Investigation of the binding pocket of human hematopoietic prostaglandin (PG) D2 synthase (hH-PGDS): A tale of two waters

Article abstract of DOI:10.1016/j.bmcl.2012.04.004

The inhibition of hH-PGDS has been proposed as a potential target for the development of anti-allergic and anti-inflammatory drugs. Herein we describe our investigation of the binding pocket of this important enzyme and our observation that two water molecules bind to our inhibitors and the enzyme. A series of compounds were prepared to the probe the importance of the water molecules in determining the binding affinity of the inhibitors to the enzyme. The study provides insight into the binding requirements for the design of potent hH-PGDS inhibitors.

Full text of DOI:10.1016/j.bmcl.2012.04.004

Bioorganic & Medicinal Chemistry Letters 22 (2012) 3795–3799
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Investigation of the binding pocket of human hematopoietic prostaglandin
(PG) D2 synthase (hH-PGDS): A tale of two waters
John I. Trujillo a, , James R. Kiefer a, , Wei Huang a, Jacqueline E. Day a, Joseph Moon d, Gina M. Jerome b,
Christine P. Bono c, Christine M. Kornmeier b, Melanie L. Williams c, Cyrille Kuhn e, Glen R. Rennie e,
Thomas A. Wynn e, Christopher P. Carron b, Atli Thorarensen a
⇑
⇑
a Departments of Medicinal Chemistry, Pfizer Global Research and Development, Chesterfield, MO 63017, United States
b Departments of Biochemical Pharmacology, Pfizer Global Research and Development, Chesterfield, MO 63017, United States
c Departments of Biochemistry/Enzymology, Pfizer Global Research and Development, Chesterfield, MO 63017, United States
d Departments of Computational Chemistry Pfizer Global Research and Development, Chesterfield, MO 63017, United States
e Research Technology Center, Cambridge, MA, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The inhibition of hH-PGDS has been proposed as a potential target for the development of anti-allergic
and anti-inflammatory drugs. Herein we describe our investigation of the binding pocket of this impor-
tant enzyme and our observation that two water molecules bind to our inhibitors and the enzyme. A ser-
ies of compounds were prepared to the probe the importance of the water molecules in determining the
binding affinity of the inhibitors to the enzyme. The study provides insight into the binding requirements
for the design of potent hH-PGDS inhibitors.
Received 9 January 2012
Revised 29 March 2012
Accepted 2 April 2012
Available online 13 April 2012
Keywords:
Ó 2012 Elsevier Ltd. All rights reserved.
Prostaglandin D2 synthase
PGDS inhibitors
Isoquinoline
Water displacement
Prostaglandin (PG) D-2 is formed from arachidonic acid by three
enzymatic reactions; the first two mediated by cyclooxygenase
(COX) to provide PGH-2, which then is transformed by PGD
synthase (PGDS) to PGD-2 (Fig. 1).1 PGD-2 is an important mediator
of allergic and inflammatory processes in mast cells and Th2
cells.2a PGD-2 serves as the ligand for two important receptors
expressed in allergic reactions and airway inflammation, DP1 and
CRTH2. When activated, DP1 causes contraction of smooth muscle
in the airway, while CRTH2 serves to mediate chemotaxis of eosin-
ophils and basophils into the lungs.2b
Two isoforms of human PGDS have been identified with several
key differences.3 H-PGDS (hematopoietic PGDS) is primarily found
in mast cells and Th2 cells, while L-PGDS (lipocalin PGDS) is
located in the central nervous system, genital organs and heart.
In addition, the enzymes differ by their dependence on the cofactor
glutathione (GSH) for activity, H-PGDS is GSH dependent while
L-PGDS does not utilize the cofactor.4 Studies have suggested
selective inhibitors of H-PGDS as potential anti-allergic and
anti-inflammatory drugs.5
Recent publications by Astra-Zeneca6 (1 and 2), Sanofi-
Aventis7(3 and 4), Evotec8 (5 and 6) Taiho9 (7) and Pfizer10 (8) have
disclosed inhibitors of hH-PGDS with
lM to nM potencies
( Fig. 2). Recently, we identified compound 9 Figure 3 through
screening our compound collection. The inhibitor was found to
possess excellent activity in an enzyme assay measuring inhibition
of PGD-2 production (IC50 = 2.34 nM). In addition, in silico property
modeling suggested that the compound would have favorable PK
and ADME characteristics. The published crystal structures of
H-PGDS revealed two solvent molecules consistently located adja-
cent to the inhibitors, the primary water molecule forming a
hydrogen bond to a nitrogen in the compound. The second water
forms part of a network of hydrogen bonds to the primary water
molecule.5b,6,8a,11
We determined the 1.8 Å resolution crystal structure of the iso-
quinolyl inhibitor (9), and its binding mode preserves the water
network and the hydrogen bond to the compound (Fig. 4a). The pri-
mary water molecule is tetra-coordinated, as in complexes with
other potent inhibitors. Additional interactions between the inhib-
itor and protein include p–p stacking of the central phenyl ring of
the inhibitor with the sidechain of Trp104. Van der Waals interac-
tions dominate the other protein–inhibitor contacts, except for a
hydrogen bond between the amide nitrogen of the inhibitor and
the oxidized glutathione species12 bound adjacent to the active
⇑
Corresponding authors. Present address: Genentech, Inc., Dept. of Structural
Biology, 1 DNA Way, MS:27, South San Francisco, CA 94080.
E-mail addresses: john.i.trujillo@pfizer.com (J.I. Trujillo), kiefer.james@gene.com
(J.R. Kiefer).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.04.004

