Structure-Based Drug Discovery of N-((R)-3-(7-Methyl-1 H-indazol-5-yl)-1-oxo-1-(((S)-1-oxo-3-(piperidin-4-yl)-1-(4-(pyridin-4-yl)piperazin-1-yl)propan-2-yl)amino)propan-2-yl)-2′-oxo-1′,2′-dihydrospiro[piperidine-4,4′-pyrido[2,3- d][1,3]oxazine]-1-carboxam

Article abstract of DOI:10.1021/acs.jmedchem.0c01003

Structure-based drug design enabled the discovery of 8, HTL22562, a calcitonin gene-related peptide (CGRP) receptor antagonist. The structure of 8 complexed with the CGRP receptor was determined at a 1.6 ? resolution. Compound 8 is a highly potent, select

Full text of DOI:10.1021/acs.jmedchem.0c01003

Subscriber access provided by ECU Libraries
Article
Structure-Based Drug Discovery of N-((R)-3-(7-Methyl-1H-indazol-5-yl)-1-
oxo-1-(((S)-1-oxo-3-(piperidin-4-yl)-1-(4-(pyridin-4-yl)piperazin-1-yl)propan-2-
yl)amino)propan-2-yl)-2'-oxo-1',2'-dihydrospiro[piperidine-4,4'-pyrido[2,3-
d][1,3]oxazine]-1-carboxamide (HTL22562): a Calcitonin Gene-Related
Peptide (CGRP) Receptor Antagonist For Acute Treatment of Migraine
Sarah Joanne Bucknell, Mark A Ator, Alastair Brown, Jason Brown, Andrew Cansfield, Julie
Cansfield, John Christopher, Miles Congreve, Gabriella Cseke, Francesca Deflorian, Christopher
Jones, Jonathan S. Mason, Alistair O'Brien, Gregory R. Ott, Mark Pickworth, and Stacey Southall
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c01003 • Publication Date (Web): 19 Jun 2020
Downloaded from pubs.acs.org on June 19, 2020
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 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure-Based Drug Discovery of N-((R)-3-(7-
Methyl-1H-indazol-5-yl)-1-oxo-1-(((S)-1-oxo-3-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(piperidin-4-yl)-1-(4-(pyridin-4-yl)piperazin-1-
yl)propan-2-yl)amino)propan-2-yl)-2'-oxo-1',2'-
dihydrospiro[piperidine-4,4'-pyrido[2,3-
d][1,3]oxazine]-1-carboxamide (HTL22562): a
Calcitonin Gene-Related Peptide (CGRP) Receptor
Antagonist For Acute Treatment of Migraine
,
‖
§,†
‖
‖
‖
Sarah J. Bucknell* , Mark A. Ator , Alastair J. H. Brown , Jason Brown , Andrew D. Cansfield ,
‖
‖
‖
‖
Julie E. Cansfield , John A. Christopher , Miles Congreve , Gabriella Cseke , Francesca
‖
‖,‡
‖
‖
§
Deflorian , Christopher R. Jones , Jonathan S. Mason , M. Alistair O’Brien , Gregory R. Ott ,
‖
‖
Mark Pickworth and Stacey M. Southall
‖Sosei Heptares, Steinmetz Building, Granta Park, Great Abington, Cambridge, CB21 6DG, UK
§Teva Pharmaceuticals, 145 Brandywine Parkway, West Chester, PA 19380, USA
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
Structure-based drug design enabled the discovery of 8, HTL22562, a CGRP receptor antagonist.
The structure of 8 complexed with the CGRP receptor was determined at 1.6 Å resolution.
Compound 8 is a highly potent, selective, metabolically stable and soluble compound suitable for
a range of administration routes that have the potential to provide rapid systemic exposures with
resultant high levels of receptor coverage (e.g. subcutaneous). The low lipophilicity coupled with
a low anticipated clinically efficacious plasma exposure for migraine also suggests a reduced
potential for hepatotoxicity. These properties have led to 8 being taken forward as a clinical
candidate for acute treatment of migraine.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Migraine is a disabling headache disorder characterized by moderate to severe pain, often
accompanied with sensory and other symptoms such as aura, photophobia, nausea and vomiting.1
The triptan class of drugs (5-HT1B/D agonists) represent an important class of abortive treatment
for migraine, reducing symptoms by activation of 5-HT1B receptors which leads to a range of
pharmacological effects including vasoconstriction of intracranial arteries. For this reason, triptans
are contraindicated in patients with hypertension or cardiovascular (CV) disease as coronary
arteries are also constricted by action of the drug.2,3
Calcitonin gene-related peptide (CGRP) is part of the calcitonin family of peptides that also
4
includes amylin and adrenomedullin. CGRP is a 37 amino acid neuropeptide whose levels are
5
elevated during migraine attacks. One potential function of CGRP in the symptoms of migraine
6
is its role as a potent vasodilator. The CGRP receptor is formed from a heterodimeric complex
between Class B G protein-coupled receptor (GPCR) calcitonin receptor-like receptor (CLR) and
ACS Paragon Plus Environment
2

Page 3 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
7
receptor activity-modifying protein 1 (RAMP1). In the activating peptide ligand binding mode
the N-terminus of the peptide is buried deep into the receptor with the C-terminus bound at the
8
interface between the extracellular domains (ECDs) of CLR and RAMP1. Superposition of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
CGRP-bound and small molecule antagonist bound structures of olcegepant (1) and telcagepant
(
2) (Chart 1) indicate that their effect is exerted by blocking key interactions of the CGRP peptide
with the receptor.
Proof of concept that CGRP receptor antagonism could reduce migraine symptoms without
inducing triptan-like cardiovascular side effects was first demonstrated by 1 via iv
1
0
11
administration. This was followed in the clinic by the orally acting 2. Subsequent work in this
area has produced a breadth of chemical series12,13 entering Phase II and III clinical trials. This
14
includes agents suitable for intranasal (IN) dosing such as vazegepant (3) (Chart 1) and second
1
5
16
17
and third generation orally dosed compounds MK-3207 (4) , rimegepant (5) , atogepant (6)
1
7
and ubrogepant (7) , the latter of which received FDA approval in late 2019 for the acute treatment
18
19
of migraine in adults followed by rimegepant in early 2020. Several monoclonal antibodies
(
mAbs) have also been developed20 with erenumab21 the first to receive approval for the
preventative treatment of migraine in adults. Of concern in the clinical studies of orally
administered CGRP antagonists has been the discontinuation of agents 2 and 4 due to the
observation of elevated liver enzyme levels on chronic dosing.22,23 For this reason, we sought to
develop a compound that was suitable for administration via a range of non-oral routes as this
would place a lower first-pass exposure burden on the liver thus reducing the chance of liver injury.
A non-oral route such as subcutaneous (sc) delivery also had the opportunity to provide a molecule
with a rapid onset of action and a high receptor coverage suitable for an acute migraine treatment
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
setting. This work has led to the novel CGRP receptor antagonist HTL22562 (8), the discovery of
which will be described below.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chart 1. CGRP receptor antagonists olcegepant 1, telcagepant 2, vazegepant 3, MK-3207 4,
rimegepant 5, atogepant 6, ubrogepant 7 and HTL22562 8.
ACS Paragon Plus Environment
4

