Discovery of Potent Cyclophilin Inhibitors Based on the Structural Simplification of Sanglifehrin A

Article abstract of DOI:10.1021/acs.jmedchem.6b01329

Cyclophilin inhibition has been a target for the treatment of hepatitis C and other diseases, but the generation of potent, drug-like molecules through chemical synthesis has been challenging. In this study, a set of macrocyclic cyclophilin inhibitors was synthesized based on the core structure of the natural product sanglifehrin A. Initial compound optimization identified the valine-m-tyrosine-piperazic acid tripeptide (Val-m-Tyr-Pip) in the sanglifehrin core, stereocenters at C14 and C15, and the hydroxyl group of the m-tyrosine (m-Tyr) residue as key contributors to compound potency. Replacing the C18-C21 diene unit of sanglifehrin with a styryl group led to potent compounds that displayed a novel binding mode in which the styrene moiety engaged in a π-stacking interaction with Arg55 of cyclophilin A (Cyp A), and the m-Tyr residue was displaced into solvent. This observation allowed further simplifications of the scaffold to generate new lead compounds in the search for orally bioavailable cyclophilin inhibitors.

Full text of DOI:10.1021/acs.jmedchem.6b01329

Subscriber access provided by Fudan University
Article
Discovery of Potent Cyclophilin Inhibitors based
on Structural Simplification of Sanglifehrin A
Victoria A. Steadman, Simon B. Pettit, Karine G. Poullennec, Linos Lazarides, Andrew J. Keats, David
K. Dean, Steven J Stanway, Carol A. Austin, Jonathan A. Sanvoisin, Gregory M. Watt, Hans G. Fliri,
Albert C. Liclican, Debi Jin, Melanie H. Wong, Stephanie A. Leavitt, Yu-Jen Lee, Yang Tian, Christian
R. Frey, Todd C. Appleby, Uli Schmitz, Petr Jansa, Richard L. Mackman, and Brian Edward Schultz
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01329 • Publication Date (Web): 11 Jan 2017
Downloaded from http://pubs.acs.org on January 11, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry 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 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Potent Cyclophilin Inhibitors based on
Structural Simplification of Sanglifehrin A
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
Victoria A. Steadman,† Simon B. Pettit,† Karine G. Poullennec,†,¶ Linos Lazarides,† Andrew J.
Keats,† David K. Dean,† Steven J. Stanway,† Carol A. Austin,† Jonathan A. Sanvoisin,†
Gregory M. Watt,†,# Hans G. Fliri,‡ Albert C. Liclican,§ Debi Jin,§ Melanie H. Wong,§
Stephanie A. Leavitt,§ YuꢀJen Lee,§ Yang Tian,§ Christian R. Frey,§ Todd C. Appleby,§ Uli
Schmitz,§ Petr Jansa,§ Richard L. Mackman,§ Brian E. Schultz§,*
†
Selcia Ltd, Fyfield Business & Research Park, Fyfield Road, Ongar, Essex, CM5 0GS, United
Kingdom
‡
Cypralis Ltd, Babraham Research Campus, Cambridge, CB22 3AT, United Kingdom
§Gilead Sciences, 333 Lakeside Drive, Foster City, CA 94404, USA
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 72
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: Cyclophilin inhibition has been a target for the treatment of hepatitis C and other
diseases, but the generation of potent, drugꢀlike molecules through chemical synthesis has been
challenging. In this study, a set of macrocyclic cyclophilin inhibitors was synthesized based on
the core structure of the natural product sanglifehrin A. Initial compound optimization identified
the valineꢀmꢀtyrosineꢀpiperazic acid tripeptide (ValꢀmꢀTyrꢀPip) in the sanglifehrin core,
stereocenters at C14 and C15, and the hydroxyl group of the mꢀtyrosine (mꢀTyr) residue as key
contributors to compound potency. Replacing the C18ꢀC21 diene unit of sanglifehrin with a
styryl group led to potent compounds that displayed a novel binding mode in which the styrene
moiety engaged in a πꢀstacking interaction with Arg55 of cyclophilin A (Cyp A), and the mꢀTyr
residue was displaced into solvent. This observation allowed further simplifications of the
scaffold, to generate new lead compounds in the search for orally bioavailable cyclophilin
inhibitors.
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
Cyclophilins are a family of proteins involved in a variety of biological processes. They
1
function as peptidylꢀprolyl isomerases but also act as mediators of the immunosuppressive
2
, 3
effects of the natural product cyclosporine A (CsA) (Figure 1),
and as chaperone proteins.
CsA, a macrocyclic oligopeptide, exerts its immunosuppressive effects through binding to the
active site of cyclophilins and simultaneously interacting with the protein calcineurin. (MeꢀIleꢀ4)
cyclosporine (NIM811), a cyclosporine isolated from Tolypocladium niveum, displayed strong
inhibition of cyclophilin but lacked the ability to bind calcineurin, and therefore, lacked
4, 5
immunosuppressive properties. This discovery prompted further studies aimed at interrogating
ACS Paragon Plus Environment
2

Page 3 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the biological role of cyclophilin inhibition without immunomodulation and connected the
function of some family members to disease states such as asthma, inflammation,
6
ꢀ8
ischemia/reperfusion, fibrosis, viral diseases, and cancer. More recently, the sanglifehrins,
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
natural products isolated from Actinomyces bacteria, have also been reported to bind to
9
ꢀ11
cyclophilins and demonstrate similar immunosuppressive properties to those of CsA.
In
contrast to the macrocyclic oligopeptide of CsA, the sanglifehrins are macrocycles containing a
valineꢀmetaꢀtyrosineꢀpiperazic acid (ValꢀmꢀTyrꢀPip) tripeptide motif as well as an extended
polyketide structure (Figure 1).
Beyond the early interest in using nonꢀimmunosuppressive cyclophilin inhibitors to target
1
2
human immunodeficiency virus (HIV), cyclosporine derivatives have played a prominent role
in the race for panꢀgenotypic hepatitis C virus (HCV) antivirals. Involvement of cyclophilins in
the life cycle of hepatitis C was first identified based on the efficacy of CsA against hepatitis C
1
3
virus in infected hepatocytes, and in HCV replicon assays. Knockdown experiments utilizing
RNA interference identified Cyp A as the key cyclophilin necessary for HCV replication, with
14, 15
no significant contribution from cyclophilins B and C.
A combination of viral resistance
selections and biophysical studies has identified the nonꢀstructural protein NS5A of HCV as the
1
6ꢀ19
principal viral protein to which Cyp A binds,
but the exact mechanism by which this
interaction facilitates viral replication remains unclear. Cyp A is a desirable target for HCV
treatment on account of the efficacy of cyclophilin inhibitors against all genotypes of hepatitis
20ꢀ22
C,
and the probable high barrier to the development of viral resistance. Importantly, clinical
proof of concept studies with the nonꢀimmunosuppressive CsA analog alisporivir (Figure 1)
demonstrated significant antiꢀHCV activity before its development was put on hold due to
23
toxicity observed when dosed in combination with ribavirin and pegylated interferon. On the
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 72
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
24ꢀ26
heels of this molecule, the CsA analog SCYꢀ635
(Figure 1) also progressed to early clinical
studies for the treatment of HCV. Not surprisingly, the sanglifehrins have also been explored as
potential antiꢀHCV agents. Degradation studies on sanglifehrin A determined that removal of the
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
27
C24 polyketide side chain eliminated the immunosuppressive properties of the molecule while
15
retaining the nanomolar binding affinity to cyclophilin A, B, and C. Moreover, a degradation
product from these studies was subjected to further chemical synthesis to yield AAE931 (1,
Figure 1), a compound that showed potent antiꢀHCV activity in the replicon assay (EC = 130
5
0
1
5, 28
nM).
Subsequently, researchers at Biotica have bioengineered the polyketide pathways to
provide a variety of sanglifehrin analogs such as NVP018 (Figure 1) as potential oral cyclophilin
2
9ꢀ32
inhibitors for the treatment of HCV and other diseases.
Our longꢀterm endeavor was to develop a small molecule cyclophilin inhibitor derived solely
through chemical synthesis that would possess excellent oral pharmacokinetics with no drugꢀ
drug interactions. Such an inhibitor would then be amenable to coꢀformulation with other direct
3
3
acting oral HCV antivirals such as sofosbuvir to generate a pan genotypic fixed dose oral
regimen for the treatment of HCV. To date, drug discovery programs based on classical smallꢀ
molecule library screening and lead optimization have not been reported to result in lateꢀstage
cyclophilin inhibitors with optimized oral pharmacokinetic properties. We conducted a pilot
screen on ~19,000 library compounds and did not discover compounds that were promising Cyp
A inhibitors in a timeꢀresolved fluorescence resonance energy transfer (TRꢀFRET)ꢀbased binding
assay. Therefore, our attention turned to a strategy of simplification of the sanglifehrinꢀderived
molecule 1 that demonstrated potent cyclophilin inhibition and antiꢀHCV activity. The total
3
4ꢀ36
synthesis of sanglifehrin A was reported by two groups
and provided the groundwork for this
effort. However, a projected total synthesis of 1 would still require almost 40 synthetic steps!
ACS Paragon Plus Environment
4

