Letter Cite This: ACS Comb. Sci. 2018, 20, 251−255
pubs.acs.org/acscombsci
Development and Synthesis of DNA-Encoded Benzimidazole Library Yun Ding,* Jing Chai, Paolo A. Centrella,† Chenaimwoyo Gondo,‡ Jennifer L. DeLorey,§ and Matthew A. Clark‡ GlaxoSmithKline, Platform Technology & Science, 200 Cambridgepark Drive, Cambridge, Massachusetts 02140, United States
Downloaded via KAOHSIUNG MEDICAL UNIV on June 27, 2018 at 12:27:12 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: Encoded library technology (ELT) is an effective approach to the discovery of novel small-molecule ligands for biological targets. A key factor for the success of the technology is the chemical diversity of the libraries. Here we report the development of DNA-conjugated benzimidazoles. Using 4-fluoro-3-nitrobenzoic acid as a key synthon, we synthesized a 320 million-member DNA-encoded benzimidazole library using Fmoc-protected amino acids, amines and aldehydes as diversity elements. Affinity selection of the library led to the discovery of a novel, potent and specific antagonist of the NK3 receptor. KEYWORDS: benzimidazole, DNA-encoded library technology
T
by nucleophilic substitution of a DNA-appended 4-fluoro-3nitrobenzamide with primary amines, followed by annulation with various aldehydes (Scheme 1). Libraries with at least two points of diversification can be synthesized based on this chemistry.
he DNA-encoded library (DEL) technology has been proven to be an effective approach to the discovery of small molecule ligands for biological targets since it was first proposed by Lerner, Brenner, and Janda.1,2 It has been widely used both in the pharmaceutical industry and academia.3−7 Potent inhibitors for various targets have been reported using this approach.8−12 A key factor to the success of DEL technology is the diversity of chemical space that DEL can cover, especially the drug-like chemotypes. This is highly dependent on the repertoire of chemical reactions that are compatible with “on-DNA” synthesis. Recently various groups have reported a number of synthetic chemistries compatible with DNA-encoding.13−18 Here, we report the development of DNA-appended benzimidazole synthesis and its application to the production of a library which has delivered ligands for many targets. Benzimidazole is a privileged chemotype found in diverse biological active compounds.19 The synthesis of benzimidazoles in solution and solid-phase conditions have been well developed.20 Recently two groups reported the development of benzimidazole chemistry on DNA in aqueous conditions. One report used Raney nickel and hydrazine to reduce DNA-bound onitroarylamine to the corresponding diamine,17 which had practical challenges to apply in library synthesis. The other reported a process to reduce aryl nitro groups to aryl amines with Na2S2O4/viologen.18 Both reports conducted the condensation to form benzimidazole ring under elevated temperatures. Here, we report a novel synthetic route to DNA-tagged benzimidazoles with FeSO4 as reducing reagent. To the broad o-nitroarylamines, FeSO4 delivered better reducing results compared with Na2S2O4, which would yield significant amount of byproduct for some amine substituents. Instead of conducting the condensation to form benzimidazole ring at 80 or 60 °C overnight,17,18 we conducted the condensation under a much milder condition at room temperature. In addition, this Letter provides insight on the library synthesis with characterization of the products and byproducts in a small size library. The library can be constructed © 2018 American Chemical Society
Scheme 1
Library Development: SNAr. Nucleophilic substitution of DNA-appended 4-fluoro-3-nitrobenzoate (HP-1) can be proceeded at room temperature with aliphatic amines such as benzylamine in aqueous solution buffered to pH 9.4.21 However, under the same condition, the more challenging amines such as aromatic amines only generated low or no yield of substitution product. To expand the diversity of amines in the library, we optimized the reaction condition by increasing the temperature to 60 °C for 16 h. A small set of amines (Table 1) were validated under such conditions. Overall the reaction proceeded smoothly with primary aliphatic amines. For some challenging amines, a dimethylamine adduct would generate if DMA was used as solvent for the amines. To minimize this side reaction, the mixture of CH3CN/H2O at 1/1 ratio was used to dissolve the Received: January 23, 2018 Revised: March 28, 2018 Published: April 12, 2018 251
DOI: 10.1021/acscombsci.8b00009 ACS Comb. Sci. 2018, 20, 251−255
Letter
ACS Combinatorial Science Table 1. Nucleophilic Substitution with Primary Aminesa
reactions in water and pH 5.5 phosphate buffer. Only 40% benzimidazole was formed under pH 9.4 condition. Likewise, Nbenzylphenylenediamine was reacted with benzaldehyde under water, pH 5.5 and pH 9.4 conditions. After proceeding at room temperature overnight, similar results were obtained. We concluded the formation of target benzimidazole best proceeded under slightly acidic conditions, either in water or pH 5.5 phosphate buffer. Next, we sought to demonstrate that the optimized condition would be suitable for wide scope of aldehydes. We chose to test the cyclization of benzylamine substituted diamine (HP-3b) with a scope of aldehydes (Table 3). In aqueous solution buffered to Table 3. Benzimidazole Formation with Scope of Aldehydesa
a
Reaction conditions: 1 equiv of HP-1 (1 mM in pH 9.4 borate buffer (250 mM)), 40 equiv of amines (400 mM in DMA), 60 °C for 16 h.
