pyrrolizines as Novel Bruton's Tyrosine Kinase (BTK)

Subscriber access provided by Kaohsiung Medical University

Article

Discovery of 4,7-Diamino-5-(4-phenoxyphenyl)-6-methylene-pyrimido[5,4b]pyrrolizines as Novel Bruton’s Tyrosine Kinase (BTK) Inhibitors Yu Xue, Peiran Song, Zilan Song, Aoli Wang, Linjiang Tong, MeiYu Geng, Jian Ding, Qingsong Liu, Liping Sun, Hua Xie, and Ao Zhang J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00441 • Publication Date (Web): 01 May 2018 Downloaded from http://pubs.acs.org on May 1, 2018

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 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 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.

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 61 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

Journal of Medicinal Chemistry

Discovery of 4,7-Diamino-5-(4-phenoxyphenyl)-6-methylene-pyrimido[5,4-b]pyrro lizines as Novel Bruton’s Tyrosine Kinase (BTK) Inhibitors Yu Xue,†,‡,ǁ Peiran Song,ǂ,∆,ǁ Zilan Song,‡,ǁ Aoli Wang,∫ Linjiang Tong,ǂ Meiyu Geng,ǂ,§,∆ Jian Ding,ǂ,§,∆ Qingsong Liu,∫ Liping Sun,†,* Hua Xie,ǂ,§,* Ao Zhang‡,§,∆,* †

Department of Medicinal Chemistry, China Pharmaceutical University, Nanjing 210009, China

‡

CAS Key Laboratory of Receptor Research, and the State Key Laboratory of Drug Research,

Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, China ǂ

Division of Anti-tumor Pharmacology, the State Key Laboratory of Drug Research, Shanghai Institute

of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, China ∫

High Magnetic Field Laboratory, Chinese Academy of Sciences, Hefei 230031, China

§

College of Pharmacy, University of Chinese Academy of Sciences, China

∆

School of Life Science and Technology, ShanghaiTech University, Shanghai 201210, China

ǁ

All these authors contributed equally to this work.

∗To whom correspondence should be addressed. For A.Z.: phone: +86-21-50806035; fax:

86-21-50806035;

E-mail:

[email protected];

For

H.X.:

phone:

+86-21-50805897; fax: 86-21-50805897; E-mail: [email protected]; For L.S.: phone: +86-25-83271414; fax: 86-25-83271414; Email: [email protected]

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 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

Abstract: An alternative medicinal chemistry approach was conducted on BTK inhibitor 1 (ibrutinib) by merging the pyrazolo[3,4-d]pyrimidine component into a tricyclic skeleton. Two types of compounds were prepared, and their biochemical activities on BTK as well as stereochemistry effects were determined. Structural optimization focusing on the reactive binding group to BTK Cys481 and on the metabolic site guided by metabolic study were conducted. 7S was identified as the most potent showing an IC50 value of 0.4 nM against BTK and 16 nM against BTK-dependent TMD8 cells. Compared to 1, 7S was slightly more selective with strong inhibition on B-cell receptor signaling pathway. In TMD8 cell-derived animal xenograft model, 7S showed Relative Tumor Volume of 5.3 at 15 mg/kg QD dosage that was more efficacious than 1 (RTV 6.6) at higher dose of 25 mg/kg QD. All these results suggest 7S as a new BTK inhibitor worthy of further profiling.


