Conformational analysis of N,N-disubstituted-1,4-diazepane orexin receptor antagonists and implications for receptor binding

Article abstract of DOI:10.1016/j.bmcl.2009.04.026

NMR spectroscopy, X-ray crystallography, and molecular modeling studies indicate that N,N-disubstituted-1,4-diazepane orexin receptor antagonists exist in an unexpected low-energy conformation that is characterized by an intramolecular π-stacking interact

Full text of DOI:10.1016/j.bmcl.2009.04.026

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2997–3001
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Conformational analysis of N,N-disubstituted-1,4-diazepane orexin receptor
antagonists and implications for receptor binding
a
a
a
Christopher D. Cox a, , Georgia B. McGaughey , Michael J. Bogusky , David B. Whitman ,
*
Richard G. Ball b, Christopher J. Winrow c, John J. Renger c, Paul J. Coleman a
a Department of Medicinal Chemistry, Merck Research Laboratories, WP14-2, PO Box 4, Sumneytown Pike, West Point, PA 19486, USA
b Department of Pharmaceutical Research, Merck Research Laboratories, PO Box 2000, 126 E. Lincoln Highway, Rahway, NJ 07065, USA
c Department of Depression and Circadian Disorders, Merck Research Laboratories, PO Box 4, Sumneytown Pike, West Point, PA 19486, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
NMR spectroscopy, X-ray crystallography, and molecular modeling studies indicate that N,N-disubsti-
Received 19 March 2009
Revised 8 April 2009
Accepted 9 April 2009
Available online 14 April 2009
tuted-1,4-diazepane orexin receptor antagonists exist in an unexpected low-energy conformation that
is characterized by an intramolecular p-stacking interaction and a twist-boat ring conformation. Synthe-
sis and evaluation of a macrocycle that enforces a similar conformation suggest that this geometry mim-
ics the bioactive conformation.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Orexin receptor antagonist
Diazepane
Pi-stacking
Macrocyclicization
Antagonism of the orexin (or hypocretin) system has recently
been identified as a novel mechanism for the treatment of insom-
nia.1 In a recent Communication, we described the discovery of a
novel series of HTS-derived dual (OX1R/OX2R) orexin receptor
antagonists based on a 1,4-diazepane core.2 This effort culminated
in the discovery of 1, a potent and brain-penetrant compound that
demonstrates efficacy for silencing orexin signaling in freely loco-
moting rats, including the induction of REM and non-REM sleep.
During our synthetic efforts leading to 1, examination of 1H
NMR spectra revealed that compounds in this chemical class exist
in multiple conformations in solution, and that several proton res-
onances appear with unexpected chemical shifts that are indicative
of an unusual conformational bias. We therefore undertook an
investigation to elucidate the nature and origin of these conforma-
tional effects, and to determine if they have ramifications for
receptor binding. Herein, we describe the results of our study that
line widths, but upon cooling the sample to À40 °C in CD3OD, the
resonances narrow significantly to reveal four components in equi-
librium (ꢀ5:4:1:1) which are likely due to amide and 2-aminoqui-
nazoline rotamers.3 Though the spectrum is complex with many
overlapping signals, proton resonance assignments and NOE corre-
lations of the two major rotamers can be determined.
N
N
N
O
N
N
N
N
Me
1
OX1R Ki = 1.2 nM
OX2R Ki = 0.6 nM
suggest an intramolecular p-stacking interaction and the adoption
of a twist-boat ring conformation favor an unexpected low-energy
conformation in orexin receptor antagonists such as 1, and that
this low-energy structure resembles the bioactive conformation.
As a first step to investigate the conformational preferences of
these novel molecules, we performed detailed 600 MHz 1H NMR
studies to appraise their three-dimensional structure in solution.
Analysis of 1 at room temperature is precluded by broad resonance
In the major conformers, rotamer A and rotamer B, NOE’s are
observed from the proton at the 4-position of the quinazoline ring
to the aromatic methyl group as well as the proton at the 6-posi-
tion of the phenyl ring, as illustrated in Figure 1A and B. Addition-
ally, the 1H NMR resonances due to the phenyl and methyl protons
in rotamer A, 6.25 and 1.69 ppm, respectively, are shifted signifi-
cantly upfield of their expected positions of approximately 7.23
and 2.43 ppm due to the shielding effect of the quinazoline ring.4
Taken together, this data suggests that the predominant solution
conformation is one in which the molecule adopts a horseshoe-
* Corresponding author. Tel.: +1 215 652 2411; fax: +1 215 652 7310.
