Expanding Reactivity in DNA-Encoded Library Synthesis Via Reversible Binding of DNA to an Inert Quaternary Ammonium Support Dillon T

Home , Reductive amination

Expanding Reactivity in DNA-Encoded Library Synthesis via Reversible Binding of DNA to an Inert Quaternary Ammonium Support Dillon T. Flood,1⊥ Shota Asai,1⊥ Xuejing Zhang,1 Jie Wang,1 Leonard Yoon,1 Zoë C. Adams,1 Blythe C. Dillingham,1 Brittany B. Sanchez,§ Julien C. Vantourout,1 Mark E. Flanagan,2 David W. Piotrowski,2 Paul Richardson,3 Samantha A. Green,1 Ryan A. Shenvi,1 Jason S. Chen,§ Phil S. Baran,1* Philip E. Dawson.1 *

1Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, United States 2Pfizer Medicinal Chemistry, Eastern Point Road, Groton, CT 06340, United States 3Pfizer Medicinal Chemistry, 10578 Science Center Drive, San Diego, CA 92121, United States §Automated Synthesis Facility, Scripps Research, 10550 North Torrey Pines Road, La Jolla, CA 92037, United States ABSTRACT: DNA Encoded Libraries have proven immensely powerful tools for lead identification. The ability to screen billions of compounds at once has spurred increasing interest in DEL development and utilization. Although DEL provides access to libraries of unprecedented size and diversity, the idiosyncratic and hydrophilic nature of the DNA tag severely limits the scope of applicable chemistries. It is known that biomacromolecules can be reversibly, non-covalently adsorbed and eluted from solid supports, and this phenomenon has been utilized to perform synthetic modification of biomolecules in a strategy we have described as reversible adsorption to solid support (RASS). Herein, we present the adaptation of RASS for a DEL setting, which allows reactions to be performed in organic solvents at near anhydrous conditions opening previously inaccessible chemical reactivities to DEL. The RASS approach enabled the rapid development of C(sp2)-C(sp3) decarboxylative cross-couplings with broad substrate scope, an electrochemical amination (the first electrochemical synthetic transformation performed in a DEL context), and improved reductive amination conditions. The utility of these reactions was demonstrated through a DEL-rehearsal in which all newly developed chemistries were orchestrated to afford a compound rich in diverse skeletal linkages. We believe that RASS will offer expedient access to new DEL reactivities, expanded chemical space, and ultimately more drug-like libraries.

Introduction mM, due to solubility considerations) making bimolecular Sydney Brenner and Richard Lerner’s seminal 1992 reactions sluggish, and (3) a high degree of report established a profound, new type of combinatorial chemoselectivity is required so as not to disturb the functional-group rich nucleotide backbone.2,5,6,8 The chemistry.1 They postulated that individual chemical transformations could be encoded in DNA, resulting in pragmatic result of these factors is that most modern DEL libraries, while exhibiting broad diversity from the libraries of unprecedented size and chemical diversity.1 monomers employed, are comprised of a severely limited Since their proposal, many groups and pharmaceutical 2,9 companies have invested heavily into DEL research and set of skeletal linkages. To be sure, these are mostly comprised of amides, biaryls, and C–N linkages through technology.2–4 Modern, industrialized DEL libraries routinely contain billions of compounds that are screened 1,3,5-triazine “hubs” which create planar libraries lacking significant 3D shape.2,9 Thus, although great numbers of for biological activity, all at once.2–4 Although many compounds can be generated, often, drug likeness and success stories have resulted from DEL-based discovery 9 campaigns, including multiple therapeutic candidates in implicit diversity suffer. Even with those caveats, such clinical trials, the synthetic pathways employed during libraries have shown some success for lead identification, DEL construction lag severely behind the unconstrained fueling a resounding call for the development of more methodologies of modern organic and medicinal interesting DEL compatible chemistries and ultimately more drug-like libraries.11 chemistry.2,5–7 This glaring disparity can be attributed to the idiosyncratic reaction requirements of the encoding Numerous labs have chosen to directly address this molecule, DNA, which manifests in three confounding challenge by adapting reaction conditions to fit the unusually ways (Figure 1A): (1) as DNA is insoluble in most organic demanding requirements of DEL synthesis.2,12–14 Although solvents, reactions need to be conducted in the presence this approach has encountered some success, of water, (2) highly diluted conditions are required (<1 1

