Battling Btk Mutants With Noncovalent Inhibitors ... - ACS Publications

Aug 29, 2016 - Battling Btk Mutants With Noncovalent Inhibitors That Overcome. Cys481 and Thr474 Mutations. Adam R. Johnson,*,†. Pawan Bir Kohli,. â...
0 downloads 0 Views 2MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Battling Btk Mutants With Non-Covalent Inhibitors That Overcome Cys481 and Thr474 Mutations Adam R. Johnson, Pawan Bir Kohli, Arna Katewa, Emily Gogol, Lisa D. Belmont, Regina Choy, Elicia Penuel, Luciana Burton, Charles Eigenbrot, Christine Yu, Daniel Fred Ortwine, Krista Bowman, Yvonne Franke, Christine Tam, Alberto Estevez, Kyle Mortara, Jiansheng Wu, Hong Li, May Lin, Philippe Bergeron, James J. Crawford, and Wendy B. Young ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b00480 • Publication Date (Web): 29 Aug 2016 Downloaded from http://pubs.acs.org on August 30, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

TITLE Battling Btk Mutants With Non-Covalent Inhibitors That Overcome Cys481 and Thr474 Mutations

Adam R. Johnson,*,† Pawan Bir Kohli,† Arna Katewa,‡ Emily Gogol,‡ Lisa D. Belmont,§ Regina Choy,§ Elicia Penuel,# Luciana Burton,# Charles Eigenbrot,¶ Christine Yu,¶ Daniel F. Ortwine,|| Krista Bowman,¶ Yvonne Franke,¶ Christine Tam,¶ Alberto Estevez,¶ Kyle Mortara,¶ Jiansheng Wu,¶ Hong Li,¶ May Lin,¶ Philippe Bergeron,% James J. Crawford,% and Wendy B. Young %



Biochemical and Cellular Pharmacology, Genentech, 1 DNA Way, South San Francisco,

California 94080 ‡

Discovery Immunology, Genentech, 1 DNA Way, South San Francisco, California

94080 §

Discovery Oncology, Genentech, 1 DNA Way, South San Francisco, California 94080

#

Biomarker Development, Genentech, 1 DNA Way, South San Francisco, California

94080 ¶

Protein Chemistry and Structural Biology, Genentech, 1 DNA Way, South San

Francisco, California 94080 ||

Computational Chemistry, Genentech, 1 DNA Way, South San Francisco, California

94080 %

Discovery Chemistry, Genentech, 1 DNA Way, South San Francisco, California 94080

*To whom correspondence should be addressed ([email protected])

ACS Paragon Plus Environment

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 37

ABSTRACT The Bruton's tyrosine kinase (Btk) inhibitor ibrutinib has shown impressive clinical efficacy in a range of B-cell malignancies, however acquired resistance has emerged and second generation therapies are now being sought. Ibrutinib is a covalent, irreversible inhibitor that modifies Cys481 in the ATP binding site of Btk and renders the enzyme inactive, thereby blocking B-cell receptor signal transduction. Not surprisingly, Cys481 is the most commonly mutated Btk residue in cases of acquired resistance to ibrutinib. Mutations at other sites, including Thr474, a gatekeeper residue, have also been detected. Herein we describe non-covalent Btk inhibitors that differ from covalent inhibitors like ibrutinib in that they do not interact with Cys481, they potently inhibit the ibrutinibresistant Btk C481S mutant in vitro and in cells, and they are exquisitely selective for Btk. Non-covalent inhibitors such as GNE-431 also show excellent potency against the C481R, T474I, and T474M mutants. X-ray crystallographic analysis of Btk provides insight into the unique mode of binding of these inhibitors that explains their high selectivity for Btk and their retained activity against mutant forms of Btk. This class of non-covalent Btk inhibitors may provide a treatment option to patients, especially those who have acquired resistance to ibrutinib by mutation of Cys481 or Thr474.

KEYWORDS: Bruton's tyrosine kinase, Btk, mutant, Cys481, C481S, C481R, Thr474, T474I, T474M, ibrutinib, resistance, kinase inhibitor

ACS Paragon Plus Environment

2

Page 3 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

INTRODUCTION Btk is a non-receptor Tec family cytoplasmic tyrosine kinase that is expressed in hematopoietic cells, including myeloid cells and B cells, but not T cells. Btk activity is integral to signaling through both the B-cell receptor (BCR) in B cells (1,2) and the Fcγ receptor in myeloid cells (2). The Btk inhibitor ibrutinib (PCI-32765, Imbruvica®) (3,4) has demonstrated clinical efficacy in a range of B-cell malignancies including chronic lymphocytic leukemia (CLL) (5,6), relapsed or refractory mantle-cell lymphoma (MCL) (7), and Waldenström's macroglobulinemia (WM) (8). Pre-clinically, Btk inhibitors have shown activity in models of BCR and Fcγ receptor driven autoimmune diseases such as collagen-induced arthritis, adjuvant induced arthritis, and lupus, which are thought to involve pathogenic myeloid or B cells (2,4,9–11). A salient facet of ibrutinib's molecular mechanism is its covalent binding to Btk. Ibrutinib (1a) (Figure 1) contains an electrophilic acrylamide moiety that has been shown to react irreversibly with Cys481 of Btk (3). However, ibrutinib also inhibits other kinases that possess this analogous cysteine residue including Blk, Bmx, Egfr, ErbB2, ErbB4, Itk, Tec, Txk, and Jak3 (4,12). Interestingly, ibrutinib also potently inhibits kinases that lack this cysteine, such as Csk, Fgr, Lck, Brk, Hck, Yes1, Frk, Ret, Flt3, Abl, Fyn, Lyn, and Src (4,12). Several other covalent inhibitors of Btk have been reported that share a similar electrophilic warhead as ibrutinib (Figure 1). CC-292 (AVL-292, spebrutinib) (2) binds to Btk in a different manner than ibrutinib, and therefore displays a different yet still broad kinase activity profile (9), whereas GS-4059 (ONO-4059) (3c) and ACP-196 (acalabrutinib) (4) are reported to be more selective than ibrutinib (12,13). Nevertheless, Btk inhibitors with improved selectivity are desired. While covalent inhibition of Btk is clinically effective, covalent inhibitors have liabilities that stem from their chemical reactivity, potential to form reactive metabolites, and potential ability to generate haptens that are recognized by the immune system as foreign and result in an unpredictable, idiosyncratic toxicity (14). Besides these safety concerns unique to covalent inhibitors, both covalent and noncovalent inhibitors in cancer indications face the inevitable reality of drug resistance. Unfortunately, acquired resistance to ibrutinib with disease progression has been reported (15–20). In one study, 5.3% of CLL patients relapsed over a median observation of 14 months (16,17). Mechanisms of resistance included mutations in Btk that render it

ACS Paragon Plus Environment

3

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 37

insensitive to ibrutinib (15,18,19) and activating mutations in phospholipase Cγ2, the substrate of Btk (15,20). The first ibrutinib-resistant Btk mutation reported was the change of Cys481 to serine (C481S) (15,16,18). In one study, 5 of 6 CLL patients who progressed while on ibrutinib therapy were found to harbor the C481S mutation (15). In a later study, 9 of 11 relapsed CLL patients had the C481S mutation (19), making it the most prevalent resistance mutant detected. Mutation of Cys481 precludes reaction of Btk with ibrutinib and other covalent inhibitors (18). Although ibrutinib can still bind noncovalently to C481S (15), its pharmacokinetics and reversible mechanism are likely not sufficient to allow increased dosages to maintain efficacy in patients with this mutation (15,18). In addition to C481S, ibrutinib resistant patients have also presented with C481R, C481F, C481Y, T474I, and T474S mutations (19). Besides the problem of resistance, ibrutinib therapy carries safety concerns, highlighted by reports of atrial fibrillation (21), drug interaction with verapimil (22), and hair and nail changes (23). Due to acquired resistance, emerging safety concerns, and the poor prognosis of patients whose disease progresses (24,25), physicians and patients need additional therapeutic options (26). We have had a long-standing interest in the development of Btk inhibitors for immune disorders such as rheumatoid arthritis (2, 27–29). During our immunology focused medicinal chemistry efforts we generated over a thousand non-covalent Btk inhibitors covering a broad range of chemical space and physicochemical properties. We have previously reported the discovery and structure-activity relationships (SAR), for wildtype (WT) Btk, of such chemical matter derived from these efforts (28,29). From our large library of proprietary Btk inhibitors, we selected a diverse, yet representative, set of molecules to profile against several of the clinically relevant Btk mutants that surfaced from ibrutinib treatment (30,31). Herein we report exquisitely selective non-covalent Btk inhibitors that have nanomolar potency in vitro, in cells, and in whole blood. Whereas covalent inhibitors lose potency against Cys481 mutants of Btk, the non-covalent inhibitors, described herein, retain potent inhibition of mutants C481S and C481R in vitro, and in transfected cells expressing C481S. Additionally, non-covalent inhibitors such as GNE-431 show pan-Btk activity, potently inhibiting all Cys481 mutants and Thr474 gatekeeper mutants tested. It