Page 5 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
OH
H
N
Br
Br
O
O
NH
N
N
N
O
F
H
N
H
N
F3C
N
N
N
N
F
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
N
HN
O
N
1
2
O
NH2
HN N
H
N
O
O
O
HN
O
NH
H
N
N
N
N
N
H
O
N
O
3
4
N
F
F
N
O
NH
O
N
N
N
O
NH
N
O
H
N
N
F3C
N
N
H N
O
2
O
5
6
F
F
F
F
F
HN N
O
O
O
NH
N
O
NH
H
N
N
H
N
N
N
F3C
N
N
O
O
N
HN
O
N
8
7
O
N
H
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS AND DISCUSSION
Design and SAR. Making use of the previously published crystal structure of the CGRP receptor
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
ectodomain complex bound to 1 and also in-house X-ray crystal structures with 1 (Figure 1) and
other antagonist ligands bound to the receptor, we were able to take a structure-guided approach
2
4
to rapidly develop a novel compound series suitable for sc administration. Insights from the high-
resolution structures obtained pointed towards operating outside of oral drug space being
beneficial as the binding site is 18 Å wide with key hydrogen bonding interactions between ligand
and receptor needing to be fulfilled at either end. The design of larger, high molecular weight
compounds with a number of hydrogen bond donors would therefore be necessary, however the
properties of these molecules would mitigate against an oral agent on permeability grounds, whilst
facilitating the aqueous solubility required for sc delivery.
We sought a compound possessing picomolar affinity and potency (i.e. pK and pK >10) with
i
b
low in vivo clearance to maximize receptor coverage and also lower dose requirements. A
thermodynamic solubility of >1 mM was also desired for sc administration therefore incorporation
of a basic center was thought advantageous.
2
5
Compound 1, a high molecular weight, non-orally available molecule , was chosen as a start
point for design. Our in-house crystal structure of 1 in complex with the CGRP receptor (Figure
1
), although of higher resolution (1.7 Å vs 2.1 Å), shows no significant differences in the ligand
9
binding site to that previously reported (PDB 3N7S). Key hydrogen bonding interactions to the
receptor include a donor/acceptor interaction between the dihydroquinazolinone urea and Thr122
CLR, the terminal pyridine nitrogen and Asp94 CLR, the lysine moiety and Asp71 RAMP1 and a
water mediated hydrogen bond from the phenol at the back of the pocket to Arg38 CLR.
ACS Paragon Plus Environment
6

Page 7 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Crystal structure of compound 1 bound to the CGRP receptor. CLR and RAMP1 in
cartoon representation colored green and cyan respectively with selected residues in stick
representation with nitrogen and oxygen atoms in blue and red respectively. 1 in stick
representation with carbon, nitrogen, oxygen and bromine atoms colored magenta, blue, red and
brown respectively. Crystallographic water molecules represented as red spheres with selected
hydrogen bonds shown as yellow dashed lines.
Our structural studies showed that reported CGRP antagonists explore a number of vectors in
the receptor, and that the molecular shape of 1 allowed maximum exploration of the available
vectors that might be exploited to assist with the drive towards a compound with high affinity,
potency and solubility. In order to achieve the desired high affinity and potency we recognized the
need to interact with the whole of the 18 Å wide binding site and the necessity of maintaining the
large number of hydrogen bonds with the receptor as made by compound 1 (Figure 1). Compound
1 can be broken down into four fragments that correspond to the available vectors for investigation
(Table 1). Due to the pseudo-peptidic nature of the compound these fragments also broadly
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
correspond to the building blocks required for synthesis, which we recognized would allow for
rapid modular construction of molecules for evaluation. The opportunity for incorporation of
fragments from literature compounds known from our structural studies to be exploring a particular
vector existed, and additionally, structural insights could be used to design new fragments to probe
particular ligand-receptor interactions.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The 1-(pyridin-4-yl)piperazine (A) fragment of compound 1 contributes to the activity of the
molecule by the pyridine moiety forming a π-stacking interaction with Phe92 CLR and a hydrogen
bond with Asp94 CLR (Figure 1). We found little benefit in terms of increasing activity and
potency when altering this fragment found in 1 (data not shown) and so this (A) was fixed as
fragment 1 in subsequent SAR investigations (Table 1).
The lysine moiety present as fragment 2 in 1 forms a hydrogen bond with Asp71 RAMP1 (Figure
1
). We hypothesized that constraining the flexibility of the lysine chain could be beneficial in terms
of enhancing this interaction with the receptor, however the 3-(4-piperidinyl)-L-alanine (PipAla)
amino acid (C) afforded no significant change to affinity and potency. The rat in vivo clearance
was however reduced relative to 1 which was desirable for the compound profile which we sought
to achieve (compound 9) (Table 1).
We were also concerned about the effect that the 3,5-dibromo-D-tyrosine present as fragment 3
(E) in 1 would have on the aqueous solubility due to its lipophilicity. This was replaced with the
lower molecular weight and less lipophilic indazole amino acid (F) found in 3 as we had observed
from structural studies that this overlaid the desired vector and position in the receptor. Whilst the
tyrosine moiety forms a water mediated hydrogen bond to Arg38 CLR via a water that is also
within hydrogen bonding distance to the backbone carbonyl of Arg67 RAMP1 in the compound 1
bound crystal structure (Figure 1), we hypothesized that the indazole group would instead interact
ACS Paragon Plus Environment
8

