Identification and development of an irreversible monoacylglycerol

Subscriber access provided by Nottingham Trent University

Article

Identification and development of an irreversible monoacylglycerol lipase (MAGL) positron emission tomography (PET) radioligand with high specificity Lei Zhang, Christopher R. Butler, Kevin P Maresca, Akihiro Takano, Sangram Nag, Zhisheng Jia, Ryosuke Arakawa, Justin Piro, Tarek Samad, Deborah Lynn Smith, Deane M Nason, Steven V. O'Neil, Laura A. McAllister, Klaas Schildknegt, Patrick Trapa, Timothy McCarthy, Anabella Villalobos, and Christer Halldin J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 04 Sep 2019 Downloaded from pubs.acs.org on September 4, 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 34 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

Identification and development of an irreversible monoacylglycerol lipase (MAGL) positron emission tomography (PET) radioligand with high specificity Lei Zhang,1* Christopher R. Butler,1 Kevin P. Maresca,2 Akihiro Takano,4 Sangram Nag,4 Zhisheng Jia,4 Ryosuke Arakawa,4 Justin R. Piro,6 Tarek Samad,6 Deborah L. Smith,6 Deane M. Nason,3 Steven O’Neil,3 Laura McAllister,1 Klaas Schildknegt,7 Patrick Trapa,5 Timothy J. McCarthy,2 Anabella Villalobos,8 and Christer Halldin4 1Medicine 2Clinical

Design, Medicinal Chemistry, Pfizer Inc., Cambridge, Massachusetts 02139, United States;

& Translational Imaging, Early Clinical Development, Pfizer Inc., Cambridge, Massachusetts

02139, United States; 3Medicine Design, Medicinal Chemistry, Pfizer Inc., Groton, Connecticut 06340, United States; 4Department of Clinical Neuroscience, Center for Psychiatry Research, Karolinska Institutet and Stockholm County Council, SE-17176 Stockholm, Sweden;

5Medicine

Design,

Pharmacokinetics, Dynamics and Metabolism, Pfizer Inc., Groton, Connecticut 06340, United States; 6Internal

Medicine, Pfizer Inc., Cambridge, Massachusetts 02139, United States; 7Pharmaceutical

Sciences, Pfizer Inc., Groton, Connecticut 06340, United States; 8Biogen, Cambridge, Massachusetts 02142, United States. RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required Keywords: Positron emission tomography, PET, MAGL, irreversible MAGL inhibitor, CNS PET MPO, PET imaging

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 2 of 34

ABSTRACT:

Monoacylglycerol lipase (MAGL), a serine hydrolase extensively expressed throughout the brain, serves as a key gatekeeper regulating the tone of endocannabinoid signaling. Pre-clinically, inhibition of MAGL is known to provide therapeutic benefits for a number of neurological disorders. Availability of a MAGL-specific positron emission tomography (PET) ligand would considerably facilitate the development and clinical characterization of MAGL inhibitors via non-invasive and quantitative PET imaging. Herein, we report identification of the potent and selective irreversible MAGL inhibitor 7 (PF06809247) as a suitable radioligand lead, which upon radiolabeling was found to exhibit a high level of MAGL specificity; this enabled cross-species measurement of MAGL brain expression (Bmax), assessment of in vivo binding in rat, and non-human primate (NHP) PET imaging. INTRODUCTION Monoacylglycerol lipase (MAGL) is a serine hydrolase target of growing interest due to its bifunctional nature of signaling, both through the endocannabinoid pathway via 2-arachidonoylglycerol (2-AG) as well its influence on arachidonic acid (AA) production in the brain.1 MAGL catalyzes the hydrolysis of 2-AG to liberate AA; approximately 50% of AA produced in the brain derives from 2-AG. Peripherally, however, phospholipid hydrolysis by cytosolic phospholipase A2 (PLA2) acts as the predominant source of AA. The arachidonate pathway has a significant role in the inflammatory response, as AA is not only an inflammatory signal itself, but moreover can be converted downstream to pro-inflammatory eicosanoids. Thus, the dual impact of MAGL on brain inflammation occurs through the simultaneous modulation of both the endocannabinoid and arachidonate systems:2 inhibition of MAGL decreases pro-inflammatory eicosanoids and cytokines in the brain while also increasing 2-AG signaling at CB1/2 receptors.3 Mouse models of Alzheimer’s disease that invoke disruption of MAGL activity, either through genetic alteration or pharmacological inhibition, exhibit decreased neuroinflammation biomarkers, including cytokines and gliosis.4 Currently, steroids are the primary intervention for pronounced brain inflammation, but are limited in their duration and frequency of dosing due to a lack of precision in their mode of action. As such, there is a strong interest in the ACS Paragon Plus Environment