Page 2 of 61

Page 3 of 61 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


INTRODUCTION The Bruton’s tyrosine kinase (BTK) protein is a nonreceptor tyrosine kinase belonging to the Tec kinase family (the other members are BMX, ITK, TEC and TXK).1 It plays a critical role in B-cell receptor (BCR) signaling by mediating B cell development and functioning. Therefore, dysregulation of BTK generally causes severe leukemias and B-cell related lymphomas.2 BTK is the upstream activator of multiple antiapoptotic signaling molecules and networks, and is primarily expressed in hematopoietic cells, particularly in B cells, but not in T cells or normal plasma cells.3 The essential role of BTK in BCR signaling pathway and its restricted expression pattern warrant it as a viable and attractive therapeutic target for the treatment of B-cell malignancies.4 The first-generation multi-targeted BTK inhibitor ibrutinib5 (1, PCI-32765) (Figure 1A) has been approved by FDA in 2013 to treat mantle cell lymphoma (MCL) and chronic lymphocytic leukemia (CLL), and various clinical trials are ongoing for new indications.6,7 Compound 1 is an irreversible inhibitor by covalently binding to Cys481 of BTK with its acrylamido moiety. In addition, 1 also irreversibly binds to other kinases such as epidermal growth factor receptor (EGFR), ITK and TEC which might be attributed to the clinically observed adverse effects in ibrutinib-treated patients, including rash, diarrhea, arthralgias, atrial fibrillation, ecchymosis and major hemorrhage.7 Though it is not for sure the off-targets also contribute to the clinical antitumor efficacy of 1, the objective of the second-generation BTK inhibitors is to achieve better BTK on-target selectivity to minimize the side effects.8 As shown in Figure 1, several compounds such as acalabrutinib9 (2, ACP-196), tirabrutinib10 (3, ONO/GS-4059) and spebrutinib11 (4, CC-292) are being extensively studied in the clinic both for evaluation of their therapeutical efficacies and for their safety profiles. Very recently, compound 2



possessing higher selectivity and inhibitory effect against BTK received the FDA’s approval as a best-in-class treatment for CLL with a 100% response rate for patients positive for the 17p13.1 gene deletion - a subgroup of patients that typically results in a poor response to therapy and low expected clinical outcomes.12 Intriguingly, compound 4, another second-generation BTK inhibitor also irreversibly binding the Cys481 of BTK in high selectivity shows a clinical activity (in particular, durability of response) inferior to that of 1 or 2.7, 11 In addition to covalent inhibitors, several non-covalent BTK inhibitors have also been reported.13-15 These non-covalent inhibitors could not form irreversible binding to BTK Cys481 because of absence of the reactive binding group. However, they could target the binding pocket of inactive BTK conformations through conformational changes, which offers reversible interaction and binding to BTK.16 Although the precise reason is unclear for the different clinical outcomes of these compounds, more diverse inhibitors with both novel structures and selective on-target binding profile are needed.17


Page 4 of 61

Page 5 of 61 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


A. Clinical BTK inhibitors bearing an acrylamido or butynamido moiety to covalently interact with Cys481

O

NH

N

N N N

O

N

NH

N

F

O

O N

N

1 (ibrutinib, PCI-32765)

N

O N

N

N

N

N

NH2

N

O

HN NH2

NH2

N

O

O

N

O

N H

O 2 (acalabrutinib, ACP-196) 3 (triabrutinib, ONO/GS-4059) 4 (spebrutinib, CC-292)

B. Design of our new BTK inhibitors I and II O X

R X

R

NH2 ring-merging

N

N N

N

H2N

H2N

N N

1

O

N N

O N H

I 6,7,8,9-tetrahydropyrimido [5,4-b]indolizine

O

N

N N

N H

II 6-methylene-7,8-dihydro-6Hpyrimido[5,4-b]pyrrolizine

Figure 1. Clinical BTK inhibitors 1-4 (A) and our design of new BTK inhibitors I and II (B).

Rather than replacing the central hinge binder pyrazolo[3,4-d]pyrimidin-4-amine template of 1 with other bicyclic (e.g. imidazo[1,5-a]pyrazin-8-amine in 2 and 6-amino-7H-purin-8(9H)-one in 3) or monocyclic (e.g. 2,4-diaminopyrimidine in 4) bioisosteres, we recently conducted a different medicinal chemistry approach by merging the pyrazolo[3,4-d]pyrimidine component and the piperidine ring of 1 into a tricyclic skeleton I (Figure 1B). Interestingly, during the synthesis of I, we found that a change of the cyclization reaction condition led to a different cyclization product II. Both compound series I and II contain brand new core structures that are rarely seen in other categories of therapeutical drugs.18 Herein, we report the design, synthesis, and pharmacological evaluations of both series of compounds as novel BTK inhibitors.



Figure 2. Binding modes of compounds 1 (A), 5 (B) and 7 (C) with BTK (PDB id: 3GEN).