Page 9 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with Arg38 CLR, either directly as a hydrogen bond acceptor or via a water mediated hydrogen
bonding interaction. As this would maintain a key hydrogen bond interaction with the receptor this
switch was thought likely to be well tolerated, and indeed this change in fragment 3 had the effect
of increasing activity and potency slightly and importantly trebling the aqueous solubility, getting
this property over the desired >1 mM threshold at both pHs tested (compounds 10 and 11) (Table
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
).
With fragment 3 fixed as F, we further investigated fragment 2, still in conjunction with the
2
6
CONH “privileged structure” (H) found in 2 as fragment 4. Molecular docking studies using the
compound 1 bound crystallographic structure suggested that incorporation of the aromatic 3-(4-
pyridyl)-L-alanine (L-4-Pal) (D) along this vector giving compound 13 (Table 1), may result in
additional -stacking interactions with Trp74 in the RAMP1 portion of the receptor (Figure 1) or
allow intramolecular hydrophobic collapse with the indazole fragment F. Despite the potential loss
of a hydrogen bonding interaction by using this group, it was hypothesized that the additional -
stacking interactions or shielding of a lipophilic group from solvent (in the case of hydrophobic
collapse) would compensate for this with the result being increased activity and potency.
Comparison of amino acids B, C and D showed that this did not appear to be the case (compounds
1
2, 11 and 13 respectively) (Table 1), with 13 having lower affinity and potency than 11 and 12.
Incorporation of fragment D also offered no advantages in terms of solubility or clearance
compared to fragment C when comparing compounds 13 and 11 and did not advance the
compound towards the desired profile.
With these parts of the molecule optimized using A, C and F as fragments 1, 2 and 3 we turned
to investigating fragment 4 using a number of “privileged structures” found in literature
compounds. These are all able to maintain the hydrogen bond donor/acceptor interaction with
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Thr122 CLR as seen with fragment G in compound 1 (Figure 1). Fragment H from 2 gave a
compound (11) (Table 1) with excellent affinity, potency and solubility, but was found to be
rapidly cleared in a rat iv PK study. Use of fragment I from 3 did not offer the desired levels of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
potency and solubility (pK <10 and solubility <1 mM) (compound 14) and whilst the “privileged
b
structure” G from 1 initially looked attractive in 15 with a low rat iv clearance of 6 mL/min/kg,
we found this fragment to have an oxidative liability (previously reported in the literature27,28) in
2
9
the context of our scaffold which was not desirable. Incorporation of spiro fragment J gave the
compound (8) with the best balance of affinity, potency, solubility and rat iv clearance out of those
studied and was taken forward into further studies including determination of the co-crystal
structure by X-ray crystallography.
Table 1. Pharmacological, solubility and PK profiles of compounds 1, 8 and 9 - 15.
ACS Paragon Plus Environment
1
0

Page 11 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
OH
H
Br
Br
O
N
N
H
N
N
4
N
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
N
HN
O
N
1
O
2
NH2
Fragment 1:
Fragment 2:
Fragment 3:
Fragment 4:
OH
H
N
HN
N
Br
Br
O
N
N
O
N
H
N
G
N
A
NH2
O
O
B
O
E
HN
O
NH
N
N
N
HN
N
H
N
H
C
D
H
N
H
N
O
HN
O
O
F
I
N
O
O
N
O
NH
N
N
J
Fragment
CGRP CGRP TD sol (M,
Rat iv PK Cl
pKi
pKb
pH 6, pH 7.4) (mL/min/kg)
1
2
3
4
38a
1
A
B
E
G
10.5
11.2
150, 431
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
NDd
21a
9
A
A
A
A
A
A
A
A
C
C
C
B
D
C
C
C
E
E
F
F
F
F
F
F
G
H
H
H
H
I
10.1
10.4
10.7
10.6
10.0
10.1
10.2
10.5
11.2
10.8
11.2
11.0
10.0
9.8
1
0
1
2
3
4
5
1029, 890
3387, 3222
NDd
28a
55b
NDd
87c
31a
6a
1
1
1
1
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1772, 230
10, 509
G
J
10.1
10.2
1171, 1205
2922, 2462
8
17a
a
b
iv dosing at 2 mg/kg using 10% DMAC + 10% Solutol HS15 + 80% saline as vehicle. iv dosing
at 0.5 mg/kg using 100% acidified saline (pH 6.6) as vehicle. iv dosing at 2 mg/kg using 100%
c
d
acidified saline (pH 4.1) as vehicle. Not determined.
Synthesis. Synthesis of 8 begins with the ester hydrolysis of the commercially available 16 and
coupling of the resultant acid with 1-(4-pyridyl)piperazine to give amide 17. Removal of the Cbz-
protecting group under flow hydrogenation conditions then gave amine intermediate 18. Synthesis
of the right-hand side of compound 8 utilized building block 19 for which we developed a more
30
robust large scale synthesis during the course of this work. This was coupled with 20 using DSC
giving a urea of which the ester functionality was then hydrolyzed to give acid 21. A HATU
mediated amide coupling of this with intermediate 18 united the two halves of the molecule, with
a Boc-deprotection giving compound 8 (Scheme 1).
Scheme 1. Synthesis of compound 8a
ACS Paragon Plus Environment
1
2