amines and similar substitution results were obtained. Under this optimized condition, a total of 2423 amines were validated. Of those, 1252, or 51%, passed at a yield of 70% or greater, among which 800 were primary amines, including 200 primary aromatic ones. Library Development: Nitro Reduction. Many reagents, such as Zn, Na2S2O4, FeSO4, and SnCl2, can be used to reduce the nitro group. After exploring a few reducing reagents, which will be described elsewhere, we found FeSO4 to be the best reducing reagent for the reduction of DNA-conjugated nitrobenzenes. With 40 equiv of FeSO4 and 80 equiv of NH4OH or NaOH, the nitro group can be reduced to amine at room temperature within minutes. For some electron-deficient HP-2 substrates, heating at higher temperature is necessary for the complete reduction. We determined that 80 °C for 2 h was the optimal condition for the nitro reduction in the library production. Library Development: Benzimidazole Ring Formation. To explore the formation of the benzimidazole core by condensation and oxidation, we initially explored the effects of heat and pH on the annulation of N-methylphenylenediamine (HP-3a) with benzaldehyde (Table 2). The reactions were set up
a
Reaction conditions: 1 equiv of HP-3b (1 mM in pH5.5 phosphate buffer (250 mM)), 60 equiv of aldehyde (400 mM in CH3CN or 200 mM in CH3CN/DMA 1/1), RT for 16 h.
pH 5.5, all the aldehydes, including aromatic aldehydes and aliphatic aldehydes, yielded the target benzimidazoles in excellent yield. The reactions in water delivered similar results. Since some of the aldehydes may contain some functional groups that would affect the pH of reaction, to better control the pH of the reaction, we decided to use the condition of pH 5.5 phosphate buffer instead of water as our standard condition for benzimidazole ring formation. We also explored the effect of solvent that was used to dissolve aldehydes on the ring formation. Both DMA and CH3CN generated similar results with excellent yield of the desired benzimidazoles. With this data in hand we turned to the validation of a large number of aldehydes for benzimidazole formation with HP-3b. Using the condition described above, a total of 1287 aldehydes were evaluated. Of those, 930, or 72%, passed at a yield of greater than 70%. In general, sterically hindered aldehydes yielded the desired benzimidazoles in relatively lower yield. However, there was no strong structural trend to predict pass/fail. In some cases, we observed byproducts which were proposed to be HP-4c and HP-4d (Scheme 2). The formation of HP-4c was independent of the solvent used to dissolve aldehydes. We predicted it was formed from the degradation of condensation intermediate with aldehyde.22 The mechaism to form byproduct HP-4d was still unclear. Library Development and Mock Library. With the chemistry developed for the derivatized benzimidazole library, we had the option to include another point of diversity before
Table 2. Benzimidazole Formation under Different pH Conditions
conditions
yield (80 °C)a (%)
yield (RT)b (%)
water pH 5.5 pH 9.4
60 70 30
∼100 ∼100 40
a
Reaction conditions: 1 equiv of HP-3a (1 mM in buffer or water), 60 equiv of benzaldehyde (400 mM in CH3CN), 80 °C for 16 h. bRT for 16 h.
