Selective Inhibition of Histone Deacetylase 10: Hydrogen Bonding to

Furthermore, docking into human HDAC10 homology models indicated that a hydrogen-bond between a cap group nitrogen and the gatekeeper residue Glu272 ...
0 downloads 0 Views 1MB Size
Subscriber access provided by IDAHO STATE UNIV

Article

Selective Inhibition of Histone Deacetylase 10: Hydrogen Bonding to the Gatekeeper Residue is Implicated Magalie Géraldy, Michael Morgen, Peter Sehr, Raphael Steimbach, Davide Moi, Johannes Ridinger, Ina Oehme, Olaf Witt, Mona Malz, Mauro S. Nogueira, Oliver Koch, Nikolas Gunkel, and Aubry K. Miller J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01936 • Publication Date (Web): 09 Apr 2019 Downloaded from http://pubs.acs.org on April 9, 2019

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 90 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

Selective Inhibition of Histone Deacetylase 10: Hydrogen Bonding to the Gatekeeper Residue is Implicated Magalie Géraldy1,†, Michael Morgen1, Peter Sehr2, Raphael R. Steimbach1,3, Davide Moi1, Johannes Ridinger3,4,5,6, Ina Oehme4,5,7, Olaf Witt4,5,6,7, Mona Malz1, Mauro S. Nogueira8, Oliver Koch8,†, Nikolas Gunkel1,7, Aubry K. Miller1,7,*

1 Cancer

Drug Development Group, German Cancer Research Center (DKFZ), 69120 Heidelberg, Germany

2

Chemical Biology Core Facility, European Molecular Biology Laboratory, 69117 Heidelberg, Germany

3

4

Biosciences Faculty, University of Heidelberg, 69120 Heidelberg, Germany

Hopp Children’s Cancer Center Heidelberg (KiTZ), 69120 Heidelberg, Germany

1 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

5

Page 2 of 90

Clinical Cooperation Unit Pediatric Oncology, German Cancer Research Center (DKFZ), 69120 Heidelberg, Germany

6

Department of Pediatric Oncology, Hematology and Immunology, University Hospital Heidelberg, 69120 Heidelberg, Germany.

7

8

German Cancer Consortium (DKTK), 69120 Heidelberg, Germany

Faculty of Chemistry and Chemical Biology, TU Dortmund University, 44227 Dortmund, Germany

ABSTRACT

The discovery of isozyme-selective histone deacetylase (HDAC) inhibitors is critical for understanding the biological functions of individual HDACs and for validating HDACs as drug targets. The isozyme HDAC10 contributes to chemotherapy resistance and has recently been described to be a polyamine deacetylase, but no studies toward selective HDAC10 inhibitors have been published. Using two complementary assays, we found tubastatin A, an HDAC6 inhibitor, to potently bind HDAC10. We synthesized tubastatin A 2 ACS Paragon Plus Environment

Page 3 of 90 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

derivatives and found that a basic amine in the cap group was required for strong HDAC10 binding. HDAC10 inhibitors mimicked knockdown by causing dose-dependent accumulation of acidic vesicles in a neuroblastoma cell line. Furthermore, docking into human HDAC10 homology models indicated that a hydrogen-bond between a cap group nitrogen and the gatekeeper residue Glu272 was responsible for potent HDAC10 binding. Taken together, our data provide an optimal platform for the development of HDAC10selective inhibitors, as exemplified with the tubastatin A scaffold.

INTRODUCTION

The discovery of histone deacetylase 1 (HDAC1) by Taunton and Schreiber in 19961 provided the sought-after enzyme target for substances like trichostatin A (TSA (1), Figure 1) and SAHA (2). At the time, 1 and 2 were reported to increase histone lysine acetylation levels, thereby inducing cellular differentiation, but their mechanism(s) of action were unknown.2 In the intervening time, it has become clear that HDACs, a family of 18 functionally related isozymes, have a broader role than catalyzing the hydrolysis of acetylated histone lysines: not only do they act in both the nucleus and the cytoplasm, 3 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

Page 4 of 90

but they catalyze the removal of acyl groups from a variety of different proteins, and therefore regulate numerous biological processes.3-7 Misregulation of acetylation mechanisms has been linked to multiple human diseases and there are four HDAC inhibitor drugs approved by the U.S. FDA (2–5) and one in China (6) (Figure 1).8

O

O Me

Me2N

OH

N H

Me

O

O

H N

N H

O

OH HN

SAHA (2)

TSA (1)