ACS Paragon Plus Environment

4

Page 5 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

is important to note that, during our medicinal chemistry program efforts (focused on immunology indications), we did not a priori design analogs for improved potency against Btk mutations (as they had not yet been clinically identified at that time). Based on extensive structural analysis of Btk in complex with our non-covalent inhibitors and covalent inhibitors, such as ibrutinib, we can easily retrospectively rationalize the SAR. This new class of non-covalent inhibitors may provide a potentially effective treatment option to patients, including those with malignancies that have become resistant to ibrutinib by mutation of Cys481 or Thr474, or even provide front line therapy to naïve patients.

RESULTS AND DISCUSSION A model of ibrutinib (1a) docked and covalently bound to Btk (Figure 2) shows Cys481, the covalent attachment point of 1a and the other known covalent Btk inhibitors 2, 3c and 4. From this covalent tether, 1a extends to the backbone amide groups of the hinge residues Glu475 and Met477 where it forms two hydrogen bonds. The phenoxyphenyl portion of 1a extends past the Thr474 gatekeeper residue and occupies a back pocket common to many kinases that is adjacent to the α-C-helix. Knowing that drug resistance in oncology is highly likely, we proactively initiated an effort to generate and characterize Btk proteins harboring mutations that could arise by single nucleotide variations of the BTK sequence. We focused on mutations of Cys481 and Thr474 as we expected these would be most likely to adversely affect interactions with ibrutinib, as illustrated by the model (Figure 2). Mutation of Cys481 is an obvious change that could confer resistance to ibrutinib and other covalent inhibitors. We mutated Cys481 to the isosteric serine residue, C481S, and also generated a less conservative mutant, C481R. In other kinases, resistance to small molecule inhibitors has been caused by mutation of the gatekeeper residue. For instance, mutation of the Bcr-Abl gatekeeper Thr315 to isoleucine (T315I) causes resistance to imatinib and dasatinib (32,33). Similarly, mutation of the Egfr kinase gatekeeper Thr790 to methionine (in the oncogenic L858R mutant) causes gefitinib resistance (34). The Btk gatekeeper residue Thr474 aligns with Thr315 in Bcr-Abl and Thr790 in Egfr. Based on the proximity of ibrutinib to Thr474 in Btk (Figure 2), we expected that mutations at this site might perturb binding of ibrutinib

ACS Paragon Plus Environment

5

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 37

and other structurally similar inhibitors. Therefore, we also generated and characterized the T474I and T474M mutants of Btk. While the C481S, C481R, and T474I mutants have all been detected clinically (19), T474M has not yet been reported in patients. Because the T474M mutation requires two nucleotide changes to the BTK gene, we expect that this mutation will be much less prevalent than those that require only a single nucleotide change (C481S, C481R, and T474I). Nevertheless, given that a Thr to Met gatekeeper mutation confers drug resistance in other kinases, such as Egfr (34), we generated and characterized the Btk T474M mutant to gain further insight to the Btk system.

Cys481 and Thr474 mutations do not increase Btk activity. The purified C481S, C481R, T474I, and T474M Btk proteins were active in the in vitro activity assay we routinely use for WT Btk (see Methods). They were profiled in order to determine their kinetic constants: Km (Michaelis constant) for ATP, kcat (turnover number), and kcat/Km (catalytic efficiency) (Table 1). As these studies were carried out using a low peptide substrate concentration (1 µM) that we expected would be well below the Km, the kinetic constants reported here should be considered apparent values. WT Btk showed a Km = 49 µM ATP, a kcat of 0.25 s-1, and a catalytic efficiency kcat/Km of 5,100 M-1s-1 (Table 1 and Figure S1). The C481S mutant had nearly identical kinetic constants (Table 1), demonstrating that the C481S mutation did not alter the catalytic activity of the enzyme. In contrast, an earlier study reported that the C481S mutant showed enhanced kinase activity over WT (15). The C481R mutant displayed reduced catalytic activity with a kcat (0.017 s-1) that was 15-fold lower than WT. C481R showed a reproducible sigmoidal initial velocity vs. substrate plot (Figure S1) in two separate experiments. Several mechanisms can cause sigmoidal kinetics such as allosteric subunit regulation, substrate cooperativity, or a kinetically preferred order of substrate binding (35). While an in depth kinetic analysis of this mutant is not reported here, C481R had sufficient catalytic activity for inhibitor testing. It showed half-maximal activity at a concentration of 48 µM ATP, similar to the Km for WT, and was used as isolated as were the other enzymes. The T474I mutant showed a slightly higher ATP Km (63 µM) and lower kcat (0.18 s-1) than WT. The T474M mutant had an ATP Km (12 µM) that was 4-fold lower than WT and had a slightly reduced kcat (0.15 s-1). The lower Km for T474M may be due to a methionine

ACS Paragon Plus Environment

6

Page 7 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

residue having a more flexible side chain that enables it to adopt a conformation that allows ATP to bind and turn over with higher efficiency. Regardless, the Cys481 and Thr474 mutations did not appear to increase the catalytic activity over WT in a way that would be consistent with stronger oncogenic signaling.

Only covalent inhibitors lose potency against Cys481 mutants. In this work we have profiled covalent and non-covalent inhibitors. As there is no unifying metric to directly compare the potency of covalent and non-covalent inhibitors, we report apparent potency as the IC50 for the specific 30-minute duration of the inhibition assay (see Supporting Information). We acknowledge that 'potency' as described here for covalent inhibitors is a combination of binding affinity and chemical reactivity, and note that for a non-covalent inhibitor the IC50 may underestimate true potency as an IC50 does not account for the potential of tight-binding. Nevertheless, the IC50 values we have determined provide a sufficiently quantitative assessment of potency under the standard assay conditions we applied to both the covalent and non-covalent inhibitors. We first assessed the potency of several known covalent Btk inhibitors and noncovalent analogs that we synthesized as tool molecules (Table 2). The covalent inhibitor ibrutinib (1a) showed an IC50 = 0.72 nM against WT Btk, comparable to the published potency (IC50 = 0.5 nM) (4). When tested against mutant C481S that can no longer react covalently, 1a was 6-fold less potent (IC50 = 4.6 nM), as seen by the shift in the inhibition curves between WT and C481S (Figure 3a). The covalent inhibitor CC-292 (2) also lost potency against C481S, but it experienced a larger 40-fold shift in potency between WT (IC50 = 22 nM) and the C481S mutant (IC50 = 908 nM) (Figure 3b). We were interested to know how much the covalent reaction of molecules like ibrutinib contributed to the apparent potency, so we synthesized 1b, a reversible analog of 1a that was saturated at the reactive motif (see Figure 1). Inhibitor 1b was 7-fold less potent (IC50 = 4.9 nM) than 1a was against WT Btk, but it showed equivalent potency against the C481S mutant (IC50 = 4.7 nM) (Table 2 and Figure 3c). Inhibitor 1b also displayed the same potency as 1a against the C481S mutant (Figures 3a and c). The covalent inhibitor GS-4059 (3c) exhibited relatively weak inhibitory activity against WT (IC50 = 52 nM) and showed a 5fold loss in potency against the C481S mutant (IC50 = 268 nM) (Table 2). We were