3796
J. I. Trujillo et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3795–3799
Figure 1. Role of PGD-2 in airway inflammation.
Figure 2. Structures of selected H-PGDS inhibitors in the literature.
protein–ligand structures can stabilize the complex, as found in
the structure of human dihydrofolate reductase with methotrex-
ate.14 Alternatively, the water molecules can be displaced by the li-
gand, leading to an enhancement in binding affinity due to the
entropic gain from the displacement of water to the bulk solvent.
The displacement of water has been observed to also cause a de-
crease in binding affinity, for example, in the binding of tripeptide
inhibitors to OppA.14 Indeed the decision to engage or displace
water in inhibitor design is an area of much investigation. Recently,
attempts have been made to computationally classify water mole-
cules in the protein binding sites with the outcome being the selec-
tive targeting of water molecules for displacement or engagement
to achieve greater ligand affinity.15,16 Thus far, published struc-
tures for H-PGDS with nanomolar potency inhibitors all showed
the hydrogen bond to the primary water molecule. To probe
whether this interaction was strictly required or whether instead
the water(s) could be displaced or replaced by inhibitor atoms,
we synthesized a series of analogs (Fig. 4) designed to perturb
the solvent molecules. The analogs probe the region near the struc-
tural water molecules, as well as the hydrophobic region occupied
Figure 3. Structure of screening lead and relevant data.
site. The morpholino tail of the compound partially contacts
the protein and otherwise remains solvent exposed. These
observations are analogous to those observed by researchers at
Astra-Zeneca for compound 16 and Evotec for compound 6.8
It is well established that solvent mediated protein–inhibitor
interactions can be critical for inhibitor potency, as seen in early
HIV protease inhibitors.13 For example, water molecules in

J. I. Trujillo et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3795–3799
3797
Figure 4. Crystal structures of isoquinoline and naphthalene inhibitors in complex with H-PGDS. In each case, the complex with 9 is shown as gray carbons and surface as a
reference, and the comparator complex is orange. (a) Compound 9 coordinates to the primary solvent molecule. (b) Addition of a methyl group (11) slightly shifts the inhibitor
position, but hydroxymethyl (13, c) and aminomethyl (14, d) substitutions replace the primary water molecule.
by the B-ring of the isoquinoline, attempting to displace or replace
the primary solvent molecule.
Crystal structures of the H-PGDS complexes with compounds 11,
13, and 14 were determined to 2.0 Å or better resolution
(Fig. 4b–d).
The syntheses of the compounds (10–14 Fig. 5) are described in
Scheme 1. Thus N-(2-morpholinoethyl)-4-(4,4,5,5-tetramethyl-
1,3,2-dioxaborolan-2-yl)benzamide 15 was prepared by an amide
coupling of 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)ben-
zoic acid with 2-morpholinoethanamine in 95% yield. The halo-
heterocycles (16–20) were obtained from commercial sources,
while compound (20) was prepared from 1-bromo-2-(bromo-
methyl)naphthalene by displacement of the benzylic bromine with
the anion generated from di-tert-butyl iminodicarbonate. The
boronate (15) was coupled to the halo-heterocycles 16–20 under
Suzuki conditions (PdCl2(dppf), dioxane/H2O, K3PO4, 100 °C)17 to
give the compounds 10–14. For compound 14 the Boc-group was
removed with TFA/DCM to give the desired compound in 88% yield.
Compounds 10–14 were tested for their ability to inhibit the
production of PGD-2 (see Supplementary data) in an enzyme assay
(see Table 1).18 Removal of the B-ring of the isoquinoline, com-
pound 10, resulted in a 100-fold reduction in activity, suggesting
the loss of a significant hydrophobic contribution to binding. Incor-
poration of a methyl group at the 3-position of the isoquinoline
ring (11) resulted in a modest reduction in activity. The crystal
structure of this complex showed displacement of the auxiliary
water molecule by the methyl group (see Fig. 4b). In addition,
the compound rotated clockwise within the binding site approxi-
mately 5° to accommodate the methyl group. Conversion of the
isoquinoline ring system to a naphthalen-2-ol ring system (12)