in water, pH 5.5 phosphate buffer and pH 9.4 borate buffer, and allowed to react at room temperature or 80 °C for 16 h in all cases. We observed that at the elevated temperature, the reactions in water and pH 5.5 generated benzimidazole in moderate yield (60−70%), but with some unknown byproducts (20−30%). The reaction in pH 9.4 was much cleaner but with only ∼30% yield of desired condensation product. The room temperature conditions yielded benzimidazole quantitatively for 252
DOI: 10.1021/acscombsci.8b00009 ACS Comb. Sci. 2018, 20, 251−255
Letter
ACS Combinatorial Science Scheme 2. Byproducts in Benzimidazole Formation
coupling 4-fluoro-3-nitrobenzoic acid. Besides Fmoc amino acids as we applied in our earlier reported library,23 we also evaluated primary amines and secondary diamines that can be added to electron deficient acrylamide headpiece to form secondary amine headpiece HP-5 (Table 4). The introduced secondary amines Table 4. Validation of Cycle 1 Aminesa
Figure 1. Synthetic scheme for mock library.
products attached to the linker and a single nucleotide (Figure 2), which could be analyzed by LCMS. All 6 of the expected
a Reaction conditions: 1 equiv of HP-5 (1 mM in pH 9.4 borate buffer (250 mM)), 40 equiv of core (400 mM in CH3CN), 40 equiv of DMT-MM (400 mM in H2O), RT for 3 h.
were further acylated with 4-fluoro-3-nitrobenzoic acid with DMT-MM as activating reagent.23 Total 355 amines passed at a yield of greater than 50% after two steps of chemistry. Overall only the generated secondary amines without α-substitution can afford decent yield of acylation product. Now that we had a large set of validated building blocks, we then turned to the effect of library molecular biology step on the chemistry behavior. To ensure that all the steps and operations in a library do not interfere with one another, we synthesized a library with a number of members small enough that it can be rigorously characterized. We designed a library of 6 compounds (Figure 1), with benzylamine at cycle 1, benzylamine, 2methylpropan-1-amine and 3-chloro-4-methoxyaniline at cycle 2, benzaldehyde and 2,4-dimethylthiazole-5-carbaldehyde at cycle 3. In the mock library synthesis, we observed the ligation proceeded efficiently after each step of chemistry. The chemical steps were not affected with the length of DNA as well. After completion of the test library synthesis, the DNA portion of the library was digested enzymatically using a combination of DNase and S1 Nuclease (Invitrogen). This process left the library
Figure 2. Digestion results of mock library.
products were observed, as well as the byproducts, which had been observed during the library development. For example, condensation with benzaldehyde yielded the desired 3 benzimidazole products (P1, P2, and P3) with retention time of 12.2 and 12.8 min at the LCMS of its digested sample. Two smaller peaks at retention time of 10.4 and 10.8 min were 253
DOI: 10.1021/acscombsci.8b00009 ACS Comb. Sci. 2018, 20, 251−255
Letter
ACS Combinatorial Science accounted for the byproducts (P4, P5, P6, and P7) which have been observed during aldehydes evaluation. 2,4-Dimethylthiazole-5-carbaldehyde generated similar results as benzaldehyde. These data gave us confidence that the library scheme was feasible, that the chemical steps did not interfere with ligation or vice versa, that the chemistry was well-behaved on reactants with longer DNA. Library Synthesis. Confident that we could enter production phase, we designed a library based on the same scheme as mock library (Figure 3). We conducted the library
The ELT selection for cycle 2 and cycle 3 disynthons allowed for truncation at cycle 1 position which was consistent with off-DNA compound activity. The compound has good potency, specificity and ligand efficiency properties.26
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscombsci.8b00009. The structure of the headpiece, experimental procedures, analytical methods, and LCMS analysis of the data (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*Phone: 1-781-795-4290. E-mail:
[email protected]. ORCID
Yun Ding: 0000-0002-9650-9332 Present Addresses †
X-Chem Inc., 100 Beaver Street, Waltham, MA 02453, USA. McKinsey & Company, 1 Jermyn Street, London, SW1Y4UH, United Kingdom. § Tedor Pharma, Inc., 400 Highland Corporate Drive, Cumberland, RI 02864, USA. ‡
Notes
The authors declare no competing financial interest.