ACS Paragon Plus Environment

7

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 37

surprised at the somewhat low potency of this clinically used alkyne 3c (GS-4059) so we prepared an analog (3a) that contains a more commonly used acrylamide warhead. Inhibitor 3a was 9-fold more potent against WT Btk (IC50 = 5.7 nM) than 3c, suggesting that the different alkyne warhead of 3c may be the cause of the reduced efficacy against Btk relative to the acrylamide analog, 3a. We also prepared 3b as a non-covalent analog of 3a. Compound 3b was 38-fold weaker against WT than 3a, and inhibited WT and C481S with nearly equivalent potencies (IC50 = 219 and 191 nM, respectively). As we saw with the covalent/non-covalent 1a/1b analog pair, the potencies of 3b for the reversible inhibition of WT and C481S were similar to each other, and to the reversible potency of 3a (and 3c) against C481S. The data for 1b illustrate that covalent interaction is not required to generate a potent Btk inhibitor, but having the ability to trap the enzyme in a covalent dead end complex, as 1a does, clearly enhances the potency over analogs that cannot react with Cys481. Thus, chemical removal of the covalent warhead and genetic deletion of the Cys481 nucleophile by mutation had the same outcome: a significant loss in potency. Finally, ACP-196 (4) inhibited WT with modest potency (IC50 = 18 nM) and, like the other covalent inhibitors, it was much less potent against C481S (IC50 = 247 nM). The results against the C481R mutant were similar to those for C481S (Table 2). The potency of 1a against C481R (IC50 = 23 nM) was reduced 30-fold relative to its potency against WT. Surprisingly, the non-covalent analog 1b showed similarly weak inhibition of C481R (IC50 = 34 nM). All of the other covalent inhibitors, 2 (IC50 = 854 nM), 3a (IC50 = 703 nM), 3c (IC50 >1000 nM) and 4 (IC50 >1000 nM), and the noncovalent analog 3b (IC50 >1000 nM) were also less potent against C481R than C481S. These data are consistent with the notion that, in addition to loss of the Cys481 nucleophile interaction, the mutation to Arg481 may introduce steric and/or electronic repulsion that also reduces binding affinity, consistent with the predicted binding orientation of 1a (Figure 2).

Non-covalent inhibitors retain potency against Cys481 mutants. Inhibitors 5–9 include a wide range of chemical structures that differ structurally and mechanistically from all reported covalent inhibitors. These inhibitors are non-covalent and reversible, yet all are highly potent (Table 2). For example, 6 (GNE-431) showed IC50 = 3.2 nM

ACS Paragon Plus Environment

8

Page 9 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

against WT and similar potency against the C481S mutant (IC50 = 2.5 nM) (Figure 3d and Table 2). In contrast to the covalent inhibitors, inhibitors 5–9 all retained equivalent potency against the C481S mutant, the most prevalent ibrutinib-resistant Btk mutation found in patients (15,18,19). Furthermore, inhibitors 5–9 all retained high potency against the C481R mutant as well (IC50 = 7.5–10 nM) (Table 2). A mechanistic framework can explain the differences in observed potency for covalent and non-covalent inhibitors against WT and C481S (Figure 3e). Against WT, noncovalent inhibitors display potency that is directly related to the reversible binding inhibition constant, Ki. The observed potency (IC50) of a covalent inhibitor against WT Btk is enhanced relative to its reversible binding affinity, Ki, due to the added influence of the covalent reaction described by the inactivation rate constant, kinact. On the other hand, the covalent reaction cannot occur with C481S. Cys481-targeting inhibitors forfeit their covalent advantage and default to being non-covalent inhibitors, and potency is then directly influenced by the reversible binding affinity, Ki. Recently, the concept of designing reversible covalent inhibitors of Btk was proposed as a way to reduce reaction with off targets and enable rapid target disengagement in case of mechanism-based toxicity (36). While this approach may address certain safety concerns, any covalent inhibitor that relies on Cys481—irreversible or reversible—will lose potency against Cys481 mutants. A recent clinical study of the covalent inhibitor ACP-196 (4) in relapsed CLL reported that this molecule has higher kinase selectivity than ibrutinib, and thus may have the potential to be safer (12). Regardless, since ACP-196 (4) is also a covalent inhibitor, it also loses potency against Cys481 mutants (Table 2).

Thr474 mutations have mixed effects on covalent and non-covalent inhibitors. Inhibitor 1a showed 8-fold lower potency against T474I (IC50 = 5.6 nM) than WT, whereas 1b was 100-fold weaker against T474I (IC50 = 468 nM) than WT. By contrast, 2 inhibited T474I with an IC50 = 48 nM, similar to its potency against WT, whereas 3a and 4 both showed weak activity against T474I (IC50 = 345 and 207 nM, respectively). Both 3c and the non-covalent analog 3b were ineffective (IC50 >1000 nM) against the T474I mutant. The impact of the T474M mutation was even greater than the effect of T474I for 1a (IC50 = 66 nM) and for 3a, 3c, and 4 (all IC50 >1000 nM). Interestingly, 2 was only

ACS Paragon Plus Environment

9

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 37

slightly weaker against T474M (IC50 = 75 nM) than T474I (IC50 = 48 nM), likely reflecting its different binding interactions relative to those for 1a and 3c. Regardless, the covalent inhibitors 1a, 3a, 3c and 4 and the non-covalent analogs 1a and 3b were all negatively impacted by the T474M and T474I mutations. Non-covalent inhibitors 5–9 spanned a range of potencies against T474I. Whereas 9 was weakly active against T474I (IC50 = 439 nM) relative to its potency against WT (IC50 = 7.2 nM), 6 retained impressive potency against T474I (IC50 = 11 nM). The varied effects on potency of these covalent and non-covalent inhibitors are consistent with the idea that the change of the gatekeeper residue from Thr474 to a larger and less flexible isoleucine side chain induced steric interference against some, but not all, inhibitors and that the presence of Cys481 could not rescue potent inhibition by covalent inhibitors. By contrast, the T474M mutation had a less dramatic effect on the potencies of inhibitors 5–9. For example, 5 and GNE-431 (6) both retained impressive potency against T474M, displaying IC50 values of 9.9 and 8.8 nM, respectively. Furthermore, GNE-431 is noteworthy in that it retains high potency against all Btk forms tested: WT, three clinical mutants (C481S, C481R, and T474I), and T474M. 'Pan-Btk' inhibitors such as GNE-431 may be able to overcome the acquired resistance to ibrutinib caused by mutation of Cys481 or Thr474.

Non-covalent inhibitors are exquisitely selective for Btk. With encouraging data for inhibitors 5–9 against the Btk mutants, we compared their in vitro kinase selectivity to that of the covalent inhibitors. Inhibitors were tested at a concentration of 1 µM across a panel of 221 kinase assays (activity or binding). The results are illustrated in a heat map (Figure 4a), while the full kinase selectivity data set is detailed in the Supporting Information (Figure S2). Compound 1a was the least selective molecule tested; it showed >50% inhibition of 31 of the 220 off-target (i.e., non-Btk) kinases tested (Figures 4a and b). In agreement with published results (4,12), nine of the off targets 1a inhibited were kinases that contain a cysteine residue at the homologous position as Cys481 in Btk (Blk, Tec, Bmx, Txk, ErbB2, ErbB4, Egfr, Itk, Jak3) (Figure S2). The selectivity profile of the non-covalent analog 1b was instructive: its 25 off targets are also off targets of 1a. These kinase panel results for the covalent/non-covalent 1a/1b analog pair clearly showed that selectivity did not depend on with whether or not the molecule acted by a covalent

ACS Paragon Plus Environment

10

Page 11 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

mechanism. The diaminopyrimidine chemical core of CC-292 (2) differs from the pyrazolopyrimidine core of 1a, and the two molecules have slightly different warheads. Similar to the low selectivity of 1a, compound 2 also inhibits a large number of off targets. Not surprisingly, many of the 20 off target kinases that 2 inhibits are different from those that 1a inhibits (Figures 4a and S2). This obvious difference in off targets between two covalent inhibitors supports the idea that sharing a common inhibitory mechanism does not pre-dispose two covalent inhibitors to the same selectivity profile. Covalent acrylamide inhibitor 3a inhibited 14 off targets, whereas the weaker alkyne derivative 3c (GS-4059) inhibited only 7 off target kinases. Of the covalent inhibitors we profiled, 4 (ACP-196) was the most selective, inhibiting only 4 of 220 off targets (Figures 4a and b, and S2). However, like GS-4059 (3c), the Btk potency we measured for 4 was relatively weak (IC50 = 18 nM) compared to 1a. Thus, the selectivity of 3c and 4 may be over-estimated by these selectivity panel data, which were collected at a fixed 1-µM test concentration for all inhibitors regardless of their Btk potency. Claims have been made that covalent inhibitors are in general less selective than noncovalent inhibitors, or that focusing a covalent warhead on cysteine-containing kinases will make a molecule selective only toward kinases that contain a properly placed cysteine (37–40). On the contrary, our kinase selectivity data for 1a and the non-covalent analog 1b and the very different yet broad selectivity profile for 2 clearly showed that inhibitor selectivity is defined not by the presence or absence of a reactive moiety, but rather by the non-covalent interactions the molecule makes with binding sites of target proteins. This is consistent with the conclusions from recent studies of covalent inhibitors (41,42). Selectivity is an important factor influencing the long-term safety of a drug, but it is impossible to predict every off-target protein that a covalent inhibitor may bind to and irreversibly modify. For example, a recent proteomics report showed that ibrutinib (1a) could covalently react with non-kinase proteins in cells (43). Thus, its off targets are not limited to kinases, and more selective molecules are desirable. Encouragingly, the noncovalent molecules 5–9 all showed higher selectivity for Btk than any of the covalent inhibitors or non-covalent analogs profiled. Impressively, the most selective non-covalent inhibitor 9 did not inhibit any off-target kinase by >50% (Figures 4a and b, and S2).