A DEL Synthesis the use of organic solvents and reaction conditions that would be otherwise incompatible for the biomolecule.16,17 Aqueous Reaction VS RASS Reaction Applied to DEL, the RASS strategy could dramatically expand the toolkit of available organic reactions for library construction by bringing DNA into organic solvents and protecting the backbone from reagent- induced degradation. To identify an appropriate solid support, we screened a series of commercially available inert resins/solid supports for their ability to selectively bind and elute a small DNA-fragment which mimics that [water requisite] [high dilution] [organic solvents] [high concentration] typically employed in DEL (Figure 1). An important [limited reactivity space] [expanded reactivity space] additional requirement was that the resin itself should not Me R Me B Et N N Me interject its own reactivity profile (eg. acids, bases or N R Et 2 nucleophiles). Hydrophobic resins such as reversed phase O 16 Si O O O silica could competently bind DNA, but unfortunately, the O HO HO n n OH O weak binding led to premature elution in organic media n n Reversed Phase Weak Anion Exchanger Strong Anion Exchanger (Figure 1B). Weak anion exchange resins were ruled out [Hydrophobic Interactions] [Ion-Ion Interactions] [Ion-Ion Interactions] as well, because most bear basic, nucleophilic, or [Silica] [Polystyrene] [Sugar] [Polystyrene] [Polystyrene] 15 [High % Organic [Basic Solution [Salt Solution otherwise reactive moieties (Figure 1B). The potential of Solvent] (pH > 8)] ([NaClO4] = 1M)] DNA immobilization is supported by the work of Harbury, Condition Elute EtOH ppt C Bind DNA Wash who demonstrated peptide and peptoid synthesis on DNA Resin 1 M NaClO4, Cold EtOH, PBS, 5 min 1:1 ACN:H2O 18,19 1:1 ACN:H2O Buffer Spin, Dry using a weak anion exchange resin (DEAE Sepharose). However, the presence of abundant hydroxylated Binding Eluant functionality and tertiary amines has limited the scope of Supernatant employable reactions (vide infra). In contrast, a mixed mode polystyrene strong anion Absorbance (Au) Absorbance Absorbance (Au) Absorbance exchange resin, (Phenomenex, Strata-XA) which 5 10 15 20 5 10 15 20 Time (Mins) Time (Mins) contains a butyl quaternary ammonium moiety, proved to be an excellent platform for RASS on DNA (Figure 1B). Figure 1. DEL Synthesis via RASS. (A) Aqueous vs RASS Such resins incorporate both hydrophobic interactions reactions for DEL. (B) Resins selection considerations. (C) (polystyrene, butyl substituents) and electrostatic Basic DEL RASS workflow. DNA binding and elution of DNA interactions (quaternary amine) and effectively anchor by HPLC the DNA in a polyvalent manner. When model DEL it has often proven to be a time consuming and laborious headpiece with a pendent free amine, SI-1, in PBS (100 endeavor. For instance, the recently developed DNA- µM DNA), was added to the resin, it efficiently adsorbs to compatible Giese reaction developed by our lab (PSB) for the surface, as indicated by the lack of DNA detected in use in the construction of high value sp3-sp3 linkages in the binding supernatant (Figure 1C). The bound DNA SI- DEL required a unique method for kinetic evaluation and 1 could then be eluted into a high salt elution buffer, an optimization involving hundreds of optimization approach that is widely applied in molecular biology.20 experiments.13 Clearly, adapting organic reactions for use Importantly, the Strata-XA resin allows the DNA-resin in dilute water presents many difficulties as many complex to be transferred from aqueous solution into neat interesting bond forming reactions invoke water- organic solvents. Following removal of the solvent from incompatible reagents or intermediates. Thus, the the resin, the DNA SI-1 can be washed with aqueous dominant paradigm in this field is to bring organic buffer, eluted from the resin with salt and isolated reactions into water, whereas the simplest approach through ethanol precipitation. Failures encountered with might just be to bring DNA into organic solvents. numerous other resins that were tested can be attributed to a lack of strong initial DNA binding or lack of retention An Organic Approach to DEL-Chemistry in organic solvents. This represents a robust platform for Recent work in our lab (PED) exploited peptide and the controlled, reversible binding and elution of DNA protein immobilization as a tool for synthetic chemistry. fragments to facilitate manipulation in organic media in The Reversible Adsorption to Solid Support (RASS) ways that are not possible using conventional aqueous approach, leverages the multivalent binding kinetics of DEL techniques. As an initial proof of concept, a simple biomacromolecules to selectively bind to and elute from amide bond formation reaction SI-2 was performed on a an inert solid support, such as reversed phased silica.16,17 model DEL headpiece SI-1 with organic solvents (CH2Cl2, Critically, the differences in binding kinetics of THF, DMF, Dioxane, DMSO, and MeCN, see SI) biomacromolecules and small molecules allow for commonly used in small molecule chemistry (>75% adsorbed biomolecules to react with small molecule yield). Interestingly the observed yields were consistent reagents with the same logic as employed in conventional with solvent depenedencies for small molecue 21 (covalently bound) solid phase organic synthesis carbodiimide couplings. strategies.16,17 Thus, excess small molecule reagents are simply washed away while the polyvalent macromolecule Expanding the Reaction Space of DEL remains adsorbed to the solid support.16,17 This allows for To demonstrate the advantage of utilizing an inert support via RASS, a reaction that was previously 2 recalcitrant in a DEL-setting was enabled: forging highly binding (weak anion exchangers and reversed phase) desirable sp2-sp3 linkages from ubiquitous building blocks resins (Entry 10, See SI). and reagents. Specifically, the cross coupling between With optimized conditions in hand, the utility of the abundant (het)aryl halides and redox active esters (RAEs) reaction was demonstrated with >40 different relevant derived from simple, ubiquitous, carboxylic acids was carboxylic acids. Excellent to satisfactory yields were pursued. Numerous attempts at this reaction built off the obtained by utilizing either isolated RAEs or RAEs success of the Giese reaction, by enlisting aryl iodides and generated in-situ without altering conditions (Figure 2C). RAEs in aqueous solution using Zn as a reductant under Multiple different (het)aryl iodides (>5), substituted at Ni-catalysis (SI).13 Recent findings from the Molander various positions, are also competent coupling partners group demonstrated the feasibility of such a bond affording similar yields. Finally, the scope could be formation, albeit only in the case of carboxylic acids further expanded to aryl bromides which are more widely bearing a radicle stabilizing a-heteroatom.22 Aiming for a available. DNA-bound aryl bromide could be employed transformation without this inherent limitation, 4- simply upon addition of 250 mM NaI forming coupled iodobenzoic acid was coupled to a common DEL head products, albeit in slightly attenuated yields. In accord piece5,8 resulting in DNA-Ar-I 1 (Figure 2A), setting the with standard practice in DEL synthesis, yields are stage for a coupling with unstabilized RAE 2. determined by integration of the total UV absorbance at Conventional DEL-based conditions were interrogated 260 nm via HPLC/MS of the crude reaction mixture.13,28,29 (see SI for details) resulting in a maximum conversion of This protocol provides an accurate measure of relative ca. 5% to 3 (entry 2). It was reasoned that a more soluble yield on nanomole scale as DNA is the dominant reductant capable of reducing both the RAE and the NiII chromophore.28,29 To confirm the robustness of our precatalyst to its active low valent Ni state would be measurements, all of experiments using our preferred superior. Reports from the Weix23 group and ours24 Protocol B (Figure 2B, Entry 9) were conducted in demonstrated that low valent Ni is competent to reduce triplicate. RAEs and form transient alkyl radicals while reports from Within the known realm of DNA-compatible chemistry, the Shenvi group show that the combination of various oxidative reactions are rare. This is because chemical silanes and base can reduce NiII to low valent Ni.25–27 oxidants lack chemoselectivity, indiscriminately attacking “RubenSilane” (isopropoxy(phenyl)silane, RS), the sensitive functional groups found on the purine developed by the Shenvi group, was therefore employed portion of the DNA backbone leading to degradation.30–34 but unfortunately did not improve the conversion (entry Such an event in the context of DEL is deleterious and 3). Numerous other attempts were pursued changing irreversible as critical encoding information can be various reagent concentrations, solvent systems and permanently lost.31 Synthetic organic electrochemistry surfactants to no avail. offers an opportunity to precisely control redox-potentials As reported previously, the protocols used to improve and provide superior selectivity but has never been the yield of a DEL-based reaction are quite different due applied in such a context.35,36 This is likely due to the to the kinetics of highly diluted conditions in water.13 In charged nature of DNA that, upon exposure to an stark contrast, the RASS-based approach enabled a more electrode surface, will irreversibly be adsorbed. RASS traditional optimization strategy that ultimately led to offers a unique opportunity to site-isolate DNA thus success. Thus, although application of Zn-based rendering it amenable to redox transformations conditions to this system resulted in a heterogeneous dual accessible electrochemically that might otherwise destroy surface reaction that could not be readily optimized (see it.37 To test this hypothesis we turned to our recently SI), the homogeneous solution using RS as a soluble reported Ni-catalyzed electrochemical aryl amination due reductant proved immediately fruitful (entry 6) providing to its highly modular nature and potential to improve trace product 3 (Figure 3B, entry 6). From this initial hit, upon the harsh conditions that are currently used in the reaction was optimized to yield 3 in 84% after 18-24 h analogous Pd- and Cu-based reactions (Table 2A).38–40 (entry 8) by increasing the concentration of Ni catalyst Indeed, Ullmann-Buchwald-Hartwig type aminations are and reductant (see SI for further details). The reaction workhorse reactions to generate diversity in medicinal time could be dramatically reduced by simply subjecting chemistry and drug discovery yet robust applications in the substrate to two cycles of reagent addition (82% yield, the context of DEL are lacking.9,41 The DEL-compatible 3 h total reaction time, entry 9). In striking contrast, variants require exotic ligand systems, scavengers and extremely limited DNA recovery was observed (<5%) and high temperatures due to the dilute aqueous conditions only trace product was detected when utilizing sepharose employed.40 The Ni-catalyzed system was therefore or polystyrene-based weak anion exchange systems, with investigated to probe both the DEL-compatibility of the optimized reaction conditions. Most likely, the basic electrochemistry and to explore potential synergies with nature of the reaction led to deprotonation of the tertiary the RASS approach. amino groups of the DEAE sepharose, which are critical As predicted, attempts to employ electrochemical to DNA retention, and ultimately led to premature DNA amination with a traditional aqueous system led to no release. This trend held true with a variety of other weakly recoverable DNA (Figure 3B, entry 1). In striking contrast, the site-isolation garnered by the RASS approach led to