Page 3 of 34 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


development of potent, selective MAGL inhibitors for the treatment of acute and chronic diseases associated with neuroinflammation;5 this will simultaneously require the identification of quantifiable and translational biomarkers to facilitate clinical development of such inhibitors. Specifically, we are interested in developing a positron emission tomography (PET) radioligand6 with high target specificity to enable preclinical and clinical characterizations of MAGL inhibitors. PET is a well-established, non-invasive imaging method that has proven to be particularly valuable for central nervous system (CNS) drug discovery, as it provides quantitative biomarker information, such as target expression and target occupancy measurements, that is otherwise unattainable in the clinical setting.7 Our group has previously disclosed a set of physicochemical and in vitro ADME parameters to guide rational design of novel CNS PET tracers.8 Application of these CNS PET ligand design parameters has facilitated our research team’s discovery of several first-in-class PET ligands, targeting PDE2,8 PDE4B,9 and recently BACE1.10 Specifically, we focus our PET ligand discovery effort within the favorable physicochemical property space defined by CNS PET multi-parameter optimization score (MPO) > 3. This optimization increases the likelihood of the desired alignment between passive permeability [Ralph-Russ canine kidney (RRCK) Papp AB > 5 x 10-6 cm/s] / low efflux (MDR1 BA/AB ≤ 2.5) and fraction unbound in brain [(Fu_b) > 0.05], for brain permeability and low risk of non-specific binding (NSB), respectively. Herein, we describe the discovery, guided by the aforementioned CNS PET ligand design parameters, of the highly specific MAGL PET radioligand lead PF-06809247 (7). This PET ligand enabled measurement of cross-species MAGL expression, rat in vivo binding study and, subsequently, a key non-human primate (NHP) PET imaging study, which demonstrated rapid brain uptake and high MAGL specificity for 7. RESULTS AND DISCUSSION



A) O 11 C N R

O O

CF3 R=

CF3

O

O

HO

11 1 [ C]KML29

O

HO

2-AG

OH R=

O O

N N

N

2 [11C]JJKK-0048 O F3C

R=

Cl H N

O

O 11 C R N

O

H

11

3 [ C]SAR127303 R=

S O

CF3

N N

N

5

F

B) Colored by ClogD ClogD ≤ 3 ClogD > 3

0.08

N

N

4 [11C]TZPU

0.10

OH

O (R)

N

H

CF3

cFu_b

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 34

JJKK-0048 (2)

Comp.

ClogD

CNS PET MPO

cFu_b

Serine hydrolase selectivity

1

5.04

2.00

0.002

yes

2

1.34

3.95

0.09

no

3

4.66

2.45

0.003

yes

4

0.86

4.00

0.08

no

5

3.09

3.81

0.02

yes

TZPU(4)

PET-specific analogs

0.06

0.04

0.02

5 KML29 (1) SAR127303(3)

0 1

2

3

4

CNS PET MPO

5

ClogD ≤ 3 ClogD > 3

CNS PET MPO > 3 CNS PET MPO ≤ 3

cFu_b > 0.05 cFu_b ≤ 0.05

Figure 1. A) Structures of representative literature MAGL inhibitor chemotypes; B) analysis of CNS PET parameters and serine hydrolase selectivity.

Our PET ligand discovery effort started with a property survey on available literature and Pfizer inhouse MAGL inhibitor chemotypes, with a primary goal of defining an optimal starting point for PETspecific SAR. As shown in Figure 1A, previously disclosed MAGL PET radioligands bear two different ACS Paragon Plus Environment

Page 5 of 34 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


types of leaving groups: a hexafluoroisopropanol (HFIP) moiety, such as [11C]KML29 (1)11,12 and [11C]SAR127303 (3),12,13 or a 1,2,4-triazole moiety, as found in [11C]JJKK-0048 (2)14 and [11C]TZPU (4).12 Ligands incorporating the polar 1,2,4-triazole clearly demonstrate better properties for PET considerations, including low ClogD, favorable CNS PET MPO and high cFu_b (Figure 1B). The high reactivity of this moiety, however, leads to unfavorable selectivity versus other serine hydrolases,15 and this hampers their usage as PET ligands. Unlike 1,2,4-triazole, HFIP exhibits sufficient reactivity toward Ser122 at the MAGL active site while retaining sufficient stability to minimize off-target serine hydrolase inhibitions, resulting in a desirable level of MAGL potency and selectivity. Unfortunately, the highly lipophilic nature of the HFIP moiety results in sub-optimal CNS PET MPO scores (2.00 for compound 1 and 2.45 for compound 3) and exceptionally low fractions unbound in brain (0.002–0.003 for these two ligands), suggesting a high risk of non-specific binding (Figure 1B). In contrast, a new irreversible MAGL inhibitor 5 bearing a fine-tuned trifluoromethyl glycol leaving group, recently disclosed by our team,16 exhibits a well-balanced profile that not only retains the potency and selectivity of HFIP, but also lower lipophilicity; this translates to both favorable physicochemical properties (CNS PET MPO = 3.81) and a much improved in silico calculated fraction unbound in brain (cFu_b= 0.02). While compound 5 did not fully satisfy our PET design criteria, it served as an encouraging starting point relative to the other irreversible chemotypes. In an effort to mitigate PET-specific shortcomings, a design strategy was employed that included the following: 1) further reduction in lipophilicity while maintaining desirable MAGL potency/selectivity; and 2) introduction of a suitable functional group amenable to late-stage [3H]- or [11C]-radiolabeling, to enable saturation binding and PET imaging studies (e.g., a methoxy group). Based on the SAR previously reported around 516, our chemistry effort focused on the R1 vector due to its toleration of structural variations (Table 1). Significant prioritization was applied to potential analogs of interest by using in silico calculated PET-specific properties. This provided a minimalistic approach that advanced only the top targets to synthesis, through application of the following criteria: 1) improved PET parameters compared to compound 5, particularly ClogD and cFu_b; 2) the presence of structural moieties amenable to facile [3H]- and [11C]-labeling; and 3) the ready availability of coupling ACS Paragon Plus Environment