■
Figure 3. Synthetic scheme for benzimidazole library.
ACKNOWLEDGMENTS The authors acknowledge Svetlana Belyanskaya for closing the library and Neil Carlson for analyzing the naiv̈ e library sequencing data.
synthesis using a split-and-pool strategy in 96-well plates. At cycle 1, two sets of building blocks were applied. 167 Fmoc amino acids and one blank were coupled to the headpiece via DMT-MM activating, followed by deprotection. The second set of building blocks was 321 primary amines, 2 secondary diamines and one blank that were reacted with the DNA-acrylamide conjugate by addition to electron-deficient double bond. After pooling the two sets, 4-fluoro-3-nitrobenzoic acid was acylated under the activation of DMT-MM. At cycle 2, the fluoride was substituted with 710 primary amines (plus one blank) through the nucleophilic aromatic substitution. At cycle 3, the diamine substrates underwent annulation with 922 aldehydes (plus one blank) to form benzimidazoles. The final library which contained 320 763 800 benzimidazole components was further ligated with cycle 4 tags and closing primer before handing off for the affinity selection. Benzimidazole chemistry has been demonstrated to be robust for the reaction on-DNA. Along with this 3-cycle library, a 2-cycle library without cycle 1 diversity was also synthesized. To date, both libraries have been screened against over 100 biological targets and delivered active hits against various targets.24,25 In many cases, we observed the same enriched features from both libraries which hinted the cycle 1 was just functioning as a linker part. For example, in selections of the 3-cycle library against NK3, a member of the tachykinin family of G-protein coupled receptors (GPCRs), a novel antagonist 1 was discovered.26
■ ■
ABBREVIATIONS DEL: DNA-encoded library HP: covalently linked DNA duplex−the “headpiece” with a free amine warhead (SI Figure 1) ELT: DNA-encoded library technology REFERENCES
(1) Brenner, S.; Lerner, R. A. Encoded combinatorial chemistry. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 5381−5383. (2) Nielsen, J.; Brenner, S.; Janda, K. D. Synthetic methods for the implementation of encoded combinatorial chemistry. J. Am. Chem. Soc. 1993, 115, 9812−9813. (3) Lerner, R. A.; Brenner, S. DNA-encoded compound libraries as open source: a powerful pathway to new drugs. Angew. Chem., Int. Ed. 2017, 56, 1164−1165. (4) Franzini, R. M.; Randolph, C. Chemical space of DNA-encoded libraries. J. Med. Chem. 2016, 59, 6629−6244. (5) Clark, M. A.; Acharya, R. A.; Arico-Muendel, C. C.; Belyanskaya, S. L.; Benjamin, D. R.; Carlson, N. R.; Centrella, P. A.; Chiu, C. H.; Creaser, S. P.; Cuozzo, J. W.; Davie, C. P.; Ding, Y.; Franklin, G. J.; et al. Design, synthesis and selection of DNA-encoded small-molecule libraries. Nat. Chem. Biol. 2009, 5, 647−654. 254
DOI: 10.1021/acscombsci.8b00009 ACS Comb. Sci. 2018, 20, 251−255
Letter
ACS Combinatorial Science (6) Kleiner, R. E.; Dumelin, C. E.; Liu, D. R. Small-molecule discovery from DNA-encoded chemical libraries. Chem. Soc. Rev. 2011, 40, 5707− 5717. (7) Wrenn, S. J.; Weisinger, R. M.; Halpin, D. R.; Harbury, P. B. Synthetic ligands discovered by in vitro selection. J. Am. Chem. Soc. 2007, 129, 13137−13143. (8) Ding, Y.; O’Keefe, H.; DeLorey, J. L.; Israel, D. I.; Messer, J. A.; Chiu, C. H.; Skinner, S. R.; Matico, R. E.; Murray-Thompson, M. F.; Li, F.; Clark, M. A.; Cuozzo, J. W.; Arico-Muendel, C.; Morgan, B. A. Discovery of potent and selective inhibitors for ADAMTS-4 through