RESULTS AND DISCUSSION Structure-Based Drug Design. From the covalent docking model of 1 to BTK (PDB id: 3GEN),19 the acrylamido moiety covalently binds to the Cys481 in the active site of BTK to achieve a potent and irreversible inhibition (Figure 2A). In addition, the pyrazolo[3,4-d]pyrimidine backbone of 1 forms three critical hydrogen bonds with hinge residues Met477, Glu475 and Thr474, respectively, through an edge-to-face π-π interaction between the terminal phenyl group and Phe540 in the hydrophobic pocket. The pyrazolopyrimidine backbone of 1 proves to be effective in developing BTK inhibitors, and a dozen of mimetics (e.g. 2-4) of this bicyclic framework have been reported.20 A close examination of the binding mode of 1 with BTK reveals to us that there is a large space between the nearly orthogonal pyrazolyl and piperidyl groups, indicating that occupying this space with an appropriate ring might be well tolerated. Initially, we considered to directly cyclize the pyrazolyl and


Page 6 of 61

Page 7 of 61 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


piperidyl moieties of 1 to form a tetracyclic artwork. However, consideration of the potential ring tension and the synthetic difficulty of the proposed tetracyclic backbone, a cut-off of the piperidyl ring is necessary, thus leading to our first design of the tricyclic compound series I (Figure 1B). Preliminary Evaluation of Compounds 5-7. The tricyclic compound 5 was synthesized as a prototypic representative of compound series I. The tricycle pyrimido[5,4-b]pyrrolizine 7 (compound series II) was also obtained during optimization of the cyclization reaction condition. A docking of 5 to BTK was shown in Figure 2B, in which the critical irreversible binding, key hydrogen bonds and hydrophobic interaction as that showed in 1 were all retained. Interestingly, compound 7 bearing an exocyclic double bond also maintained the key interactions with BTK (Figure 2C) and the exocyclic double bond had no significant contacts in the catalytic domain. This analysis suggested that compounds 5 and 7 might be potent BTK inhibitors.



Figure 3. Characterizations of compounds 5-7 as potent BTK inhibitors. (A) Chemical structures of 5-7. (B) In vitro kinase inhibitory activities of compounds against wild-type BTK. (C) In vitro kinase inhibitory activities of compounds 7 and 1 against BTKC481S mutant.

BTK inhibitory activity of compounds 5 and 7 were evaluated by using


Page 8 of 61

Page 9 of 61 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


well-established ELISA assays. The saturated compound 6 was prepared as a comparison and compound 1 (ibrutinib) was included as a positive control (Figure 3A). As shown in Figure 3B, compounds 5 and 7 bearing an acrylamido moiety were highly potent BTK inhibitors with IC50 values of 5.3 nM and 2.8 nM, respectively, which were only 13-fold and 7-fold less potent than 1. Compound 6 bearing a nonreactive propylamido moiety was approximately 1040-fold less potent (IC50 of 417.4 nM) than 1. The significant discrepancy in biochemical activity between 5 and 6 illustrated the necessity of the acrylamido warhead for irreversible covalent binding with BTK Cys481. The effects of compound 7 against BTKC481S mutant were further examined (Figure 3C). Compared to the high potency against wild-type BTK, compound 7 exhibited 165-fold reduction of potency against BTKC481S (IC50 of 463.6 vs 2.8 nM). These results suggested that the new tricyclic compounds 5 and 7 were potent BTK inhibitors and might covalently bind to the Cys481 leading to irreversible inhibition on BTK kinase activity, a similar interaction profile to that of 1.5 Kinase Selectivity Study of Compounds 5 and 7. To determine possible off-targets, we further analyzed the kinase selectivity of the two new BTK inhibitors 5 and 7 by the KINOMEScan™ screening platform.21 Both compounds were tested against a panel of 468 kinases and mutants at 1 µM concentration and the results were showed in Figure 4. Compound 7 was found to possess a better selectivity (S-score (1) = 0.007), whereas the S(1) value of 5 was 0.027. Not surprisingly, the major off-targets of both compounds were those kinases bearing a cysteine residue in the ATP binding pocket similar to BTK, which could be covalently bound by the acrylamido warhead, including BLK, EGFR, EebB4, TEC and so on. More specifically, compound 7 had only three major off-targets (BLk, ErbB4, MEK) with a percent control number less than 1 at 1 µM concentration, a selectivity profile much



Page 10 of 61

better than that of compounds 5 and 1.9b

Compd.