Page 13 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
N
N
O
N
HN
O
O
N
HN
O
O
N
NH2
O
N
N
O
a, b
O
c
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
N
O
O
O
O
1
6
17
18
HN N
O
O
NH
N
+
NH2
HCl
HN
2
O
O
19
20
HN N
d, e
O
O
NH
HN N
O
H
N
N
N
N
O
NH
f, g
O
N
HN
O
H
N
N
N
N
O
O
HO
O
21
8
N
H
a
Reagents and conditions: (a) LiOH, 1,4-dioxane/H O, rt; (b) 1-(4-pyridyl)piperazine, HBTU,
iPr EtN, DMF, rt; (c) H , Pd/C, EtOH, 50 °C; (d) DSC, NEt , DMF, rt; (e) 1M LiOH, THF/H O,
2
2
2
3
2
rt; (f) Intermediate 18, HATU, iPr EtN, DMF, rt; (g) TFA, DCM, rt.
2
Structure determination of the CGRP receptor in complex with 8. To determine the structure
of CGRP receptor with the antagonist 8, an expression construct containing the ectodomain
complex was generated, similar to that used in the determination of the structure of the CGRP
31
receptor bound to the peptide antagonist CGRPmut. An MBP-RAMP1 ECD-CLR ECD fusion
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
protein was generated in which the two ECDs were covalently tethered with a (Gly-Ser-Ala)3
linker. A CLR-RAMP1 ECD only complex has been shown to be a valid functional surrogate for
9
the full-length receptor complex in antagonist binding. Crystals of the protein bound to 8 were
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
grown in vapour diffusion experiments and diffraction data were collected to 1.6 Å resolution. The
structure was solved by molecular replacement (MR) using the separate MBP and receptor
3
1
ectodomain complex portions of PDB ID 4RWG as search models. Details of data collection and
refinement statistics are listed in Supporting Information (SI) Table 1. Strong and unambiguous
electron density was observed for 8 in sigma-A weighted 2m|Fo| − |DFc|, m|Fo| − D|Fc| maps
3
2
33
calculated using Phenix following molecular replacement using the program Phaser from the
3
4
CCP4 Suite facilitating model building and refinement using Phenix and CCP4 (Figure 2). In
general, the protein backbone of this structure is in close agreement with both the published and
our in-house olcegepant co-structure, with Cα rmsd values of 1.49 Å and 0.70 Å respectively
between the CLR regions of those and CLR in this structure, and Cα rsmd of 0.45 Å for the RAMP1
component in both cases.
ACS Paragon Plus Environment
1
4

Page 15 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Panel A: Crystal structure of compound 8 bound to the CGRP receptor. CLR and
RAMP1 in cartoon representation colored green and blue respectively with selected residues in
stick representation with nitrogen and oxygen atoms in blue and red respectively. 8 in stick
representation with carbon, nitrogen and oxygen atoms colored magenta, blue and red respectively.
Crystallographic water molecules represented as red spheres with selected hydrogen bonds shown
as yellow dashed lines. Panels B-E: Highlight key interactions of compound 8 in different areas of
the receptor. Colors as panel A. GRID maps are contoured and colored in the following manner:
Lipophilic - C1 probe in yellow transparent solid at –2.8 kcal/mol to highlight lipophilic hotspots
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
in the pocket; Shape - CH methyl group probe in grey mesh at 1 kcal/mol to define the pocket
3
surface in terms of how close a ligand carbon atom can reside. Panel D is a close-up of the indazole
cavity therefore part of the ligand and receptor is hidden for clarity. The figure panels B-E were
made in VIDA (OpenEye).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Despite binding mode similarity and very similar, rich hydrogen bond networking, the two
crystallographic structures show that compounds 1 and 8 induce different movements to optimize
the binding site around the ligand. (Figure 2, panel A and Figure 3). The druggability assessment
of the pseudo-apo binding site of 8 was conducted using the GRID molecular interaction field
(
MIF)35,36 (Figure 2, panels B – E). Compound 8 adopts an extended conformation in the ligand
binding site, with hydrogen bond interactions anchoring the spiro fragment to Thr122 CLR, the
indazole moiety to Asp71 RAMP1, and the pyridine ring of the 1-(pyridin-4-yl)piperazine group
to Asp94 CLR (Figure 2, panel A). The ligand binding site is wide and shallow, very exposed to
the solvent with water molecules contributing to secure the ligand to the binding site.
The cavity containing the indazole moiety of 8 is mainly formed by residues in alpha helices
from both RAMP1 and CLR, allowing simple adjustments in side chain orientations, rather than
of the backbone with its consequently energetically expensive movements of the alpha helix. To
the left of this pocket, the hydrogen bond anchoring to CLR is formed by residues on an
unstructured loop, making this region more adaptable than the other more structured regions. There
are conserved hydrogen bond interactions between the amino group of both the 3,5-dibromo-D-
tyrosine amino acid of compound 1 and the indazole amino acid moiety of 8 and the backbone
carbonyl group of Gly71 CLR. Although this residue is located on a loop, it is next to the bulky
Trp72 CLR that is embedded in interactions with a carbonyl group of the ligands and therefore
quite constrained. The superposition of the two structures (Figure 3) shows that the region
ACS Paragon Plus Environment
1
6

Page 17 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
adjusting to the slightly shorter “privileged structure” found in compound 8 verses 1 is the loop
region containing Thr122 CLR. The backbone carbonyl and amino groups of Thr122 CLR face
the ligand, offering the crucial hydrogen bond interactions to the hydrogen bond donor/acceptor
privileged structure of the spirocyclic group. The aromatic core of the spirocyclic fragment in 8 is
stabilized in a lipophilic sub-pocket (highlighted by the Lipophilic – C1 GRID map) which is an
aromatic cage formed by Trp121, Tyr124, Trp72 of the CLR domain (Figure 2, panel B). Notably,
the Trp121 CLR side chain folds on top of this group, rather than flipping outwards as in the
compound 1 bound structure (Figure 1).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Minor movements of the receptor side chains are visible around the common extended 1-
(pyridin-4-yl)piperazine group of compounds 1 and 8. The carbonyl group here is stabilized by
interactions with the water network, with water molecules bridging the carbonyl group of the
spirocyclic moiety and the residues at the top of the pocket, i.e. Arg119 CLR (Figure 2, panel B).
The 1-(pyridin-4-yl)piperazine group is extended toward the opening of the pocket, forming an
ionic interaction between the pyridine moiety and Asp94 CLR and closely following the surface
of the receptor with favorable interactions with the aromatic moiety of Phe92 CLR (Figure 2, panel
C).
A key finding of significant interest for the design of novel drug candidate ligands is that whilst
the secondary structure elements from RAMP1 and CLR forming the major part of the lipophilic
indazole containing cavity occupy the same overall position in the two ligand binding poses, the
plasticity of the cavity is conferred by ligand-induced, specific rotameric states of the side chains
of Asp71 RAMP1 and Arg38 CLR (Figure 3). This allows adaptation of the size, shape, lipophilic
and hydrogen bond capabilities of the pocket toward the different moieties of the ligands. The 3,5-
dibromo-D-tyrosine fragment of compound 1 induces an upward movement of Asp71 RAMP1 side
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
chain to avoid repulsion with the negatively charged oxygen with the consequent engagement of
Asp71 in the ionic interaction with the lysine moiety of compound 1. Water molecules are also
needed to fill the cavity and mediate the interactions with Arg38 CLR (Figure 1). On the contrary,
for compound 8, the Asp71 RAMP1 side chain adjusts to optimize the hydrogen bonding
interactions with the indazole moiety and define the lipophilic hotspot of the sub-pocket, again
highlighted by the GRID map (Figure 2, panel D) for the indazole methyl group. At the same time,
the Arg38 CLR side chain moves outward to optimize the water mediated network and stabilize
the new orientation of the Asp71 RAMP1 side chain and the indazole moiety (Figure 2, panel D).
The piperidine ring of the PipAla group in 8 folds toward the RAMP side of the binding site,
shielding the hydrophobic cavity containing the indazole moiety from the solvent. However,
whereas the lysine side chain of 1 forms an ionic interaction with Asp71 RAMP1, the piperidine
ring of 8 cannot reach Asp71 RAMP1, now engaged in a hydrogen bond interaction with the
indazole moiety (vide supra). Instead, the piperidine ring adopts a boat conformation in order to
better optimize the interactions between the positive charge and the solvent. In fact, a structured
water network stabilizes the PipAla side chain, with water molecules bridging between the charge
of the piperidine and Asp71 RAMP1 carboxyl group (Figure 2, panel E).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
1
8