Page 5 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Sedrani et al. synthesized a series of sanglifehrin analogs with reduced complexity via ringꢀ
2
7, 37, 38
closing metathesis.
Multiple chemical changes were made in each of the reported analogs
compared to 1 and the result was an approximate 1000ꢀfold drop in potency. Despite this result,
we reasoned based on structural analysis of both the sanglifehrin A complex and compound 2, a
close analog of 1, that the generation of potent simplified sanglifehrins should be feasible (Figure
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
2). A key part of the strategy which differed from the previous report was retaining the Valꢀmꢀ
TyrꢀPip tripeptide that forms several hydrogen bonding interactions within the active site.
Iterative simplifications of 1, guided by a combination of biological, biophysical, and
crystallographic data, led to the generation of compounds 3ꢀ9. Compounds 8 and 9 are
surprisingly simple, low molecular weight, synthetically tractable molecules that have
demonstrated potent inhibition of cyclophilin A and antiviral activity in the HCV replicon
assays.
RESULTS AND DISCUSSION
2
7
The crystal structure of compound 2 bound to cyclophilin A showed a binding geometry
nearly identical to that of the macrocyclic core of sanglifehrin A (Figure 3). Both of the piperazic
acid and mꢀTyr amino acids make hydrogen bonds with the cyclophilin protein, suggesting that
both are likely critical for achieving potency. In contrast, the C23 and C14 side chains do not
form hydrogen bonds suggesting they may be dispensable. Therefore, the first simplified target
to validate this structural hypothesis was compound 3, containing the diene at C18ꢀC21 and the
four chiral centers found at C14ꢀC17 in sanglifehrin (Scheme 1). Because the two hydroxyls at
C15 and C17 of 2 appeared to not make direct interactions with the protein, we chose to cap
them with methyl groups to simplify the protecting group strategy. Our synthetic approach to
27
compound 3 centered on RCM at C20ꢀC21 unlike the simplified analogs reported earlier
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 72
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
(
Scheme 1). The synthesis of 3 began with known diol 10, which was prepared using Paterson’s
3
9
methodology in 5 steps in high yield. Methylation of 10 followed by debenzylation and
reprotection of the resulting alcohol provided the pivaloyl ester 11. Ozonolysis of the alkene and
condensation of the resulting aldehyde with diethyl allyl phosphonate afforded the terminal diene
12. Hydrolysis of the ester followed by a 2 step oxidation furnished the carboxylic acid 13.
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
7
Peptide 15 was prepared from the ValꢀmꢀTyrꢀPip tripeptide 14 by hydrolysis of the TCE ester
and reꢀesterification with homoallyl alcohol via an acid anhydride. Deprotection of the Boc
group followed by an HATU mediated coupling with acid 13 then afforded the RCM precursor
1
6. Ring closing metathesis using Grubbs I followed by TBAF mediated silyl group removal
provided 3.
The removal of the C16 and C17 stereocenters for target compounds 4 and 6ꢀ7 shortened the
synthesis of this region of the macrocycle to only a few steps (Scheme 2). Silyl enol ether
formation on Oppolzers sultam 17 was followed by TiCl mediated antiꢀaldol reaction with 4ꢀ
4
pentenal to provide alcohol 18 with complete stereocontrol. Methylation of the alcohol and
hydrolysis of the chiral auxiliary afforded the desired acid 19. In contrast to the route described
for compound 3, we preferred the C18ꢀC19 ring closure for these targets since this enabled the
evaluation of different diene replacements e.g. aryl alkenes 6ꢀ7. With 19 in hand, the RCM
precursors 25ꢀ27 were readily generated from the tripeptide TCE ester 14. First, removal of the
TCE group was followed by coupling with (E)ꢀhexaꢀ3,5ꢀdienꢀ1ꢀol to provide 22, while EDC
mediated coupling with benzyl alcohol 21 provided 24. Intermediate 23 was prepared directly
from the TCE ester 14 by base mediated transesterification with 3ꢀvinyl benzyl alcohol. Boc
removal on intermediates 22ꢀ24 and either HATU or EDC mediated coupling with carboxylic
acid 19, generated the desired precursors 25ꢀ27 for the RCM. Grubbs I catalyst mediated closure
ACS Paragon Plus Environment
6

Page 7 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of the ring followed by silyl deprotection resulted in targets 4, 6, and 7 respectively, in moderate
to good yield without significant optimization.
Compound 5 lacking all C14ꢀC17 functionality was first described by Novartis and was
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
27
prepared following the literature procedure (compound 47d ).
Lastly, compounds 8 and 9 with an alanine in place of mꢀTyr (Scheme 3) required a different
sequence of steps but ultimately converged to the same C18ꢀC19 RCM reaction. Coupling of
3
7
piperazic TCE ester 28 with Boc protected alanine, followed by Boc deprotection and
subsequent coupling with Bocꢀvaline afforded the protected tripeptide 29 in good yield.
Deprotection of the terminal Boc group and HATU coupling with acid 19 or heptꢀ6ꢀenoic acid
gave intermediates 30 and 31 respectively. Base hydrolysis of the TCE ester and subsequent
EDC coupling with chiral alcohol 21 furnished the RCM precursors 32 and 33. In this example,
HoveydaꢀGrubbs II catalyst proved higher yielding than Grubbs I to give the targets 8 and 9 in
28% and 22% yield respectively.
The binding affinity of compounds 3ꢀ9 for cyclophilin A was measured using a competitive
binding TRꢀFRET assay, and was corroborated with a functional peptidylꢀprolyl isomerase
(PPIase) assay. Antiviral activity was measured using a hepatitis C genotype 1b replicon assay in
Huh7 cells (Table 1). For all compounds, antiviral activity generally paralleled compound
binding affinity to Cyp A.
Compound 3 showed strong binding to Cyp A (K = 25 nM) and submicromolar antiꢀviral
d
activity in the HCV replicon cellular assay (EC50 = 600 nM). Capping the C15 and C17
hydroxyls was therefore tolerated, and the truncation of the side chains at C14 and C23 had a
minimal effect on biochemical potency (the reported K values for Cyp A binding for
d
27
sanglifehrin A, 1, and 2 are singleꢀdigit nanomolar ). The activity of the additional diene
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 72
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
compounds 4 and 5 demonstrated the effect of sequential removal of the stereocenters in the
C14ꢀC17 region. Compound 4, with only the C16 and C17 substituents removed, showed
potency nearly equal to that of compound 3 in the cyclophilin binding assay (K = 65 nM), with
d
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
slightly better antiꢀviral activity (EC = 240 nM). This result suggested that the chiral centers at
5
0
C16 and C17 in compound 3 are not essential for activity, an observation that enabled us to
reduce the complexity of subsequent target molecules. Compound 5, which lacks all four C14ꢀ
C17 stereocenters present in compound 3, showed a dramatic decrease in potency, yielding a Kd
value of 2.6 ꢀM against Cyp A. Compound 5 was previously reported by Novartis to have
37
singleꢀdigit micromolar activity consistent with our findings. The low biochemical potency also
translated to a low cellular potency (EC50 = 4.5 ꢀM). The retained potency of compound 4
bearing only 2 stereocenters at C14 and C15 was surprising when considering the set of very
similar macrocycles reported earlier that showed micromolar binding affinities when the three
2
7
C14ꢀC16 stereocenters were retained. However, compared to those synthetic macrocycles,
compound 4 contained the mꢀTyr residue, suggesting that the mꢀTyr residue and the C14ꢀC15
stereocenters are sufficient to obtain good binding potency and submicromolar replicon activity
in the diene containing compounds.
To elucidate the structural basis for the effects rendered by the C14ꢀC17 sidechains, we
determined Xꢀray crystal structures for compounds 3 and 5 bound to cyclophilin A. Not
surprisingly, the more potent compound 3 exhibits a binding mode almost identical to that
observed for sanglifehrin A and 2. The RMSD for all shared heavy atoms vs sanglifehrin A is
0.44 Å and 0.56 Å vs 2, respectively (Figure 4). This suggests that more elaborate C14 and C23
sidechains as well as the methylated C15 and C17 hydroxyls have little influence on the
2
7
macrocycle’s conformation itself. As described previously, the ValꢀmꢀTyrꢀPip tripeptide
ACS Paragon Plus Environment
8