5

7

Selectivity Score Type

Number of Hits

Number of Non-Mutant Kinases

Selectivity Score

S(35)

47

403

0.117

S(10)

30

403

0.074

S(1)

11

403

0.027

S(35)

33

403

0.082

S(10)

19

403

0.047

S(1)

3

403

0.007

Figure 4. Human kinome wide selectivity profiling of compounds 5 and 7 in DiscoveRx KINOMEScan™ screening platform. Measurements were performed at 1 µM concentration of the compounds. Upper panel: TREEspot™ interaction maps for 5 and 7 in 468 kinase targets. Lower panel: S-scores of 5 and 7 with percent control numbers less than 35, 10 and 1, respectively.

Structure Optimization on Lead Compound 7. In view of the structural novelty, high potency and better kinase selectivity profile, we selected compound 7 as our tricyclic lead BTK inhibitor for further structural optimization. First, to identify the


Page 11 of 61 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


optimal covalent warhead binding to BTK Cys481 residue, a panel of reactive acrylamido groups was tested (Table 1). First, we inserted variant amino acid moieties into the acrylamido fragment and the corresponding compounds 8-11 showed reduced potencies against BTK. Compared to compound 7, insertion of smaller amino acid moieties such as 2-aminoacetyl (8, IC50 of 23.6 nM), 2-aminopropanoyl (9, IC50 of 40.1 nM) or 1-aminocyclopropanecarbonyl (10, IC50 of 23.8 nM) led to 8- to 14-fold loss of potency, and insertion of a longer amino acid moiety such as 3-aminopropanoyl (11, IC50 of 725.2 nM) significantly reduced the inhibitory potency. Substitution on the terminal vinyl carbon with a dimethylaminomethyl yielded compound 12, which retained high potency with an IC50 value of 4.3 nM. Notably, replacing the acrylamido moiety in 7 with a but-2-ynamido moiety provided compound 13 showing an IC50 value of 2.1 nM, a potency even higher than that of 7 (IC50 of 2.8 nM). Compound 14 lacking the acryloyl warhead completely lost the potency (IC50 > 1000 nM), whereas saturation of the acrylamido with nonreactive propionyl moiety led to 22-fold reduction of potency (15, IC50 of 62.6 nM). These results further confirmed that the existence of a Michael acceptor warhead was essential for BTK’s covalent binding, and the but-2-ynamido (13, IC50 of 2.1 nM) was slightly better than the prototypic acrylamido moiety (7, IC50 of 2.8 nM). Table 1. The in vitro BTK inhibition of compounds bearing different covalent or noncovalent reactive groups.

Compd.

R1


BTK IC50 (nM)a


Page 12 of 61

7

2.8 ± 1.1

8

23.6 ± 4.5

9

40.1 ± 4.7

10

23.8 ± 5.0

11

725.2 ± 182.5

12

4.3 ± 0.7

13

2.1 ± 0.2

14

> 1000

15

62.6 ± 13.4

a

IC50 values are reported as means of duplicates.

Chiral Resolution of Potent BTK inhibitors 5, 7 and 13. Since compound 1 is an optically pure R-enantiomer in the stereogenic carbon center, its S-enantiomer as well as the racemate are less potent.5a To determine where this was also the case in our new tricyclic BTK inhibitors, the R- and S-enantiomers of the three potent racemic compounds 5, 7 and 13 were prepared and tested. As shown in Table 2, enantiomers 5R and 5S showed nearly identical IC50 values (3.0 vs 2.8 nM). Surprisingly, significant discrepancy was observed between the two enantiomers of 7 and 13 both bearing the pyrimido[5,4-b]pyrrolizine framework. In both cases, the


Page 13 of 61 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


S-enantiomers were more potent than their corresponding R-enantiomers. Both 7S and 13S showed high potency with an identical IC50 value of 0.4 nM, which was equal to that of 1 (IC50 of 0.4 nM), whereas the IC50 values of R-enantiomers 7R and 13R were 6.1 and 3.3 nM, respectively. The similar biochemical activity of the enantiomers 5S and 5R on BTK binding is likely ascribed to the fact that the newly formed piperidine ring in 5 could torture to adapt a configuration change upon covalently binding to the active site of BTK. However, the newly formed pyrrole ring in 7 and 13 have more ring and steric strain that is disadvantageous for ring tortuosity. Table 2. The in vitro BTK inhibition of the stereoisomers of potent compounds 5, 7 and 13. Compd.

