Discovery of 2-(3-Benzamidopropanamido)thiazole-5-carboxylate Inhibitors of the Kinesin HSET (KIFC1) and the Development of Cellular Target Engagement Probes

Abstract

The existence of multiple centrosomes in some cancer cells can lead to cell death through the formation of multipolar mitotic spindles and consequent aberrant cell division. Many cancer cells rely on HSET (KIFC1) to cluster the extra centrosomes into two groups to mimic the bipolar spindle formation of non-centrosome-amplified cells and ensure their survival. Here, we report the discovery of a novel 2-(3-benzamidopropanamido)thiazole-5-carboxylate with micromolar in vitro inhibition of HSET (KIFC1) through high-throughput screening and its progression to ATP-competitive compounds with nanomolar biochemical potency and high selectivity against the opposing mitotic kinesin Eg5. Induction of the multipolar phenotype was shown in centrosome-amplified human cancer cells treated with these inhibitors. In addition, a suitable linker position was identified to allow the synthesis of both fluorescent- and trans-cyclooctene (TCO)-tagged probes, which demonstrated direct compound binding to the HSET protein and confirmed target engagement in cells, through a click-chemistry approach.

Figure 1

(A) Determination of the binding affinity (K_d) to FL HSET for probe 37 in a fluorescence polarization assay. (B) Binding of 37 to FL HSET is minimally affected by the addition of MTs. (C) ADP, ATP, and compound 13 displace the binding of the FP probe from the FL HSET protein.

Scheme 1. Anticipated IEDDA Reaction between TCO Probe 40 and Tetrazine-Cy5 41

Figure 2

(A) Images showing the distribution of Cy5-induced fluorescence from tetrazine-Cy5 (Txz-Cy5) with and without the addition of TCO probe 40 compared to the HSET intensities identified by indirect immunofluorescence using an Alexa-488 fluorophore in DLD1 4N and DLD1 HSET KO cell lines. The overlay shows the mitotic pole areas identified by staining for pericentrin (red), the HSET intensity as measured by indirect immunofluorescence using an Alexa-488 fluorophore (green) and the nucleus stained with DAPI (blue). (B) Images showing the effects on the fluorescence of increasing the concentration of 36, which outcompeted the TCO probe 40. (C) Plot showing the decreasing Cy5 signal intensity measured at the mitotic pole areas (defined by pericentrin staining), normalized to % control, with the increasing concentration of 36. (D) Plot showing the increasing HSET 488 signal measured at the mitotic pole areas (defined by pericentrin staining), normalized to % control, with the increasing concentration of 36. (E) Plot showing the ratio of the corrected click probe Cy5 signal intensity (subtracting KO cell signal) the over total HSET protein 488 signal, normalized to % control, with the increasing concentration of 36.

Scheme 2. Synthesis of Compounds 2, 4–7, 9–18, 22, and 23

Reaction conditions are as follows: (a) Boc-β-Ala-OH (for 45 and 47) or Boc-GABA–OH (for 46), HOBt, EDC·HCl, DMF, N₂, 50 °C, overnight, 80–95%; (b) 4 N HCl in 1,4-dioxane, EtOH, rt, 3 h, 34–97%; (c) R¹CO₂H (see Table 1), HATU, DIPEA, DMF, rt, overnight, 7–79%; (d) NaOH, MeOH/H₂O (2:5), 55 °C, 1 h, 27%; and (e) EtNH₂·HCl, HOBt, EDC, DIPEA, DMF, rt, 2 h, 43%.

Scheme 3. Synthesis of Compounds 19–21, 24, 26–29, and 32

Reaction conditions are as follows: (a) HATU, DIPEA, DMF, rt, overnight, 81–97%; (b) LiOH·H₂O, THF/H₂O (1:1), rt, 1.5 h, 72–97%; (c) R¹R³NH, HOBt, EDC, DMF, 60 °C, 18 h, 5–75%; and (d) TFA, DCM, rt, 2 h, 74–100% over two steps.

Scheme 4. Synthesis of Compound 25

Reaction conditions are as follows: (a) iPrMgCl·LiCl, THF, −78 °C, 10 min, then N-formylmorpholine, 25 min, 69%; (b) (i) 2-methyl-2-butene, THF, t-BuOH, rt, 5 min, (ii) NaH₂PO₄, NaClO₂, H₂O, rt, 1.5 h, 72% over two steps; (c) N-Boc-ethylenediamine, HOBt, EDC.HCl, DMF, rt, 48 h, 37%; (d) 4 N HCl in 1,4-dioxane, 1,4-dioxane, rt, overnight; and (e) 3-(5-methyl-1,2,4-oxadiazol-3-yl)benzoic acid 56, HOBt, EDC, DMF, rt, overnight, 51% over two steps.

Scheme 5. Synthesis of Compounds 33–36

Reaction conditions are as follows: (a) 55, DIPEA, HATU, DMF, rt, 16 h; (b) propyl 2-amino-4-methyl-thiazole-5-carboxylate, EDC·HCl, HOBt, DMF, 60 °C, 18 h, 22% after two steps; (c) HCl in dioxane, propanol, rt, 50 min, 46%; (d) formaldehyde, NaBH(OAc)₃, AcOH/DCE, rt, 16 h, 34%; (e) DMP, DCM, rt, 2 h; (f) Ph₃P = CHCO₂^tBu, toluene, 120 °C, 16 h, 70% over two steps; (g) 79 or 80, ⁿBuLi, THF, −78 °C, 1 h, 42–69%; (h) Pd(OH)₂, HCO₂·NH₄, HCO₂H, MeOH, 60 °C, 16 h, 51–78%; (i) 3-(5-Methyl-1,2,4-oxadiazol-3-yl)benzoic acid, Et₃N, T3P, DMF, rt, 2 h, 79–80%; (j) KOH, 50 °C, 5 h, THF/MeOH/H₂O, 44%; (k) propyl 2-amino-4-methyl-thiazole-5-carboxylate, EDC·HCl, HOBt, DMF, 45 °C, 16 h, 66-80%; and (l) TFA, DCM, rt, 1 h, 74%–80%.

Scheme 6. Synthesis of FP Probe 37

Reaction conditions are as follows: (a) TEA, DMF, rt, 16 h, 62%.

Scheme 7. Synthesis of TCO Probes 38–40

Reaction conditions are as follows: (a) 87, DMF, rt, 7 d, 64%; (b) HCl in dioxane, rt, 4–6 h, 44–65%; (c) 89, DIPEA, DMF, rt, 16 h, 67–96%; and (d) 90, DMF, rt, 24 h, 46%.