A. On-DNA Decarboxylative sp2—sp3 Cross Coupling

OH RASS Reaction Bz O O N 3' C C C T C A G T O P O [Ni], Reductant 3 O O H Bz N O N N OH H NHPI DNA 5' G A G T C A P O O RAE (100 mM) (2) O P I OH 3 O (1) O (3) B. Optimization table t Entry NiBr2•diglyme (mM) bubpy (mM) Conditions Yield (%) 1 80 40 Zn, TPPTS (40 mM), AcOH, DMA/H2O (4/1) 0 a 2 N/Aa N/Aa Zn, DMA, (Representative of >100 Reactions, see SI) <5 3 100 200 RS (300 mM), K2CO3 (300 mM), DMA/H2O (4/1), 18 h No DNA 4 10 20 DEMS (300 mM), K2CO3 (300 mM), DMA, 18 h 0 5 10 20 PhSiH3 (300 mM), K2CO3 (300 mM), DMA, 18 h 0 6 10 20 RS (300 mM), K2CO3 (300 mM), DMA, 18 h 3 RS (300 mM), K CO (300 mM), DMA, 18 h 75 7 100 200 2 3 RS (900 mM), K CO (900 mM), DMA, 18 h (Protocol A) 8 100 200 2 3 84 RS (900 mM), K CO (900 mM), DMA, 2 x 1.5 h (Protocol B) 9 100 200 2 3 82 10 100 200 Reversed Phase and Weak Anion Exchange Resins Trace DNA 11 100 200 Other Strong Anion Exchange Resins up to 70% C. Scope of the RASS reaction From aryl iodide R S N R n N N O H H H Boc H H N N N N Boc N DNA DNA DNA DNA DNA O O O O O b,c a,c b,c a,c b a b a b R = Bz, 3: 78%±3.8, 84% R = CH2, 6: 62%±6.8, 65% n = 1, 9: 83.3%±7.2 , 78% 11: 75%±2.6, 80% 12: 61%±10.6 b,c a,c b,c b a R = Boc, 4: 67%±4.6, 77% R = CF2, 7: 67%±4.9 n = 2, 10: 80%±12.5, 87% R = Ts, 5: 59%±7.2,b,c 56%a,c R = O, 8: 52%±14,b,c Me Me Me Br Me O H H H Me H N H N N N N DNA DNA DNA Cl DNA DNA O O O O O 13: 65%±11b,c 14: 68%±3.5,b 65%a 15: 60%a 16: 50%±12.2b,c 17: 69%±8.6b,c Cl

