), 1.42 (m, 1H, H11_a), 1.41 (m, 2 H, H2_b, H21_b), 1.40 (m, 1H, H6_b), 1.39 (m, 2 H, H7_a, H7_b), 1.28 (m, 1H, H9), 1.22 (m, 1H, H16_b), 1.21 (m, 1H, H11_b), 1.06 (m, 1H, H12_b), 1.04 (m, 2 H, H15_b, H22_b), 1.02 (s, 3 H, H26), 0.98 (s, 3 H, H27), 0.97 (s, 3 H, H23), 0.85 (m, 1H, H1_b), 0.84 (s, 3 H, H24), 0.78 (s, 3 H, H25), 0.72 (d, 1H, J = 9.77 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 109.7 (C29), 92.9 (C1’), 81.5 (C3), 69.6 (C5’), 67.1 (C4’), 64.9 (C6’), 60.6 (C28), 55.6 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17), 47.8 (C19), 42.7 (C14), 41.0 (C8), 38.6 (C4), 38.4 (C1), 37.3 (C10), 37.1 (C13), 34.3 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.6 (C23), 27.0 (C12), 25.5 (C3’), 25.2 (C15), 24.1 (C2’), 22.0 (C2), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.4 (C25), 16.2 (C24), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₇H₆₁O₇, 617.4423; found, 617.5220, [M + CH₅O]⁺ calcd for C₃₇H₆₅O₆, 605.4776; found, 605.4763.

Betulin-3-yl 2,3,6-trideoxy-α-L-erythro-hexopyranoside (15α)

Hydrogenation of 9α (170 mg, 0.306 mmol) gave 15α (126 mg, 74%); R_f 0.54 (eluent D); ¹H NMR (CDCl₃, 500 MHz): δ 4.76 (s, 1H, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 3.80 (d, 1H, J = 10.99 Hz, H28_a), 3.73 (dq, 1H, J = 9.14 Hz, J = 6.41 Hz, H5’), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.26 (br, 1H, H4’), 3.03 (dd, 1H, J = 11.60/10.99 Hz, J = 4.88/4.27 Hz, H3), 2.39 (td, 1H, J = 10.99 Hz, J = 5.79 Hz, H19), 1.96 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.87 (m, 1H, H2_a’), 1.86 (m, 1H, H22_a), 1.84 (m, 1H, H3_a’), 1.78 (m, 1H, H3_b’), 1.75 (m, 1H, H2_a), 1.71 (m, 2 H, H2_b’, H12_a), 1.69 (s, 3 H, H30), 1.67 (m, 1H, H2_b), 1.65 (m, 3 H, H1_a, H13, H15_a), 1.60 (m, 1H, H18), 1.50 (m, 1H, H6_a), 1.42 (m, 1H, H11_a), 1.41 (m, 1H, H21_b), 1.39 (m, 3 H, H6_b, H7_a,H7_b), 1.27 (m, 1H, H9), 1.22 (d, 3 H, J = 6.41 Hz, H6’), 1.22 (m, 1H, H16_b), 1.21 (m, 1H, H11_b), 1.06 (m, 1H, H12_b), 1.05 (m, 2 H, H15_b, H22_b), 1.02 (s, 3 H, H26), 0.98 (s, 3 H, H27), 0.89 (m, 1H, H1_b), 0.88 (s, 3 H, H23), 0.84 (s, 3 H, H24), 0.77 (s, 3 H, H25), 0.69 (d, 1H, J = 10.07 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 109.7 (C29), 99.1 (C1’), 88.2 (C3), 72.4 (C4’), 69.5 (C5’), 60.6 (C28), 55.5 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 39.1 (C4), 38.8 (C1), 37.3 (C13), 36.9 (C10), 34.2 (C7), 34.0 (C22), 30.3 (C2’), 29.8 (C21), 29.2 (C16), 28.1 (C23), 27.8 (C3’), 27.1 (C12), 25.6 (C2), 25.2 (C15), 20.9 (C11), 19.1 (C30), 18.3 (C6), 17.8 (C6’), 16.3 (C25), 16.2 (C24), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₇H₆₁O₆, 601.4474; found, 601.5243, [M + CH₅O]⁺ calcd for C₃₇H₆₅O₅, 589.4827; found, 589.4791.