ACS Paragon Plus Environment

11

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 37

Furthermore, unlike 1a these non-covalent molecules do not inhibit Egfr, and thus potentially may not induce skin rash or diarrhea side effects that are characteristic of Egfr inhibitors (44,45). Overall, these non-covalent inhibitors showed markedly increased selectivity for Btk inhibition relative to the covalent inhibitors we tested.

Structural data clarify the basis for the kinase selectivity of 9. To understand the selectivity of the most selective non-covalent inhibitor, we determined the co-crystal structure of 9 with the human Btk kinase domain. The single Btk protein kinase domain in the crystallographic asymmetric unit adopts a canonical bi-lobate form with the Asp539-Phe540-Gly541 (DFG) motif in a DFG-in conformation and a well-ordered glycine-rich loop (G-loop) in close contact with 9 (Figure 5a). The ligand is found in an extended conformation, similar to that of previously reported inhibitors (2), occupying the hinge region that is normally found in contact with the adenine moiety of ATP and projecting into close proximity of the DFG motif. Notably, 9 binds with its long axis approximately orthogonal to that of 1a (Figure 5b) and lacks any close contacts with Cys481. The activation loop is well ordered in the structure with 9 (Figure 5a) and includes ligand contacts closer than 4 Å at Tyr551. The substituted phenylpyridazinone of 9 occupies a unique pocket previously reported as the H3 selectivity pocket (2), the base of which is formed by Tyr551. The H3 pocket volume is depicted as a green surface in Figures 5a, 5b, and 5c. Inhibitors such as 1a do not occupy this pocket, and an H3 pocket has only been reported in relatively few kinases (46). Thus, inhibitors that occupy this binding site can thereby exhibit extremely high selectivity across the kinome. Although Src and Hck have been reported to have 'equivalent pockets' to the H3 pocket in Btk (46), our kinase selectivity data show only modest inhibition of Src by compounds 5– 9 and little to no inhibition of Hck (Figure S2). Selectivity for Btk versus other related kinases depends primarily on differences in residues that line the H3 site (2) (see Figures S3a and S3b). Thus, while other kinases reportedly can form an H3 pocket, none of the off target kinases that were detected possesses the exact same residues at these positions (Figure S3b). The precise threedimensional architecture and electrostatics of this site are, therefore, unique to Btk. Because our inhibitors are conformationally rigid when bound, the H3 groups are

ACS Paragon Plus Environment

12

Page 13 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

constrained to occupy a precise location that is complementary to the topography presented by these site-lining residues. Other residue differences among adjacent activation loop amino acids in other kinases may also hinder this loop from attaining the necessary low energy conformation to form the required H3 site to complement our inhibitors. While further investigation will be needed to rigorously define the existence and molecular topologies and properties of H3 pockets in other kinases, occupancy of the H3 pocket of Btk is the source of the high kinase specificity for these non-covalent molecules in general, and 9 in this example.

Modeling mutants helps to understand the pan-Btk inhibition by 6. Based on the Btk/9 co-crystal structure, we built a structural model of Btk where Cys481 was mutated to serine and arginine, and Thr474 was mutated to isoleucine and methionine (Figure 5c). We replaced 9 in the inhibitor-binding site with GNE-431 (6), a molecule that potently inhibits WT Btk as well as the C481S, C481R, T474I, and T474M mutants. The Thr474 mutations increase the steric bulk and hydrophobicity in that region of the protein (Figure 5c). This is an area where covalent inhibitors such as 1a, 2, 3c, and 4 may encounter steric repulsion from Ile474 and Met474 side chains in mutant forms of Btk (Figure 2). Even though the Cys481 nucleophile is present in the Thr474 mutants, steric interference of Ile474 or Met474 with the covalent inhibitors may perturb their binding orientation and prevent optimal alignment of the electrophile with Cys481, thereby reducing the overall potency of these inhibitors. On the other hand, GNE-431 (6) binds in an approximately orthogonal mode relative to the covalent inhibitors and does not experience steric overlap with the Thr474 residue side chain. Based on this model, the imidazopyridazine core of 6 would not be negatively affected by mutation of Thr474 to Ile or Met (Figure 5c), consistent with the results from our biochemical inhibition assays. In fact, the imidazopyridazine core of inhibitor 6 has a hydrophobic carbon in the imidazole ring that points toward the side chain of the gatekeeper residue in the computational model (Figure 5c). We hypothesize that this hydrophobic carbon may be able to favorably interact with hydrophobic gatekeeper residues such as in the T474I and T474M mutants. In summary, the orthogonal binding mode of the non-covalent inhibitors and their occupation of the unique H3 selectivity pocket confers a highly selective mode

ACS Paragon Plus Environment

13

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 37

of inhibition of Btk over other kinases that cannot form this type of binding site. In addition, the pan-Btk inhibitor GNE-431 potently inhibits WT Btk and all four of the mutants we profiled, which may allow it to overcome the ibrutinib-resistant Btk mutations found in patients.

Non-covalent inhibitors potently inhibit BCR signaling in stimulated human blood. We assessed the ability of Btk inhibitors to perform in a complex, physiological cell based environment by testing them in a human whole blood assay that measures BCRmediated activation of B cells. In the assay, blood is stimulated overnight in the presence of anti-IgM. After incubation, we assessed the cell surface expression of the B-cell activation marker, CD69, on CD19+CD27– B cells by fluorescence activated cell sorting (FACS). Compound 1a was a potent inhibitor in this assay (IC50 = 12 nM), whereas its non-covalent analog 1b was less active (IC50 = 589 nM) (Figure 6a). CC-292 (2) also had weak activity in this assay (IC50 = 392 nM). GS-4059 (3c) showed moderate potency (IC50 = 103 nM), but the non-covalent analog 3b was very weak (IC50 = 3860 nM). ACP196 (4) was much weaker than ibrutinib in this whole blood assay (IC50 = 198 nM). The large differences in potency between 1a and 1b and between 3c and 3b may mimic the impact that the C481S mutation has on the potency of covalent inhibitors in patients. The non-covalent inhibitors showed varying potencies in this whole blood assay, with IC50 values ranging from 13.5–221 nM. Compounds 8 and 9 showed modest activity (IC50 = 41 and 43 nM, respectively), while 5 and 7 were more potent (IC50 = 19 and 13.5 nM, respectively), and 6 was comparable in potency to ACP-196 (IC50 = 221 nM). While an in depth analysis of all of the molecular properties (e.g., solubility, permeability, protein binding, polar surface area, etc.) that may affect cellular activity and influence the biochemical-to-cellular potency shift is beyond the scope of the present work, the high potency of 7 is comparable to that of ibrutinib (1a), and is consistent with the activity that may be needed for a molecule to advance to human studies.