Page 9 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
moiety is the most buried portion of the inhibitor and is responsible for making all of the
hydrogen bonding interactions observed between the ligand and enzyme based on heavy atom
distances (2.7ꢀ3.0 Å). The four carbon atoms of the piperazic acid bind in a deep hydrophobic
pocket formed by Phe 60, Met 61, Leu 122, and Phe113. The backbone carbonyl oxygen of the
Val residue forms a hydrogen bond with the side chain nitrogen (NE2) of Gln 63. A pair of
hydrogen bonds is observed between backbone atoms of the tripeptide mꢀTyr and Asn 102 of the
enzyme. The side chain carbonyl oxygen (OE2) of the Gln63 also participates in hydrogen
bonding with the αꢀnitrogen of the piperazic acid. The most solventꢀexposed hydrogen bond is
formed by the hydroxyl oxygen of the mꢀTyr and the His 126 (NE) side chain atoms. In addition
to these direct proteinꢀligand interactions, there is an additional solventꢀmediated hydrogen bond
between the Val carbonyl oxygen of the tripeptide and a structurally conserved water molecule
which is in turn coordinated by three different protein residues (His 54, Arg 55, and Gly 72). For
the 40ꢀfold less potent compound 5 that lacks the C14ꢀC17 stereocenters, the binding mode of
the ValꢀmꢀTyrꢀPip tripeptide portion is very well conserved, indicated by a heavy atom RMSD
of 0.27 Å to sanglifehrin A. However, the conformation of the nonꢀpeptidic C14ꢀC23 moiety is
different, with an RMSD of 1.05 Å to sanglifehrin A, compared to an RMSD of only 0.44Å for
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
2
compound 2. (This difference results in a loss of ~40Å of total buried surface area compared to
compound 2).
With respect to the cyclophilin structure itself for the crystal structures of compounds 3 and 5,
there are no remarkable differences when compared to the complex of 2 (heavy atom RMSD for
protein residues within 5 Å of ligands is 0.33 and 0.36 Å, respectively) (Figure 4).
One molecular feature that sets compound 3, along with sanglifehrin A and 2, apart from
compound 5 is the observation of the intramolecular hydrogen bond between the amide nitrogen
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 72
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
of the valine residue and the oxygen at C15 (Figure 5). Analysis of in silico conformational
models for compounds 3 and 5 showed that the intramolecular hydrogen bond is maintained in
43% of low energy conformers (<10 kcal) while the corresponding backbone conformation for
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
compound 5 is seen in less than 5% of conformers (Supporting Information Figure S1).
With this potential conformational bias for compound 3, we probed the thermodynamic basis
for the trend in potency using isothermal calorimetry (ITC) and variableꢀtemperature surface
plasmon resonance (SPR). Results are shown in Table 2. The binding of compound 3 was driven
predominantly by enthalpy, with a contribution of ꢀ10.0 kcal/mol from enthalpy and ꢀ0.3
kcal/mol from entropy, based on the ITC data. The SPR data showed a similar pattern. For
compound 4, a ꢁH value of ꢀ8.6 kcal/mol was measured by ITC, with the ꢀ(TꢁS) contribution of
ꢀ0.9 kcal/mol. For compound 5, the decrease in biochemical potency was driven by ꢁH, with a
value of ꢀ7.0 kcal/mol from the ITC experiments. The ꢀ(TꢁS) component of binding was ꢀ1.2
kcal/mol. The thermodynamic values from the SPR experiments for compounds 3ꢀ5 were
consistent with those from the ITC measurements. That the differences in ꢁG for binding of
compounds 3ꢀ5 to Cyp A primarily reflected changes in enthalpy might seem surprising given
that conformational bias is typically related to an entropic contribution. However, an increased
enthalpic contribution for compound 3 is most likely due to transferring nonꢀhydrogen bonded
solution conformers into the hydrogenꢀbonded, bound conformation.
We then explored further rigidification of the nonꢀpeptidic portion by modifying the diene
moiety. Modeling studies suggested that a metaꢀsubstituted styryl group could be introduced into
the C18ꢀC22 region, leading to compound 6 as a target. Indeed compound 6 was found to
maintain high potency in the biochemical and cellular assays (K = 64 nM; EC =320 nM).
d
50
ACS Paragon Plus Environment
10

Page 11 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The coꢀcrystal structure of compound 6 with cyclophilin A shows some significant changes in
binding mode driven by the presence of the styrene unit (Figure 6). Besides a significant mꢀTyr
side chain rotation of approximately 100º (about χ ) that eliminated the hydrogen bond with His
1
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
1
26, the styrene moiety makes a different interaction with Arg 55. The overlay with the
compound 3 coꢀcrystal structure in Figure 5 shows that the edgeꢀon hydrophobic interaction of
the diene moiety has been replaced with a tight π–stacking interaction of the styrene with the π–
face of the guanidiniumꢀgroup of Arg 55. In light of the very similar potencies of compounds 3,
4, and 6, it seems that the contribution of the mꢀTyr side chain interaction with His 126 can be
compensated by this new interaction between the styrene group and Arg 55. The position of the
mꢀTyr side chain seems to represent a hydrophobic collapse of the ligand itself, defined by
intramolecular hydrophobic packing of mꢀTyr phenyl moiety against the edge of the styrene
group and the apolar side of the C13ꢀcarbonyl. This binding mode represents maximal
hydrophobic surface burial for the ligand, leaving the mꢀTyr hydroxyl available to interact with
bulk solvent.
For further compound optimization, we chose to introduce a C23ꢀmethyl substituent as a
simplified version of the side chains present in sanglifehrin and 1, synthesizing compound 7.
Compound 7 showed improved biochemical potency (K = 11 nM) as well as greater cellular
d
potency (EC = 36 nM) than compound 6. The crystal structures for compounds 6 and 7 did not
5
0
offer a clear explanation for this potency improvement, showing practically identical binding
modes (Figure 7A), including the side chain conformation for the mꢀTyr. It appears that the C23
methyl group engages in a hydrophobic interaction with the side chains of Ile 57 and Phe 60,
which may contribute to the increase in potency. The C23 methyl group was incorporated into all
subsequent analogs due to its improved potency. It is noteworthy that all of the reported potent
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 72
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
10, 11, 15, 27, 29ꢀ31, 40
sanglifehrins and sanglifehrin derivatives
possess the unusual amino acid mꢀTyr,
suggesting that this moiety is necessary for potent binding. However, the Xꢀray structure of
compounds 6 and 7 strongly suggested that the mꢀTyr side chain may be dispensable for binding
when the styryl group is present, prompting the synthesis of compound 8 containing an alanine
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
residue instead of mꢀTyr. Compound 8 had a remarkable biochemical K value of 24 nM and a
d
cellular EC value of 87 nM, confirming that the mꢀTyr sidechain was dispensable in the context
5
0
of the novel styrene binding mode. The importance of this switch, beyond providing a simpler,
more efficient synthesis using a natural amino acid, is the removal of possible metabolic
liabilities of the mꢀTyr group such as glucuronidation and aromatic oxidation.
The coꢀcrystal structure of compound 8 bound to cyclophilin A is shown in Figure 7B. The
ligand binding mode of compound 8 is remarkably similar to that of compounds 6 and 7.
However, the absence of the mꢀTyr sidechain leads to a significant change in the C16ꢀC19 linker
(
aromatic) region. The styrene group is now flipped down toward the bottom of the pocket,
making a tighter hydrophobic interaction with the πꢀface of the Arg55 guanidinium group.
Notably, the C23ꢀmethyl group has moved slightly away from the Ile57/Phe60 hydrophobic
shelf. The coꢀcrystal structure of compound 7 shows that the hydrophobic interactions of the
nonꢀpeptidic linker moiety are essential for binding to the cyclophilin A pocket with Arg55
emerging as a key residue.
To verify that this simplified scaffold lacked immunosuppressive properties, compounds 4 and
8
were chosen for testing in a fourꢀday FACSꢀbased T cell proliferation assay. In this assay, the
control compound CsA inhibited proliferation with an EC value of 6.4 ± 2.5 nM, consistent
50
1
1
with reported literature values using a radiometric T cell proliferation assay. However,
ACS Paragon Plus Environment
12

Page 13 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds 4 and 8 yielded EC50 values >10 ꢂM, suggesting that these compounds are not
immunosuppressive.
The final key question remaining was whether the stereocenters at C14 and C15 were also
essential for potency in the styrene series. Compound 9 was prepared and showed only a 2ꢀfold
drop in the binding affinity compared to 8 and equivalent functional activity. Thus by iterative
structure based SAR the synthetic complexity and challenges associated with compound 3
containing the diene, C14ꢀC17 stereocenters, and the mꢀTyr were all removed in arriving at
compound 9. These realizations and the new crystallographic binding mode opened up many
more design possibilities to optimize the absorption and metabolic properties of these simplified
sanglifehrins to afford good oral bioavailability. In addition, the πꢀstacking interactions between
the styrene residues and Arg55 reveal an additional avenue for compound optimization.
CONCLUSION
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
The potential utility of nonꢀimmunosuppressive cyclophilin inhibitors for the treatment of a
variety of diseases has been recognized for over twenty years, but to date it has been very
challenging to generate potent, drugꢀlike cyclophilin inhibitors through chemical synthesis.
Utilizing the sanglifehrin natural product, we have applied focused medicinal chemistry
informed by critical insights gained from crystallography and biophysical studies to address this
problem. The synthesis and systematic analysis of compounds modifying the macrocyclic core of
sanglifehrin has resulted in the first potent and synthetically tractable inhibitors of cyclophilins.
For macrocycles retaining the the diene unit of sanglifehrin, we were able to reduce the
complexity of the molecules and determine that the presence of the mꢀTyr along with the C14
and C15 stereocenters was sufficient to maintain potency. The dramatically different binding
mode observed for macrocycles containing the styrene unit was a further breakthrough in
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 72
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
generating simple, potent cyclophilin inhibitors. This new binding mode, based on the πꢀstacking
of the aromatic unit at C20ꢀ22 with R55 of cyclophilin A, enabled a second generation of
analogs in which the mꢀTyr was replaced with alanine and all of the C14 to C17 stereocenters
were removed without loss of potency. These findings have provided new, more synthetically
tractable leads for the design of potent, orally bioavailable smallꢀmolecule cyclophilin inhibitors
that lack immunosuppressive properties. Indeed, more than 400 synthetic macrocyclic
cyclophilin inhibitors that were potent in biochemical and antiꢀviral assays were generated as a
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
4
1ꢀ44
result of this discovery.
With these compounds, the biology of cyclophilin inhibition as it
relates to disease states may now be interrogated more effectively.
EXPERIMENTAL SECTION
ACS Paragon Plus Environment
14