Configuration

BTK IC50 (nM)a

5 (racemate)

5.3 ± 0.2

5R (R-enantiomer)

3.0 ± 1.9

5S (S-enantiomer)

2.8 ± 0.3

7 (racemate)

2.8 ± 1.1

7R (R-enantiomer)

6.1 ± 1.4

7S (S-enantiomer)

0.4 ± 0.1

13 (racemate)

2.1 ± 0.2

13R (R-enantiomer)

3.3 ± 1.1

13S (S-enantiomer)

0.4 ± 0.1

a




Page 14 of 61

Pharmacokinetic (PK) Parameters of the Selected BTK Inhibitors 7S and 13S. Since both 7S and 13S showed high potencies against BTK kinase, they were selected for pharmacokinetic study in rats following intravenous (1 mg/kg) and oral (3 mg/kg) administration. As a comparison, 1 was also tested in the same conditions. As shown in Table 3, both 7S and 13S have similar half-lives (T1/2 = 0.57 h and 0.54 h, respectively), which were slightly longer than that of 1 (T1/2 = 0.42 h). Meanwhile, better plasma exposure after oral administration was observed for both tricyclic compounds, and the AUC0-∞ values for 7S and 13S were 4- and 10-fold higher than that of 1, respectively. The same trend was observed in their Cmax values. In addition, both 7S and 13S showed lower clearances than 1. Finally, the oral bioavailabilities of 7S and 13S were calculated as 18.3% and 35.8%, respectively, which were much higher than that of 1 (F = 4.00%). Table 3. PK parameters of the new BTK inhibitors 7S and 13S in rats.a iv (1 mg/kg) Compd.

po (3 mg/kg)

CL (L/h/Kg)

Vss (L/kg)

T1/2 (h)

Cmax (ng/mL)

Tmax (h)

AUC0-∞∞ (ng·h/mL)

F (%)

1

61.2

1838

0.42

39

0.25

57

4.00

7S

46.3

1299

0.57

223

0.25

211

18.3

13S

32.8

1390

0.54

340

0.25

557

35.8

a

Values are the average of three determinations. Vehicle: DMSO, Tween 80, normal saline. CL,

clearance; Vss, volume of distribution; T1/2, half-life; Cmax, maximum concentration; Tmax, time of maximum concentration; AUC0-∞, area under the plasma concentration time curve; F, oral bioavailability.

Metabolic Stability Study of the Selected BTK Inhibitors 7S and 13S. To evaluate the metabolic stability of the new tricyclic BTK inhibitors 7S and 13S, we


Page 15 of 61

investigated the potential metabolites of both compounds by incubating with liver microsomes of different species (mouse, rat and human) for three hours. The major metabolites were tested by LC/MS (supporting information, Table S1). Both compounds showed moderate to good stability with similar metabolic products in the liver microsomes. Compound 7S was more stable in mouse liver microsomes (MLM) than in human liver microsomes (HLM) and rat liver microsomes (RLM). Differently, compound 13S was much more stable in HLM and MLM, but not stable in RLM. In general, the major metabolic pathway for both compounds was the para-oxidation of the phenyl group in the diphenyl ether component leading to para-hydroxylated compound M-1, accounting for 16.5% and 18.8%, respectively, of the total metabolites (including parent compound M-0) in HLM (Table S1). In addition, substantial dihydroxylation occurred in both the vinyl (7S) and ethynyl (13S) moieties across the three liver microsomes. The proposed metabolic pathways of both compounds were shown in Figures 5A and 5B.

hy y la ox tio n

O-dearylation

dr

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



ox dr hy

ati on de ary l O-

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

yla tio n


Figure 5. Metabolic study of compounds 7S and 13S in liver microsomes of different species.