Non-covalent inhibitors block Btk autophosphorylation in C481S-transfected cells. In order to test whether the effects of C481S mutation on inhibitors we observed at the enzymatic level translated to a cellular context, we transfected human HEK293T cells

ACS Paragon Plus Environment

14

Page 15 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

with plasmids expressing full-length constructs of WT or C481S Btk and then monitored the effect of inhibitors (1 µM) on Btk Tyr223 autophosphorylation. In the absence of plasmid expressing exogenous Btk, endogenous Btk total protein and phospho-Btk are undetectable. Transfection of WT Btk or C481S mutant expressing plasmid resulted in detectable signals for both total and phospho-Btk, and the pBtk/Btk ratio in the absence of inhibition was defined as 100% activity (0% inhibition). Ibrutinib (1a) effectively inhibited autophosphorylation of WT Btk, but was ineffective at inhibiting autophosphorylation in cells transfected with the C481S mutant (Figure 6b). The loss of efficacy of 1a in C481S expressing cells was consistent with its loss in potency in the C481S enzyme assay relative to WT. The non-covalent analog 1b was ineffective at reducing Btk phosphorylation in both WT and C481S mutant expressing cell lines. The cellular results align with the biochemical inhibition data for these two inhibitors with Btk and the C481S mutant: 1b is equipotent against WT and C481S, and its potency in enzyme and activity in cell assays is about the same as the that of 1a with the C481S enzyme and in the C481S-expressing cells. Inhibitor 2 only showed 66% inhibition of WT phosphorylation, consistent with its moderate potency in the enzyme assay. Like 1a, 2 failed to inhibit Btk phosphorylation in cells expressing C481S. In contrast, the noncovalent inhibitors blocked Btk phosphorylation with equal potency in WT and C481S expressing cells. While compounds 8 and 9 modestly inhibited both WT and C481S, 5 and 7 were much more potent against both WT and C481S (Figure 6b). Thus, the covalent inhibitors were unable to inhibit phosphorylation of the C481S mutant in transfected cells, whereas the non-covalent inhibitors showed equivalent activity in WT and C481S mutant Btk transfected cells.

CONCLUSIONS Acquired resistance to ibrutinib is a growing concern to patients and physicians (17,19,20,24,25,47). While researchers have proposed salvage approaches such as using alternate kinase inhibitors to counter the problem of acquired resistance (17,47), the stark reality is that outcomes remain poor and survival duration is extremely short for patients experiencing relapse (24–26). Thus, more selective inhibitors and molecules that will retain efficacy against mutated Btk forms are needed. The non-covalent Btk inhibitors 5–

ACS Paragon Plus Environment

15

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 37

9 offer several advantages over existing covalent inhibitors. Most importantly, these molecules retain potency in biochemical and cellular assays against the C481S mutant, the most prevalent Btk mutation detected clinically. The non-covalent inhibitors are more selective for Btk than most of the covalent inhibitors we profiled. The co-crystal structure of Btk with the exquisitely selective inhibitor 9 provides insight into the orthogonal binding mode of these molecules relative to ibrutinib, and clarifies how this difference contributes to the selectivity of the non-covalent molecules. Further, non-covalent panBtk inhibitors like GNE-431 (6) also retain high potency against other clinical Btk mutants, C481R and T474I, as well as against the T474M mutant. Finally, as the noncovalent inhibitors can also potently block B-cell activation in whole blood, pan-Btk inhibitors such as these may be suitable for development as a first-line therapy and/or as a rescue therapy for patients who develop ibrutinib resistance as a result of Btk mutations.

METHODS Methods and procedures are provided in the Supporting Information.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: Methods, Figures S1, S2 and S3, and Table S1. Accession Code The crystallographic coordinates for Btk bound to 9 are deposited in the Protein Data Bank (PDB) under accession ID code 5KUP. AUTHOR INFORMATION Corresponding Author [email protected] Notes Authors were employees of Genentech, a Member of the Roche Group, at the time of this work.

ACS Paragon Plus Environment

16

Page 17 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

ACKNOWLEDGMENTS We thank W. Liang for compiling the synthetic methods of the inhibitor molecules used in this work. The U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357, supported use of the Advanced Photon Source. The Michigan Economic Development Corporation and the Michigan Technology Tri-Corridor (Grant 085P1000817) supported use of the LS-CAT Sector 21. REFERENCES 1. Cheng, S., Ma, J., Guo, A., Lu, P., Leonard, J. P., Coleman, M., Liu, M., Buggy, J. J., Furman, R. R., and Wang Y. L. (2014) Btk inhibition targets in vivo CLL proliferation through its effects on B-cell receptor signaling activity, Leukemia 28, 649–657. 2. Di Paolo, J. A., Huang, T., Balazs, M., Barbosa, J., Barck, K. H., Bravo, B. J., Carano, R. A. D., Darrow, J., Davies, D. R., DeForge, L. E., Diehl, L., Ferrando, R., Gallion, S. L., Giannetti, A. M., Gribling, P., Hurez, V., Hymowitz, S. G., Jones, R., Kropf, J. E., Lee, W. P., Maciejewski, P. M., Mitchell, S. A., Rong, H., Staker, B. L., Whitney, J. A., Yeh, S., Young, W. B., Yu, C., Zhang, J., Reif, K., and Currie, K. S. (2011) Specific Btk inhibition suppresses B cell- and myeloid cell-mediated arthritis, Nat. Chem. Biol. 7, 41– 50. 3. Pan, Z., Scheerens, H., Li, S.-J., Schultz, B. E., Sprengeler, P. A., Burrill, L. C., Mendonca, R. V., Sweeney, M. D., Scott, K. C. K., Grothaus, P. G., Jeffery, D. A., Spoerke, J. M., Honigberg, L. A., Young, P. R., Dalrymple, S. A., and Palmer, J. T. (2007) Discovery of Selective Irreversible Inhibitors for Bruton's Tyrosine Kinase, ChemMedChem 2, 58–61. 4. Honigberg, L. A., Smith, A. M., Sirisawad, M., Verner, E., Loury, D., Chang, B., Li, S., Pan, Z., Thamm, D. H., Miller, R. A., and Buggy, J. J. (2010) The Bruton tyrosine kinase inhibitor PCI-32765 blocks B-cell activation and is efficacious in models of autoimmune disease and B-cell malignancy, Proc. Natl. Acad. Sci. U. S. A. 107, 13075– 13080. 5. Advani, R. H., Buggy, J. J., Sharman, J. P., Smith, S. M., Boyd, T. E., Grant, B., Kolibaba, K. S., Furman, R. R., Rodriguez, S., Chang, B. Y., Sukbuntherng, J., Izumi, R., Hamdy, A., Hedrick, E., and Fowler, N. H. (2012) Bruton Tyrosine Kinase Inhibitor

ACS Paragon Plus Environment

17

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 37

Ibrutinib (PCI-32765) Has Significant Activity in Patients With Relapsed/Refractory BCell Malignancies, J. Clin. Oncol. 31, 88–94. 6. Byrd, J. C., Furman, R. R., Coutre, S. E., Flinn, I. W., Burger, J. A., Blum, K. A., Grant, B., Sharman, J. P., Coleman, M., Wierda, W. G., Jones, J. A., Zhao, W., Heerema, N. A., Johnson, A. J., Sukbuntherng, J., Chang, B. Y., Clow, F., Hedrick, E., Buggy, J. J., James, D. F., and O'Brien, S. (2013) Targeting Btk with Ibrutinib in Relapsed Chronic Lymphocytic Leukemia, N. Engl. J. Med. 369, 32–42. 7. Wang, M. L., Rule, S., Martin, P., Goy, A., Auer, R., Kahl, B. S., Jurczak, W., Advani, R. H., Romaguera, J. E., Williams, M. E., Barrientos, J. C., Chmielowska, E., Radford, J., Stilgenbauer, S., Dreyling, M., Jedzejczak, W. W., Johnson, P., Spurgeon, S. E., Li, L., Zhang, L., Newberry, K., Ou, Z., Cheng, N., Fang, B., McGreivy, J., Clow, F., Buggy, J. J., Chang, B. Y., Beaupre, D. M., Kunkel, L. A., Blum, K. A. (2013) Targeting Btk with Ibrutinib in Relapsed or Refractory Mantle-Cell Lymphoma, N. Engl. J. Med. 369, 507– 516. 8. Treon, S. P., Tripsas, C. K., Meid, K., Warren, D., Varma, G., Green, R., Argyropolous, K. V., Yang, G., Cao, Y., Xu, L., Patterson, C. J., Rodig, S., Zehnder, J. L., Aster, J. C., Harris, N. L., Kanan, S., Ghobrial, I., Castillo, J. J., Laubach, J. P., Hunter, Z. R., Salman, Z., Li, J., Cheng, M., Clow, F., Graef, T., Palomba, M. L., and Advani, R. H. (2015) Ibrutinib in Previously Treated Waldenström's Macroglobulinemia, N. Engl. J. Med. 372, 1430–1440. 9. Evans, E. K., Tester, R., Aslanian, S., Karp, R., Sheets, M., Labenski, M. T., Witowski, S. R., Lounsbury, H., Chaturvedi, P., Mazdiyasni, H., Zhu, Z., Nacht, M., Freed, M. I., Petter, R. C., Dubrovsky, A., Singh, J., and Westlin, W. F. (2013) Inhibition of Btk with CC-292 Provides Early Pharmacodynamic Assessment of Activity in Mice and Humans, J. Pharmacol. Exp. Ther. 346, 219–228. 10. Xu, D., Kim, Y., Postelnek, J., Vu, M. D., Hu, D.-Q., Liao, C., Bradshaw, M., Hsu, J., Zhang, J., Pashine, A., Srinivasan, D., Woods, J., Levin, A., O'Mahony, A., Owens, T. D., Lou, Y., Hill, R. J., Narula, S., DeMartino, J., and Fine, J. S. (2012) RN486, a Selective Bruton's Tyrosine Kinase Inhibitor, Abrogates Immune Hypersensitivity Responses and Arthritis in Rodents, J. Pharm. Exp. Ther. 34, 90–103.