Page 15 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Proton Magnetic Resonance (NMR) spectra were recorded on a Bruker instrument at 300, 400 or
500 MHz. Chemical shifts are reported in ppm (δ) using the residual solvent line as internal
standard. Splitting patterns are designed as s, singlet; d, doublet; t, triplet; q, quartet; m,
multiplet; b, broad.
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
LC/MS Data were generated on a Waters ZQ Mass Spectrometer, operating in switched ES+ and
ESꢀ ionization modes coupled to an Agilent 1100 Series HPLC system with in line Aglient 1100
UVꢀDAD and Sedere SEDEX 75 ELSD Detection. Instrument control and data acquisition is
mediated through the Waters MassLynx – OpenLynx software suite. Retention times are denoted
as Tr.
All isolated characterized compounds were determined to have a purity of >95% by LCꢀMS and
NMR.
27
39
27
37
Compounds 5, , 10, 14, and 28 were prepared as described in the literature. Compounds 17
and 20 were purchased from commercial sources.
(13E,15E)ꢀ(3S,6S,9R,10R,11S,12S,21S)ꢀ3ꢀ(3ꢀHydroxyꢀbenzyl)ꢀ6ꢀisopropylꢀ10,12ꢀdimethoxyꢀ
9,11ꢀdimethylꢀ19ꢀoxaꢀ1,4,7,25ꢀtetraazaꢀbicyclo[19.3.1]pentacosaꢀ13,15ꢀdieneꢀ2,5,8,20ꢀ
tetraone (3)
A solution of 16 (228.7 mg, 0.291 mmol) in dry dichloromethane (150 mL) was treated with
Grubbs’ first generation catalyst (47.9 mg, 0.058 mmol). After stirring at reflux for 18 h, more
Grubbs’ first generation catalyst was added (47.9 mg, 0.058 mmol). After stirring for 2 h at
reflux, the reaction was cooled to room temperature and polymerꢀsupported methylthiourea was
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 72
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
added (1.5 g). After stirring for 24 h at room temperature, the mixture was filtered through Celite
and the volatiles were removed in vacuo. The residue was subjected to silica gel chromatography
using a 25 g Isolute cartridge eluted with a continuous gradient of isoꢀhexane/ethyl acetate 1:0 to
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
1:4 to afford the crude macrocycle (71.8 mg) which was dissolved in anhydrous tetrahydrofuran
(8 mL) and treated with a solution of tetrabutylammonium fluoride (1 M in tetrahydrofuran, 0.5
mL, 0.5 mmol). After stirring at room temperature for 40 min more tetrabutylammonium fluoride
was added (1 M in tetrahydrofuran, 0.5 mL, 0.5 mmol). After stirring at room temperature for 18
h, the volatiles were removed in vacuo and the residue was purified by silica gel chromatography
using a 10 g Isolute cartridge eluted with a continuous gradient of isoꢀhexane/ethyl acetate 1:0 to
1
0
:1 to afford the title compound (26.3 mg, 14% over 2 steps) as a yellow solid. H NMR (300
MHz, d ꢀDMSO) δ 9.14 (s, 1H), 8.04 (d, J = 7.3 Hz, 1H), 7.06–6.95 (m, 2H), 6.66–6.48 (m, 4H),
6
6
.19–6.04 (m, 2H), 5.56–5.27 (m, 3H), 4.84 (d, J = 11.3 Hz, 1H), 4.29–4.19 (m, 1H), 4.18–3.99
m, 3H), 3.18 (s, 3H), 3.04 (s, 3H), 2.81–2.61 (m, 3H), 2.45–2.37 (m, 1H), 2.35–2.25 (m, 1H),
.94–1.74 (m, 2H), 1.73–1.55 (m, 3H), 1.53–1.39 (m, 1H), 1.38–1.26 (m, 1H), 1.13 (d, J = 7.7
(
1
Hz, 3H), 0.86 (d, J = 6.8 Hz, 3H), 0.80 (d, J = 6.8 Hz, 3H), 0.67 (d, J = 6.8 Hz, 3H). LC/MS
(m/z) 665.4 [M+Na], Tr = 4.78 min.
(13E,15E)ꢀ(3S,6S,9R,10R,21S)ꢀ3ꢀ(3ꢀHydroxyꢀbenzyl)ꢀ6ꢀisopropylꢀ10ꢀmethoxyꢀ9ꢀmethylꢀ19ꢀ
oxaꢀ1,4,7,25ꢀtetraazaꢀbicyclo[19.3.1] pentacosaꢀ13,15ꢀdieneꢀ2,5,8,20ꢀtetraone (4)
st
A solution of 25 (133 mg, 0.180 mmol) in dichloromethane (85 mL) was prepared and Grubbs 1
generation catalyst (45 mg, 0.054 mmol) was added. The stirred reaction mixture was heated at
reflux for 36 h then cooled to room temperature. Silica gel was added and the mixture was
evaporated to dryness. The residue was purified by silica gel chromatography eluting with a
ACS Paragon Plus Environment
16

Page 17 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
stepwise gradient of isoꢀhexane/ethyl acetate 1:0 to 0:1 to yield (13E,15E)ꢀ(3S,6S,9R,10R,21S)ꢀ
3ꢀ[3ꢀ(tertꢀbutylꢀdimethylꢀsilanyloxy)ꢀbenzyl]ꢀ6ꢀisopropylꢀ10ꢀmethoxyꢀ9ꢀmethylꢀ19ꢀoxaꢀ
1,4,7,25ꢀtetraaza bicyclo[19.3.1]pentacosaꢀ13,15ꢀdieneꢀ2,5,8,20ꢀtetraone (38 mg, 30%) as a dark
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
1
brown solid. H NMR (300 MHz, CDCl ) 7.20–7.05 (m, 1H), 6.97 (d, J = 9.1 Hz, 1H), 6.80 (d, J
3
= 7.6 Hz, 1H), 6.74–6.52 (m, 3H), 6.47 (d, J = 8.0 Hz, 1H), 6.20–5.98 (m, 2H), 5.80–5.49 (m,
3
H), 4.55–4.44 (m, 1H), 4.41–4.20 (m, 2H), 4.19–4.02 (m, 2H), 3.52 (d, J = 12.0 Hz, 2H), 3.47
s, 1H), 3.41–3.09 (m, 2H), 2.96–2.72 (m, 2H), 2.70–2.29 (m, 4H), 2.28–1.94 (m, 4H), 1.88–1.39
(m, 6H), 1.38–1.20 (m, 4H), 0.99 (s, 9H), 0.94 (d, J = 6.7 Hz, 3H), 0.19 (s, 6H). LC/MS (m/z)
13.4 [M+H], Tr = 5.6 min.
(
7
A stirred solution of (13E,15E)ꢀ(3S,6S,9R,10R,21S)ꢀ3ꢀ[3ꢀ(tertꢀbutylꢀdimethylꢀsilanyloxy)ꢀ
benzyl]ꢀ6ꢀisopropylꢀ10ꢀmethoxyꢀ9ꢀmethylꢀ19ꢀoxaꢀ1,4,7,25ꢀtetraaza bicyclo[19.3.1]pentacosaꢀ
1
3,15ꢀdieneꢀ2,5,8,20ꢀtetraone (38 mg, 0.053 mmol.) in anhydrous tetrahydrofuran (10 mL) was
cooled to 0 °C under a nitrogen atmosphere before adding tetrabutylammonium fluoride (1 M in
tetrahydrofuran, 265 µL, 0.265 mmol). The reaction mixture was warmed to room temperature
and was stirred under a nitrogen atmosphere for 2.5 h. The reaction mixture was then treated
with saturated sodium bicarbonate solution (20 mL) and the aqueous layer was extracted with
ethyl acetate (3 × 15 mL). The combined extracts were dried over anhydrous sodium sulfate,
filtered, and evaporated. The residue was purified by silica gel chromatography eluting with a
stepwise gradient of isoꢀhexane/ethyl acetate 1:1 to 0:1 to yield the title product (15 mg, 47%) as
1
a colorless solid. H NMR (300 MHz, CDCl ) 8.38–8.15 (m, 1H), 7.41 (d, J = 9.6 Hz, 1H), 7.08
3
(
t, J = 7.7 Hz, 1H), 6.85 (d, J = 8.7 Hz, 1H), 6.78–6.68 (m, 2H), 6.58–6.51 (d, J = 7.4 Hz, 1H),
6.00–5.91 (m, 3H), 5.69–5.43 (m, 3H), 4.55 (d, J = 12.7 Hz, 1H), 4.40–4.09 (m, 4H), 3.56 (d, J
=
12.5 Hz, 1H), 3.52 (s, 3H), 3.30–3.18 (m, 2H), 2.92–2.75 (m, 2H), 2.71–2.55 (m, 2H), 2.53–
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 72
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
2
.38 (m, 2H), 2.19 (s, 1H), 2.10–1.45 (m, 6H), 1.36 (d, J = 7.4 Hz, 3H), 1.01–0.93 (m, 4H),
.92–0.81 (m, 3H). LC/MS (m/z) 599.4 [M+H], Tr = 4.59 min.
0
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
(E)ꢀ(5S,11S,14S,17R,18R)ꢀ11ꢀ(3ꢀHydroxyꢀbenzyl)ꢀ14ꢀisopropylꢀ18ꢀmethoxyꢀ17ꢀmethylꢀ3ꢀ
5
,9
oxaꢀ9,12,15,28ꢀtetraazaꢀtricyclo[21.3.1.1 ]octacosaꢀ1(27),21,23,25ꢀtetraeneꢀ4,10,13,16ꢀ
tetraone (6)
st
To 26 (280 mg, 0.36 mmol) in anhydrous dichloromethane (120 mL) was added Grubbs 1
generation catalyst (89 mg, 0.11 mmol). The reaction was heated to reflux for 16 h. The reaction
was cooled and concentrated in vacuo. The product was purified by silica gel chromatography
eluting with a stepwise gradient of isoꢀhexane/ethyl acetate 1:1 to 1:2 to afford (E)ꢀ
(5S,11S,14S,17R,18R)ꢀ11ꢀ[3ꢀ(tertꢀButylꢀdimethylꢀsilanyloxy)ꢀbenzyl]ꢀ14ꢀisopropylꢀ18ꢀmethoxyꢀ
5
,9
17ꢀmethylꢀ3ꢀoxaꢀ9,12,15,28ꢀtetraazaꢀtricyclo[21.3.1.1 ]octacosaꢀ1(27),21,23,25ꢀtetraeneꢀ
4,10,13,16ꢀtetraone (213 mg, 75%) as a brown oil that solidifies under vacuum. To the resulting
residue (70 mg, 0.09 mmol), was added anhydrous tetrahydrofuran (1 mL) and cooled to 0 °C. A
solution of tetrabutylammonium fluoride (0.5 mL, 0.47 mmol, 1.0 M in tetrahydrofuran) was
added and the reaction was stirred at 0 °C for 0.5 h. The reaction was quenched with aqueous
saturated ammonium chloride and extracted with dichloromethane (2×). The combined organic
layers were filtered through a hydrophobic frit and concentrated in vacuo. The product was
purified by silica gel chromatography eluting with a stepwise gradient of isoꢀhexane/ethyl acetate
1
1:1 to 1:2 to afford the title product (25 mg, 44%) as a white solid. H NMR (300 MHz, d ꢀ
6
DMSO) δ 8.10 (br s, 1H), 7.41–7.21 (m, 3H), 7.21–7.16 (m, 1H), 7.12–7.05 (m, 1H), 6.94–6.89
m, 1H), 6.85–6.76 (m, 1H), 6.72–6.63 (m, 2H), 6.50–6.43 (m, 1H), 6.10–6.04 (m, 2H), 5.78–
(
5
.68 (m, 1H), 5.21 (s, 2H), 4.62–4.52 (m, 1H), 4.27–4.18 (m, 1H), 3.67 (d, J = 12.2 Hz, 1H),
ACS Paragon Plus Environment
18