F Boc NHBoc HN N Cl H H H H H N C F N N N N DNA 6 5 DNA DNA DNA DNA O O O b,c O b,c O b,c b a 18: 60%±3.5 19: 73%±6.0 O 20: 66% 21: 73.3%±5.5 22: 72% OBn H R N H H O H 2 H H N N N N Boc N DNA DNA DNA DNA DNA Br O O O a,c O a Bz O a,d R = CH2CO2Me, 25: 49% a a 23: 97% 24: 70% a,c 27: 60% 28: 60% N R = Ph, 26: 67% Bz N Boc N m m O N H o o O DNA N Boc H H H N DNA N O N N H DNA DNA N 4 DNA H O b,c a O O O 34: 82%±8.1 35: 55% o, 29: 62%a o, 31: 60%±13.1b,c 33: 60%±13.1b,c a From aryl bromide m,, 30: 53% m,, 32: 82%±2.11b,c R Boc N Bz Bz N N N Boc n H N H N N H DNA Ph N N DNA N H N H N Me DNA O H N Me N O N DNA N DNA O DNA R = Bz, 39: 81%±4.5b,c n=1, 42: 67.3%±4.0b 44: 70%a,c O O O R = Boc, 40: 62%±8.1b,c n=2, 43: 84%b a b,c b,c 36: 68% 37: 45%±3 38: 43%±3 R = Ts, 41: 50%a,c Figure 2. On-DNA Decarboxylative sp2-sp3 Cross-Coupling. (A) Reaction Scheme. (B) Optimization Table; a Reactions include small molecule reactions under DEL conditions and on-DNA DEL reactions (C) Scope Table; a Protocol A (18 hr), b Protocol B (2 x 3 hr), c Isolated RAE, d 1:1 desired rroduct:reduced product. 4 promising results on the first attempt. Thus, adapting the as well as one amide proved competent coupling partners published conditions employing a Ni(bpy)3Br2 precatalyst with yields ranging from good, to useable (Figure 3C). (50 mM), DBU (300 mM), and 4 mA of current for 3 Interestingly, aniline and most of the piperazines tested hours, furnished 37% of the alkyl-aryl amine 45 from aryl proved to be incompetent coupling partners (See SI). The iodide 1 (entry 2 ).38 Although this initial result was observed yields are in line (20-70%) with those previously encouraging, the major side product was the seen using the harsh conditions mentioned above.40 More corresponding phenol, likely a result of hydroxide importantly, this study provides a proof of concept for the formation in situ. Removing DBU reduced the use of electrochemical methodologies in the demanding hydroxylated side product at the expense of slightly context of DEL. attenuated yields of 45 (34%, entry 3). The addition of 4 The RASS-based DEL platform presented above not Å molecular sieves to the base-free protocol reduced the only enables access to new reaction manifolds, it also can hydroxylated side product and moderately boosted improve the scope of known DEL-compatible reactions. conversion (50%, entry 5). Finally, increasing Towards this end, reductive amination, listed as among concentration of the Ni precatalyst (100 mM) brought the top-three reactions used in medicinal, is often yields to acceptable levels (74%, entry 6). With optimized conditions in hand, various alkyl, and heteroaryl amines

A. On-DNA Electrochemical Amination

OH O O RASS Reaction nBu 3' C C C T C A G T O P O NH O [Ni], Electrolyte H O N OH H H2N Me N 5' G A G T C A P O O DNA O P I OH Aqueous Reaction O O (1) (45) B. Optimization table

Entry NiBr2(bipy)3 (mM) TBAB (mM) Conditions Yield (%) 1 100 50 Amine (100 mM), Graphite Electrodes, 1:1 DMA:H2O, 4 mA, 3 h No DNA 2 50 50 DBU (300 mM), Amine (100 mM), Graphite Electrodes, DMA, 4 mA, 3 h 37 3 50 50 Amine (100 mM), Graphite Electrodes, DMA, 4 mA, 3 h 34 4 50 50 Amine (500 mM), Graphite Electrodes, DMA, 4 mA, 3 h 7 5 50 50 Amine (100 mM), Graphite Electrodes, DMA, 4 mA, 3 h, 4 Å MS 50 6 100 50 Amine (100 mM), Graphite Electrodes, DMA, 4 mA, 3 h, 4 Å MS 74 7 100 50 Amine (100 mM), RVC Electrodes, DMA, 4 mA, 3 h, 4 Å MS 42 8 100 50 Amine (100 mM), Glassy Carbon Electrodes, DMA, 4 mA, 3 h, 4 Å MS 42 C. Scope of the RASS reaction

From amines H H H H H N N N N N H H H H H N N N DNA N N N n DNA DNA Me Me DNA MeO DNA O CF3 O Me O O O a 50 (34%a) 45 (74%a) 46 (51%±1a) 47 (41%a) n =1, 48 (54% ) n = 2, 49 (48%a) t O O Bu R R Me H O O N N N N NH H H H H N N Ph H N N DNA DNA N DNA DNA DNA O O Ph O O O R = O, 55 (51%) 57 (47%a) R = Me, 52 (55%a) a 54 (43% ) R = S, 56 (49%a) 51 (31%b) R = OH, 53 (46%a)

