Synthesis of a Non-Symmetrical Disorazole C₁-Analogue and Its Biological Activity

Abstract

The synthesis of a novel disorazole C₁ analogue is described, and its biological activity as a cytotoxic compound is reported. Based on our convergent and flexible route to the disorazole core, we wish to report a robust strategy to synthesize a non-symmetrical disorazole in which we couple one half of the molecule containing the naturally occurring oxazole heterocycle and the second half of the disorazole macrocycle containing a thiazole heterocycle. This resulted in a very unusual non-symmetrical disorazole C₁ analogue containing two different heterocycles, and its biological activity was studied. This provided exciting information about SAR (structure-activity-relationship) for this highly potent class of antitumor compounds.

1. Introduction

The disorazoles are a family of 39 macrolides isolated so far, showing macrolide ring sizes between 26 and 32 (Figure 1) [1,2,3]. They are secondary metabolites from the Myxobacterium Sorangium cellulosum (So ce12) and were isolated by the research groups of Höfle and Reichenbach in 1994 [4].

All these natural products show very potent antitumor activity due to inhibition of tubulin polymerization combined with a very powerful cytotoxicity up to picomolar activity against various human cancer cell lines [5,6]. This exciting biological profile generated a tremendous interest in the scientific community in both total synthesis and biology [7]. In addition, their enormous biological potency makes them very attractive in personalized medicine as potesntial payloads for antibody-drug conjugates (ADCs) in targeted cancer therapy [8].

We recently published a flexible and robust new route to synthesize (—)-disorazole C₁, which involved, at the endgame, a coupling of the building blocks via Yamaguchi esterification and a final Yamaguchi macrolactonization [9].

The advantage of this powerful strategy to construct the disorazole core is that it offers high diversity in each building block before the desired coupling takes place and provides great opportunities to design a variety of disorazole analogues to study SAR (structure-activity relationships) [10].

Based on this strategy, we wish to show our efforts along these lines and report an efficient synthesis to construct a non-symmetrical disorazole C₁ analogue with potent antitumor activity. Most of the published analogues of this highly active natural product family are based on symmetrical compounds, and to the best of our knowledge, only one non-symmetrical disorazole synthesis has been published in the past by K. C. Nicolaou [11].

2. Results

In a recent study, we focused on the structure-activity relationship of the disorazole C₁ core by exchanging the oxazole ring with thiazole, and we studied the influence of chiral centers within the disorazole framework [10]. Employing the same strategy we successfully used in our total synthesis of disorazole C₁, we have prepared an analogue with the oxazole ring replaced by a thiazole unit. To our surprise, this replacement resulted in an increase in IC₅₀ values between 200- and 800-fold. Even more dramatic was the inversion of configuration at C(14)/C(14’) stereocenters, with a drop in activity of several thousand-fold [10].

Much less influence had the inversion of configuration at C(5)/C(5’) stereocenters, with a loss of activity of only between 40- and 160-fold [10].

Based on these findings, it would be of great interest to synthesize a kind of hybrid structure connecting one half of the C2-symmetric macrocycle containing the natural oxazole heterocycle and the other half containing a thiazole unit.

Our published strategy offers the possibility of synthesizing a large variety of disorazole analogues. In order to construct hybrid 16, one just has to construct the two halves 9 and 10. Fragment 10 can be used from the total synthesis of disorazole C₁, and the thiazole fragment 9 can be used from the synthesis of bis-thiazole 17.

Alcohol 9 was esterified using the Yamaguchi protocol with acid 10 to obtain compound 11 at 81% as a single compound (Scheme 1). Mild deprotection of TES-protected compound 11 with CSA provided the desired alcohol 12, which was directly hydrolyzed with LiOH, providing the starting material 13 for the Yamaguchi macrolactonization. Macrocycle 14 was isolated in 58% yield as a single isomer, which was transferred to the required (E,Z,Z)-triene derivative 15 in 62% using the Boland hydrogenation protocol.