Page 19 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
.50 (s, 3H), 3.43–3.24 (m, 2H), 2.74–2.58 (m, 2H), 2.26–1.91 (m, 5H), 1.82–1.55 (m, 6H), 1.37
(d, J = 7.1 Hz, 3H), 1.00 (2d, J = 6.7 Hz, 6H). LC/MS (m/z) 635.4 [M+H], Tr = 5.01 min.
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
(E)ꢀ(2R,5S,11S,14S,17R,18R)ꢀ11ꢀ(3ꢀHydroxyꢀbenzyl)ꢀ14ꢀisopropylꢀ18ꢀmethoxyꢀ2,17ꢀ
5
,9
dimethylꢀ3ꢀoxaꢀ9,12,15,28ꢀtetraazaꢀtricyclo[21.3.1.1 ]octacosaꢀ1(26),21,23(27),24ꢀtetraeneꢀ
,10,13,16ꢀtetraone (7)
4
HoveydaꢀGrubbs II catalyst (25 mg, 0.04 mmol) was added to a solution of 27 (318 mg, 0.402
mmol) in 1,2ꢀdichloroethane (150 mL) and the reaction was heated at reflux under nitrogen for 3
h. The reaction was cooled and silica was added. The solvent was evaporated and the resultant
residue purified by silica gel chromatography eluting with isoꢀhexane/ethyl acetate 1:2 to afford
the macrocycle (214 mg, 70%) as a brown foam which was dissolved in tetrahydrofuran (10 mL)
and tetrabutylammonium fluoride (1.0 M solution in THF, 0.45 mL, 0.45 mmol) was added. The
reaction was stirred at room temperature for 1 h. Silica was added and the solvent was
evaporated. The residue was purified by silica gel chromatography eluting with a stepwise
gradient of isoꢀhexane/ethyl acetate 1:2 to 1:3 by preparative TLC eluting with isoꢀhexane/ethyl
1
acetate 1:3 to afford the title product (115 mg, 63%) as a colorless solid. H NMR (300 MHz,
CD CN) δ 8.25 (s, 1H), 7.34 (d, J = 8.0 Hz, 1H), 7.27–7.19 (m, 3H), 7.12–7.01 (m, 3H), 6.98–
3
6.91 (m, 2H), 6.69–6.60 (m, 1H), 6.10–5.98 (m, 1H), 5.81–5.67 (m, 2H), 5.48 (dt, J = 9.2, 6.2
Hz, 1H), 4.36 (br d, J = 13.4 Hz, 1H), 4.20 (t, J = 8.5 Hz, 1H), 3.97 (d, J = 12.3 Hz, 1H), 3.65
(
dt, J = 11.8, 3.1 Hz, 1H), 3.45 (s, 3H), 3.38–3.28 (m, 1H), 2.89 (d, J = 6.3 Hz, 2H), 2.73–2.58
(m, 2H), 2.10–1.60 (m, 9H), 1.57 (d, J = 6.7 Hz, 3H), 1.31 (d, J = 7.4 Hz, 3H), 0.97 (d, J = 6.7
Hz, 3H), 0.96 (d, J = 6.9 Hz, 3H). LC/MS (m/z) 649.3 [M+H], Tr = 4.96 min.
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 72
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
(
E)ꢀ(2R,5S,11S,14S,17R,18R)ꢀ14ꢀIsopropylꢀ18ꢀmethoxyꢀ2,11,17ꢀtrimethylꢀ3ꢀoxaꢀ9,12,15,28ꢀ
5
,9
tetraazaꢀtricyclo[21.3.1.1 ]octacosaꢀ1(27),21,23,25ꢀtetraeneꢀ4,10,13,16ꢀtetraone (8)
A solution of 32 (195 mg, 0.334 mmol) in 1,2ꢀdichloroethane (110 mL) was treated with
HoveydaꢀGrubbs II catalyst (20.9 mg, 0.033 mmol). After stirring at reflux for 40 min, the
reaction was cooled to room temperature, the volatiles were removed in vacuo and the residue
was purified by silica gel chromatography using a 25 g Isolute cartridge eluting with a
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
continuous gradient of isoꢀhexane/ethyl acetate 1:1 to 0:1 to provide the title compound (52.5
1
mg, 28%) as a white solid along with mixed fractions (33.3 mg). H NMR (300 MHz, d ꢀDMSO)
6
δ 8.04 (d, J = 7.6 Hz, 1H), 7.37–7.13 (m, 4H), 7.06 (d, J = 8.9 Hz, 1H), 6.34–6.14 (m, 2H), 5.81
(
q, J = 6.7 Hz, 1H), 5.25–5.10 (m, 2H), 4.10 (app t, J = 8.7 Hz, 1H), 3.53 (app t, J = 9.4 Hz, 1H),
3.31 (s, 3H), 2.93–2.77 (m, 1H), 2.62 (dd, J = 7.1, 3.3 Hz, 1H), 2.29–2.15 (m, 1H), 2.07–1.58
m, 4H), 1.75–1.51 (m, 4H), 1.51–1.42 (m, 4H), 1.30 (d, J = 7.1 Hz, 3H), 1.11 (d, J = 7.1 Hz,
H), 0.87 (d, J = 6.7 Hz, 3H), 0.82 (d, J = 6.7 Hz, 3H). LC/MS (m/z) 557.3 [M+H], 579.3
(
3
[M+Na], Tr = 5.20 min.
3
1
2
3
4
5
6
(1 S,4R,14S,17S,E)ꢀ14ꢀIsopropylꢀ4,17ꢀdimethylꢀ1 ,1 ,1 ,1 ,1 ,1 ꢀhexahydroꢀ3ꢀoxaꢀ13,16ꢀ
diazaꢀ1(3,1)ꢀpyridazinaꢀ5(1,3)ꢀbenzenacyclooctadecaphanꢀ6ꢀeneꢀ2,12,15,18ꢀtetraone (9)
A solution of 33 (330 mg, 0.61 mmol) in 1,2ꢀdichloroethane (200 mL) was treated with
HoveydaꢀGrubbs II catalyst (52 mg, 0.06 mmol). After stirring at reflux for 60 min under argon
atmosphere, the reaction was cooled down to room temperature, the volatiles were removed in
vacuo and the residue was purified by silica gel chromatography using a gradient from 20 to 80%
ethyl acetate + methanol (4/1) in hexane to provide the title compound (70 mg, 22%) as a white
1
solid. H NMR (400 MHz, DMSOꢀd ) δ 7.94–7.86 (m, 2H), 7.36–7.21 (m, 3H), 7.16 (dt, J = 7.6,
6
ACS Paragon Plus Environment
20