O O

O O S N H

N H

O

OH

HN

N H

N

HN

O O

O

F

Chidamide (6)

Cap group

N S

N O

O

Me

N H N

O N H

OH

N

N

O HO

NH2

H N

Romidepsin (5)

Ph

O

Panobinostat (3)

N H

O

Belinostat (4) Ph

Me

OH

O

S S NH O

N H

H N

H N O

H N

OH Linker

OH

ACY-738 (9)

O Zinc Binding Group

Tubacin (7) Tubastatin A (8)

Figure 1: Well-known HDAC inhibitors

The approved drugs are known to cause severe side effects, which is attributed in part to their lack of isozyme selectivity.9-10 More selective inhibitors are expected to overcome 4 ACS Paragon Plus Environment

Page 5 of 90 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

these liabilities and are likely to improve the clinical value of this target class.11 Moreover, the development of isozyme-selective chemical probes will be critical to further disentangle the biological role(s) of individual HDAC members.12 A number of selective HDAC6 (Class IIB) inhibitors have been described, and tubacin (7), tubastatin A (8), and ACY-738 (9) are all designated as HDAC6 chemical probes in the Chemical Probes Portal (Figure 1).13 Most HDAC6-selective inhibitors, like 8 and 9, achieve selectivity over Class I enzymes by incorporating a relatively bulky phenyl hydroxamate “linker” moiety in addition to a “cap group” that can make specific interactions with the HDAC6 protein surface (see 8, Figure 1). HDAC10,14-16 the only other HDAC Class IIB member, has received little attention from the medicinal chemistry community,17-19 and to the best of our knowledge no study dedicated to the development of HDAC10-selective inhibitors has been published. On the other hand, a number of studies have highlighted HDAC10 as a potential cancer drug target.20-23 In one study, high HDAC10 expression levels were found to correlate with poor clinical outcome for advanced stage 4 neuroblastoma patients who received chemotherapy.23 Consistent with these findings, HDAC10 depletion in neuroblastoma 5 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

Page 6 of 90

cells interrupts autophagic flux and sensitizes cells for chemotherapy, and enforced HDAC10 expression protects neuroblastoma cells against doxorubicin treatment.18 Recently, Christianson and co-workers solved the x-ray crystal structure of zebrafish HDAC10 (zHDAC10), and elegantly demonstrated that both zHDAC10 and human HDAC10 (hHDAC10) are highly active polyamine deacetylases (PDACs), while being poor lysine deacetylases.4 Two specific structural features near the active site of the enzyme(s) were identified as being responsible for PDAC activity. First, the negatively charged Glu272 (hHDAC10 numbering) amino acid was demonstrated to be a unique gatekeeper residue, which establishes specificity for cationic polyamine substrates over acetylated lysines. Second, the L1 loop of HDAC10 contains a four residue 310-helix (Pro21-Glu22-Cys23-Glu24, hHDAC10 numbering) that constricts the active site, causing an acetylated lysine side chain to be too short to reach the active site zinc atom. In the present study, we report our initial studies to discover selective HDAC10 inhibitors. Using two ligand displacement assays, we demonstrate that tubastatin A (8) binds to HDAC10 even more potently than to HDAC6. Structure–activity relationship (SAR) studies on 8, supported with selectivity testing against Class I enzymes, functional 6 ACS Paragon Plus Environment

Page 7 of 90 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

cell-based experiments, and docking into human HDAC10 homology models, suggest that binding to the Glu272 gatekeeper residue is critical for potent HDAC10 inhibition. Collectively, our results present the first systematic medicinal chemistry study of HDAC10 inhibition, and lay the foundation for the development of isozyme-selective HDAC10 inhibitors.

RESULTS AND DISCUSSION Testing of known substances in two displacement assays reveals unexpectedly potent HDAC10 binding. Although it is known that HDAC10 enzymatic activity is challenging to assay with acetylated lysine substrates,12 some contract research organizations offer enzymatic profiling of all HDACs, and IC50 values measured in this way against HDAC10 are sometimes reported in the literature. It has also been claimed that residual co-purified Class I HDACs from HDAC preparations of Class IIA HDACs are largely responsible for the activity observed when testing these proteins in acetyl-lysine-based enzymatic assays.24 In light of this claim, and the recent finding that HDAC10 has significant PDAC activity,4 we analyzed a selection of literature data, where enzymatic activities across 7 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

Page 8 of 90

multiple HDAC isozymes were reported.25-33 Interestingly, strong correlations between HDAC10 and Class I HDAC inhibition values were found, while HDAC6 gave a completely distinct activity profile, despite its high sequence homology to HDAC10 as a Class IIB enzyme (Figure 2, Supplementary Figures 1 and 2).