Very simple deprotection of the bis-MOM compound 15 with two drops of HBr at −15 °C in CH₃CN provided the disorazole analogue 16 in 50% yield.

After completion of the synthesis of 16, the biological activity of the analogue was evaluated against several immortalized animal and human cancer cell lines and compared to the cytotoxicity of synthetic disorazole C₁ and epothilone B as internal controls (Figure 2) [12,13].

Table 1. Biological activities of Disorazole C₁ and thio analogues. Epothilone B was used as internal standard.

	Cell Line
	L929	KB3.1	A431	A549	PC-3	MCF-7
Compound	IC50 (ng/mL)
Epothilone B	0.24	0.017	0.026	0.034	0.048	0.015
Disorazole C1 4	0.25	0.23	0.32	0.60	0,11	0.28
Analogue 16	2.1	0.94	0.94	1.6	1.7	0.75
Analogue 17	180	85	82	86	120	91

shows these results and demonstrates again the very high activity of disorazole C₁ in the low sub-nanomolar range (between 0.11 and 0.6 ng/mL). Compound 16 is only one order of magnitude less active than the natural product 4, but several orders of magnitude more active than the bis-thiazole analogue 17.

3. Discussion

The hybrid disorazole 16 provides a very important piece of information to understand the biological activity of disorazoles because we can compare a series of analogues and can now offer a rationale to understand the activity differences and bring some light into the structure-activity relationship (SAR) of these fascinating molecules.

So far, only for disorazole Z, an X-ray exists to provide competitive data about the interaction of the molecule with the target protein tubulin [14,15]. Given the fact that more or less the same interactions take place with other disorazoles, only one half of disorazole C₁ is involved in protein binding to tubulin, and therefore, this can explain the activity differences with our compounds.

Disorazole C₁ is still the most active compound in our series, whereas bis-thiazole 17 showed around 200–800 fold less activity. However, hybrid 16 still retained most of the activity of the natural product 4, with an increased IC₅₀ value of about a maximum of 10. This clearly demonstrates that the natural oxazole half of 16 preferentially binds to the protein, leaving the thiazole half outside of the close interaction with tubulin. In addition, it explains the quite large drop in activity of the bis-thiazole compound 17.

4. Materials and Methods

Solvents were dried by standard procedures and redistilled under an N₂ atmosphere prior to use. All reactions were run under nitrogen, unless otherwise stated. For reactions that require heating, an oil bath was used as an external heat source. When the reactions were run at room temperature, a temperature of 22 °C ± 2 °C was implied. The products were purified by flash chromatography on Merck silica gel 60 (40–63 μm). POLYGRAM SIL G/UV254 prefabricated TLC plates with fluorescent indicators from Macherey-Nagel have been used for analytical thin layer chromatography (TLC). The separated substances were detected by irradiation with UV light with a wavelength of 254 nm, staining with vanillin or potassium permanganate reagent, and subsequent warming with a heat gun. Electrospray ionization (ESI) mass spectra were recorded on Waters Xevo G2-TOF spectrometers. ¹H and ¹³C-NMR spectra were recorded on Bruker AVIII 400 and Bruker AVI 600 spectrometers (Supplementary Materials). Chemical shifts (δ) are reported in ppm, referencing the resonance signal of the residual undeuterated solvent for ¹H- NMR and the deuterated solvent for ¹³C-NMR (¹H-NMR = CDCl₃: δ 7.26, CD₃OD: δ 3.31; ¹³C-NMR = CDCl₃: δ 77.16, CD₃OD: δ 49.00). Data are reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, t = triplet, br = broad, m = multiplet, app = apparent), coupling constants (Hz), and integration. Optical rotations were recorded on Perkin-Elmer 341 and Anton Paar MCP150 polarimeters employing the solvent and concentration indicated and are reported in units of 10⁻¹ (deg cm² g⁻¹).