Page 21 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
.5 Hz, 1H), 6.36–6.12 (m, 2H), 5.81 (q, J = 6.5 Hz, 1H), 5.19 (d, J = 11.3 Hz, 1H), 5.17–5.05
(
m, 1H), 4.18–4.05 (m, 1H), 3.52 (td, J = 11.0, 2.9 Hz, 1H), 2.85–2.71 (m, 1H), 2.29 (ddd, J =
3.2, 11.4, 4.0 Hz, 1H), 2.22–2.09 (m, 1H), 2.09–1.91 (m, 2H), 1.91–1.80 (m, 2H), 1.75–1.59
m, 3H), 1.61–1.40 (m, 4H), 1.37–1.22 (m, 4H), 0.84 (dd, J = 7.1, 7.1 Hz, 6H). LC/MS (m/z)
1
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
(
513.2 [M + H].
2,2ꢀDimethylꢀpropionic acid (E)ꢀ(2S,3S,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀoctꢀ6ꢀenyl ester
(11)
To 18ꢀcrownꢀ6 (3.8 g, 14.4 mmol) in anhydrous tetrahydrofuran (90 mL) at 0 °C and under an
atmosphere of nitrogen was added potassium hydride (30 wt% dispersion in mineral oil, 6.7 g,
50.4 mmol) and the suspension was stirred for 5 min. Iodomethane (2.5 mL, 40.3 mmol) was
then added at 0 °C and the reaction stirred for an additional 5 min. A solution of 10 (12 g, 7.2
3
9
mmol, prepared as described previously ) in anhydrous tetrahydrofuran (40 mL) was added
dropwise and the reaction was allowed to warm to room temperature and was stirred overnight.
The reaction was carefully quenched with saturated ammonium chloride and extracted with
diethyl ether (2×). The combined organic layers were dried over anhydrous magnesium sulfate,
filtered, and concentrated in vacuo. The product was purified by silica gel chromatography
eluting with isoꢀhexane/ethyl acetate 25:1 to afford ((E)ꢀ(2S,3S,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀ
1
dimethylꢀoctꢀ6ꢀenyloxymethyl)ꢀbenzene (2.0 g, 91%) as a clear oil. H NMR (300 MHz, CDCl )
3
δ 7.40ꢀ7.29 (m, 5H), 5.66 (dq, J = 15.3, 6.4 Hz, 1H), 5.21 (ddd, J = 15.3, 8.8, 1.6 Hz, 1H), 4.54
(s, 2H), 3.61 (dd, J = 8.8, 3.3 Hz, 1H), 3.53–3.46 (m, 2H), 3.40 (s, 3H), 3.36 (t, J = 9.7 Hz, 1H),
3
.23 (s, 3H), 1.98–1.87 (m, 1H), 1.77 (dd, J = 6.4, 1.6 Hz, 3H), 1.73–1.65 (m, 1H), 0.95 (d, J =
.9 Hz, 3H), 0.75 (d, J = 6.9 Hz, 3H).
6
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 72
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
To ((E)ꢀ(2S,3S,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀoctꢀ6ꢀenyloxymethyl)ꢀbenzene (5.5 g, 18.0
mmol) in anhydrous tetrahydrofuran (180 mL) at −78 °C and under an atmosphere of nitrogen,
was slowly added freshly prepared lithium diꢀtertꢀbutylbiphenyl solution until a dark green color
persisted. The temperature was maintained below −70 °C during the addition. The reaction was
quenched with saturated ammonium chloride, allowed to warm to room temperature and
extracted with diethyl ether (2×). The combined organic layers were dried over anhydrous
magnesium sulfate, filtered, and concentrated in vacuo. The product was purified by silica gel
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
chromatography eluting with a stepwise gradient of isoꢀhexane/ethyl acetate 50:1 to 1:1 to afford
1
(
E)ꢀ(2S,3S,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀoctꢀ6ꢀenꢀ1ꢀol (3.6 g, 93%) as a clear oil. H NMR
(
300 MHz, CDCl ) δ 5.66 (dq, J = 15.2, 6.4 Hz, 1H), 5.19 (dd, J = 15.2, 8.8 Hz, 1H), 3.75–3.52
3
(m, 3H), 3.49 (s, 3H), 3.31 (t, J = 9.1 Hz, 1H), 3.24 (s, 3H), 2.82 (dd, J = 8.2, 2.9 Hz, 1H), 1.96–
1.88 (m, 1H), 1.77 (d, J = 6.4 Hz, 3H), 1.71–1.63 (m, 1H), 0.84 (d, J = 6.9 Hz, 3H), 0.79 (d, J =
7.1 Hz, 3H).
To (E)ꢀ(2S,3S,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀoctꢀ6ꢀenꢀ1ꢀol (1.5 g, 6.94 mmol) in anhydrous
pyridine (20 mL) at 0 °C and under an atmosphere of nitrogen was added pivaloyl chloride (1.33
mL, 10.7 mmol). The reaction was warmed to room temperature and stirred for 2 h, after which
it was quenched with saturated ammonium chloride and extracted with dichloromethane (2×).
The combined organic layers were filtered through a hydrophobic frit and concentrated in vacuo.
The product was purified by silica gel chromatography eluting with a stepwise gradient of isoꢀ
1
hexane/ethyl acetate 20:1 to 1:1 to afford the title compound (2.0 g, 100%) as a clear oil. H
NMR (300 MHz, CDCl ) δ 5.67 (dq, J = 15.3, 6.6 Hz, 1H), 5.26–5.15 (m, 1H), 4.23 (dd, J =
3
10.6, 3.3 Hz, 1H), 4.1 (dd, J = 10.6, 6.0 Hz, 1H), 3.52 (dd, J = 10.0, 2.0 Hz, 1H), 3.42 (s, 3H),
ACS Paragon Plus Environment
22

Page 23 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
.34 (t, J = 9.3 Hz, 1H), 3.25 (s, 3H), 2.03–1.88 (m, 1H), 1.78 (dd, J = 6.4 Hz, 1.5 Hz, 3H),
.74–1.62 (m, 1H), 1.24 (s, 9H), 0.90 (d, J = 7.1 Hz, 3H), 0.76 (d, J = 7.1 Hz, 3H).
1
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
2,2ꢀDimethylꢀpropionic acid (E)ꢀ(2S,3S,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀnonaꢀ6,8ꢀdienyl
ester (12)
To 11 (2.0 g, 6.94 mmol) in anhydrous dichloromethane (60 mL) at −78 °C, ozone was bubbled
through until a pale blue color persisted. Nitrogen was then bubbled through until the solution
turned colorless. Triphenylphosphine (2.7 g, 10.4 mmol) was added and the reaction was
warmed to room temperature and stirred overnight. The reaction was concentrated in vacuo and
the product was purified by silica gel chromatography eluting with a stepwise gradient of isoꢀ
hexane/ethyl acetate 20:1 to 4:1 to afford 2,2ꢀdimethylꢀpropionic acid (2S,3S,4R,5R)ꢀ3,5ꢀ
1
dimethoxyꢀ2,4ꢀdimethylꢀ6ꢀoxoꢀhexyl ester (1.62 g, 81%) as a yellow oil. H NMR (300 MHz,
CDCl ) δ 9.51 (d, J = 4.1 Hz, 1H), 4.14 (dd, J = 10.8, 3.3 Hz, 1H), 4.00 (dd, J = 10.8, 5.9 Hz,
3
1
H), 3.36–3.29 (m, 2H), 3.33 (s, 3H), 3.32 (s, 3H), 1.97–1.87 (m, 2H), 1.20 (s, 9H), 0.84 (d, J =
.0 Hz, 3H), 0.78 (d, J = 7.0 Hz, 3H).
7
To diethylallylphosphonate (1.86 g, 10.4 mmol) in anhydrous tetrahydrofuran (25 mL), at –78 °C
and under an atmosphere of nitrogen, was slowly added nꢀbutyllithium (2.5 M in hexanes, 4.2
mL, 10.4 mmol) maintaining the temperature below −78 °C. The reaction was stirred at –78 °C
for 30 min after which a solution of 2,2ꢀdimethylꢀpropionic acid (2S,3S,4R,5R)ꢀ3,5ꢀdimethoxyꢀ
2,4ꢀdimethylꢀ6ꢀoxoꢀhexyl ester (2.0 g, 6.9 mmol) and 1,3ꢀdimethyltetrahydropyrimidinꢀ2(1H)ꢀ
one (freshly distilled over calcium hydride, 1.7 mL, 13.9 mmol) in anhydrous tetrahydrofuran (5
mL) was added. The reaction was allowed to stir at −78 °C for 1 h and then warmed to room
temperature and stirred overnight. The reaction was diluted with saturated ammonium chloride
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 72
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
and extracted with dichloromethane (2×). The combined organic layers were filtered through a
hydrophobic frit and concentrated in vacuo. The product was purified by silica gel
chromatography eluting with isoꢀhexane/ethyl acetate 20/1 to afford the title compound (1.3 g,
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
6
0%) as a clear oil. H NMR (300 MHz, CDCl ) δ 6.41 (dt, J = 16.8, 10.2 Hz, 1H), 6.24 (dd, J =
3
15.2, 10.4 Hz, 1H), 5.47 (dd, J = 15.2, 8.6 Hz, 1H), 5.25 (dd, J = 16.8, 1.5 Hz, 1H), 5.13 (dd, J =
9.7, 1.5 Hz, 1H), 4.23 (dd, J = 10.6, 3.3 Hz, 1H), 4.01 (dd, J = 10.6, 6.0 Hz, 1H), 3.52 (dd, J =
10.0, 2.0 Hz, 1H), 3.43 (s, 3H), 3.48–3.39 (m, 1H), 3.27 (s, 3H), 2.03–1.89 (m, 1H), 1.79–1.66
(m, 1H), 1.24 (s, 9H), 0.90 (d, J = 6.9 Hz, 3H), 0.77 (d, J = 7.1 Hz, 3H).
(E)ꢀ(2R,3R,4S,5S)ꢀ3,5ꢀDimethoxyꢀ2,4ꢀdimethylꢀnonaꢀ6,8ꢀdienoic acid (13)
To 12 (1.3 g, 4.2 mmol) in anhydrous methanol (10 mL) at room temperature and under an
atmosphere of nitrogen was added sodium methoxide (30wt% in methanol, 3.7 mL, 20.8 mmol).
The reaction was stirred overnight. More sodium methoxide (30wt% in methanol, 3.7 mL, 20.8
mmol) was added and the reaction was stirred for 1 h. The reaction was quenched with saturated
ammonium chloride and extracted with dichloromethane (3×). The combined organic layers were
dried by passing through a hydrophobic frit and concentrated in vacuo. The product was purified
by silica gel chromatography eluting with a stepwise gradient of isoꢀhexane/ethyl acetate 10:1 to
2:1 to afford (E)ꢀ(2S,3S,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀnonaꢀ6,8ꢀdienꢀ1ꢀol (2.0 g, 100%) as
1
a clear oil. H NMR (300 MHz, CDCl ) δ 6.40 (dt, J = 16.8, 10.2 Hz, 1H), 6.23 (dd, J = 15.2,
3
1
0.4 Hz, 1H), 5.46 (dd, J = 15.2, 8.4 Hz, 1H), 5.26 (dd, J = 17.0, 1.6 Hz, 1H), 5.14 (dd, J = 10.2,
1.6 Hz, 1H), 3.77–3.53 (m, 3H), 3.51 (s, 3H), 3.42 (t, J = 9.3 Hz, 1H), 3.27 (s, 3H), 2.75 (dd, J =
.4, 3.1 Hz, 1H), 1.99–1.82 (m, 1H), 1.77–1.65 (m, 1H), 0.85 (d, J = 7.1 Hz, 3H), 0.80 (d, J = 7.1
Hz, 3H).
8
ACS Paragon Plus Environment
24