n From heteroaryl amines or amides N H H H H N N N N N N N DNA N H H N H NH H O N N N N O DNA DNA DNA N DNA a n =1, 58 (63% ) O O O O n = 2, 59 (26%a) a b n = 3, 60 (47% ) 62 (18% ) 63 (25%b) 64 (28%b) 65 (21%b) n =4, 61 (31%a) Figure 3. On-DNA Electrochemical Amination. (A) Reaction Scheme. (B) Optimization Table (C) Scope Table, a Conditions from Entry 6, b Conditions from Entry 6 + DBU (100 mM). 5 employed in DEL-library synthesis despite the reducible imine species and is a common tactic in fundamental limitation that the amine partner must be borohydride mediated strategies.43 In classical organic used in excess.2,9,29,31 Specifically, performing reductive chemistry conditions the use of excess carbonyl is aminations between carbonyl-containing compounds and generally avoided, as dialkylated products can result.43 an amine partner loaded on-DNA has proven difficult and After significant optimization, conditions were identified inefficient. For example, within Pfizer’s DEL-based to couple 4-heptanone 68 to DNA-headpiece 66, a reductive amination screen using 218 different carbonyl commercially available scaffold that is widely employed in compounds, less than 50 provided yields above 50%. DEL studies, to deliver 71 in 75% yield (Figure 4B, entry Acyclic ketones, benzylic carbonyls, and sterically 8). Multiple ketones and aldehydes proved viable in this hindered carbonyls were all challenging substrates. transformation with yields ranging from excellent to Noticing this limitation in current DEL technology satisfactory. The conditions identified also allow for the conditions were developed to overcome this challenge. use of other on-DNA primary and secondary amine Reductive aminations are typically performed with the building blocks in good yield. The enabling effect of boric carbonyl on-DNA and a vast excess of amine in acid could also be enlisted in a purely aqueous DEL solution.31,42 These conditions force the equilibrium of the system (Entry 9).

On-DNA Reductive Amination

OH RASS Reaction O O 3' C C C T C A G T O P O 3 Acid, NaBH3CN H O O NH2 N O N O O (67) DNA NH OH H 4 5' G A G T C A P O O Pr Pr 4 O P O OH 3 (66) Aqueous Reaction (68) Pr Pr O Optimization table

Entry NaBH3CN (mM) Ketone (mM) Conditions Yield (%) 1 x x Representative of >100 Reactionsa 0-30 2 875 250 5% AcOH,NMP, RT, 20 h 22 3 875 1000 5% AcOH, NMP, 37º C, 20 h 29 4 875 1000 5% AcOH, NMP, 37º C, 70 h 48 5 875 1000 2% AcOH, NMP, 37º C, 70 h 69 6 875 500 2% AcOH, NMP, 37º C, 70 h 71

7 1000 500 B(OH)3 (400 mM), NMP, 37º C, 20 h 70 8 1000 500 B(OH)3 (400 mM), 4:1 NMP:H2O, 37º C, 20 h 77 9 1000 500 B(OH)3 (400 mM), 4:1 NMP:H2O, 37º C, 20 h 52 Scope of the RASS reaction (green) and comparison with aqueous (blue) Scope of ketone and aldehyde Pr Me Me Me Me DNA-PEG O DNA-PEG4 DNA-PEG4 DNA-PEG4 DNA-PEG4 4 N N Pr N Me N Ph N H H H H a a 68: 77% 69: 82%±3.5a 70: 68%±5.3a 71: 73%±2 72: 16%±1.5a b b b 52% 85% 41%b 52% 75%b I R EtO C Me O Me 2 DNA-PEG4 DNA-PEG DNA-PEG N 4 DNA-PEG4 4 N DNA-PEG4 N Me N H H N H H H a a b a 73: 79%±2.6a 74: 69%±2.2a 75: 72%±4 R = CH2, 76: 76%±5.5, 68% 78: 35%±8.5 b a b b 70%b 81%b 84% R = O, 77: 76%±6.6, 66% 41% R Boc Cl N H Cl DNA-PEG4 H NH DNA-PEG 4 DNA-PEG4 DNA-PEG4 N N DNA-PEG4 N N H N H H 2 H R = Me, 79: 76%±8.8,a 83%b I 81: 79%±3.5a 82: 27%±4a R = tBu, 80: 59%±3.7,a 85%b a a 86%b 11%b 83: 40%±14 84: 49%±16 7%b 82%b

O Scope of DNA-PEG4-AA H H O O DNA-PEG4 DNA-PEG N N 4 N DNA-PEG4 DNA-PEG4 H N H N N H H H N Pr I N N Pr a a 85: 56%±11 86: 23%±7 a 88: 51%±10a b b 87: 47%±9.1 Pr 39% 10% 51%b Pr 70%b Figure 4. On-DNA Reductive Amination. (A) Reaction Scheme. (B) Optimization Table, a Reactions conducted by Pfizer and results communicated through personal communications (C) Scope Table, a On Resin, b Aqueous Reaction.