DNA-encoded library technology. ACS Med. Chem. Lett. 2015, 6, 888− 893. (9) Litovchick, A.; Dumelin, C. E.; Habeshian, S.; Gikunju, D.; Guie, M. A.; Centrella, P.; Zhang, Y.; Sigel, E. A.; Cuozzo, J. W.; Keefe, A. D.; Clark, M. A. Encoded library synthesis using chemical ligation and the discovery of sEH inhibitors from a 334-million member library. Sci. Rep. 2015, 5, 10916. (10) Seigal, B. A.; Connors, W. H.; Fraley, A.; Borzilleri, R. M.; Carter, P. H.; Emanuel, S. L.; Fargnoli, J.; Kim, K.; Lei, M.; Naglich, J. G.; Pokross, M. E.; Posy, S. L.; Shen, H.; Surti, N.; Talbott, R.; Zhang, Y.; Terrett, N. K. The discovery of macrocyclic XIAP antagonists from a DNA-programmed chemistry library, and their optimization to give lead compounds with in vivo antitumor actvity. J. Med. Chem. 2015, 58, 2855−2861. (11) Gilmartin, A. G.; Faitg, T. H.; Richter, M.; Groy, A.; Seefeld, M. A.; Darcy, M. G.; Peng, X.; Federowicz, K.; Yang, J.; Zhang, S.-Y.; Minthorn, E.; Jaworski, J.-P.; Schaber, M.; Martens, S.; McNulty, D. E.; Sinnamon, R. H.; Zhang, H.; Kirkpatrick, R. B.; Nevins, N.; Cui, G.; Pietrak, B.; Diaz, E.; Jones, A.; Brandt, M.; Schwartz, B.; Heerding, D. A.; Kumar, R. Allosteric Wip1 phosphatase inhibition through flapsubdomain interaction. Nat. Chem. Biol. 2014, 10, 181−187. (12) Fernández-Montalván, A. E.; Berger, M.; Kuropka, B.; Koo, S. J.; Badock, V.; Weiske, J.; Puetter, V.; Holton, S. J.; et al. Isoform-selective ATAD2 chemical probe with novel chemical structure and unusual mode of action. ACS Chem. Biol. 2017, 12 (11), 2730−2736. (13) Gartner, Z. J.; Kanan, M. W.; Liu, D. R. Expanding the reaction scope of DNA-templated synthesis. Angew. Chem., Int. Ed. 2002, 41, 1796−1800. (14) Malone, M. L.; Paegel, B. M. What is a “DNA-compatible” reaction? ACS Comb. Sci. 2016, 18, 182−187. (15) Li, Y.; Gabriele, E.; Samain, F.; Favalli, N.; Sladojevich, F.; Scheuermann, J.; Neri, D. Optimized reaction conditions for amide bond formation in DNA-encoded combinatorial libraries. ACS Comb. Sci. 2016, 18, 438−443. (16) Ding, Y.; Clark, M. A. Robust Suzuki-Miyaura cross-coupling on DNA-linked substrates. ACS Comb. Sci. 2015, 17, 1−4. (17) Satz, A. L.; Cai, J.; Chen, Y.; Goodnow, R.; Gruber, F.; Kowalczyk, A.; Petersen, A.; Naderi-Oboodi, G.; Orzechowski, L.; Strebel, Q. DNA compatible multistep synthesis and applications to DNA encoded libraries. Bioconjugate Chem. 2015, 26, 1623−1632. (18) Du, H.-C.; Huang, H. DNA-compatible nitro reduction and synthesis of benzimidazoles. Bioconjugate Chem. 2017, 28 (10), 2575− 2580. (19) Examples of reviews: (a) Yadav, G.; Ganguly, S. Structure activity relationship (SAR) study of benzimidazole scaffold for different biological activities: A mini-review. Eur. J. Med. Chem. 2015, 97, 419− 443. (b) Kaur, G.; Kaur, M.; Silakari, O. Benzimidazoles: An ideal privileged drug scaffold for teh design of multitargeted antiinflammatory ligands. Mini-Rev. Med. Chem. 2014, 14 (9), 747−767. (c) Song, D.; Ma, S. Recent development of benzimidazole-containing antibacterial agents. ChemMedChem 2016, 11 (7), 646−659. (20) Examples of literatures: (a) Chi, Y.-C.; Sun, C.-M. Soluble polymer-supported synthesis of a benzimidazole library. Synlett. 2000, 5, 591−594. (b) Carpenter, R. D.; Deberdt, P. B.; Lam, K. S.; Kurth, M. J. Carbodiimide-based benzimidazole library method. J. Comb. Chem. 2006, 8 (6), 907−914. (c) Yun, Y. K.; Porco, J. A., Jr.; Labadie, J. Polymer-asisted parallel solution phase synthesis of substituted benzimidazoles. Synlett 2002, 5, 739−742.