Page 25 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To (E)ꢀ(2S,3S,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀnonaꢀ6,8ꢀdienꢀ1ꢀol (1.3 g, 5.70 mmol) in
anhydrous dichloromethane (58 mL) at room temperature and under an atmosphere of nitrogen
was added activated 4 Å molecular sieves (1.3 g), NꢀmethylmorpholineꢀNꢀoxide (2.0 g, 17.1
mmol) and tetrapropylammonium perruthenate (100 mg, 0.29 mmol). The reaction was stirred
for 3 h and filtered through a silica gel pad, and the pad was washed with diethyl ether. The
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
filtrate was concentrated in vacuo to afford (E)ꢀ(2R,3R,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀnonaꢀ
1
6
,8ꢀdienal (1.2 g, 93%) as a clear oil. H NMR (300 MHz, CDCl ) δ 9.83 (d, J = 2.7 Hz, 1H),
3
6.40 (dt, J = 16.8, 10.2 Hz, 1H), 6.24 (dd, J = 15.2, 10.6 Hz, 1H), 5.45 (dd, J = 15.2, 8.8 Hz, 1H),
5
.26 (dd, J = 16.4, 1.3 Hz, 1H), 5.14 (dd, J = 10.2, 1.6 Hz, 1H), 3.87 (dd, J = 9.1, 2.4 Hz, 1H),
.48–3.40 (m, 1H), 3.43 (s, 3H), 3.26 (s, 3H), 2.70–2.55 (m, 1H), 1.78–1.61 (m, 1H), 1.00 (d, J =
3
7.1 Hz, 3H), 0.82 (d, J = 7.1 Hz, 3H).
To (E)ꢀ(2R,3R,4S,5S)ꢀ3,5ꢀdimethoxyꢀ2,4ꢀdimethylꢀnonaꢀ6,8ꢀdienal (1.2 g, 5.3 mmol) in tertꢀ
butanol (24 mL) and 2ꢀmethylꢀ2ꢀbutene (5.6 mL, 53.1 mmol) at room temperature, was added a
solution of sodium chlorite (2.4 g, 26.5 mmol) and sodium dihydrogenphosphate (1.5 g, 10.6
mmol) in water (5 mL). The reaction was stirred vigorously for 1.5 h, after which brine (10 mL)
was added and the reaction acidified to pH 5 with hydrochloric acid (2 M). The aqueous was
extracted with dichloromethane (3×), the organics were combined, filtered through a
hydrophobic frit and concentrated in vacuo. The oily residue was left on a vacuum pump
1
overnight to afford the title product (1.02 g, 79%) as an orange solid. H NMR (300 MHz,
CDCl ) δ 6.39 (dt, J = 16.7, 10.3 Hz, 1H), 6.23 (dd, J = 14.9, 10.5 Hz, 1H), 5.45 (dd, J = 14.9,
3
8
.7 Hz, 1H), 5.26 (d, J = 16.3 Hz, 1H), 5.14 (d, J = 10.3 Hz, 1H), 3.91 (dd, J = 9.4, 1.8 Hz, 1H),
3.46 (s, 3H), 3.52–3.37 (m, 1H), 3.26 (s, 3H), 2.74–2.58 (m, 1H), 1.80–1.64 (m, 1H), 1.13 (d, J =
.1 Hz, 3H), 0.78 (d, J = 7.1 Hz, 3H).
7
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 72
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
(
S)ꢀ1ꢀ{(S)ꢀ2ꢀ((S)ꢀ2ꢀtertꢀButoxycarbonylaminoꢀ3ꢀmethylꢀbutyrylamino)ꢀ3ꢀ[3ꢀ(tertꢀbutylꢀ
dimethylꢀsilanyloxy)ꢀphenyl]ꢀpropionyl}ꢀhexahydroꢀpyridazineꢀ3ꢀcarboxylic acid butꢀ3ꢀ
enyl ester (15)
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
27
A solution of 14 (1.891 g, 2.562 mmol, prepared as described previously ) in tetrahydrofuran
(50 mL) was successively treated with zinc powder (3.685 g, 56.368 mmol) and a solution of
ammonium acetate (2.962 g, 38.430 mmol) in water (10 mL). After stirring for 24 h at room
temperature, the mixture was filtered through a pad of Celite and rinsed with a dilute solution of
potassium hydrogensulfate (pH 4). The aqueous layer was extracted with ethyl acetate (3 × 50
mL) and the organics were combined, dried over anhydrous sodium sulfate and filtered. The
volatiles were removed in vacuo and the residual acetic acid was azeotroped off with toluene (3
× 50 mL) to provide (S)ꢀ1ꢀ{(S)ꢀ2ꢀ((S)ꢀ2ꢀtertꢀbutoxycarbonylaminoꢀ3ꢀmethylꢀbutyrylamino)ꢀ3ꢀ
[
3ꢀ(tertꢀbutylꢀdimethylꢀsilanyloxy)ꢀphenyl]ꢀpropionyl}ꢀhexahydroꢀpyridazineꢀ3ꢀcarboxylic acid
as a white solid which was then suspended in anhydrous toluene (20 mL). Upon the addition of
triethylamine (540 ꢀL, 3.843 mmol), the mixture became clear. The mixture was subsequently
treated with 2,4,6ꢀtrichlorobenzoyl chloride (480 ꢀL, 3.074 mmol). After stirring for 50 min at
room temperature,
a
solution of allyl alcohol (330 ꢀL, 3.843 mmol) and 4ꢀ
dimethylaminopyridine (469 mg, 3.843 mmol) in toluene (20 mL) was added. After stirring
overnight at room temperature, the volatiles were removed in vacuo and the residue was purified
by silica gel chromatography using a 50 g Isolute cartridge eluting with a continuous gradient of
isoꢀhexane/ethyl acetate 1:0 to 1:2 to provide the title compound (85 mg, 14% over 2 steps) as a
1
white solid. H NMR (300 MHz, CDCl ) δ 7.13 (app t, J = 7.5 Hz, 1H), 6.81 (d, J = 7.5 Hz, 1H),
3
6
.74–6.64 (m, 2H), 6.48 (d, J = 8.4 Hz, 1H), 5.90–5.68 (m, 2H), 5.18–5.08 (m, 2H), 5.08–5.01
(m, 1H), 4.34–4.25 (br d, J = 13.9 Hz, 1H), 4.18 (app dp, J = 6.8, 4.4 Hz, 2H), 4.00–3.90 (m,
ACS Paragon Plus Environment
26