Page 19 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Cartoon representation of the superposition of the crystal structures of compounds 1 and
8
bound to the CGRP receptor using main chain atoms. Interacting residues and the ligands are
shown in stick representation with nitrogen, oxygen and bromine atoms colored blue, red and dark
red with the carbon atoms of 1 and 8 colored pink and magenta respectively, CLR colored light
green or green and RAMP colored light cyan or blue. Water molecules are represented as red
spheres.
Pharmacological, in vitro ADME and pharmacokinetic properties. The key in vitro
properties of 8 are presented in Table 2. The compound shows high affinity and functional potency
in both the human and cynomolgus (cyno) monkey binding and functional assays respectively. As
expected, a lower functional potency was observed in dog and rat assays due to residue differences
37
in the non-primate RAMP1 portion of the receptor heterodimer (tryptophan 74 to lysine mutation
3
8
39
in rat ) which affects compound binding and therefore potency in non-primate species.
8
demonstrates functional selectivity over representative members of the calcitonin receptor family
showing >8000-fold selectivity over the human amylin-1 and adrenomedullin-1 receptors (which
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
are heterodimers of the calcitonin receptor and RAMP1, and CLR and RAMP2 respectively).
Broader selectivity was also demonstrated over the hERG ion channel and in the DiscoverX
gpcrMAX panel with >1000-fold selectivity in these assays thus demonstrating limited potential
for secondary pharmacology at the therapeutically anticipated human plasma exposures. Low
MDCK-WT permeability was observed with some efflux, although this was difficult to fully
quantify due to the low permeability. The unbound fraction in plasma was also high (at least 58%)
across species, which is consistent with the high polarity of the compound and the low measured
logD of -0.4. Solubility was extremely high (in the millimolar range) across the pH range tested,
although this determination was on amorphous, rather than crystalline, material as multiple
attempts at crystallization were not successful. Solution stability testing showed that 8 was more
susceptible to hydrolysis at high pH over low pH. In P450 inhibition assays 8 demonstrates no
significant inhibition against all isoforms tested and showed minimal potential for induction in
vivo of PXR and AhR. Hepatic stability was high across all species tested and no metabolites were
observed in vivo in rat and cyno. Compound 8 also showed no evidence of genotoxicity in a
miniAMES test.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Profile of 8, HTL22562.
Human CGRP pK 10.5, pK 10.2
Human amylin-1 pK <6.3
i
b
b
Cyno CGRP pK 11.3, pK 10.5
Human adrenomedullin-1 pK <5.7
i
b
b
Dog CGRP pK 7.2, Rat CGRP pK <7
b
b
hERG 6% maximum inhibition observed at DiscoverX panel (gpcrMAX, 168 targets): 2
3
00 µM
targets > 50% activity or inhibition at 10 µM;
CHRM2 (65% inh, antagonist mode);
MRGPX2 (80% activity, agonist mode) EC50
0.9 µM, > 1000x selectivity
ACS Paragon Plus Environment
2
0

Page 21 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
-6
-1
b
Permeability 0.8x10 cm s (efflux ratio >2.7) Cross-species PPB% 41, 23, 31, 36
LogD -0.4 (pH 7.4)
Solubility (mg/mL) 9 (pH 8.2), 115 (pH 5.7),
74 (pH 4.0)
1
Solution stability at 50 ºC (loss in purity after 1
week (%)) 4.5 (pH 8.0), 0.8 (pH 5.5), 0.4 (pH
3.0)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
In vitro hepatic stability t >300 min
CYP P450 inhibition:
1
/2
pIC 3A4/2C19/2D6/2C8/2C9/1A2 < 4.3
5
0
CYP induction: PXR 2.3-fold at 1 µM, AhR Genotoxicity: AMES negatived
.4-fold at 3 µM
1
a
b
MDCK-WT, Papp A-B. Binding to rat, dog, cyno and human plasma proteins respectively
measured using a high throughput dialysis method). Rat, dog, cyno and human hepatocytes.
Mini AMES TA98, TA100, TA1535, TA1537.
c
(
d
The overall in vitro profile of 8 was very promising, with the combination of high affinity,
functional potency and extremely high aqueous solubility desired for subcutaneous dosing. In
addition, the compound has a low logD of -0.4 at pH 7.4 which would also help to mitigate any
4
0
hepatotoxicity risks which have been linked to high lipophilicities in previous compounds. This
prompted us to take the compound forward to assess the iv and sc PK parameters in several species
(
Table 3).
Table 3. Intravenous and subcutaneous PK parameters of 8.
iv
sc
Dose
Cl
t1/2
Vss
Dose
t1/2
Tmax
AUCinf
F
(
mg/kg) (mL/min/kg) (h) (L/kg) (mg/kg) (h)
(h) (ng*h/mL) (%)
Rat
2a
17
10
4.7
1.2
1.0
0.3
1b
2.6
0.7
1285
ND
123
ND
Dog
0.5c
NA
ND
ND
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Cyno
0.5d
2
5.3
0.4
0.013e
.5f
3.5
3.6
0.8
0.8
81
68
0
4207
117
a
b
Dosed using 10% DMAC + 10% Solutol HS15 + 80% saline as vehicle. Dosed using 100%
c
d
acidified saline (pH 4.9) as vehicle. Dosed using 100% acidified saline (pH 4.0) as vehicle. Dosed
using 100% acidified saline (pH 3.2) as vehicle. Dosed using 100% acidified saline (pH 6.0) as
vehicle. Dosed using 100% acidified saline (pH 3.5) as vehicle.
e
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
f
Following iv administration, 8 demonstrated a moderate to low clearance across the species
4
1
studied (31, 32 and 5% of liver blood flow in rat, dog and cyno respectively ). Upon sc
administration a short Tmax (0.8 h) and complete bioavailability was observed in rat and cyno with
dose proportional exposure at the two doses (0.013 and 0.5 mg/kg) studied in the latter. This
suggested that sc administration could deliver the requisite systemic exposure.
One PK/PD model used in the literature to determine CGRP antagonist efficacy is a capsaicin-
induced dermal vasodilation (CIDV) model. Here, after administration of a CGRP antagonist, the
reversal of increased dermal blood flow induced by topical application of capsaicin to the forearm
4
2
42
37
is monitored by laser Doppler imaging. Previous PK/PD studies in the literature of 2 , 4 and
4
3
7 suggest that for efficacy to be observed in a CIDV model in monkey, unbound plasma
concentrations of the CGRP antagonist must be above the in vitro IC of that compound, i.e. the
9
0
in vitro IC90 is approximately equal to the EC determined in the CIDV study. A similar
50
relationship has been observed in human CIDV studies with the small molecules 2, 4 and 744,45,43
4
6
and also with the mAb erenumab. Clinical studies in patients suggest that approximately 10-fold
higher plasma concentrations than the EC observed in CIDV studies are needed for effective
5
0
migraine relief44–46 and therefore unbound plasma concentrations above the in vitro IC of the
9
9
CGRP antagonist are required for in vivo efficacy in human.
ACS Paragon Plus Environment
2
2

