Structure of PDE3A–SLFN12 complex and structure-based design for a potent apoptosis inducer of tumor cells

Molecular glues are a class of small molecular drugs that mediate protein-protein interactions, that induce either the degradation or stabilization of target protein. A structurally diverse group of chemicals, including 17-β-estradiol (E2), anagrelide, nauclefine, and DNMDP, induces apoptosis by forming complexes with phosphodiesterase 3A (PDE3A) and Schlafen 12 protein (SLFN12). They do so by binding to the PDE3A enzymatic pocket that allows the compound-bound PDE3A to recruit and stabilize SLFN12, which in turn blocks protein translation, leading to apoptosis. In this work, we report the high-resolution cryo-electron microscopy structure of PDE3A-SLFN12 complexes isolated from cultured HeLa cells pre-treated with either anagrelide, or nauclefine, or DNMDP. The PDE3A-SLFN12 complexes exhibit a butterfly-like shape, forming a heterotetramer with these small molecules, which are packed in a shallow pocket in the catalytic domain of PDE3A. The resulting small molecule-modified interface binds to the short helix (E552-I558) of SLFN12 through hydrophobic interactions, thus “gluing” the two proteins together. Based on the complex structure, we designed and synthesized analogs of anagrelide, a known drug used for the treatment of thrombocytosis, to enhance their interactions with SLFN12, and achieved superior efficacy in inducing apoptosis in cultured cells as well as in tumor xenografts.


Supplementary Tables
Supplementary Table 1

Supplementary Note 1
We did not observe the linear correlation of the logP value with IC50 of all the molecules Supplementary Table 2. However, as we indicated in the text, the hydrophobic interaction is crucial for the binding site of the PDE3A/SLFN12 interface, we then analyzed the correlation of the logP value and IC50 of the molecules with the similar interaction pattern at this interface and indeed observed the same tendency that the greater the value of LogP, the lower IC50 value. We admit that the interaction situation in this binding site is much more complicated and there are certainly other parameters that might affect the interaction, such as the steric interruption, π-π-stacking interactions and so on, but likely, the hydrophobic interaction plays a key role for the gluing activity in the current study.

Supplementary Methods
Chemical Synthesis. All reactions were carried out under an atmosphere of nitrogen in flame-dried glassware with magnetic stirring unless otherwise indicated. Commercially obtained reagents were used as received. Solvents were dried by passage through an activated alumina column under argon. Liquids and solutions were transferred via syringe. All reactions were monitored by thin-layer chromatography with E. Merck silica gel 60 F254 pre-coated plates (0.25 mm). 1H and 13C NMR spectra were recorded on Varian Inova-400 or 500 spectrometers. Data for 1H NMR spectra are reported relative to CDCl3 (7.26 ppm), CD3OD (3.31 ppm), or DMSO-d6 (2.50 ppm) as an internal standard and are reported as follows: chemical shift (δppm), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, sept = septet, m = multiplet, br = broad), coupling constant J (Hz), and integration. Data for 13C NMR spectra are reported relative to CDCl3 (77.23 ppm), CD3OD (49.00 ppm) or DMSO-d6 (39.52 ppm) as an internal standard and are reported in terms of chemical shift (δ ppm). Samples preparation and purity analysis were conducted on Waters HPLC (Column: XBridge C18, 5μm, 19 x 150 mm) with 2998PDA and 3100MS detectors, and Waters UPLC (Column: BEH C18, 1.7μm, 2.1 x 50 mm) with PDA and SQD MS detectors, using ESI as ionization. HRMS data were obtained on a Thermo Q Exactive.
All new compounds were synthesized as indicated in detail in the following schemes. Compound A2 was synthesized according to the same synthetic route as compound A1 using different starting materials 3-fluoro-2-chlorobenzaldehyde. Di-substituted compounds A3, A4, A6-A14, A16, A18-A22 were synthesized using a Suzuki coupling reaction from A1. Compounds A5 were synthesized according to the same synthetic route as compound A17 using different starting materials 3,4,5-trichloroaniline. Tri-substituted compound A15 was synthesized using a Suzuki coupling reaction from A17. A14 was the by-product of the Suzuki coupling reaction for disubstituted compounds.

Synthesis of compound 2
In a flame-dried 50 mL 2-necked round-bottom flask, concentrated sulfuric acid (15mL), and 3-bromo-2-chlorobenzaldehyde (2g, 9.11mmol, 1.0equiv) was added in small portions while stirring. Conc. nitric acid (3.0mL) was added dropwise at 0 o C. The reaction mixture was stirred at 0 o C for 30min and then warmed to rt overnight. It was checked by TLC until completion. The reaction mixture was poured into 100mL ice water, the mixture was extracted with EA (3 x 50 mL), washed with brine and concentrated. The mixture was purified by flash chromatography on silica gel (PE:EA = 300:1) to give compound 2 as a white solid (1.1g, 46% yield).