Unlike the cross-coupling and e-amination described after the first encoding cycle, bound efficiently and eluted above, reductive amination is not strictly enabled using a from the solid support with the same properties as the RASS approach. Releasing the shackles of conventional smaller DNA headpiece. Figure 5 illustrates the multistep DEL reaction development, a classical organic approach process of DEL-rehearsal utilizing the three reactions inadvertently led us to a new set of B(OH)3-mediated reported herein. This three-cycle synthesis was devised conditions that can function with or without RASS. As using cross-coupling/deprotection, reductive amination, with prior examples, the reactions reported in Figure 4C and finally electrochemical amination. A mock DEL build were conducted in triplicate. was performed using three cycles of chemistry, resulting in a skeleton, rich in linkage diversity (Figure 5). The DNA RASS-Enabled DEL: Toward Creating Libraries headpiece-small molecule intermediates were eluted from In order for a new strategies to be adopted for use in resin after each cycle, as would be requisite in a DEL DEL library construction, individual diversity generating build. Starting with 1 a decarboxylative sp2-sp3 cross steps must be sufficiently chemoselective and DNA coupling with Fmoc-proline was performed and the recovery sufficiently efficient for at least 30% of the DNA resulting product was deprotected. A reductive amination to be recovered.31 While developing the above sp2-sp3 was performed on the resultant free amine, which allowed cross coupling, it became apparent that the reaction for the introduction of another aryl iodide moiety. The would result in sub-optimal DNA recovery (~20%). aryl-iodide product was then utilized in an Fortunately, this problem was addressed by optimizing electrochemical amination to produce the final compound the EtOH precipitation procedure. Increasing the ratio of 90 in 9% yield over 4 steps. This provides an example that EtOH:elute buffer ratio from 5:1 to 10:1 resulted in these diversity generating chemistries can be coupled in satisfactory DNA recovery of ~30%. The observed low series, through multiple RASS cycles, to yield on-DNA DNA recovery is likely a result of reaction workup and products that are rich in therapeutically relevant precipitation conditions, as opposed to actual reactivity functionality. based degradation. Fortunately, we did not observe any DNA recovery problems in any of the other reactions Structural Insights outlined above. In addition to facilitating reactions in neat organic To further establish compatibility with library solvents, the RASS-DEL approach is distinguished by a construction, the viability of the product DNA was significant resistance to DNA damage. This effect that was confirmed. After a completion of a full RASS cycle (bind, observed during the above studies could be a result of the reaction, elute, EtOH precipitate) 3 was ligated to a 20- DNA maintaining significant double helix structure while mer dsDNA primer, commonly used in DEL builds. The adsorbed to the support. To investigate the conformation ligation efficientcy was identical to 1, DNA that had not of the imobnilized DNA we used fluorescence microscopy undergone a RASS cycle (SI).14 This suggests that a DEL to observe the dsDNA specific interaction of a DNA library member can be enzymatically encoded after an intercalator (Figure 6). A 40-base pair dsDNA was organic reaction in the RASS cycle. designed to mimic the encoding molecule in a growing DEL library. As a control, two non-hydridizing 40-mers As a DEL library is assembled, the encoding DNA tag of single stranded DNA were used. Each DNA was increases in length. The applicability of the RASS adsorbed to Phenomenex Strata-XAL resin, and then approach with an elongated molecule, was demonstrated treated with SYBR Green which increases in fluorescence by selectively binding and eluting a double stranded 40 when intercalated into dsDNA. The resins were washed in base pair oligo from the same solid support used in acetonitrile (5x) and suspended in ethanol for reaction development (Srata-XA) as well as a related solid microscopy. The relative fluorescence observed by. support optimized for molecules larger than 10 kDa (Srata-XAL). This larger DNA, representative of DNA

DEL Rehersal—Three RASS Cycles

Bind Wash De-CO2 Cross-Coupling Reductive Amination e-Amination Decant Ni:dtbbpy (100:200 mM) 500 mM Ketone Ni(bpy)3Br2 (100 mM) Me RAE (100 mM) 400 mM B(OH)3 Amine(100 mM) K2CO3 (900 mM) 1 M NaBH3CN Graphite Electrodes Rubin’s Silane (900 mM) 20% H2O in NMP 4 mA, 3 h Enz. Encoding Organic Rxn Fmoc-Deprotection N 2 10% Piperidine in H2O 50º C, 5 h RASS Cycle [Aqueous Reaction] Me I Quench EtOH Ppt. Wash

De-CO2 Fmoc- Electro- Elute Cross- Reductive Chemical Deprotection Amination HN O HN O Coupling Amination I DNA DNA O O Me NH2 [modular] [expanded reactivity] 1 90 HO N [paralellizable] Fmoc 2 (9% over 4 steps)

Figure 5. DEL—Rehearsal, Graphical Workflow Representation and Synthesis of 90. 7 confocal fluorescence microscopy was consistent with the As shown in Figure 6, the average fluorescence intensity dsDNA construct remaining in the duplex form (strong of the double stranded was much greater than that of the relative fluorescence).44–46 All experiments were single DNA and the stained resin. The increase in normalized to the total DNA content. fluorescence between the double stranded and single stranded samples was statistically significant (p <0.0001) Structural Insights while the difference in fluorescence between the single stranded DNA and resin (without any DNA) was A insignificant (p >0.05). If the DNA was predominantly denatured upon adsorbing to the resin, a reduced fluorescence output would have been observed, similar to that of the single stranded control. The observed difference in average fluorescence intensity supports the notion that DNA is double stranded while adsorbed to the resin.44,45

Conclusion As the use of DEL in drug discovery increases, there is a growing need to expand the tool box of organic transformations available to practitioners. In this study we have outlined how strong-anion exchange resins based on quaternary ammonium moieties (RASS) can be used to facilitaite reactions in orgainic solvents. Reactions previously outside the realm of DEL including, complex sp2-sp3 cross couplings and electrochemical aminations have been demonstrated. The realization of a B(OH)3- mediated reductive amination with expanded generality B in both aqueous and RASS contexts was also discovered. Application of these three reactions in a DEL-rehearsal suggests that this technology is applicable to real-world library construction. Finally, structural studies have confirmed that DNA is still double-stranded while bound to the resin, a necessary proviso to protect the encoding molecule. As such, it is anticipated that this approach to DEL construction will find widespread utility.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website.

Detailed experimental procedures and data (PDF)

AUTHOR INFORMATION C 60 Corresponding Author dsDNA ssDNA *[email protected] *[email protected] **** Resin 40 AUTHOR CONTRIBUTIONS ⊥These authors contributed equally.

20 ACKNOWLEDGMENT Financial support for this work was provided by NIH (GM-

Relative Frequency (%) 118176) and Pfizer. We are grateful to Dr. Dee-Hua Huang and Dr. Laura Pasternack (Scripps Research) for assistance 0 with nuclear magnetic resonance (NMR) spectroscopy. Relative Fluorescence REFERENCES Figure 6. Structural Insights. Confocal Microscopy Image of (1) Brenner, S.; Lerner, R. A. Encoded Combinatorial Resin with (A) Double Stranded DNA (B) Single Stranded Chemistry. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 5381– DNA Adsorbed and Stained with SYBR Green. (C) 5383. Quantification of Resin Fluorescence. (2) Neri, D.; Lerner, R. A. DNA-Encoded Chemical Libraries: A Selection System Based On Endowing Organic 8