Page 23 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 4. Mean unbound plasma exposures of 8 over 24 hours in cyno monkey upon 0.5 mg/kg
iv and 0.013 and 0.5 mg/kg sc administration
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The mean unbound plasma exposures of 8 over time in cyno monkey following iv and sc
administration are shown in Figure 4 along with the values of IC (pK ), IC and IC in that
50
b
90
99
species (the latter extrapolated from the pK ). The high potency, low clearance and high free
b
fraction in plasma of 8 combine to give very high levels of free receptor coverage over the IC at
9
0
the doses studied subcutaneously (~40x at Cmax at 0.013 mg/kg and ~2000x at Cmax and ~400x at
8 h post dose at 0.5 mg/kg). This demonstrates that the required concentrations can be achieved by
sc dosing to fully explore the mechanism in a translatable PK/PD model such as monkey CIDV as
exposures above the IC90 are achieved. Consistent with our predictions, when dosed
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
subcutaneously at 0.001, 0.01 and 0.1 mg/kg 8 demonstrated significant effects at all doses in the
monkey CIDV model. These data will be published at a later date in support of future clinical
studies. As the human pK and PPB are comparable to that of cyno (10.2 vs 10.5 and 31 vs 36%
b
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
respectively) it can be extrapolated that this situation would be similar in man thus making 8 a
viable option in an acute migraine setting, giving rapid delivery of a high receptor coverage (i.e.
achieving plasma concentrations significantly above the IC for a portion of the dosing interval).
99
Low doses such as those studied here would also potentially reduce the risk of liver injury. In a
1
0-day repeat dose PK and tolerability study in rat (0.1 mg/kg/day and 1 mg/kg/day sc), 8 showed
dose proportional exposure and no behavioral or tolerability issues were observed.
CONCLUSION
In conclusion, discovery of HTL22562 (8) was enabled by a structure-based drug discovery
approach utilizing information from X-ray crystal structures of literature compounds bound to the
CGRP receptor to build a highly potent, selective, metabolically stable and soluble compound
suitable for sc dosing. The compound has good PK characteristics across species, with the desired
short Tmax, high free receptor coverage and good bioavailability following sc administration and a
clean profile in a 10-day repeat dose rat toxicology study. This, in combination with the potential
for a low, non-oral, dose and low lipophilicity, suggests that 8 has a low potential for hepatotoxicity
previously seen with some clinical CGRP antagonists in this area. These properties have led to 8
being selected as a clinical candidate to be taken forward for further development.
EXPERIMENTAL SECTION
ACS Paragon Plus Environment
2
4