11: NEt₃ (90 µL, 0.648 mmol, 6 eq) and 2,4,6-trichlorobenzoyl chloride (68 µL, 0.432 mmol, 4 eq) were added to a solution of carboxylic acid 10 (91 mg, 0.162 mmol, 1.5 eq) in THF (5 mL), and the resulting turbid solution was stirred for 2 h at room temperature. Toluene (3 mL) was added, and then the mixture was added dropwise to a solution of alcohol 9 (50 mg, 0.108 mmol) and DMAP (79 mg, 0.648 mmol, 6 eq) in toluene (5 mL). The mixture was stirred for 16 h at room temperature and then quenched with a saturated aqueous NH₄Cl solution (15 mL). The aqueous phase was extracted with EtOAc (3 × 10 mL). The organic layers were dried over Na₂SO₄, filtered, and concentrated in vacuo. The residue was purified by flash chromatography (Et₂O/pentane 2:1) to afford 11 (83 mg, 0.0813 mmol, 81%) as a slightly yellow oil. General Data: C₅₅H₈₀N₂O₁₂SSi; MW: 1021.39; TLC: Rf = 0.30 (pentane/Et₂O 2:1); UV (+); Vanillin: black; ${\left[\alpha \right]}_D^{20}$ = +24.44 (c = 0.9, CHCl₃).

¹H-NMR (400 MHz, CD₃OD): δ (ppm): 8.47 (s, 1H); 8.25 (s, 1H); 6.10–6.23 (m, 1H); 6.02–5.94 (m, 4H); 5.90–5.80 (m, 1H); 5.72–5.58 (m, 4H); 5.47–5.33 (m, 2H); 5.25 (app t, J = 6.6 Hz, 1H); 4.69–4.59 (m, 2H); 4.50–4.37 (m, 2H); 4.24–4.16 (m, 1H); 3.88 (s, 3H); 3.89–3.83 (m, 2H); 3.34 (s, 3H); 3.33 (s, 3H); 3.28 (s, 3H); 3.27 (s, 3H); 3.44–2.98 (m, 3H); 2.81–2.70 (m, 2H); 2.65–2.54 (m, 2H); 2.48–2.39 (m, 1H); 1.72 (ddd, J = 12.8, 6.6, 1.6 Hz, 6H); 0.984 (app t, J = 7.6 Hz, 9H); 1.01 (s, 3H); 0.96 (s, 3H); 0.94 (s, 3H); 0.87 (s, 3H); 0.64 (q, J = 7.7 Hz, 6H). ¹³C NMR (101 MHz, CD₃OD): δ (ppm): 164.41, 164.38, 162.87, 162.22, 146.08, 143.59, 142.00, 141.81, 141.66, 136.61, 133.61, 133.16, 132.34, 129.19, 128.63, 128.51, 114.76, 114.57, 112.11, 110.81, 94.64, 94.52, 92.57, 88.92, 88.49, 82.75, 81.37, 80.08, 78.59, 77.90, 66.91, 61.53, 57.12, 57.07, 56.33, 55.97, 52.48, 44.22, 42.85, 35.50, 35.04, 34.65, 32.40, 19.73, 19.54, 18.05, 18.03, 15.45, 14.47, 7.52, 6.52. HRMS (ESI) m/z: [M+H]⁺ Calcd for C₅₅H₈₁N₂O₁₂SSi: 1021.5279; found: 1021.5275.