Page 27 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1H), 3.52 (d, J = 11.0 Hz, 1H), 2.93 (qd, J = 15.5, 6.0 Hz, 2H), 2.81–2.70 (m, 1H), 2.59–2.48 (m,
1H), 2.41 (app q, J = 6.8 Hz, 2H), 2.20–2.06 (m, 1H), 1.88–1.73 (m, 2H), 1.57–1.49 (m, 2H),
1.46 (s, 9H), 0.99 (s, 9H), 0.94 (d, J = 6.8 Hz, 3H), 0.88 (d, J = 6.8 Hz, 3H), 0.20 (s, 6H). LC/MS
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
(
m/z) 661.6 [M+H], 683.5 [M+Na], Tr = 5.79 min.
(S)ꢀ1ꢀ{(S)ꢀ3ꢀ[3ꢀ(tertꢀButylꢀdimethylꢀsilanyloxy)ꢀphenyl]ꢀ2ꢀ[(S)ꢀ2ꢀ((E)ꢀ(2R,3R,4S,5S)ꢀ3,5ꢀ
dimethoxyꢀ2,4ꢀdimethylꢀnonaꢀ6,8ꢀdienoylamino)ꢀ3ꢀmethylꢀbutyrylamino]ꢀpropionyl}ꢀ
hexahydroꢀpyridazineꢀ3ꢀcarboxylic acid butꢀ3ꢀenyl ester (16)
A cooled (0 °C) solution of 15 (328 mg, 0.500 mmol) in anhydrous dichloromethane (15 mL)
was treated with trimethylsilyl trifluoromethane sulfonate (140 ꢀL, 0.750 mmol). After 1.5 h at
0
°C, N,Nꢀdiisopropylethylamine (0.3 mL, 2.0 mmol) was added and the volatiles were removed
in vacuo to provide crude (S)ꢀ1ꢀ{(S)ꢀ2ꢀ((S)ꢀ2ꢀaminoꢀ3ꢀmethylꢀbutyrylamino)ꢀ3ꢀ[3ꢀ(tertꢀbutylꢀ
dimethylꢀsilanyloxy)ꢀphenyl]ꢀpropionyl}ꢀhexahydroꢀpyridazineꢀ3ꢀcarboxylic acid butꢀ3ꢀenyl
ester as a white solid which was used without further purification.
A cooled (0 °C) solution of 13 (100 mg, 0.413 mmol) in anhydrous N,Nꢀdimethylformamide (1
mL) was successively treated with N,Nꢀdiisopropylethylamine (290 ꢀL, 1.652 mmol) and a
solution of 2ꢀ(1Hꢀ7ꢀazabenzotriazolꢀ1ꢀyl)ꢀ1,1,3,3ꢀtetramethyl uronium hexafluorophosphate
methanaminium (188 mg, 0.496 mmol) in N,Nꢀdimethylformamide (1 mL). After 40 min at 0 °C,
the reaction mixture was treated with a solution of freshly prepared (S)ꢀ1ꢀ{(S)ꢀ2ꢀ((S)ꢀ2ꢀaminoꢀ3ꢀ
methylꢀbutyrylamino)ꢀ3ꢀ[3ꢀ(tertꢀbutylꢀdimethylꢀsilanyloxy)ꢀphenyl]ꢀpropionyl}ꢀhexahydroꢀ
pyridazineꢀ3ꢀcarboxylic acid butꢀ3ꢀenyl ester in N,Nꢀdimethylformamide (2 mL). The green
solution was allowed to warm overnight to room temperature and then quenched with pH 7
buffer (10 mL). The aqueous layer was extracted with dichloromethane (3 × 10 mL) and the
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 72
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
organics were combined, dried over anhydrous sodium sulfate, filtered, and the volatiles were
removed in vacuo. The residual N,Nꢀdimethylformamide was azeotroped off with toluene (2 × 30
mL) and the solid residue was purified by silica gel chromatography using a 25 g Isolute
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
cartridge eluting with a continuous gradient of isoꢀhexane/ethyl acetate 1:0 to 1:1 to provide the
1
title compound (229 mg, 70%) as a white solid. H NMR (300 MHz, CDCl ) δ 7.12 (app t, J =
3
8
.2 Hz, 1H), 6.80 (d, J = 8.2 Hz, 1H), 6.74–6.63 (m, 2H), 6.50–6.32 (m, 2H), 6.27 (s, 1H), 6.25–
.16 (m, 1H), 5.90–5.67 (m, 2H), 5.47 (dd, J = 15.0, 8.8 Hz, 1H), 5.25 (d, J = 16.1 Hz, 1H),
6
5.18–5.08 (m, 3H), 4.29 (dd, J = 7.9, 6.0 Hz, 2H), 4.18 (app dp, J = 6.6, 5.3 Hz, 2H), 3.87 (dd, J
9.1, 2.2 Hz, 1H), 3.53 (d, J = 11.1 Hz, 1H), 3.42 (app q, J = 9.3 Hz, 1H), 3.40 (s, 3H), 3.25 (s,
H), 2.99–2.85 (m, 2H), 2.78–2.69 (m, 1H), 2.62–2.50 (m, 1H), 2.46–2.32 (m, 3H), 2.22–2.08
(m, 1H), 1.89–1.67 (m, 2H), 1.53–1.40 (m, 1H), 1.31–1.25 (m, 2H), 1.07 (d, J = 7.0 Hz, 3H),
=
3
0
.99 (s, 9H), 0.98 (d, J = 6.8 Hz, 3H), 0.96 (d, J = 6.8 Hz, 3H), 0.77 (d, J = 7.1 Hz, 3H), 0.19 (s,
H). LC/MS (m/z) 785.6 [M+H], 807.6 [M+Na], Tr = 5.97 min.
6
6
1,5
(2R,3R)ꢀ1ꢀ((1R,5S)ꢀ10,10ꢀDimethylꢀ3,3ꢀdioxoꢀ3λ ꢀthiaꢀ4ꢀazaꢀtricyclo[5.2.1.0 ]decꢀ4ꢀyl)ꢀ3ꢀ
hydroxyꢀ2ꢀmethylꢀheptꢀ6ꢀenꢀ1ꢀone (18)
A solution of 17 (3.95 g, 14.55 mmol) in toluene (50 mL) was prepared, then evaporated to
dryness. This process was repeated and then the resulting white solid was dissolved in anhydrous
dichloromethane (16 mL). A small quantity of calcium hydride was added before adding tertꢀ
butyldimethylsilyl trifluoromethanesulfonate (3.83 mL, 14.5 mmol) and anhydrous triethylamine
(2.33 mL, 16.7 mmol). The reaction mixture was stirred at room temperature under a nitrogen
atmosphere for 15 h. The resulting solution was evaporated to yield a thick paste which was
dissolved in anhydrous dichloromethane (15 mL) and added dropwise to a cooled (–78 °C)
ACS Paragon Plus Environment
28

Page 29 of 72
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
solution of 4ꢀpentenal (2.69 g, 32.0 mmol) and titanium tetrachloride (1 M in dichloromethane,
3
2 mL, 32 mmol) in anhydrous dichloromethane (20 mL). The reaction was stirred at –78 °C for
0 min before diluting with saturated aqueous ammonium chloride solution (100 mL). The layers
3
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
were separated and the aqueous layer was extracted with dichloromethane (2 × 50 mL). The
combined extracts were dried over anhydrous sodium sulfate, filtered, and evaporated to give a
brown gum which was purified by silica gel chromatography eluting with isoꢀhexane/ethyl
1
acetate 4:1 to yield the title compound (3.09 g, 60%) as a colorless gum. H NMR (300 MHz,
CDCl ) δ 5.90–5.74 (m, 1H), δ5.00–4.94 (m, 2H), 3.91 (dd, J = 7.6, 5.1 Hz, 1H), 3.67 (br s, 1H),
3
3
.50 (ABq, ꢁδAB = 0.06, JAB = 13.8 Hz, 2H), 3.26–3.15 (m, 1H), 2.53–2.08 (m, 5H), 2.01–1.84
m, 3H), 1.76–1.63 (m, 1H), 1.55–1.31 (m, 3H), 1.25 (d, J = 6.7 Hz, 3H), 1.19 (s, 3H), 0.99 (s,
3H). LC/MS (m/z) 356.2 [M+H], Tr = 3.41 min.
(
(2R,3R)ꢀ3ꢀMethoxyꢀ2ꢀmethylheptꢀ6ꢀenoic acid (19)
A solution of 18 (250 mg, 0.703 mmol) in anhydrous dichloromethane (7 mL) was prepared and
a solution of N,N,N′,N′ꢀtetramethylꢀ1,8ꢀnaphthalenediamine (452 mg, 0.703 mmol) in anhydrous
dichloromethane (7 mL) was added. Trimethyloxonium tetrafluoroborate (208 mg, 1.406 mmol)
was then added. The reaction mixture was stirred at room temperature for 15 h. The reaction
mixture was then treated with methanol (1 mL), hydrochloric acid (2 M, 20 mL) and saturated
brine (20 mL). The aqueous layer was extracted with ethyl acetate (3 × 15 mL) and the extracts
were dried over anhydrous sodium sulfate, filtered, and evaporated to give a yellow gum. The
gum was purified by silica gel chromatography eluting with isoꢀhexane/ethyl acetate 3:1 to give
6
1,5
(
2R,3R)ꢀ1ꢀ((1R,5S)ꢀ10,10ꢀdimethylꢀ3,3ꢀdioxoꢀ3λ ꢀthiaꢀ4ꢀazaꢀtricyclo[5.2.1.0 ]decꢀ4ꢀyl)ꢀ3ꢀ
1
methoxyꢀ2ꢀmethylꢀheptꢀ6ꢀenꢀ1ꢀone (223 mg, 86%) as a colorless gum. H NMR (300 MHz,
ACS Paragon Plus Environment
29