E-mail address: chris_cox@merck.com (C.D. Cox).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.04.026

2998
C. D. Cox et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2997–3001
A
B
C
1.69
2.17
H3C
H3C
9.17
7.26
7.76
7.40
7.80
8.85
7.82
7.73
NOE's
NOE's
N
4
N
7.28
6.25
6
4
7.28
6.57
6
N
N
N
N
N
N
N
N
N
N
N
N
7.65
7.65
O
O
1-rotamer B
1-rotamer A
Figure 1. Panel A: 1H NMR data for the most highly populated rotamer of compound 1, 1-rotamer A, determined by analysis of 1 at À40 °C in CD3OD. Key NOE correlations
are indicated by black arrows, and proton resonance assignments (in ppm) are illustrated; resonances highlighted in red are most relevant for deducing solution
conformation. Panel B: similar data as presented in Panel A, but for the second most populated rotamer, 1-rotamer B (Rotamer A accounts for 45% of the solution and
rotamer B is 36%). Panel C: a low-energy conformation computed for 1 that is consistent with the NMR data.5
or U-shaped structure where the quinazoline and phenyl groups are
in close proximity, likely engaged in a -stacking interaction. Molec-
ular modeling studies predict a low-energy conformation for 1 that
is consistent with these observations, as illustrated in Figure 1C.5
In 1-rotamer B, the key proton resonances appear at 6.57 and
In the major component of compound 2, rotamer A, NOE’s are
observed between the proton at the 6-position of the phenyl group
and those at the 1-, 3-, and 4-positions of the naphthyl group
(highlighted in red in Fig. 2A). Additionally, a strong NOE between
the aromatic methyl protons and the proton at the 5-position of
the naphthyl group is observed (highlighted in blue). Furthermore,
dramatic upfield shifts of the phenyl and aromatic methyl protons
to 5.99 and 1.73 ppm indicate a strong shielding effect due to their
proximity to the naphthyl ring. In contrast, no significant NOEs are
observed between the aromatic moieties in rotamer B (Fig. 2B),
and the chemical shifts of the phenyl and methyl protons at 6.87
and 2.27 ppm are closer to their expected isotropic values of
approximately 7.23 and 2.43 ppm.4 This implies rotamer B has
greater separation between the aromatic moieties, and thus a
p
2.17 ppm, suggesting a weaker p-stacking interaction than in rot-
amer A. However, in conjunction with the observed NOE’s, the
NMR data support a U-shaped structure for rotamer B as well. In
contrast, there are no significant NOE correlations or upfield reso-
nance shifts in either of the minor components that suggest a close
spatial relationship between the two aryl groups. Magnetization
transfer between the various conformers present in solution pro-
hibits a more detailed analysis of 1.
We next synthesized the 2-naphthyl analog of 1 to simplify the
NMR spectrum and permit a more complete conformational anal-
ysis. Compound 2 is a potent dual orexin receptor antagonist,
and, more importantly, it exists as a mixture of only two compo-
nents at room temperature, identified by NOE correlations to be
cis/trans amide rotamers, in a ratio of approximately 1.4–1. As pic-
tured in Figure 2, the amide rotamers are assigned by NOE’s be-
tween the proton at C6 of the phenyl ring and the protons on the
diazepan ring adjacent to this position, and all key aromatic reso-
nances can be explicitly assigned for both rotamers.
weaker p-stacking interaction, than in rotamer A.