ACS Paragon Plus Environment

18

Page 19 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

11. Rankin, A. L., Seth, N., Keegan, S., Andreyeva, T., Cook, T. A., Edmonds, J., Mathialagan, N., Benson, M. J., Syed, J., Zhan, Y., Benoit, S. E., Miyashiro, J. S., Wood, N., Mohan, S., Peeva, E., Ramaiah, S. K., Messing, D., Homer, B. L., DunissiJoannopoulos, K., Nickerson-Nutter, C. L., Schnute, M. E., and Douhan III, J. (2013) Selective Inhibition of Btk Prevents Murine Lupus and Antibody-Mediated Glomerulonephritis, J. Immunol. 191, 4540–4550. 12. Byrd, J. C., Harrington, B., O'Brien, S., Jones, J. A., Schuh, A., Devereux, S., Chaves, J., Wierda, W. G., Awan, F. T., Brown, J. R., Hillmen, P., Stephens, D. M., Ghia, P., Barrientos, J. C., Pagel, J. M., Woyach, J., Johnson, D., Huang, J., Wang, X., Kaptein, A., Lannutti, B. J., Covey, T., Fardis, M., McGreivy, J., Hamdy, A., Rothbaum, W., Izumi, R., Diacovo, T. G., Johnson, A. J., and Furman, R. R. (2016) Acalabrutinib (ACP-196) in Relapsed Chronic Lymphocytic Leukemia, N. Engl. J. Med. 374, 323–332. 13. Walter, H. S., Rule, S. A., Dyer, M. J. S., Karlin, L., Jones, C., Cazin, B., Quittet, P., Shah, N., Hutchinson, C. V., Honda, H., Duffy, K., Birkett, J., Jamieson, V., CourtenayLuck, N., Yoshizawa, T., Sharpe, J., Ohno, T., Abe, S., Nishimura, A., Cartron, G., Morschhauser, F., Fegan, C., and Salles, G. (2016) A Phase I clinical trial of the selective Btk inhibitor ONO/GS-4059 in relapsed and refractory mature B-cell malignancies, Blood 127, 411–419. 14. Singh, J., Petter, R. C., Baillie, T. A., and Whitty, A. (2011) The resurgence of covalent drugs, Nat. Rev. Drug Disc. 10, 307–317. 15. Woyach, J. A., Furman, R. R., Liu, T.-M., Ozer, H. G., Zapatka, M., Ruppert, A. S., Xue, L., Li, D. H.-H., Steggerda, S. M., Versele, M., Dave, S. S., Zhang, J., Yilmaz, A. S., Jaglowski, S. M., Blum, K. A., Lozanski, A., Lozanski, G., James, D. F., Barrientos, J. C., Lichter, P., Stilgenbauer, S., Buggy, J. J., Chang, B. Y., Johnson, A. J., and Byrd, J. C. (2014) Resistance Mechanisms for the Bruton's Tyrosine Kinase Inhibitor Ibrutinib, N. Engl. J. Med. 370, 2286–2294. 16. Chang, B. Y., Furman, R. R., Zapatka, M., Barrientos, J. C., Li, D., Steggerda, S., Eckert, K., Francesco, M., Woyach, J. A., Johnson, A. J., James, D. F., Versele, M., Byrd, J. C., Stilgenbauer, S., and Buggy, J. J. (2013) Use of tumor genomic profiling to reveal mechanisms of resistance to the Btk inhibitor ibrutinib in chronic lymphocytic leukemia (CLL), J. Clin. Oncol. 31, suppl; abstr 7014.

ACS Paragon Plus Environment

19

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 37

17. Cheng, S., Guo, A., Lu, P., Ma, J., Coleman, M., and Lynn Wang, Y. (2015) Functional characterization of BtkC481S mutation that confers ibrutinib resistance: exploration of alternative kinase inhibitors, Leukemia 29, 895–900. 18. Furman, R. R., Chang, S., Lu, P., Setty, M., Perez, A. R., Guo, A., Racchumi, J., Xu, G., Wu, H., Ma, J., Steggerda, S. M., Coleman, M., Leslie, C., and Lynn Wang, Y. (2014) Ibrutinib Resistance in Chronic Lymphocytic Leukemia, N. Engl. J. Med. 370, 2352– 2354. 19. Maddocks, K. J., Ruppert, A. S., Lozanski, G., Heerema, N. A., Zhao, W., Abruzzo, L., Lozanski, A., Davis, M., Gordon, A., Smith, L. L., Mantel, R., Jones, J. A., Flynn, J. M., Jaglowski, S. M., Andritsos, L. A., Awan, F., Blum, K. A., Grever, M. R., Johnson, A. J., Byrd, J. C., and Woyach, J. A. (2015) Etiology of Ibrutinib Therapy Discontinuation and Outcomes in Patients With Chronic Lymphocytic Leukemia, JAMA Oncol. 1, 80–87. 20. Liu, T.-M., Woyach, J. A., Zhong, Y., Lozanski, A., Lozanski, G., Dong, S., Strattan, E., Lehman, A., Zhang, X., Jones, J. A., Flynn, J., Andritsos, L. A., Maddocks, K., Jaglowski, S. M., Blum, K. A., Byrd, J. C., Dubovsky, J. A., and Johnson, A. J. (2015) Hypermorphic mutation of phospholipase C, γ2 acquired in ibrutinib-resistant CLL confers Btk independency upon B-cell receptor activation, Blood 126, 61–68. 21. McMullen, J. R., Boey, E., J. H., Ooi, J. Y. Y., Seymour, J. F., Keating, M. J., and Tam, C. S. (2014) Ibrutinib increases the risk of atrial fibrillation, potentially through inhibition of cardiac PI3K-Akt signaling, Blood 124, 3829–3830. 22. Kuhn, E. L., Levêque, D., Lioure, B., Gourieux, B., and Bilbault, P. (2016) Adverse event potentially due to an interaction between ibrutinib and verapimil: a case report, J. Clin. Pharm. Ther. 41, 104–105. 23. Bitar, C., Farooqui, M. Z., Valdez, J., Saba, N. S., Soto, S., Bray, A., Marti, G., Wiestner, A., Cowen, E. W. (2016) Hair and Nail Changes During Long-term Therapy With Ibrutinib for Chronic Lymphocytic Leukemia, JAMA Dermatol. Epub Mar 16, 2016. DOI: 10.1001/jamadermatol.2016.0225. 24. Cheah, C. Y., Chihara, D., Romaguera, J. E., Fowler, N. H., Seymour, J. F., Hagemeister, F. B., Champlin, R. E., and Wang, M. L. (2015) Patients with mantle cell