Further Modification of Compound 7S. Since the metabolic stability study in liver microsome revealed that the most active metabolic site of compound 7S might be the para-position of the terminal phenyl group. Therefore, two approaches were conducted to eliminate or ultimately reduce the para-phenyl oxidative metabolism (Table 4). First, we introduced a small series of substituents to the para-position of the terminal phenyl to block the hydroxylation liability. It was found that all the substituents, including -t-Bu (16), -F (17), -CF3 (18) and -OCF3 (19) groups, appended in the para-position caused 5- to 19-fold decrease of potency, comparing to that of 7S. The reduced potency was likely due to the steric hindrance in the hydrophobic binding pocket. In the meantime, replacement of the terminal phenoxy fragment with various N-aryl or N-arylmethyl carbamic groups was conducted to block the active metabolic site. To our delight, compared to 7S, compound 20 bearing a pyridie-2-ylcarbamic group showed compatible biochemical potency against BTK with an IC50 value of 0.5 nM. Compound 21 bearing an additional trifluoromethyl


Page 16 of 61

Page 17 of 61 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


group also retained high potency of 0.8 nM. Replacing the terminal phenoxy group of 7S with (1-(pyridin-2-yl)ethyl)carbamic moiety significantly reduced the potency (22, IC50 of 50.4 nM), whereas slightly lower potency was observed for compounds 23 and 24, both bearing a benzylcarbamic moiety and showing IC50 values of 3.1 nM and 4.3 nM, respectively. Table 4. The in vitro BTK inhibition of compounds modified in terminal phenyl to block the para-oxidative metabolism.

Compd.

R2

BTK IC50 (nM) a

7S

0.4 ± 0.1

16

3.7 ± 0.9

17

2.2 ± 1.5

18

7.4 ± 2.5

19

3.5 ± 1.7

20

0.5 ± 0.1

21

0.8 ± 0.1



Page 18 of 61

22

50.4 ± 9.4

23

3.1 ± 1.0

24

4.3 ± 1.7

a


Antiproliferative Effects of Compounds 7S, 20 and 1. Since the optimized compound 20 showed similar high potency against BTK as that of the lead compound 7S and the approved drug 1, both 7S and 20 were selected for antiproliferative effect study in the B-cell lymphoma (Ramos and TMD8) cells. Compound 1 was also tested as a comparison. Both Ramos and TMD8 cells are B-cell lymphoma cell lines expressing BTK protein, but the Ramos cells are not strongly depended on it, whereas the survival of TMD8 cells is strongly dependent on the expression of BTK protein.4b As shown in Table 5, compounds 7S and 20 suppressed the proliferation of Ramos cells in micromolar range with IC50 values of 5.03 µM and 14.3 µM, respectively, whereas the approved drug 1 showed an IC50 value of 0.92 µM. This result confirmed that 1 was a non-selective BTK inhibitor and its antitumor activity was a consequence of multiple kinase inhibition. The lower potency of the new inhibitors 7S and 20 indicated that these new tricyclic compounds were more BTK-selective. This analysis was further confirmed by the high antiproliferative potency against the BTK highly sensitive TMD8 cells. All the three compounds 1, 7S and 20 showed low nanomolar potency with IC50 values of 10 nM, 16 nM and 4 nM, respectively in this cell line. Table 5. Target inhibition and antiproliferative effects of selected compounds. Compd.

IC50 (µM)a


Page 19 of 61 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


BTK

Ramos

TMD8

1

0.0004 ± 0.0001

0.92 ± 0.1

0.010 ± 0.004

7S

0.0004 ± 0.0001

5.03 ± 0.85

0.016 ± 0.010

20

0.0005 ± 0.0001

14.3 ± 6.1

0.004 ± 0.003

a


From the results above, both the new inhibitors 7S and 20 showed high potency and selectivity against BTK, especially for compound 20. Unfortunately, further pharmacokinetic study on 20 showed a lower oral bioavailability of 8.24% (Supporting Information, Table S2). Taking together, compound 7S showed overall optimal drug candidacy and was elected for further profiling. Covalent docking mode of 7S with BTK. Similar to binding mode of compound 7 with BTK in Figure 2C, the specific S-enantiomer 7S could covalently bind to BTK Cys481. As shown in Figure 6, 7S overlapped well with 1, including the bicyclic skeleton as well as the hydrophobic moiety. The sulfhydryl of Cys481 rotated about 68o to adapt the newly formed covalent bond between the acrylamido moiety of 7S and BTK Cys481. Critical hydrogen bonds with hinge residues Met477, Glu475 and Thr474 were also observed with distances of 2.0, 1.9 and 2.8 Å, respectively. In addition, there is a π-π interaction between the terminal phenyl group and Phe540.