Betulin-3-yl 2,3,6-tri-deoxy-α-L-threo-hexopyranoside (16α)

Hydrogenation of 10α (70 mg, 0.126 mmol) gave 16α (52 mg, 74%); R_f 0.53 (eluent D); ¹H NMR (CDCl₃, 500 MHz): δ 4.82 (s, 1H, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 4.11 (q, 1H, J = 6.71/6.14 Hz, H5’), 3.80 (d, 1H, J = 10.99 Hz, H28_a), 3.56 (m, 1H, H4’), 3.33 (d, 1H, J = 10.68 Hz, H28_b), 3.03 (dd, 1H, J = 11.29/10.99 Hz, J = 4.88/4.58 Hz, H3), 2.39 (td, 1H, J = 10.99 Hz, J = 5.80 Hz, H19), 2.01 (m, 1H, H3_a’), 1.99 (m, 1H, H2_a’), 1.97 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.86 (m, 1H, H22_a), 1.74 (m, 1H, H3_b’), 1.73 (m, 2 H, H2_a, H2_e), 1.70 (m, 1H, H12_a), 1.69 (s, 3 H, H30), 1.66 (m, 1H, H1_a), 1.65 (m, 1H, H13), 1.64 (m, 1H, H15_a), 1.59 (m, 1H, H18), 1.58 (m, 1H, H2_b’), 1.52 (m, 1H, H6_a), 1.42 (m, 1H, H11_a), 1.41 (m, 1H, H21_b), 1.40 (m, 3 H, H6_b, H7_a, H7_b), 1.28 (m, 1H, H9), 1.21 (m, 2 H, H11_b, H16_b), 1.14 (d, 3 H, J = 6.41 Hz, H6’), 1.06 (m, 1H, H12_b), 1.04 (m, 2 H, H15_b, H22_b), 1.02 (s, 3 H, H26), 0.97 (s, 3 H, H27), 0.88 (s, 3 H, H23), 0.87 (m, 1H, H1_b), 0.84 (s, 3 H, H24), 0.76 (s, 3 H, H25), 0.69 (d, 1H, J = 9.76 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 109.7 (C29), 100.0 (C1’), 88.1 (C3), 67.7 (C4’), 66.2 (C5’), 60.6 (C28), 55.5 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 39.2 (C4), 38.5 (C1), 37.3 (C13), 36.9 (C10), 34.2 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.1 (C23), 27.0 (C12), 26.0 (C3’), 25.5 (C2), 25.2 (C15), 24.1 (C2’), 20.9 (C11), 19.1 (C30), 18.3 (C6), 17.1 (C6’), 16.3 (C25), 16.2 (C24), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₇H₆₁O₆, 601.4474; found, 601.5409, [M + C₃H₉O]⁺ calcd for C₃₉H₆₉O₅, 617.5145; found, 617.5158.

Betulin-3-yl 2,3-dideoxy-β-D-glycero-hexopyranoside (17β)