14: TES-protected alcohol 11 (80.3 mg, 0.0787 mmol) was dissolved in CH₂Cl₂/MeOH 1:1 (4 mL, 0.02 M) and cooled to 0 °C. CSA (3.7 mg, 0.0157 mmol, 0.2 eq) was added, and the mixture was stirred for 1 h at 0 °C under a normal atmosphere. Saturated aqueous NaHCO₃ solution (5 mL) was added, and the layers were separated. The aqueous phase was extracted with CH₂Cl₂ (3 × 5 mL), and the combined organic extracts were dried over Na₂SO₄, filtered, and concentrated in vacuo. The crude material was used in the next step without further purification. General Data: C₄₉H₆₆N₂O₁₂S; MW: 907.13; TLC: Rf = 0.30 (CH₂Cl₂/MeOH 50:1); UV (+); Vanillin: black; ${\left[\alpha \right]}_D^{20}$ = +40.4 (c = 0.5, CHCl₃). HRMS (ESI) m/z: [M+H]⁺ Calcd for C₄₉H₆₇N₂O₁₂S: 907.4415; found: 907.4415. Deprotected alcohol 12 (71.3 mg, 0.0787 mmol) was dissolved in THF (2 mL) and LiOH (1 M in H₂O, 0.236 mL, 0.236 mmol, 3 eq) was added. The mixture was stirred at room temperature for 16 h and neutralized with 1 M HCl (~1 mL). The aqueous phase was extracted with Et₂O (3 × 3 mL), and the organic extracts were dried over Na₂SO₄, filtered, and concentrated in vacuo to give crude seco-acid 13, which was used for the next step without further purification. General Data: C₄₈H₆₄N₂O₁₂S; MW: 893.10; TLC: UV (+); Vanillin: black; ${\left[\alpha \right]}_D^{20}$ = +24.4 (c = 0.5, CHCl₃). HRMS (ESI) m/z: [M+H]⁺ Calcd for C₄₈H₆₅N₂O₁₂S: 893.4258; found: 893.4261. NEt₃ (219 μL, 1.57 mmol, 20 eq) and 2,4,6-trichlorobenzoyl chloride (123 μL, 0.787 mmol, 10 eq) were added at room temperature to a solution of crude 13 (66 mg, 0.0787 mmol) in THF (5 mL), and this turbid solution was stirred for 2 h at room temperature. Toluene (2 mL) was added, and the solution was added dropwise to a solution of DMAP (385 mg, 3.15 mmol, 40 eq) in toluene (40 mL). The mixture was stirred overnight at room temperature, and then quenched with saturated aqueous NH₄Cl solution (10 mL), water (10 mL), and the aqueous phase was extracted with EtOAc (3 × 20 mL). The organic layers were dried over Na₂SO₄, filtered, and concentrated in vacuo. The residue was purified by flash chromatography (hexane/EtOAc 2:1 to 1:1) to afford the macrocycle 14 (40 mg, 0.0457 mmol, 58% from 11) as a yellow wax. General Data: C₄₈H₆₂N₂O₁₁S; MW: 875.09; TLC: Rf = 0.50 (CH₂Cl₂/MeOH 50:1); UV (+); Vanillin: black; ${\left[\alpha \right]}_D^{20}$ = +160.40 (c = 0.5, CHCl₃). ¹H-NMR (400 MHz, CDCl₃): δ (ppm): 8.05 (s, 1H); 7.98 (s, 1H); 6.03–5.89 (m, 4H); 5.84–5.71 (m, 2H); 5.70–5.57 (m, 3H); 5.52 (m, 1H); 5.46–5.27 (m, 4H); 4.65 (dd, J = 16.0, 6.5 Hz, 2H); 4.40 (dd, J = 16.2, 6.7 Hz, 2H); 4.17–4.07 (m, 2H); 3.75 (dt, J = 9.0, 3.4 Hz, 2H); 3.38 (s, 3H); 3.36 (s, 3H); 3.35 (s, 3H); 3.34 (s, 3H); 3.29 (m, 2H); 3.26–3.14 (m, 2H); 3.07–2.94 (m, 2H); 2.47–2.37 (m, 2H); 1.72 (m, 6H); 1.07 (s, 3H); 1.05 (s, 3H); 1.03 (s, 3H); 1.00 (s, 3H). ¹³C NMR (101 MHz, CDCl₃): δ (ppm): 165.71, 161.83, 160.63, 160.62, 147.20, 147.27, 143.55, 140.76, 140.70, 140.33, 133.77, 132.01, 131.89, 127.39, 127.28, 114.51, 113.42, 112.21, 112.11, 93.83, 93.78, 91.08, 91.02, 87.98, 87.96, 81.64, 81.48, 80.35, 79.39, 77.36, 76.67, 57.07, 56.93, 56.19, 56.17, 41.69, 41.64, 39.07, 34.51, 31.73, 31.60, 20.04, 19.86, 19.69, 19.64, 18.10, 18.07. HRMS (ESI) m/z: [M+H]⁺ Calcd for C₄₈H₆₃N₂O₁₁S: 875.4330; found: 875.4332.