ACS Paragon Plus Environment

20

Page 21 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

lymphoma failing ibrutinib are unlikely to respond to salvage chemotherapy and have poor outcomes, Ann. Oncol. 26, 1175–1179. 25. Martin, P., Maddocks, K., Leonard, J. P., Ruan, J., Goy, A., Wagner-Johnston, N., Rule, S., Advani, R., Iberri, D., Phillips, T., Spurgeon, S., Kozin, E., Noto, K., Chen, Z., Jurczak, W., Auer, R., Chmielowska, E., Stilgenbauer, S., Bloehdorn, J., Portell, C., Williams, M. E., Dreyling, M., Barr, P. M., Chen-Kiang, S., DiLiberto, M., Furman, R. R., and Blum, K. A. (2016) Post-ibrutinib outcomes in patients with mantle cell lymphoma, Blood 127, 1559–1563. 26. Molica, S. (2015) The clinical safety of ibrutinib in chronic lymphocytic leukemia, Expert Opin. Drug Saf. 14, 1621–1629. 27. Liu, L., Di Paolo, J., Barbosa, J., Rong, H., Reif, K., and Wong, H. (2011) Antiarthritis Effect of a Novel Bruton's Tyrosine Kinase (BTK) Inhibitor in Rat Collagen-Induced Arthritis and Mechanism-Based Pharmacokinetic/Pharmacodynamic Modeling: Relationships between Inhibition of BTK Phosphorylation and Efficacy, J. Pharm. Exp. Ther. 338, 154-163. 28. Young, W. B., Barbosa, J., Blomgren, P., Bremer, M. C., Crawford, J. J., Dambach, D., Gallion, S., Hymowitz, S. G., Kropf, J. E., Lee, S. H., Liu, L., Lubach, J. W., Macaluso, J., Maciejewski, P., Maurer, B., Mitchell, S. A., Ortwine, D. F., Di Paolo, J., Reif, K., Scheerens, H., Schmitt, A., Sowell, C. G., Wang, X., Wong, H., Xiong, J.-M., Xu, J., Zhao, Z., and Currie, K. S. (2015) Potent and selective Bruton's tyrosine kinase inhibitors: Discovery of GDC-0834, Bioorg. Med. Chem. Lett. 25, 1333-1337. 29. Young, W. B., Barbosa, J., Blomgren, P., Bremer, M. C., Crawford, J. J., Dambach, D., Eigenbrot, C., Gallion, S., Johnson, A. R., Kropf, J. E., Lee, S. H., Liu, L., Lubach, J. W., Macaluso, J., Maciejewski, P., Mitchell, S. A., Ortwine, D. F., Di Paolo, J., Reif, K., Scheerens, H., Schmitt, A., Wang, X., Wong, H., Xiong, J.-M., Xu, J., Yu, C., Zhao, Z., and Currie, K. S. (2016) Discovery of highly potent and selective Bruton's tyrosine kinase inhibitors: Pyridazinone analogs with improved metabolic stability, Bioorg. Med. Chem. Lett. 26, 575-579. 30. Crawford, J. J., Ortwine, D. F., Wei, B., and Young, W. B. (2013) Heteroaryl pyridone and aza-pyridone compounds as inhibitors of Btk activity. PCT Int. Appl. WO 2013/067274.

ACS Paragon Plus Environment

21

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 37

31. Crawford, J. J., Ortwine, D. F., Wei, B., and Young, W. B. (2013) 8Fluorophthalazin-1(2H)-one compounds as inhibitors of Btk activity. PCT Int. Appl. WO 2013/067264. 32. Shah, N. P., Nicoll, J. M., Nagar, B., Gorre, M. E., Paquette, R. L., Kuriyan, J., and Sawyers, C. L. (2002) Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia, Cancer Cell 2, 117–125. 33. Shah, N. P., Tran, C., Lee, F. Y., Chen, P., Norris, D., and Sawyers, C. L. (2004) Overriding Imatinib Resistance with a Novel ABL Kinase Inhibitor, Science 305, 399– 401. 34. Kobayashi, S., Boggon, T. J., Dayaram, T., Jänne, P. A., Kocher, O., Meyerson, M., Johnson, B. E., Eck, M. J., Tenen, D. G., and Halmos, B. (2005) EGFR Mutation and Resistance of Non-Small Cell Lung Cancer to Gefitinib, New Engl. J. Med. 352, 786– 792. 35. Segel, I. H. (1975) Enzyme Kinetics, Wiley, New York. 36. Bradshaw, J. M., McFarland, J. M., Paavilainen, V. O., Bisconte, A., Tam, D., Phan, V. T., Romanov, S., Finkle, D., Shu, J., Patel, V., Ton, T., Li, X., Loughhead, D. G., Nunn, P. A., Karr, D. E., Gerritsen, M. E., Funk, J. O., Owens, T. D., Vernier, E., Brameld, K. A., Hill, R. J., Goldstein, D. M., and Taunton, J. (2015) Prolonged and tunable residence time using reversible covalent kinase inhibitors, Nat. Chem. Biol. 11, 525–531. 37. Marcotte, D. J., Liu, Y.-T., Arduini, R. M., Hession, C. A., Miatkowski, K., Wildes, C. P., Cullen, P. F., Hong, V., Hopkins, B. T., Mertsching, E., Jenkins, T. J., Romanowski, M. J., Baker, D. P., and Silvian, L. F. (2010) Structures of human Bruton's tyrosine kinase in active and inactive conformations suggest a mechanism of activation for TEC family kinases, Prot. Sci. 19, 429–439. 38. Xing, L. and Huang, A. (2014) Bruton's TK inhibitors: structural insights and evolution of clinical candidates, Future Med. Chem. 6, 675–695. 39. Carmi, C., Mor, M., Petronini, P. G., and Alfieri, R. R. (2012) Clinical perspectives for irreversible tyrosine kinase inhibitors in cancer, Biochem. Pharmacol. 84, 1388–1399.

ACS Paragon Plus Environment

22

Page 23 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

40. Liu, F., Zhang, X., Weisberg, E., Chen, S., Hur, W., Wu, H., Zhao, Z., Wang, W., Mao, M., Cai, C., Simon, N. I., Sanda, T., Wang, J., Look, A. T., Griffin, J. D., Balk, S. B., Liu, Q., and Gray, N. S. (2013) Discovery of a Selective Irreversible BMX Inhibitor for Prostate Cancer, ACS Chem. Biol. 8, 1423–1428. 41. Schwartz, P. A., Kuzmic, P., Solowej, J., Bergqvist, S., Bolanos, B., Almaden, C., Nagata, A., Ryan, K., Feng, J., Dalvie, D., Kath, J. C., Xu, M., Wani, R., and Murray, B. W. (2014) Covalent EGFR inhibitor analysis reveals importance of reversible interactions to potency and mechanisms of drug resistance, Proc. Natl. Acad. Sci. U. S. A. 111, 173– 178. 42. Goedken, E. R., Argiriadi, M. A., Banach, D. L., Flamengo, B. A., Foley, S. E., Frank, K. E., George, J. S., Harris, C. M., Hobson, A. D., Ihle, D. C., Marcotte, D., Merta, P. J., Michalak, M. E., Murdock, S. E., Tomlinson, M. J., and Voss, J. W. (2015) Tricyclic Covalent Inhibitors Selectively Target Jak3 through an Active Site Thiol, J. Biol. Chem. 290, 4573–4589. 43. Lanning, B. R., Whitby, L. R., Dix, M. M., Douhan, J., Gilbert, A. M., Hett, E. C., Johnson, T. O., Joslyn, C., Kath, J. C., Niessen, S., Roberts, L. R., Schnute, M. E., Wang, C., Hulce, J. J., Wei, B., Whiteley, L. O., Hayward, M. M., and Cravatt, B. F. (2014) A road map to evaluate the proteome-wide selectivity of covalent kinase inhibitors, Nat. Chem. Biol. 10, 760–767. 44. Lynch, Jr., T. J., Kim, E. S., Eaby, B., Garey, J., West, D. P., and Lacouture, M. E. (2007) Epidermal Growth Factor Receptor Inhibitor-Associated Cutaneous Toxicities: An Evolving Paradigm in Clinical Management, The Oncologist 12, 610–621. 45. Harandi, A., Zaidi, A. S., Stocker, A. M., and Laber, D. A. (2009) Clinical Efficacy and Toxicity of Anti-EGFR Therapy in Common Cancers, J. Oncol., http://dx.doi.org/10.1155/2009/567486. 46. Lou, Y., Owens, T. D., Kuglstatter, A., Kondru, R. K., and Goldstein, D. M. (2012) Bruton's Tyrosine Kinase Inhibitors: Approaches to Potent and Selective Inhibition, Preclinical and Clinical Evaluation for Inflammatory Diseases and B Cell Malignancies, J. Med. Chem. 55, 4539-4550.