Synthesis of compound 3
In a flame-dried 50 mL round-bottom flask, 3-bromo-2-chloro-6-nitrobenzaldehyde (1.1g, 4.18mmol, 1.0equiv)was dissolved in methanol (20.0mL). The reaction mixture was cooled to 0 o C. NaBH4 (316mg, 8.36mmol, 2.0equiv) was added in small portions. The reaction mixture was stirred at 0 o C for 30min. TLC showed the starting material was consumed. The reaction mixture was quenched with ice water (50mL), extracted with EA (3 x 50 mL), washed with brine and concentrated. The mixture was purified by flash chromatography on silica gel (PE:EA = 15:1) a to give compound 3 as a white solid (1.1g, 99% yield).

Synthesis of compound 7
In a flame-dried 50 mL round-bottom flask, ethyl 2-((6-amino-3-bromo-2chlorobenzyl)amino)acetate (913mg, 2.85mmol, 1.0equiv) was dissolved in dry toluene (20.0mL ), and then cyanogen bromide (453mg, 4.28mmol, 1.5equiv) added in small portions while stirring at rt. The reaction mixture was stirred at 110 o C for 14h. It was checked by TLC until completion. The reaction mixture was filtered by sand core funnel to get the crude product 7 as a solid (1.02g).
Then the reaction mixture was cooled to room temperature and poured upon ten to twelve times its volume of cracked ice. The mixture was filtered, washed 5 times with cold water to remove the sulfuric acid, and then dried. The crude reaction mixture was purified by flash chromatography on silica gel (PE:EA = 15:1) to give the compound 10 as a brown solid (0.93g, 72% yield).

Synthesis of compound 11
A flame-dried 50 mL round-bottom flask, compound 10 (0.93g, 3.15mmol, 1.0equiv), 5 M aqueous sodium hydroxide solution (3.0 mL) were added, and then hydrogen peroxide (30%) solution (1.2 mL) was dropwise over 5min. The reaction mixture was stirred at room temperature for 4h. The concentrated hydrochloric acid was added to adjust the pH to 5-6. The reaction mixture extracted with EA (3 x 100 mL), washed with brine and then concentrated to give the crude product 11 (870mg) was used directly without further purification.

Synthesis of compound 12
A flame-dried 25 mL round-bottom, compound 11 (770mg, 2.71mmol, 1.0equiv) was dissolved in tetrahydrofuran (10 mL), and then lithium aluminum hydride (141mg, 3.71mmol, 1.4equiv) in 20 mL tetrahydrofuran was added in portion at 0 o C. The reaction mixture was stirred at 0 o C for 30min before it was gradually warmed to rt overnight. The reaction mixture was poured into 10 mL ice saturated sodium bicarbonate solution, extracted with EA (3 x 100 mL) washed with brine and then concentrated. The mixture was purified by flash chromatography on silica gel (PE:EA = 10:1) to give compound 12 as a light yellow solid (510mg, 70%yield).

Synthesis of compound 13
A flame-dried 25 mL round-bottom flask, compound 12 (40mg, 0.15mmol, 1.0equiv) was dissolved in dichloromethane (5 mL), and then manganese oxide (80mg, 0.92mmol, 6.1equiv) was added. The reaction mixture was stirred at rt overnight. The reaction mixture was filtered by diatomaceous earth with suction and washed by dichloromethane. The crude mixture was purified by flash chromatography on silica gel to give the compound 13 as a light yellow solid (36mg, 90%yield).

Synthesis of compound 15
A flame-dried 25 mL round-bottom flask, compound 14 (47mg, 0.13mmol, 1.0equiv) was dissolved in dry toluene (1.3 mL), and then cyanogen bromide (46mg, 0.40mmol, 3.0equiv) was added in small portions while stirring at rt. The reaction mixture was stirred at rt overnight. The reaction mixture was filtered by sand core funnel (washed with dry toluene) to give the compound 15 as a white solid (36mg, 71% yield).

Synthesis of compound A17
A flame-dried 50 mL 2-necked round-bottom flask, compound 15 (36mg, 0.09mmol, 1.0equiv) and triethylamine (17µL, 0.12mmol, 1.3equiv) were dissolved in ethanol (1.5 mL) at rt. The reaction mixture was stirred at 80 o C for 2.5 h. The reaction mixture was filtered by sand core funnel (washed by ethanol) to give the product A17 as a light yellow solid (12mg, 38% yield).