A computational analysis of 2 was performed, and a Boltzmann
average was calculated over all low-energy conformations within
5 kcal/mol of the minima using the MMFFs forcefield.5 The lowest
energy conformer has a structure consistent with the NMR data
obtained for rotamer A, and is pictured in Figure 3. The aromatic
groups are engaged in a
p-stacking interaction with a geometry
that is in between that of well defined edge-to-face (or T-shaped)
and parallel-displaced6 orientations. The centroid separation dis-
tance is 5.2 Å, versus calculated distances in the optimized benzene
dimer of 4.96 Å for edge-to-face or 3.4–3.6 Å for parallel-displaced
interactions.7 A structure consistent with rotamer B, though pop-
ulated to a lesser degree, is also identified among the low-energy
conformations and has a centroid separation distance of 7.3 Å.8
To further our understanding of the conformational preferences
of compounds in this chemical class, we determined the structure
of 1 by X-ray crystallography.9 As pictured in Figure 4, the solid
state structure of 1 is very similar to the predominant low- energy
conformations determined by 1H NMR analysis and molecular
modeling studies on 1 and 2, wherein the central seven-membered
N
N
O
N
N
N
Me
OX1R Ki = 2 nM
OX2R Ki = 2 nM
2
A
B
7.17
7.73
7.21
1.73
H3C
5.99
7.22
7.73
7.68
7.68
O
6
N
N
N
7.09
7.33
N
N
7.00
3
1
7.18
7.32
6
7.63
5
N
7.83
N
N
N
6.87
6.87
7.54
N
O
7.36
2.27 H3C
2-rotamer A
2-rotamer B
Figure 2. Panel A: 1H NMR data for the major rotamer of compound 2, 2-rotamer A, determined by analysis of 2 at 25 °C in CD3OD. The NOE used to assign the amide rotamer
is indicated by a black arrow, and the NOE’s used to establish the U-shaped conformation are illustrated by the proton resonance assignments (in ppm) highlighted in red and
blue. Panel B: similar data as presented in Panel A, but for the minor rotamer, 2-rotamer B. Rotamer A accounts for 58% of the solution and rotamer B is 42%.

C. D. Cox et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2997–3001
2999
chose to quantitate the
p-stacking ability of the quinazoline and
phenyl rings in compound 1 beginning with the crystal structure
coordinates and optimizing to a local minima using the MP2/
aug-cc-pVDZ level of theory.19 Using the Counterpoise keyword
in Gaussian, this interaction is calculated to be À7.5 kcal/mol. Rel-
ative to the parallel-displaced benzene dimer at the same level of
theory (À4.2 kcal/mol), the quinazoline–phenyl dimer interaction
is calculated to be stronger.20 When we employ the crystal coordi-
nates for only the quinazoline and phenyl rings and perform a sin-
gle point calculation (thus, keeping the coordinates fixed at the
crystal structure solid state conformation), the two-body binding
interaction is calculated to be À3.5 kcal/mol, less than an opti-
mized benzene dimer interaction, but consistent with an energet-
ically favorable
p-stacking interaction.
N
N
Figure 3. Predominant, low-energy structure of 2 using the MMFFs forcefield in
chloroform at a constant dielectric constant of 4. The structure is consistent with
the 1H NMR for 2-rotamer A, and has a centroid separation distance of 5.2 Å.
N
N
O
N
N
N
N
N
N
N
ring adopts a twist-boat conformation10 and the aryl moieties are
in close proximity with a centroid separation distance of 5.1 Å.
Me
Within the crystal lattice there are also
p p stacking interactions
À
OX1R Ki = 930 nM
OX2R Ki = 240 nM
OX1R Ki = 9,800 nM
OX2R Ki = 550 nM
4
3
in symmetry-related instances of the quinazoline rings which pre-
sumably assist in stabilizing the low-energy conformer in the so-
lid-state.11
To better understand the driving force for enforcing the U-shaped
conformation, we performed studies to investigate the energetic
To investigate if the
ing force for adopting the U-shaped structure, we synthesized
cyclohexyl amide 3 that can not form a -stack, as well as tertiary
p-stacking interaction is the primary driv-
contribution of the
p
-stacking interaction. Along with hydrogen-
p
bonding, -stacking is a critical non-covalent interaction that helps
p
amine 4 which lacks the amide carbonyl. As expected from previ-
ous SAR, both 3 and 4 display significantly reduced binding affinity
for the orexin receptors.2a NMR analysis indicates that compound 3
exists as a mixture of four components in solution in approxi-
mately a 1:1:1:1 ratio, but in one or more conformations, weak
to moderate NOE’s are evident between the cyclohexyl and quinaz-
oline rings. This suggests a U-shaped structure is among the low-
energy conformations present in solution. In contrast, the 1H
NMR spectrum of 4 consists of a single conformation at room tem-
to control macromolecular structure, including the helical nature of
DNA,12 the secondary and tertiary structure of proteins,13 crystal
packing of aromatic molecules,14 and architectural organization in
supramolecular chemistry.15 In drug discovery, there has been much
attention given to intermolecular
p-stacking between a pharmaceu-
tical agent and its biological target.16 However, much less focus has
been placed on intramolecular p-stacking interactions that control
the conformational preferences of drug molecules.17
Quantum mechanical studies have reproduced experimental
binding energies for the benzene dimer with relatively high accu-
racy depending on the level of theory employed.18 Therefore, we
perature, with no evidence of
NOE correlations or shielding effects on the chemical shift of key
protons. Taken together, this suggests that -stacking is neither re-
p-stacking as indicated by a lack of
p
quired, nor solely sufficient, to favor the U-shaped structure as a
low-energy conformation.