15: Nitrogen was bubbled for 15 min through a suspension of zinc dust (5.00 g, 45.88 mmol) in H₂O (30 mL), and then Cu(OAc)·2H₂O (500 mg, 2.5 mmol) was added at room temperature. After 15 min, AgNO₃ (500 mg, 2.94 mmol) was added carefully. The mixture was stirred for 30 min at room temperature, filtered by suction, and finally washed with H₂O (40 mL), MeOH (30 mL), acetone (30 mL), and Et₂O (30 mL). This activated zinc solid was added to a solution of 14 (31.5 mg, 0.0360 mmol) in MeOH/H₂O 1:1 (20 mL), nitrogen was once again bubbled through the suspension for 10 min, and then the flask was sealed. The mixture was stirred for 24 h at 50 °C, then filtered on a small pad of silica with MeOH washes. The filtrate was concentrated in vacuo and the residue was purified by flash chromatography (CH₂Cl₂/MeOH 70:1) to afford 15 (19 mg, 0.0216 mmol, 62%) as a colorless wax. General Data: C₄₈H₆₆N₂O₁₁S; MW: 879.12; TLC: Rf = 0.40 (CH₂Cl₂/MeOH 50:1); UV (+); Vanillin: dark green; ${\left[\alpha \right]}_D^{20}$ = −19.9 (c = 1.0, CHCl₃). ¹H-NMR (600 MHz, CDCl₃): δ (ppm): 7.89 (s, 1H); 7.82 (s, 1H); 6.47–4.26 (m, 5H); 6.18 (dd, J = 19.7, 10.4 Hz, 1H); 6.95–6.83 (m, 2H); 5.67–5.58 (m, 3H); 5.57–5.47 (m, 3H); 5.43–5.36 (m, 2H); 5.31–5.25 (m, 2H); 4.68–4.63 (m, 2H); 4.43–4.36 (m, 2H); 4.16–4.08 (m, 2H); 3.72 (dd, J = 18.9, 8.9 Hz, 2H); 3.42–3.36 (m, 2H); 3.33 (s, 3H); 3.32 (s, 3H); 3.28 (s, 3H); 3.27 (s, 3H); 3.19–3.09 (m, 2H); 2.99 (dd, J = 14.0, 7.6 Hz, 1H); 2.81 (dd, J = 14.7, 8.1 Hz, 1H); 2.50–2.43 (m, 1H); 2.40–2.31 (m, 1H); 1.72 (ddd, J = 9.0, 5.3, 4.5 Hz, 6H); 1.03 (s, 3H); 1.01 (s, 3H); 0.98 (s, 3H); 0.94 (s, 3H). ¹³C NMR (151 MHz, CDCl₃): δ (ppm): 166.53, 162.41, 160.75, 160.74, 147.20, 143.24, 133.53, 133.22, 132.99, 132.11, 132.05, 130.17, 129.75, 129.61, 129.43, 128.18, 128.10, 127.38, 127.22, 127.14, 126.70, 125.61, 125.52, 125.44, 93.62, 81.87, 81.69, 81.63, 81.11, 80.14, 77.57, 56.67, 56.58, 56.13, 56.11, 41.72, 41.69, 39.74, 35.14, 29.84, 28.39, 20.32, 20.16, 20.00, 19.90, 18.08, 18.06. HRMS (ESI) m/z: [M+H]⁺ Calcd for C₄₈H₆₇N₂O₁₁S: 879.4466; found 879.4461.