Hydrogenation of 11β (75 mg, 0.139 mmol) gave 17β (55 mg, 73%); R_f 0.47 (eluent D); ¹H NMR (CDCl₃, 500 MHz): δ 4.70 (s, 1H, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 4.05 (d, 1H, J = 11.60 Hz, H5_a’), 3.80 (d, 1H, J = 11.29 Hz, H28_a), 3.76 (m, 1H, H4’), 3.39 (dd, 1H, J = 11.90 Hz, J = 3.66 Hz, H5_b’), 3.33 (d, 1H, J = 10.99 Hz, H28_b), 3.04 (dd, 1H, J = 11.60/11.29 Hz, J = 4.89/4.58 Hz, H3), 2.39 (td, 1H, J = 10.99 Hz, J = 6.10 Hz, H19), 2.04 (m, 1H, H3_a’), 1.98 (m, 1H, H2_a’), 1.96 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.85 (dd, 1H, J = 11.90 Hz, J = 7.94 Hz, H22_a), 1.78 (m, 1H, H2_a), 1.70 (m, 1H, H12_a), 1.66 (m, 1H, H2_b), 1.68 (s, 3 H, H30), 1.64 (m, 3 H, H1_a, H13, H15_a), 1.61 (m, 1H, H18), 1.60 (m, 2 H, H2_b’, H3_b’), 1.51 (m, 1H, H6_a), 1.42 (m, 2 H, H11_a, H21_b), 1.40 (m, 3 H, H6_b, H7_a, H7_b), 1.27 (m, 1H, H9), 1.22 (m, 1H, H16_b), 1.21 (m, 1H, H11_b), 1.06 (m, 1H, H12_b), 1.05 (m, 1H, H22_b), 1.04 (m, 1H, H15_b), 1.02 (s, 3 H, H26), 0.97 (s, 3 H, H27), 0.89 (m, 1H, H1_b), 0.89 (s, 3 H, H23), 0.83 (s, 3 H, H24), 0.77 (s, 3 H, H25), 0.69 (d, 1H, J = 9.46 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 109.7 (C29), 100.9 (C1’), 88.4 (C3), 65.9 (C5’), 65.3 (C4’), 60.6 (C28), 55.5 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17, C19), 42.7 (C14), 41.0 (C8), 39.2 (C4), 38.8 (C1), 37.3 (C13), 36.9 (C10), 34.3 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.1 (C23), 27.0 (C12), 26.7 (C3’), 26.3 (C2’), 25.7 (C2), 25.2 (C15), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.3 (C25), 16.2 (C24), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₆H₅₉O₆, 587.4317; found, 587.5049, [M + C₃H₉O]⁺ calcd for C₃₈H₆₇O₅, 603.4988; found, 603.5369.

Betulin-3-yl 2,3-dideoxy-β-L-glycero-hexopyranoside (18β)