Compounds with Amplifiable Information. Annu. Rev. (20) Budelier, K.; Schorr, J. Purification of DNA by Anion- Biochem. 2018. Exchange Chromatography. Cur. Prot. in Molecular (3) Clark, M. A.; Acharya, R. A.; Arico-Muendel, C. C.; Biology 1998, 42, 2.1.11-2.1.18. Belyanskaya, S. L.; Benjamin, D. R.; Carlson, N. R.; (21) Applications of Carbodiimides. In Chemistry and Centrella, P. A.; Chiu, C. H.; Creaser, S. P.; Cuozzo, J. W.; Technology of Carbodiimides; John Wiley & Sons, Ltd, et al. Design, Synthesis and Selection of DNA-Encoded 2007; pp 259–281. Small-Molecule Libraries. Nat. Chem. Bio. 2009, 647– (22) Phelan, J. P.; Lang, S. B.; Sim, J.; Berritt, S.; Peat, A. J.; 654. Billings, K.; Fan, L.; Molander, G. A. Open-Air Alkylation (4) Nielsen, J.; Brenner, S.; Janda, K. D. Synthetic Methods Reactions in Photoredox-Catalyzed DNA-Encoded for the Implementation of Encoded Combinatorial Library Synthesis. J. Am. Chem. Soc. 2019, 141, 3723– Chemistry. J. Am. Chem. Soc. 1993, 115, 9812–9813. 3732. (5) Goodnow, R. A. A Brief History of the Development of (23) Huihui, K. M. M.; Caputo, J. A.; Melchor, Z.; Olivares, A. Combinatorial Chemistry and the Emerging Need for M.; Spiewak, A. M.; Johnson, K. A.; DiBenedetto, T. A.; DNA-Encoded Chemistry. In A Handbook for DNA- Kim, S.; Ackerman, L. K. G.; Weix, D. J. Decarboxylative Encoded Chemistry; Wiley-Blackwell, 2014; pp 19–43. Cross-Electrophile Coupling of N-Hydroxyphthalimide (6) Goodnow, R. A. The Changing Feasibility and Economics Esters with Aryl Iodides. J. Am. Chem. Soc. 2016, 138, of Chemical Diversity Exploration with DNA-Encoded 5016–5019. Combinatorial Approaches. In A Handbook for DNA- (24) Cornella, J.; Edwards, J. T.; Qin, T.; Kawamura, S.; Wang, Encoded Chemistry; Wiley-Blackwell, 2014; pp 417–426. J.; Pan, C.-M.; Gianatassio, R.; Schmidt, M.; Eastgate, M. (7) Needels, M. C.; Jones, D. G.; Tate, E. H.; Heinkel, G. L.; D.; Baran, P. S. Practical Ni-Catalyzed Aryl–Alkyl Cross- Kochersperger, L. M.; Dower, W. J.; Barrett, R. W.; Coupling of Secondary Redox-Active Esters. J. Am. Chem. Gallop, M. A. Generation and Screening of an Soc. 2016, 138, 2174–2177. Oligonucleotide-Encoded Synthetic Peptide Library. (25) Obradors, C.; Martinez, R. M.; Shenvi, R. A. Ph(i- Proc. Natl. Acad. Sci. U S A 1993, 90, 10700–10704. PrO)SiH2: An Exceptional Reductant for Metal-Catalyzed (8) Fraley, A. W. The DNA Tag. In A Handbook for DNA- Hydrogen Atom Transfers. J. Am. Chem. Soc. 2016, 138, Encoded Chemistry; Wiley-Blackwell, 2014; pp 153–169. 4962–4971. (9) Franzini, R. M.; Randolph, C. Chemical Space of DNA- (26) Shevick, S. L.; Obradors, C.; Shenvi, R. A. Mechanistic Encoded Libraries. J. Med. Chem. 2016, 59, 6629– Interrogation of Co/Ni-Dual Catalyzed Hydroarylation. J. (10) Neri, D. Twenty-Five Years of DNA-Encoded Chemical Am. Chem. Soc. 2018, 140, 12056–12068. Libraries. ChemBioChem 2017, 18, 827–828. (27) Green, S. A.; Vásquez-Céspedes, S.; Shenvi, R. A. Iron- (11) Podolin, P. L.; Bolognese, B. J.; Foley, J. F.; Long, E.; Nickel Dual-Catalysis: A New Engine for Olefin Peck, B.; Umbrecht, S.; Zhang, X.; Zhu, P.; Schwartz, B.; Functionalization and the Formation of Quaternary Xie, W.; et al. In Vitro and in Vivo Characterization of a Centers. J. Am. Chem. Soc. 2018, 140, 11317–11324. Novel Soluble Epoxide Hydrolase Inhibitor. (28) Ding, Y.; Franklin, G. J.; DeLorey, J. L.; Centrella, P. A.; Prostaglandins Other Lipid Mediat. 2013, 104–105, 25– Mataruse, S.; Clark, M. A.; Skinner, S. R.; Belyanskaya, S. 31. Design and Synthesis of Biaryl DNA-Encoded Libraries. (12) Buller, F.; Zhang, Y.; Scheuermann, J.; Schäfer, J.; ACS Comb. Sci. 2016, 18, 625–629. Bühlmann, P.; Neri, D. Discovery of TNF Inhibitors from (29) Satz, A. L.; Cai, J.; Chen, Y.; Goodnow, R.; Gruber, F.; a DNA-Encoded Chemical Library Based on Diels-Alder Kowalczyk, A.; Petersen, A.; Naderi-Oboodi, G.; Cycloaddition. Chemistry & Biology 2009, 16, 1075– Orzechowski, L.; Strebel, Q. DNA Compatible Multistep 1086. Synthesis and Applications to DNA Encoded Libraries. (13) Wang, J.; Lundberg, H.; Asai, S.; Martín-Acosta, P.; Chen, Bioconjug. Chem. 2015, 26, 1623–1632. J. S.; Brown, S.; Farrell, W.; Dushin, R. G.; O’Donnell, C. (30) Gates, K. S. The Chemical Reactions of DNA Damage and J.; Ratnayake, A. S.; et al. Kinetically Guided Radical- Degradation. In Reviews of Reactive Intermediate Based Synthesis of C(Sp3)−C(Sp3) Linkages on DNA. Chemistry; Wiley-Blackwell, 2006; pp 333–378. Proc. Natl. Acad. Sci. U.S.A. 2018, 201806900. (31) Malone, M. L.; Paegel, B. M. What Is a “DNA-Compatible” (14) Gerry, C. J.; Yang, Z.; Stasi, M.; Schreiber, S. L. DNA- Reaction? ACS Comb Sci 2016, 18, 182–187. Compatible [3 + 2] Nitrone–Olefin Cycloaddition Suitable (32) Cadet, J.; Davies, K. J. A. Oxidative DNA Damage & for DEL Syntheses. Org. Lett. 2019, 21, 1325–1330. Repair: An Introduction. Free Radical Biology and (15) Saudek, V.; Drobník, J.; Kálal, J. Immobilization of DNA Medicine 2017, 107, 2–12. on Poly(Glycidyl Methacrylate-Co-Ethylene (33) Cooke, M. S.; Evans, M. D.; Dizdaroglu, M.; Lunec, J. Dimethacrylate), Bead Cellulose and Sepharose. Polymer Oxidative DNA Damage: Mechanisms, Mutation, and Bulletin 1980, 2, 7–14. Disease. FASEB J. 2003, 17, 1195–1214. (16) Cistrone, P. A.; Dawson, P. E. Click-Based Libraries of (34) Kanvah, S.; Joseph, J.; Schuster, G. B.; Barnett, R. N.; SFTI-1 Peptides: New Methods Using Reversed-Phase Cleveland, C. L.; Landman, U. Oxidation of DNA: Damage Silica. ACS Comb. Sci. 2016, 18, 139–143. to Nucleobases. Acc. Chem. Res. 2010, 43, 280–287. (17) Flood, D. T.; Yan, N. L.; Dawson, P. E. Post-Translational (35) Yan, M.; Kawamata, Y.; Baran, P. S. Synthetic Organic Backbone Engineering through Selenomethionine- Electrochemical Methods Since 2000: On the Verge of a Mediated Incorporation of Freidinger Lactams. Angew. Renaissance. Chem. Rev. 2017, 117, 13230–13319. Chem. Int. Ed. 2018, 57, 8697–8701. (36) Walton, D. J.; Heptinstall, J. Electrochemical (18) Halpin, D. R.; Lee, J. A.; Wrenn, S. J.; Harbury, P. B. DNA Modification of Proteins. A Review. Prep. Biochem. Display III. Solid-Phase Organic Synthesis on Biotechnol. 2000, 30, 1–14. Unprotected DNA. PLOS Biology 2004, 2, e175. (37) Tajima, T.; Nakajima, A. Direct Oxidative Cyanation (19) Franzini, R. M.; Samain, F.; Abd Elrahman, M.; Mikutis, Based on the Concept of Site Isolation. J. Am. Chem. Soc. G.; Nauer, A.; Zimmermann, M.; Scheuermann, J.; Hall, 2008, 130, 10496–10497. J.; Neri, D. Systematic Evaluation and Optimization of (38) Kawamata, Y.; Vantourout, J. C.; Hickey, D. P.; Bai, P.; Modification Reactions of Oligonucleotides with Amines Chen, L.; Hou, Q.; Qiao, W.; Barman, K.; Edwards, M. A.; and Carboxylic Acids for the Synthesis of DNA-Encoded Garrido-Castro, A. F.; et al. Electrochemically Driven, Ni- Chemical Libraries. Bioconjugate Chem. 2014, 25, 1453– Catalyzed Aryl Amination: Scope, Mechanism, and 1461. Applications. J. Am. Chem. Soc. 2019.