There are no pertinent examples of 1,4-diazepane structures in
the Cambridge Structural Database that contain an exocyclic amide
carbonyl attached to a ring nitrogen. However, there are several
non-amide containing structures, and in each case, the central ring
exists as a chair or twist-chair.21 In contrast, the diazepane ring of
1 exists in a twist-boat conformation, as illustrated in Figure 4.
It is not clear why an exocyclic amide might trigger a change in
ring conformation from chair to boat, but the increased p-stacking
potential of 1 and 2 relative to 4, and the ability of 3 to display the
cyclohexyl group in proximity to the quinazoline ring, is consistent
with this supposition. Accordingly, in the boat conformation, the
ring nitrogen substituents occupy flagpole positions and are there-
fore in close proximity, readily available to engage in a p-stacking
interaction. However, if the ground state is a chair conformation,
an energetic penalty must be paid to change the ring conformation
in order to bring the aryl moieties into proximity. The NMR data for
compound 4 suggest that in this case, the energetic penalty of
adopting a boat conformation is not overcome by a favorable
stacking interaction.
p-
In summary, the data generated by NMR spectroscopy, X-ray
crystallography, and molecular modeling suggest that the U-
shaped structure is a low-energy conformation of orexin receptor
antagonists 1 and 2. Although this provides an interesting perspec-
tive on a novel class of biologically active compounds, a key unan-
Figure 4. X-ray crystal structure of 1 illustrating the twist-boat conformation of the
central diazepan ring, as well as the overall U-shaped structure present in the solid
state.

3000
C. D. Cox et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2997–3001
swered question for the rational design of next generation orexin
receptor antagonists is whether the bioactive conformation is sim-
ilar to the U-shaped structure. If so, rational design of antagonists
with geometric constraints to enforce this conformation may lead
to increased potency and improved pharmaceutical properties.22
To test our hypothesis that the bioactive conformation of diaze-
pane orexin receptor antagonists is similar to the low-energy con-
formation observed spectroscopically for 1 and 2, we designed and
synthesized a macrocyclic analog that is locked in a U-shaped
geometry, as illustrated in Scheme 1.
Preparation of the macrocycle begins with a copper-catalyzed
C–N cross-coupling between 5-bromo-2-iodobenzoic acid (5) and
1,2,3-triazole.23 After esterification, the resulting aryl bromide (6)
is allylated under standard Stille conditions and the ester hydro-
lyzed to provide lithium salt 7. In a parallel sequence, the coupling
partner is prepared beginning with a cross-coupling of BOC-homo-
piperazine (8) with 2,6-dibromonaphthalene to provide 9. A Negi-
shi coupling with pentenylzinc bromide yields olefin 10. Following
deprotection, this intermediate is coupled with 7 to provide ring-
closing metathesis (RCM) precursor 11. Treatment of 11 with the
Zhan-1B catalyst, followed by hydrogenation of the resulting al-
kene, provides macrocycle 12 in low yield.24
two major conformers in a 1:1 ratio, with two minor components
making up less than 4% of the mixture. NOE correlations are diffi-
cult to identify and assign due to rapid magnetization transfer, but
the C6 phenyl protons of the major conformers are observed at
5.00 and 5.49 ppm, supporting the fact that 12 exists in a confor-
mation similar to the low-energy conformations of 1 and 2 in
which the aromatic groups are in close proximity.