Page 25 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
All animal experiments performed in the manuscript were conducted in compliance with
institutional guidelines.
Chemistry. Active compounds have been examined for known classes of assay interference
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
47
compounds and do not fall into this category. Compounds were assessed for purity by LC-MS
and were ≥95% except for compound 12 which was found to be 85%.
General. Where no preparative routes are included, the relevant intermediate is commercially
available. Commercial reagents were utilized without further purification. Room temperature (rt)
1
refers to approximately 20-27 °C. H NMR spectra were recorded at 400 MHz on Bruker, Varian
or JEOL instruments. Chemical shift values are expressed in parts per million (ppm), i.e. (δ)-
values. The following abbreviations are used for the multiplicity of the NMR signals: s=singlet,
br=broad, d=doublet, t=triplet, q=quartet, quin=quintet, h=heptet, dd=doublet of doublets,
dt=double of triplets, m=multiplet. Coupling constants are listed as J values, measured in Hz. NMR
and mass spectroscopy results were corrected to account for background peaks. Chromatography
refers to column chromatography performed using silica and executed under positive pressure
(flash chromatography) conditions. LC-MS experiments were carried out using electrospray
conditions under the conditions below. LC-MS data are given in the format: Mass ion, electrospray
mode (positive or negative), retention time (intermediates); Mass ion, electrospray mode (positive
or negative), retention time, approximate purity (final compounds). Method A. Instruments:
Hewlett Packard 1100 with G1315A DAD, Micromass ZQ; Column: Waters X-Bridge C-18, 2.5
micron, 2.1 x 20 mm or Phenomenex Gemini-NX C-18, 3 micron, 2.0 x 30mm; Gradient [time
(
min)/solvent D in C (%)]: 0.00/2, 0.10/2, 8.40/95, 10.00/95; Solvents: solvent C = 2.5 L H O +
2
2
.5 mL 28% aqueous ammonia solution; solvent D = 2.5 L MeCN + 135 mL H O + 2.5 mL 28%
2
aqueous ammonia solution; Injection volume 1 µL; UV detection 230 to 400 nM; column
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
°
temperature 45 C; Flow rate 1.5 mL/min. Method B. Instruments: Agilent Technologies 1260
Infinity LC with Chemstation software, DAD, Agilent 6120B Single Quadrupole MS with API-
ES Source; Column: Phenomenex Gemini-NX C-18, 3 micron, 2.0 x 30 mm; Gradient [time
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
min)/solvent D in C (%)]: 0.00/5, 2.00/95, 2.50/95, 2.60/5, 3.00/5; Solvents C and D are as
described above in Method A; Injection volume 0.5 µL; UV detection 190 to 400 nM; column
temperature 40 °C; Flow rate 1.5 mL/min. Method C. As detailed for method A, except with
Gradient [time (min)/solvent D in C (%)]: 0.00/2, 0.10/2, 2.50/95, 3.50/95. Method D. Instruments:
Acquity UPLC coupled with SQD mass spectrometer; Column: Acquity UPLC BEH C18, 1.7
micron, 2.1 x 50 mm; Gradient [time (min)/solvent B in A (%)]: 0.00/3, 1.50/100, 1.90/100, 2.00/3;
Solvents: solvent A = 10 mM aqueous solution of NH HCO (adjusted to pH 10 with ammonia);
4
3
solvent B = MeCN; Injection volume 1 µL; UV detection 210 to 350 nM; column temperature 40
C; Flow rate 0.9 mL/min. Preparative reverse phase HPLC conditions: Instruments: Gilson Semi
°
Preparative HPLC system- including Dual Piston Pumps 331 and 332, a 171 DAD and a GX-271
Liquid Handler; Column: Gemini-NX, 5 micron, C18, 100 x 30 mm; Solvents: A = H O with 0.2%
2
of 28% aqueous ammonia solution, B = MeCN; UV detection 205 nM; Flow rate 30 mL/min; see
individual experiments for gradient and run time used.
tert-Butyl 4-{(2S)-2-{[(benzyloxy)carbonyl]amino}-3-oxo-3-[4-(pyridin-4-yl)piperazin-1-
yl]propyl}piperidine-1-carboxylate (17). Step 1. To a solution of tert-butyl 4-[(2S)-2-
{
[(benzyloxy)carbonyl]amino}-3-methoxy-3-oxopropyl]piperidine-1-carboxylate (16) (4.29 g,
10.2 mmol) in 1,4-dioxane (50 mL) was added a solution of LiOH·H O (856 mg, 20.4 mmol) in
2
H O (50 mL) and the mixture was stirred for 5 h. The reaction mixture was concentrated in vacuo
2
to near-dryness and re-dissolved in H O before being acidified with 0.5 M NaHSO (aq). The
2
4
resulting thick white precipitate was extracted into EtOAc and the combined organic layers were
ACS Paragon Plus Environment
2
6

Page 27 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
washed with brine, dried (MgSO ) and concentrated in vacuo to yield N-[(benzyloxy)carbonyl]-3-
4
1
[
1-(tert-butoxycarbonyl)piperidin-4-yl]-L-alanine as a white foam (4.01 g, 9.87 mmol, 97%). H
NMR: (400 MHz, DMSO-d ) δ: 0.85-1.07 (m, 2H), 1.39 (s, 9H), 1.45-1.68 (m, 5H), 2.55-2.71 (m,
6
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
H), 3.86-3.94 (m, 1H), 4.85-4.98 (m, 2H), 5.04 (s, 2H), 7.28-7.39 (m, 5H), 7.53 (d, J = 8.3, 1H),
2.5 (br s, 1H). Step 2. To a solution of N-[(benzyloxy)carbonyl]-3-[1-(tert-
1
butoxycarbonyl)piperidin-4-yl]-L-alanine (4.00 g, 9.84 mmol) in DMF (80 mL) was added HBTU
(
4.10 g, 10.81 mmol) followed by iPr EtN (3.74 mL, 21.47 mmol) and 1-(4-pyridyl)piperazine
2
(1.69 g, 10.35 mmol) and the mixture was stirred at rt for 3 h. The reaction mixture was
concentrated in vacuo to near-dryness and the residue dissolved in EtOAc, washed twice with H O,
2
twice with NaHCO (aq) and brine, dried (Na SO ) and concentrated in vacuo. Purification by
3
2
4
flash chromatography, eluting with MeOH/DCM (5:95), yielded the title compound (4.51 g, 8.18
1
mmol, 83%) as an off-white foam. H NMR: (400 MHz, CDCl ) δ: 1.07-1.16 (br m, 2H), 1.44 (s,
3
9H), 1.54-1.60 (br m, 2H), 1.88-1.91 (m, 1H), 2.62-2.65 (br m, 2H), 3.47-3.65 (br m, 6H), 3.81-
3
.86 (br m, 2H), 4.05 (br m, 2H), 4.26 (br m, 2H), 4.70-4.74 (m, 1H), 5.04-5.12 (m, 2H), 5.70 (d,
J = 8.8, 1H), 6.82 (d, J = 6.0, 2H), 7.26-7.34 (m, 5H), 8.10 (br s, 2H).
tert-Butyl 4-{(2S)-2-amino-3-oxo-3-[4-(pyridin-4-yl)piperazin-1-yl]propyl}piperidine-1-
carboxylate (18). A solution of tert-butyl 4-{(2S)-2-{[(benzyloxy)carbonyl]amino}-3-oxo-3-[4-
(pyridin-4-yl)piperazin-1-yl]propyl}piperidine-1-carboxylate (17) (557 mg, 1.01 mmol) in EtOH
(20 mL) was eluted five times through a Pd/C cartridge at 50 ºC using a continuous flow
hydrogenation reactor (H-Cube, ThalesNano Nanotechnology Inc., Budapest, Hungary) in the
presence of H (full H mode), monitoring conversion to desired product by LC-MS. Upon > 95%
2
2
conversion the reaction mixture was concentrated in vacuo to yield the title compound (455 mg,
1.09 mmol, quant.) as a yellow glassy solid. LC-MS (Method A): m/z 318.3, 362.2, 418.3 (ES+),
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 55
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
at 2.70 min. H NMR: (400 MHz, DMSO-d ) δ: 0.89-1.09 (m, 2H), 1.38 (s, 9H), 1.43-1.69 (m,
6
5
H), 1.80 (d, J=10.9, 1H), 2.67 (br s, 2H), 3.42-3.79 (m, 7H), 3.84-4.00 (m, 2H), 4.11-4.25 (m,
1H), 4.34 (br s, 2H), 6.84 (d, J = 6.6, 2H), 8.19 (d, J = 6.2, 2H).
2R)-3-(7-Methyl-1H-indazol-5-yl)-2-{[(2'-oxo-1',2'-dihydro-1H-spiro[piperidine-4,4'-
pyrido[2,3-d][1,3]oxazin]-1-yl)carbonyl]amino}propanoic acid (21). Step 1. Et N (2.26 mL,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
3
1
6.3 mmol) was added to a solution of (R)-methyl 2-amino-3-(7-methyl-1H-indazol-5-yl)
propanoate dihydrochloride (20) (995 mg, 3.3 mmol) and DSC (917 mg, 3.6 mmol) in DMF (20
mL) and the mixture stirred at rt for 30 min. Spiro[piperidine-4,4'-[4H]pyrido[2,3-d][1,3]oxazin]-
2
'(1'H)-one (19) (785 mg, 3.6 mmol) was then added portionwise and the reaction mixture stirred
at rt for 18 h before concentration in vacuo. The residue was partitioned between H O and
2
MeOH/DCM (1:9), the phases were separated and the aqueous layer was washed with H O.
2
Residual solid from the separation step was dissolved in MeOH and the combined organic layers
were concentrated in vacuo and purified by flash chromatography, eluting with EtOAc in MeOH
(
20:1), to yield methyl (2R)-3-(7-methyl-1H-indazol-5-yl)-2-{[(2'-oxo-1',2'-dihydro-1H-
spiro[piperidine-4,4'-pyrido[2,3-d][1,3]oxazin]-1-yl)carbonyl]amino}propanoate (1.06 g, 2.22
1
mmol, 67%) as a white solid. LC-MS (Method A): m/z 479.3 (ES+), at 2.61 min, 100%. H NMR:
(
400 MHz, DMSO-d ) δ: 1.59-1.75 (m, 2H), 1.78-1.90 (m, 2H), 2.45 (s, 3H), 2.90-3.08 (m, 4H),
6
3
.59 (s, 3H), 3.86-3.96 (m, 2H), 4.28-4.38 (m, 1H), 6.94-7.06 (m, 3H), 7.32 (dd, J = 7.4, 1.2, 1H),
.39 (s, 1H), 7.95 (s, 1H), 8.18 (dd, J = 5.1, 1.6, 1H), 10.79 (s, 1H), 13.04 (s, 1H). Step 2. Methyl
7
(2R)-3-(7-methyl-1H-indazol-5-yl)-2-{[(2'-oxo-1',2'-dihydro-1H-spiro[piperidine-4,4'-
pyrido[2,3-d][1,3]oxazin]-1-yl)carbonyl]amino}propanoate (1.06 g, 2.22 mmol) was dissolved in
THF (15 mL) and MeOH (3 mL) and an aqueous solution of LiOH (1M, 4.4 mL, 4.4 mmol) was
added dropwise. After stirring at rt for 3.5 h further aqueous LiOH (1M, 2.2 mL, 2.2 mmol) was
ACS Paragon Plus Environment
2
8