Figure 6. Covalent docking mode of compound 7S (green) with BTK (PDB id: 3GEN)



Page 20 of 61

overlapping with 1 (purple).

Kinase Selectivity Profile of 7S. On the basis of the KINOMEScan™ screening results of the racemic compound 7 (Figure 4), we further tested the biochemical activity of compound 7S against a panel of 14 kinases. Similar to 1, the major off-target kinases of 7S were those with cysteine residue in the ATP binding pocket, which could be irreversibly bound by the covalent reactive group (Table 6). As such, compound 7S showed a high potency against EGFR (IC50 of 1.3 nM), but variant potency against the other members of the Tec kinase family with IC50 values of 174.1, 15.1, 1.4 and 2.5 nM for BMX, ITK, TEC and TXK, respectively. Although the overall selectivity was similar to 1, 7S showed greater BTK-selectivity than 1 against BMX (174.1 vs 5.8 nM), RET (20.3 vs 5.2 nM) and ErbB2 (23.2 vs 1.5 nM). It was not clear whether such difference in kinase selectivity between 7S and 1 was beneficial to the antitumor efficacy or clinical safety, a careful balance of the clinical outcomes should be taken if compound 7S finally proceeds to clinical trials. Table 6. Kinase selectivity profile of compound 7S. IC50 (nM)a Kinase 7S

1

BTK

0.4 ± 0.1

0.4 ± 0.2

BLK

0.2 ± 0.1

0.5 ± 0.2

ErbB4

0.6 ± 0.2

1.8 ± 1.0

EGFR

1.3 ± 0.3

1.2 ± 0.3

TEC

1.4 ± 0.6

0.5 ± 0.3

TXK

2.5 ± 0.5

3.0 ± 1.8

ITK

15.1 ± 3.2

13.9 ± 1.6

RET

20.3 ± 6.6

5.2 ± 2.7

ErbB2

23.2 ± 1.5

1.5 ± 0.9


Page 21 of 61 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


Flt-3

55.0 ± 11.7

>1000

BMX

174.1 ± 93.3

5.8 ± 2.3

PDGFR-β

214.5 ± 126.0

> 1000

EPH-A2

> 1000

> 1000

CSF1R

> 1000

> 1000

a


Effects of Compound 7S on TMD8 cells. BTK plays an imperative role in BCR signaling pathway, which is relevant to several severe leukemias and lymphomas. BCR is first activated by CD79, which causes activation of upstream kinases, such as SYK (Spleen tyrosine kinase). Then BTK is recruited to cell membrane and phosphorylated in its SH2 domain. The activated BTK then phosphorylates its substrate phospholipase γ2 (PLCγ2), leading to activation of the downstream signaling pathways, such as NF-κB and STAT3.22 In the BTK-dependent diffuse large B-cell lymphoma (DLBCL) TMD8 cells, both compound 7S and 1 were found to significantly inhibit the autophosphorylation of BTK at the Y223 site. Particularly, 7S showed a dose-dependent inhibition with an IC50 value less than 10 nM. Furthermore, 7S inhibited downstream phosphorylation of Y759 of PLCγ2 comparable to that of 1 at 1 µM (Figure 7A).



Figure 7. Effects of compound 7S on TMD8 cells. (A) Inhibitory activity of 7S on BTK and downstream signaling pathway in cells. Cells were starved in serum-free medium before treated with the indicated concentrations of compound 7S or 1 for 4 h and stimulated by anti-IGM. (B) Washout experiment of 7S in TMD8 cells. (C) Effects of 7S on cell cycle in TMD8 cells. (D) 7S induced apoptosis of TMD8 cells. Data are shown as mean ± SD. Each experiment was conducted independently for three times. *P

pyrrolizines as Novel Bruton's Tyrosine Kinase (BTK)

Recommend Documents