In the orexin receptor binding assays, the RCM precursor 11
displays activity against OX1R and OX2R of 63 nM and 120 nM,
respectively, whereas the corresponding values for macrocycle
12 are 51 nM and 42 nM. Thus, closing the macrocycle and
enforcing the U-shaped conformation is potency neutral on
OX1R and provides a slight potency increase on OX2R. The obser-
vation that 12 does not obtain the levels of potency observed for
1 and 2 may be a result of the macrocycle geometry not perfectly
mirroring the bioactive conformation, or simply that the satu-
rated carbon linker is not fully accommodated by the receptor.
Evidence in support of the latter conclusion is provided by the
observation that 11 lost significant potency relative to its simpli-
fied analog 2. Nevertheless, the fact that the geometrically con-
strained macrocycle 12 binds to both receptors with significant
levels of potency supports our hypothesis that the U-shaped con-
COSY, HMBC, and NOE correlations, as well as HRMS, confirm
the structure of 12. In DMSO-d6 solution at 75 °C, 12 exists as
formation determined for
conformation.
1
and 2 mimics the bioactive
N
N
O
I
N
N
O
N
O
N
c, d
HO
a, b
LiO
O
Br
Br
7
5
6
O
O
N
O
N
N
O
f
e
HN
Br
9
8
O
N
N
N
N
O
N
O
N
g, h
N
11
10
6
N
N
i, j
N
N
O
N
12
Scheme 1. Synthesis of the macrocycle. Reagents and conditions: (a) 1H-1,2,3-triazole, CuI, K3PO4, DMF, 60 °C; (b) HCl(g), MeOH (36% for two steps); (c) tetraallyltin,
tetrakis(triphenylphosphine) Pd(0), LiCl, DMF, 125 °C (83%); (d) LiOH, THF/MeOH/H2O, 45 °C (quant.); (e) 2,6-dibromonaphthalene, NaOtBu, BINAP, Pd2(dba)3, PhMe, 100 °C
(46%); (f) 4-pentenylzinc bromide, Pd(PtBu3)2, THF, 75 °C (38%); (g) HCl, Et2O; (h) 7, EDC, HOBt, TEA, DMF, 50 °C (63% for two steps); (i) Zhan IB, DCE, 40 °C (12%); (j) H2, Pd/C,
EtOAc (87%).

C. D. Cox et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2997–3001
3001
9. Compound 1:
b = 19.7509(4), c = 15.4723(3) Å, b = 91.3620(10)°, V = 4063.80(13) Å3, Z = 8,
Dx = 1.352 g cmÀ3
monochromatized radiation k(Mo) = 0.71073 Å,
= 0.09 mmÀ1, F(0 0 0) = 1744, T = 100 K. Data were collected on a Bruker
CCD diffractometer to a h limit of 28.31° which yielded 63565 reflections. There
are 10056 unique reflections with 5738 observed at the 2 level. The structure
was solved by direct methods (SHELXS-97, Sheldrick, G.M. Acta Crystallogr., 1990,
A46, 467–473) and refined using full-matrix least-squares on F2
C23H23N7O, Mr = 413.475, monoclinic, P21/c, a = 13.3019(2),
In conclusion, we found that N,N-disubstituted-1,4-diazepane
orexin receptor antagonists exist in a U-shaped conformation as
,
a result of favorable
p-stacking interactions and the adoption of
l
a twist-boat ring conformation. These effects result in a low-energy
conformation that resembles the bioactive conformation, reducing
the conformational entropy required to mold the structure into a
binding orientation. The lessons learned herein provide inspiration
for the synthesis of ‘bridged’ diazepane analogs in which a confor-
mational constraint is installed within the core to enforce or lock-
in the U-shaped conformation. Results of this effort will be the sub-
ject of a future report from our group.
r
(
SHELXL-97,
Sheldrick, G.M. SHELXL-97. Program for the Refinement of Crystal Structures. Univ.
of Göttingen, Germany). There are two independent molecules in the
asymmetric unit, related by a pseudo inversion center, with no significant
conformational differences between them. The final model was refined using
561 parameters and all 10056 data. All non-hydrogen atoms were refined with
anisotropic thermal displacements. The final agreement statistics are: R = 0.074
(based on 5738 reflections with I > 2
max = 0.12. The maximum peak height in a final difference Fourier map is
1.284 eÅÀ3
located within the quinazoline ring, and is without chemical
r(I)), wR = 0.165, S = 1.02 with (D/
r
)
,
Acknowledgments
significance. CCDC 723977 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
10. Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds; John Wiley &
Sons: New York, 1994.