Page 29 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
added dropwise and the mixture stirred for 1 h at rt before concentration under a stream of nitrogen.
The residue was dissolved in a minimum volume of H O and cooled to 0 ºC. Aqueous 1M HCl
2
was added dropwise to adjust the pH to ≤ 3 and the resulting precipitate was isolated by filtration,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
washed with cold H O and Et O to yield the title compound (877 mg, 1.89 mmol, 85%) as a pale
2
2
1
yellow solid. LC-MS (Method A): m/z 463.5 (ES-), 465.3 (ES+), at 0.10 min. H NMR (400 MHz,
DMSO-d ) δ: 1.53-1.91 (m, 4H), 2.44 (s, 3H), 2.89-3.14 (m, 5H), 3.89 (t, J = 11.5, 2H), 4.23 (br
6
s, 1H), 6.73 (d, J = 7.8, 1H), 6.93-7.06 (m, 2H), 7.31 (d, J = 7.4, 1H), 7.38 (s, 1H), 7.93 (s, 1H),
8
.17 (dd, J = 5.1, 1.2, 1H), 10.78 (s, 1H), 13.00 (br s, 1H).
N-((R)-3-(7-Methyl-1H-indazol-5-yl)-1-oxo-1-(((S)-1-oxo-3-(piperidin-4-yl)-1-(4-(pyridin-
4-yl)piperazin-1-yl)propan-2-yl)amino)propan-2-yl)-2'-oxo-1',2'-dihydrospiro[piperidine-
,4'-pyrido[2,3-d][1,3]oxazine]-1-carboxamide (8). Step 1. HATU (4.57 g, 12.0 mmol) was
4
added to a solution of (2R)-3-(7-methyl-1H-indazol-5-yl)-2-{[(2'-oxo-1',2'-dihydro-1H-
spiro[piperidine-4,4'-pyrido[2,3-d][1,3]oxazin]-1-yl)carbonyl]amino}propanoic acid (21) (4.65 g,
1
0.0 mmol) in DMF (150 mL), followed after 15 min by the addition of tert-butyl 4-{(2S)-2-
amino-3-oxo-3-[4-(pyridin-4-yl)piperazin-1-yl]propyl}piperidine-1-carboxylate (18) (4.60 g, 11.0
mmol) and iPr EtN (6.86 mL, 40.1 mmol). The mixture was stirred at rt for 17 h before the addition
2
of H O (600 mL). The resulting precipitate was isolated by filtration, washed with H O, and
2
2
dissolved in a small amount of MeOH. Co-evaporation twice with toluene yielded crude tert-butyl
-{(2S)-2-{[(2R)-3-(7-methyl-1H-indazol-5-yl)-2-{[(2'-oxo-1',2'-dihydro-1H-spiro[piperidine-
4
4,4'-pyrido[2,3-d][1,3]oxazin]-1-yl)carbonyl]amino}propanoyl]amino}-3-oxo-3-[4-(pyridin-4-
yl)piperazin-1-yl]propyl}piperidine-1-carboxylate (7.82 g) which was used in the next step
without further purification. A second batch of material (3.53 g) was prepared using this method.
+
1
LC-MS (Method D): m/z 864.7 (ES ) at 0.88 min. H NMR: (400 MHz, DMSO-d ) δ: 0.83-1.03
6
ACS Paragon Plus Environment
2
9