(21) The reaction also contained certain amount of organic solvent, which was used to dissolve the small molecule reagents. The percentage of organic solvent is dependent on the amount of small molecule reagents used for each reaction. (22) Leyden, R. N.; Loonat, M. S.; Neuse, E. W.; Sher, B. H.; Watkinson, W. J. Thermally induced degradation of 2,3,5,6tetrachloroterephthalylidenebis(o-aminoanline). J. Org. Chem. 1983, 48, 727−731. (23) Ding, Y.; Franklin, G. J., Jr.; DeLorey, J. L.; Centrella, P. A.; Mataruse, S.; Clark, M. A.; Skinner, S. R.; Belyanskaya, S. Design and synthesis of biaryl DNA-encoded libraries. ACS Comb. Sci. 2016, 18, 625−629. (24) Lewis, H. D.; Liddle, J.; Coote, J. E.; Atkinson, S. J.; Barker, M. D.; Bax, B. D.; Bicker, K. L.; Bingham, R. P.; Campbell, M.; Chen, Y. H.; Chung, C.; Craggs, P. D.; Davis, R. P.; Eberhard, D.; Joberty, G.; Lind, K. E.; Locke, K.; Maller, C.; Martinod, K.; Patten, C.; Polyakova, O.; Rise, C. E.; Rüdiger, M.; Sheppard, R. J.; Slade, D. J.; Thomas, P.; Thorpe, J.; Yao, G.; Drewes, G.; Wagner, D. D.; Thompson, P. R.; Prinjha, R. K.; Wilson, D. M. Inhibition of PAD4 activity is sufficient to disrupt mouse and human NET formation. Nat. Chem. Biol. 2015, 11, 189−191. (25) Wood, E. R.; Bledsoe, R.; Chai, J.; Daka, P.; Deng, H.; Ding, Y.; Harris-Gurley, S.; Kryn, L. H.; Nartey, E.; Nichols, J.; Nolte, R. T.; Prabhu, N.; Rise, C.; Sheahan, T.; Shotwell, J. B.; Smith, D.; Tai, V.; Taylor, J. D.; Tomberlin, G.; Wang, L.; Wisely, B.; You, S.; Xia, B.; Dickson, H. The Role of Phosphodiesterase 12 (PDE12) as a Negative Regulator of the Innate Immune Response and the Discovery of Antiviral Inhibitors. J. Biol. Chem. 2015, 290, 19681−19696. (26) Wu, Z.; Graybill, T. L.; Zeng, X.; Platchek, M.; Zhang, J.; Bodmer, V. Q.; Wisnoski, D. D.; Deng, J.; Coppo, F. T.; Yao, G.; Tamburino, A.; Scavello, G.; Franklin, G. J.; Mataruse, S.; Bedard, K. L.; Ding, Y.; Chai, J.; Summerfield, J.; Centrella, P. A.; Messer, J. A.; Pope, A. J.; Israel, D. I. Cell-based selection expands the utility of DNA-encoded smallmolecule library technology to cell surface drug targets: identification of novel antagonists of the NK3 tachykinin receptor. ACS Comb. Sci. 2015, 17 (12), 722−731.
255
DOI: 10.1021/acscombsci.8b00009 ACS Comb. Sci. 2018, 20, 251−255