The authors would like to thank Ms. Nancy Tsou for growing
diffractable crystals of 1, Dr. Rodney Bednar, Ms. Wei Lemaire,
and Mr. Joe Bruno for determining binding potencies of key com-
pounds, and Dr. Charles Ross and Ms. Joan Murphy for high resolu-
tion mass spectral analyses.
11. It is well known that intermolecular p-stacking can influence supramolecular
structure in the solid state. See: Goswami, S.; Gupta, V. K.; Sharma, A.; Gupta, B.
D. Bull. Mater. Sci. 2005, 28, 725.
12. Kool, E. T. Annu. Rev. Biophys. Biomol. Struct. 2001, 30, 1.
13. (a) Burley, S. K.; Petsko, G. A. Science 1985, 229, 23; (b) Thomas, A.; Meurisse,
R.; Charloteaux, B.; Brasseur, R. Proteins 2002, 48, 628.
14. Desiraju, G. R.; Gavezzotti, A. J. Chem. Soc., Chem. Commun. 1989, 621.
15. Roesky, H. W.; Andruh, M. Coord. Chem. Rev. 2003, 236, 91.
16. (a) Kryger, G.; Silman, I.; Sussman, J. L. Structure 1999, 7, 297; (b) Rosenfeld, R.
J.; Garcin, E. D.; Panda, K.; Andersson, G.; Aberg, A.; Wallace, A. V.; Morris, G.
M.; Olson, A. J.; Stuehr, D. J.; Tainer, J. A.; Getzoff, E. D. Biochemistry 2002, 41,
13915.
17. Ala, P. J.; Gonneville, L.; Hillman, M.; Becker-Pasha, M.; Yue, E. W.; Douty, B.;
Wayland, B.; Polam, P.; Crawley, M. L.; McLaughlin, E.; Sparks, R. B.; Glass, B.;
Takvorian, A.; Combs, A. P.; Burn, T. C.; Hollis, G. F.; Wynn, R. J. Biol. Chem. 2006,
281, 38013.
18. Sinnokrot, M. O.; Sherrill, C. D. J. Phys. Chem. A 2006, 110, 10656.
19. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.;
Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M.
C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.;
Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.;
Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. GAUSSIAN 98, Gaussian, Inc.: Pittsburgh, PA, 1998.
20. Sanders, J. The value provided in Ref. 18 is—4.1 kcal/mol, but was determined at
a slightly lower level of theory.
References and notes
1. (a) Bingham, M. J.; Cai, J.; Deehan, M. R. Curr. Opin. Drug. Dis. Dev. 2006, 9, 551;
(b) Roecker, A. J.; Coleman, P. J. Curr. Top. Med. Chem. 2008, 8, 977; (c) Boss, C.;
Brisbane-Roch, C.; Jenck, F.; Aissaoui, H.; Koberstein, R.; Sifferlen, T.; Weller, T.
Chimia 2008, 62, 974; (d) Boss, C.; Brisbane-Roch, C.; Jenck, F. J. Med. Chem.
2009, 52, 891.
2. (a) Whitman, D. B.; Cox, C. D.; Breslin, M. J.; Brashear, K. M.; Schreier, J. D.;
Bogusky, M. J.; Bednar, R. A.; Lemaire, W.; Bruno, J. G.; Hartman, G. D.; Reiss, D.
R.; Harrell, C. M.; Kraus, R. L.; Li, Y.; Garson, S. L.; Doran, S. M.; Prueksaritanont,
T.; Li, C.; Winrow, C. J.; Koblan, K. S.; Renger, J. J.; Coleman, P. J. ChemMedChem,
in press. doi: 10.1002/cmdc.200900069.; (b) These analogs were first disclosed
in: Brashear, K. M; Coleman, P. J.; Cox, C. D.; Smith, A. M.; Whitman, D. B. PCT
WO2007/126935 A2, 2007.; (c) Some of these results were disclosed in a recent
poster presentation: Breslin, M. J.; Cox, C. D.; Whitman, D. B.; Brashear, K. M.;
Bogusky, M. J.; Bednar, R. A.; Lemaire, W.; Hartman, G. D.; McGaughey, G.;
Reiss, D. R.; Harrell, C. M.; Winrow, C. J.; Koblan, K. S.; Renger, J. J.; Coleman, P. J.