Hydrogenation of 12β (100 mg, 0.185 mmol) gave 18β (86 mg, 86%); R_f 0.48 (eluent D); ¹H NMR (CDCl₃, 500 MHz): δ 4.70 (d, 1H, J = 3.66 Hz, H1’), 4.68 (s, 1H, H29_a), 4.58 (s, 1H, H29_b), 4.05 (d, 1H, J = 11.60 Hz, H5_a’), 3.80 (d, 1H, J = 11.29 Hz, H28_a), 3.77 (m, 1H, H4’), 3.38 (dd, 1H, J = 11.60 Hz, J = 4.27 Hz, H5_b’), 3.33 (d, 1H, J = 10.99 Hz, H28_b), 3.04 (dd, 1H, J = 11.59/11.30 Hz, J = 4.57/4.28 Hz, H3), 2.38 (td, 1H, J = 10.99 Hz, J = 5.80 Hz, H19), 2.04 (m, 1H, H3_a’), 1.98 (m, 1H, H2_a’), 1.97 (m, 1H, H21_a), 1.93 (m, 1H, H16_a), 1.86 (dd, 1H, J = 12.52/12.12 Hz, J = 8.55/8.24 Hz, H22_a), 1.78 (m, 1H, H2_a), 1.70 (m, 1H, H12_a), 1.68 (s, 3 H, H30), 1.67 (m, 1H, H2_b), 1.65 (m, 3 H, H1_a, H13, H15_a), 1.60 (m, 2 H, H2_b’, H3_b’), 1.59 (m, 1H, H18), 1.51 (m, 1H, H6_a), 1.42 (m, 1H, H11_a), 1.41 (m, 1H, H21_b), 1.40 (m, 3 H, H6_b, H7_a, H7_b), 1.27 (m, 1H, H9), 1.22 (m, 1H, H16_b), 1.20 (m, 1H, H11_b), 1.06 (m, 1H, H12_b), 1.05 (m, 1H, H15_b), 1.04 (m, 1H, H22_b), 1.02 (s, 3 H, H26), 0.97 (s, 3 H, H27), 0.89 (m, 1H, H1_b), 0.89 (s, 3 H, H23), 0.83 (s, 3 H, H24), 0.77 (s, 3 H, H25), 0.69 (d, 1H, J = 10.07 Hz, H5); ¹³C NMR (CDCl₃, 125 MHz): δ 150.5 (C20), 109.7 (C29), 100.9 (C1’), 88.4 (C3), 65.9 (C5’), 65.3 (C4’), 60.6 (C28), 55.6 (C5), 50.4 (C9), 48.8 (C18), 47.8 (C17), 47.8 (C19), 42.7 (C14), 41.0 (C8), 39.2 (C4), 38.8 (C1), 37.3 (C13), 36.9 (C10), 34.3 (C7), 34.0 (C22), 29.8 (C21), 29.2 (C16), 28.1 (C23), 27.1 (C12), 26.7 (C3’), 26.3 (C2’), 25.7 (C2), 25.2 (C15), 20.9 (C11), 19.1 (C30), 18.3 (C6), 16.3 (C25), 16.2 (C24), 16.0 (C26), 14.8 (C27). HRMS (ESI/Q-TOF) m/z: [M + HCOO]⁻ calcd for C₃₆H₅₉O₆, 587.4317; found, 587.5082, [M + C₃H₈O] calcd for C₃₈H₆₆O₅, 602.4910; found, 602.5288.

Cytotoxicity assay

An MTT test was performed to assess an impact of studied compounds on the amount of viable cells. Human cancer cells of breast (MCF7 line) and prostate (PC3 line) as well as normal keratinocytes (HaCaT line) were provided by dr M. Zdrowowicz-Żamojć (University of Gdańsk, Poland) and obtained from American Type Culture Collection (ATCC; MCF7 and PC3 lines) and Cell Line Service (CLS; HaCaT line), respectively. MCF7, PC3 and HaCaT cells were culture in RPMI 1640, F12K and high-glucose DMEM culture medium, respectively (Corning). Each medium was supplemented with 10% fetal bovine serum (Corning), L-glutamine and antibiotics mixture (penicillin and streptomycin; Sigma Aldrich). Cells were passaged in wells of 96-well plate and allowed to attach overnight. Next, cells were treated with indicated concentrations of tested compounds or equal amount of pure vehicle (0.5% DMSO; control cells). After 48 h incubation, 25 µl of MTT salt solution ((3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, Sigma Aldrich; concentration of 4 mg/ml) was added to each well. Following 3 h of incubation at 37 °C, an obtained product (formazan) was dissolved in DMSO. Absorbance of obtained solution was measured at 570 nm and 660 nm (reference wavelength) in a PerkinElmer Enspire microplate reader. The viability of control cells was taken as 100%. Three independent experiments were conducted in triplicate. Statistical analysis of the results and IC₅₀ value estimation were performed in the R statistical computing environment⁴⁹ using drc package⁵⁰. Data visualization (Fig. S1) was carried out with ggplot2 package⁵¹. Morphology of cells was examined under a light microscopy (magnification 200×) using an Olympus IX51 light microscope with a CCD camera and CellSens Software.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Supplementary Material 1 ^{(4.7MB, pdf)}

Author contributions

Conceptualization (GD, AH, BL), syntheses (GD), anticancer assays (AH), analysis of results (GD, AH, DG, BL), HRMS spectra (PCz), reviewing the manuscript (GD, AH, DG, PCz), writing an original draft (BL).

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Competing interests

The authors declare no competing interests.