Figure 2. Hierarchical clustering reveals a correlation between HDAC Class I and HDAC10 activities. A selectivity profile of each compound was calculated relative to HDAC2. After log2 transformation, the values were subjected to hierarchical clustering using the complete-linkage agglomeration rule. Red and green colors indicate higher and lower reported IC50 values, respectively, of a given drug on an HDAC isozyme relative to HDAC2. Superscripts indicate the reference from where the data was taken.

8 ACS Paragon Plus Environment

Page 9 of 90 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

On the basis of this analysis, and due to the low-throughput format of Christianson’s PDAC assay, we decided to assay HDAC10 in a non-enzymatic format. To eliminate any complications that could arise from co-purified HDAC impurities, two ligand displacement assays that employ genetically tagged HDAC proteins for signal generation were utilized, thereby ensuring reliable identification of HDAC10 binders. For screening against pure recombinant HDAC10, a time-resolved fluorescence resonance energy transfer (TR-FRET) assay that measures displacement of a fluorescently-labeled ligand was used (Figure 3A).34 A conceptually similar bioluminescence resonance energy transfer (BRET) assay was also utilized with cells that overexpress nano-luciferase-tagged HDAC10 (Figure 3B, Supplementary Figure 3).35 The BRET assay can independently confirm the TR-FRET assay and provides target engagement values in living cells.

9 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

A

340 nm

340 nm 615 nm FRET

Eu

tag

665 nm HDAC10 binder

tracer

HDAC10

B

450 nm BRET nano luc

Page 10 of 90

tracer

HDAC10

615 nm

Eu

tag

tracer

HDAC10 610 nm

450 nm HDAC10 binder

nano luc

tracer

HDAC10

Figure 3: HDAC10 ligand displacement assays. A) TR-FRET assay: A Europium labelled anti-GST tag antibody is coupled to GST-tagged HDAC10 protein. When HDAC10 is bound by a dye-labelled tracer molecule, irradiation with 340 nm light produces productive FRET between the Europium (615 nM emission) and the dye (665 nM emission). In the presence of an HDAC10 binder, which competes with the tracer ligand, FRET is disrupted and a change in the signal is observed. B) BRET assay: When cells expressing nanoluciferase-tagged HDAC10 are treated with a dye-labelled tracer and nano-luciferase substrate, productive BRET is observed. In the presence of an HDAC10 binder, BRET is disrupted.

A series of commercially available HDAC inhibitors, which have been reported to have high HDAC6 activity, were tested in both HDAC10 assays (Table 1, Supplementary Table 1, Supplementary Figure 1). Ricolinostat (10) showed almost a 10-fold difference in activity between the two assays, but the majority of compounds had pIC50 values that

10 ACS Paragon Plus Environment

Page 11 of 90 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

were within 0.5 log units of each other. The BRET assay typically gave lower potencies, which can be expected when comparing cellular and biochemical assay formats. It was not surprising that some of the pan-HDAC inhibitor compounds (2, 10–15) showed quite high activity in the HDAC10 displacement assays. Indeed, all of these compounds, except pracinostat (13), had sub-micromolar IC50 values (pIC50 ≥ 6.0) in at least one of the assays. Most interesting were the results with tubastatin A (8) HPOB (16)33, and nexturastat (17),36 compounds that have been described as highly selective HDAC6 inhibitors with poor HDAC10 inhibitory activities. While both HPOB and nexturastat were found to be moderately potent HDAC10 binders, Tubastatin A was active in the low nanomolar range (pIC50 = 7.9) in both the FRET and BRET assays. A control experiment with PCI-34051 (18), a hydroxamate-containing HDAC8 selective inhibitor, gave low binding values in both assays.

Table 1. TR-FRET and BRET HDAC10 binding data with commercial inhibitors.

Cmpd.

Substance

HDAC10

HDAC10

11 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

a

Page 12 of 90

FRET

BRET

pIC50a

pIC50b

2

SAHAc

6.7

5.9

10

Ricolinostatc

6.9

5.8

11

CAY-10603c

6.5

6.1

12

CUDC-101c

6.9

6.4

13

Pracinostatc

5.8

5.5

14

Quisinostatc

8.0

7.4

15

Abexinostatc

8.4

8.1

8

Tubastatin Ad

7.9

7.9

16

HPOBd

7.1

6.5

17

Nexturastatd

7.0

6.3

18

PCI-34051e

4.4