(39) Li, C.; Kawamata, Y.; Nakamura, H.; Vantourout, J. C.; Liu, Z.; Hou, Q.; Bao, D.; Starr, J. T.; Chen, J.; Yan, M.; et al. Electrochemically Enabled, Nickel-Catalyzed Amination. Angew. Chem. Int. Ed. 2017, 56, 13088– 13093. (40) Lu, X.; Roberts, S. E.; Franklin, G. J.; Davie, C. P. On-DNA Pd and Cu Promoted C–N Cross-Coupling Reactions Medchemcomm 2017, 8, 1614–1617. (41) Brown, D. G.; Boström, J. Analysis of Past and Present Synthetic Methodologies on Medicinal Chemistry: Where Have All the New Reactions Gone? J. Med. Chem. 2016, 59, 4443–4458. (42) Luk, K.-C.; Satz, A. L. DNA-Compatible Chemistry. In A Handbook for DNA-Encoded Chemistry; Wiley- Blackwell, 2014; pp 67–98. (43) Lee, O.-Y.; Law, K.-L.; Ho, C.-Y.; Yang, D. Highly Chemoselective Reductive Amination of Carbonyl Compounds Promoted by InCl3/Et3SiH/MeOH System. J. Org. Chem. 2008, 73, 8829–8837. (44) Singer, V. L.; Lawlor, T. E.; Yue, S. Comparison of SYBR® Green I Nucleic Acid Gel Stain Mutagenicity and Ethidium Bromide Mutagenicity in the Salmonella/Mammalian Microsome Reverse Mutation Assay (Ames Test). Mutation Research/Genetic Toxicology and Environmental Mutagenesis 1999, 439, 37–47. (45) Zipper, H.; Brunner, H.; Bernhagen, J.; Vitzthum, F. Investigations on DNA Intercalation and Surface Binding by SYBR Green I, Its Structure Determination and Methodological Implications. Nucleic Acids Res 2004, 32, e103. (46) Mackay, I. M.; Arden, K. E.; Nitsche, A. Real-Time PCR in Virology. Nucleic Acids Res. 2002, 30, 1292–1305.