Page 6 of 34

reagents, plus high synthetic doability. From this exercise, analogs 6 and 7 clearly emerged as the top targets, each bearing a methoxy pyridyl moiety suitable for radiolabeling. The calculated PET properties of 6 and 7 are given in Table 1, showing decreased lipophilicity (ClogD < 3), favorable predicted passive permeability (cRRCK), low efflux (cMDR1 BA/AB), and improved cFu_b. Compounds 6 and 7 were synthesized and evaluated in our primary human MAGL pharmacology assay. Encouragingly, both compounds retained suitable MAGL potency, with an approximately 4-fold drop from the reference compound 5. Although the pharmacology profiles of these two compounds were indistinguishable, compound 7 (PF-06809247) was selected for further advancement, primarily based on structural considerations: this constitutional isomer is free of the potential competition between O- and Nmethylation that could complicate radiolabeling of 6. Table 1. PET-specific SAR around compound 5 O F3C N

H

OH

O

H N R

Compound

1

ClogD

CNS PET MPOa

cRRCK Papp ABb

cMDR1 BA/ABc

cFu_bd

hMAGL IC50 (nM)e

Kinact/Ki (M-1s-1)f

F

3.1

3.8

21

1.3

0.02

3

29724

O

2.2

3.6

14

1.2

0.04

11

4621

O

2.7

3.2

16

1.2

0.05

13

7806

R1

5

N

N

6

N

7 (PF06809247)

aCNS

PET MPO represents the summation of scores of six commonly used individual physicochemical properties, using the transformed functions described in reference 8. bIn silico calculated passive permeability as a rate in 1 x 10-6 cm/sec. ACS Paragon Plus Environment

Page 7 of 34 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 cIn

silico calculated MDR1 efflux ratio. silico calculated free fractions in brain. eIC 50 values measured from human recombinant MAGL enzyme inhibition assay (30 min preincubation). Values represent the geometric mean of at least three experiments (for detailed assay condition, please see reference 16). fK inact/Ki ratio in human recombinant MAGL enzyme inhibition assay (for detailed assay condition, please see reference 16). dIn

Additional pharmacology and selectivity data, as well as the in vitro PK profile of 7, are summarized in Table 2. The results reveal minimal cross-species potency differences between human, rat, and mouse MAGL and a high level of selectivity versus fatty acid amide hydrolase (FAAH) (>1000-fold). The experimentally derived in vitro PK parameters align with the in silico predicted values and reside in favorable PET property space, with good passive permeability (RRCK Papp AB = 13), low efflux (MDR1 BA/AB = 1.3), and improved unbound fractions in brain (rat Fu_b = 0.041) and plasma (NHP Fu_p = 0.059 and human Fu_p = 0.044). To ascertain the selectivity of 7 beyond FAAH, it was screened against a panel of serine hydrolases, as summarized in Figure 2. At a concentration of 10 µM, 7 demonstrated modest inhibition of carboxyesterase enzyme CES1 (76%) and alpha/beta hydrolase domain 6 enzyme ABHD6 (67%), with minimal or no inhibition observed for the remainder of the panel, in line with the selectivity expected for this leaving group.16 Furthermore, compound 7 exhibited an exquisite selectivity profile in the CEREP® panel (Table 2),17 showing no significant activities at the test concentration of 10 µM, except for the CB1 receptor (EC50 = 1.5 µM). Further follow-up with an internal CB1 functional agonism assay revealed no appreciable CB1 agonism activity (EC50 > 10 µM, see supporting information). Prompted by its favorable potency, in vitro PK, and good selectivity profile, compound 7 was advanced to [3H]- and [11C]-labeling to further assess its potential as a MAGL radioligand lead in both in vitro and in vivo settings.

Table 2: Potency, selectivity, and in vitro PK profile of PET radioligand lead 7.



O F3C N

H

O

N

N O

Potency and Selectivity

In vitro PK Properties

hMAGL IC50a: 13 nM rMAGL IC50b: 7.6 nM mMAGL IC50c: 12 nM FAAH IC50d: 16 M

RRCK Papp ABe = 13 MDR1 BA/ABf = 1.3 Fu_bg = 0.041 Fu_p (NHP)h = 0.059 Fu_p (Human)i = 0.044

OH

H N

Page 8 of 34

7

[3H] and [11C] labeling site

CEREP panel: CB1 agonism: 93% @ 10 M; all others

Identification and development of an irreversible monoacylglycerol

Recommend Documents