16: Compound 15 (17.5 mg, 19.91 μmol) was dissolved in acetonitrile (2 mL) and cooled to −15 °C. HBr (two drops, 48% in H₂O) was added dropwise, and the mixture was stirred for 2 h 30 min at −15 °C. EtOAc (5 mL) was added, and the mixture was washed with saturated aqueous NaHCO₃ solution (5 mL). The aqueous phase was extracted with EtOAc (3 × 5 mL), and the organic extracts were dried over Na₂SO₄, filtered, and concentrated in vacuo. The residue was purified by flash chromatography (CH₂Cl₂/MeOH 50:1) to give thiazolyl-disorazole C₁ (16) (7.9 mg, 9.95 μmol, 50%) as a colorless wax. General Data: C₄₄H₅₈N₂O₉S; MW: 791.01; TLC: Rf = 0.20 (CH₂Cl₂/MeOH 50:1); UV (+); Vanillin: dark green; ${\left[\alpha \right]}_D^{20}$ = −85.9 (C = 0.191 in MeOH). ¹H-NMR (600 MHz, CD₃OD): δ (ppm): 8.26 (s, 1H); 8.21 (s, 1H); 6.54–6.45 (m, 2H); 6.43–6.34 (m, 2H); 6.27 (dd, J = 21.4, 10.5 Hz, 2H); 5.93 (app t, J = 11.0 Hz, 1H); 5.83 (dd, J = 10.7, 5.9 Hz, 1H); 5.71–5.62 (m, 2H); 5.62–5.53 (m, 2H); 5.53–5.42 (m, 4H); 5.26 (app dt, J = 14.0, 2.7 Hz, 2H); 4.14 (dd, J = 13.7, 7.4 Hz, 2H); 3.86 (dd, J = 20.7, 8.1 Hz, 2H); 3.25 (s, 3H); 3.22 (s, 3H); 3.04–2.96 (m, 2H); 2.78 (dd, J = 15.5, 5.7 Hz, 2H); 2.72–2.62 (m, 2H); 2.48–2.39 (m, 2H); 1.70 (ddd, J = 11.8, 6.4, 1.3 Hz, 6H); 1.03 (s, 3H); 1.01 (s, 3H); 0.98 (s, 3H); 0.95 (s, 3H). ¹³C NMR (151 MHz, CD₃OD): δ (ppm): 168.90, 164.27, 162.25, 147.55, 145.76, 134.42, 134.10, 133.80, 131.70, 131.66, 130.85, 130.60, 130.18, 129.88, 129.67, 129.62, 129.43, 129.13, 129.03, 127.32, 127.15, 126.77, 126.55, 81.66, 80.77, 79.13, 78.78, 77.91, 77.83, 56.89, 56.78, 42.81, 42.75, 40.42, 35.96, 29.28, 29.09, 19.46, 19.42, 19.31, 18.04. HRMS (ESI) m/z: [M+H]⁺ Calcd for C₄₄H₅₉N₂O₉S: 791.3941; found 791.3938.

5. Conclusions

The disorazoles are a fascinating group of natural products that attracted both the synthetic chemistry community and biologists. Within the last decade, quite a few research groups have reported new total synthesis efforts and strategies to synthesize these attractive molecules. In addition, several SAR studies have been reported to shed some light on the biology of these interesting compounds. Our results provide some more rationale for these findings for the future because still no in vivo data is reported about the therapeutic usefulness of these highly active molecules. Based on the outstanding cytotoxicity of the compounds, it is still very reasonable that they could be powerful payloads for ADCs.