Abstracts of Papers, 236th ACS National Meeting, Philadelphia, PA, United States,
August 17–21, 2008 (2008); MEDI 176.
3. It is known that restricted rotation around exocyclic partial double bonds to
heterocyclic six-membered rings can be observed on the NMR time scale; see
Kleinpeter, E.; Spitzner, R.; Schroth, W. Magn. Reson. Chem. 1987, 25, 688.
Additionally, conformational studies on compound 2 further supports hindered
rotation about the exocyclic bond to the quinazoline as a complicating factor in
the analysis of 1..
21. A search of the CSD for 1,4-diazepanes with no substitution on the ring carbons
and an exocyclic amide carbonyl at one of the ring nitrogens yields no hits. The
same search, but with the exocyclic amide constraint relaxed, retrieves several
structures, including ADASUU, ALAFUO, and QAFGUA in which the central ring
adopts a chair conformation. This analysis excludes organometallic complexes
in which the ring nitrogens are involved in a metal complex, forcing the
structure to adopt a boat conformation. See for example: Guo, Y. M.; Du, M.; Bu,
X.-H. J. Mol. Struct. 2002, 610, 27.
4. To determine the inherent isotropic shift of the C6 phenyl and aromatic methyl
protons, we synthesized and analyzed compound A in which no
p-stacking
interaction is available. The C6 phenyl and aromatic methyl protons appear at
7.23 and 2.43 ppm, respectively.
N
O
N
N
N
22. For instance, decreased molecular flexibility, as measured by the number of
rotatable bonds, may lead to improved bioavailability. See: Veber, D. F.;
Johnson, S. R.; Cheng, H.-Y.; Smith, B. R.; Ward, K. W.; Kopple, K. D. J. Med.
Chem. 2002, 45, 2615.
23. Antilla, J. C.; Baskin, J. M.; Barder, T. E.; Buchwald, S. L. J. Org. Chem. 2004, 69,
5578.
7.87
7.23
A
7.42
Me
2.43
5. Molecular modeling studies utilized the conformational searching algorithm
implemented in Maestro v8.5 (mixed torsional/low-mode sampling) with a
constant dielectric constant of 4 in chloroform using the MMFFs force field. The
conformation pictured in Figure 1C is 0.2 kcal/mol higher in energy than the
global minima, a structure with a similar overall conformation, but in which
24. Interestingly, direct RCM of 11 is expected to result in a macrocycle with six
alkyl carbons bridging the aryl groups, rather than the five carbon bridge
present in 12, but this product is not observed. A well known side reaction in
difficult RCM reactions is olefin isomerization, possibly due to the presence of a
ruthenium hydride species formed by decomposition of the catalyst, see, for
example: (a) Hong, S. H.; Day, M. W.; Grubbs, R. H. J. Am. Chem. Soc. 2004, 126,
7414; (b) Hong, S. H.; Sanders, D. P.; Lee, C. W.; Grubbs, R. H. J. Am. Chem. Soc.
2005, 127, 17160. Apparently, the six carbon-bridged structure is unable to
form under these conditions. On the other hand, the five-carbon bridged
macrocycle is accessible, but its formation requires isomerization of one of the
double bonds prior to ring closing, likely accounting for the low yield for this
process. Attempts to cyclize analogs of 11 that would have led to a four- or
seven-carbon bridged structure (i.e., one carbon-extended or two carbon-
deleted versions of 11) led to no isolatable RCM products, further supporting
the conclusion that the five-carbon bridged structure is optimal.
the triazole ring is
support the existence of the triazole
p
-stacking with the quinazoline ring. There is no evidence to
-stacked conformation in solution by
p
either NOE correlations or resonance shifts.
6. McGaughey, G. B.; Gagne, M.; Rappe, A. K. J. Biol. Chem. 1998, 273, 15458.
7. Meyer, E. A.; Castellano, R. K.; Diederich, F. Angew. Chem., Int. Ed. 2003, 42,
1211.
8. Examination of both computational and hand-held models of 2-rotamer B
indicate that geometrical constraints of this amide rotamer prevent the aryl
rings from getting into the close proximity seen in 2-rotamer A, in line with
NMR data suggesting a weaker p-stacking interaction in 2-rotamer B.