ACS Paragon Plus Environment

23

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 37

47. Zhang, S. Q., Smith, S. M., Zhang, S. Y., and Lynn Wang, Y. (2015) Mechanisms of ibrutinib resistance in chronic lymphocytic leukaemia and non-Hodgkin lymphoma, Br. J. Haematol. 170, 445–456.

ACS Paragon Plus Environment

24

Page 25 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

FIGURE LEGENDS Figure 1 Chemical structures of the Btk inhibitors used in this work.

Figure 2 Structural model of ibrutinib (1a) (orange) covalently bound to Cys481 in the Btk active site. Residues that have been found mutated in ibrutinib-resistant patients, Cys481 and Thr474, are labeled as are two hinge residues (Glu475 and Met477) that form hydrogen bonds (dashed lines) to the pyrazolopyrimidine core of 1a. Relevant protein residues are depicted in green. The α-C-helix at the back of the image is labeled, while the backbone trace for the front portion of the G-loop has been removed for clarity.

Figure 3 Covalent Btk inhibitors lose potency against C481S, whereas non-covalent inhibitors retain potency with C481S. a) Ibrutinib (1a) loses potency against the C481S mutant. b) CC-292 (2) also loses potency against C481S. c) The non-covalent analog of ibrutinib, 1b, inhibits WT with the same potency as it does C481S. d) Non-covalent inhibitor 6 is a potent inhibitor of WT Btk, and it retains equivalent potency against C481S. The % of Control activity data plotted are the means ± SD from four titrations. e) Kinetic inhibition mechanisms for non-covalent and covalent inhibitors against WT and C481S illustrate how the loss of covalent reaction with C481S diminishes the observed potency for covalent inhibitors, but does not affect that for non-covalent inhibitors.

Figure 4 Kinase selectivity of the Btk inhibitors. a) Each inhibitor was tested at 1 µM concentration against a panel of 221 kinase activity and binding assays (SelectScreen Kinase Profiling Services, Thermo Fisher Scientific). Each line denotes one kinase and the color represents the % Inhibition: black, ≥ 80%; gray, 50–80%; white, < 50%. b) Total number of off target (i.e., non-Btk) kinases inhibited by the molecules in Figure 4a.

Figure 5 A Btk/9 X-ray co-crystal structure and a docking model with 6 can be used to rationalize the Btk potency, selectivity, and activity against mutant forms of Btk for inhibitors 5–9. a) The orientation of 9 (magenta) in the Btk kinase domain structure (gray) illustrates the approximately orthogonal binding mode of this molecule in contrast to that of 1a (see Figure 2). The G-loop packs tightly over 9. Tyr551 is located in the

ACS Paragon Plus Environment

25

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 37

middle of the activation loop and forms a close interaction with the t-butyl group of 9 at the base of the H3 selectivity pocket (green surface). b) A close-up view of the different binding orientations of 1a (orange, modeled) and 9 (magenta) in the Btk active site and the locations of Cys481, Thr474, and Tyr551 (green). The H3 selectivity pocket is depicted as a green surface. c) Computational docking model of Btk with 6 illustrates its binding pose that enables it to inhibit WT Btk and all of the mutants tested in this work: C481S, C481R, T474I, and T474M. The mutant residues (magenta) do not overlap with 6 bound in the active site. The backbone trace for the front portion of the G-loop has been removed from b) and c) for clarity and the H3 selectivity pocket volume is represented by a green surface.

Figure 6 Whole blood potency and inhibition of cellular Btk autophosphorylation by covalent and non-covalent inhibitors. a) Activity of covalent Btk inhibitors as well as non-covalent molecules in a whole blood assay of BCR-dependent cell activation and surface expression of the CD69 marker of B-cell activation. IC50 values shown are the means ± SD of at least three determinations using blood from different donors. b) Covalent Btk inhibitors 1a and 2 fail to inhibit C481S autophosphorylation in transfected cells, whereas the non-covalent inhibitors 5 and 7 potently block cellular autophosphorylation of both WT Btk and C481S. The % Inhibition values are the means ± SD from two to five measurements. Statistical significance in the % inhibition between WT and C481S was assessed by unpaired t-tests using Prism 5.0 software (****, P0.05).

ACS Paragon Plus Environment

26

Page 27 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Table 1. Steady state kinetics constants for WT Btk and the mutants studied in this work. Values listed are apparent kinetic constants and are the means ± SD of four or more titrations, except C481S and C481R, which were tested in two titrations.

kcat (s-1)

Km (µM) for ATP

kcat/Km (M-1s-1)

0.25 ± 0.1

49 ± 13

5,100

C481S

0.25 ± 0.007

47 ± 0.3

5,300

C481R

0.017 ± 0.001

48 ± 0.4a

350

T474I

0.18 ± 0.02

63 ± 16

2,900

T474M

0.15 ± 0.006

12 ± 0.2

12,500

Btk Form WT

a

C481R displayed sigmoidal kinetics. The substrate concentration at which its rate is half

of the maximal velocity is formally termed the [S]0.5 valueb, rather than Km. For the purposes of this work, however, such a distinction is not germane. b

35.

ACS Paragon Plus Environment

27

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 37

Table 2. Biochemical potency of inhibitors against WT Btk and the C481S, C481R, T474I, and T474M mutants. Molecules that act as covalent inhibitors of WT are denoted with an asterisk (*). Each IC50 is the mean ± SD of three or more experiments, each performed with duplicate titrations. Molecules that inhibit 1000 nM.

IC50 (nM) Inhibitor

WT

C481S

C481R

T474I

T474M

0.72 ± 0.06

4.6 ± 1.0

23 ± 0.2

5.6 ± 1.8

66 ± 16

4.9 ± 2.0

4.7 ± 0.5

34 ± 2.8

468 ± 138

>1000

22 ± 17

908 ± 87

854 ± 108

48 ± 17

75 ± 21

3a *

5.7 ± 1.3

174 ± 46

703 ± 31

345 ± 75

>1000

3b

219 ± 63

191 ± 65

>1000

>1000

>1000

3c GS-4059 *

52 ± 15

268 ± 61

>1000

>1000

>1000

ACP-196 *

18 ± 1.0

247 ± 29

>1000

207 ± 7.8

>1000

2.6 ± 1.5

3.0 ± 0.5

10 ± 1.4

38 ± 12

9.9 ± 1.2

3.2 ± 2.0

2.5 ± 0.3

8.4 ± 0.4

11 ± 3

8.8 ± 0.7

7

5.9 ± 4.3

6.3 ± 1.5

7.8 ± 0.4

37 ± 9

21 ± 0.8

8

6.0 ± 4.3

5.3 ± 0.9

8.0 ± 0.9

36 ± 14

20 ± 1.4

9

7.2 ± 4.6

7.4 ± 0.5

7.5 ± 0.9

439 ± 201

42 ± 4

1a Ibrutinib * 1b 2

4

CC-292 *

5 6

GNE-431

ACS Paragon Plus Environment

28

Page 29 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Figure 1.

ACS Paragon Plus Environment

29

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 37

Figure 2.

ACS Paragon Plus Environment

30

Page 31 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Figure 3.

e

WT BTK Ki

E+I

C481S BTK Ki

kinact

E I

Non-covalent (Potency = Ki) Covalent (Potency 50%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 37

35 30 25 20 15 10 5 0 1a 1b 2 3a 3c 4

5

6

7

8

9

Inhibitor

ACS Paragon Plus Environment

32

Page 33 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Figure 5a.

α-C-helix

G-loop

Tyr551

H3 pocket

ACS Paragon Plus Environment

33

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 37

Figure 5b.

Thr474

Tyr551

Cys481

ACS Paragon Plus Environment

34

Page 35 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Figure 5c.

Thr474 Ile474 Met474

Tyr551

Cys481 Ser481 Arg481

ACS Paragon Plus Environment

35

ACS Chemical Biology

Figure 6.

a

IC50, nM

10000

1000

100

10

1a 1b 2 3b 3c 4

5

6

7

8

9

Inhibitor

b %Inhibition of pBTK

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 37

120

WT C481S ns

****

ns

100

*

80 60

ns

ns

ns

40 20 0

1a 1b

2

5

7

8

9

Inhibitor

ACS Paragon Plus Environment

36

Page 37 of 37

ACS Chemical Biology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



ACS Paragon Plus Environment