3798
J. I. Trujillo et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3795–3799
Figure 5. Structures of compounds 10–14.
Compounds 13 and 14 were prepared to displace the primary
water molecule and engage the protein by a direct H-bond to the
inhibitor. The crystal structures of both compounds (Fig. 4 c and
d) clearly revealed that the design goal had been achieved, and that
the hydroxyl and amino groups, respectively, had indeed replaced
the primary solvent molecule. In fact, both complexes retained
three apparent H-bonds to the inhibitor heteroatom (the fourth
bond from the original water being replaced by the covalent bond
to the methylene linker), mimicking the role of the original water
molecule. In spite of the apparent design success represented in
the crystal structures, the inhibitor potency worsened by 630-
and 360-fold, respectively. Closer inspection of the geometry of
the inhibitor in the crystal structure shows that both dihedral an-
gles linking the pendant heteroatom to the naphthalene ring are
nearly in-plane (21° and 27°, respectively) compared to the pre-
ferred angle of approximately 90°. Similarly, the dihedral angle
bridging the naphthalene and phenyl rings deviates ꢀ20° from
its low energy conformation (117° vs 97° for 14). The internal
strain in the compound in these cases appears to substantially out-
weigh the benefit of the entropic gain from solvent displacement.
Our assay and structural data strongly suggest that maximal
affinity for hH-PGDS requires inhibitors to form a hydrogen bond
to the primary water molecule buried in the active site. Further,
it appears that the geometric and chemical constraints on the envi-
ronment of that solvent molecule preclude facile replacement with
inhibitor atoms. By contrast, the displacement of the auxiliary
water molecule is well tolerated by the ligand–protein complex
with only a modest drop in inhibitor affinity. In conclusion, com-
pound 9 has been identified as a potent inhibitor of hH-PGDS. Its
H-bond to the primary water molecule appears key to its binding
affinity.19 The analogs probing the contribution of the solvent
molecules to inhibitor affinity serve as important tools elucidating
the binding requirements for potent inhibitors of hH-PGDS. The
information presented serves as a basis on which to design future
potent inhibitors of hH-PGDS.20
Scheme 1. Synthesis of compounds 9–14.
Table 1
Calculated properties and biological activity of compounds 9–14
Compd
hH-PGDS IC50, (nM)a
LEb
clogPb
LipEb
HLMc
9
2.34
274
8.26
662
0.44
0.39
0.39
0.30
0.27
0.29
3.18
2.01
3.68
3.42
2.98
2.97
5.45
4.60
4.40
2.76
2.85
3.10
90%
20%
76%
94%
–
10
11
12
13
14
1480
845
–
a
Values are means of three experiments.
b
LE (ligand efficiency) = 1.36 Â pIC50/# of heavy atoms; clogP = calculated par-
tition coefficient; LipE = pIC50–clogP.
Acknowledgements
c
Percent remaining after 30 min (human liver microsomes).
X-ray structure coordinates for all complexes have been depos-
ited into the RCSB with accession codes 4EEO (9), 4EDZ (11), 4EDY
(13), 4ECO (14). Data for two crystal structures described herein
were collected at IMCA-CAT (17-ID) beamline at the Advanced
resulted in a 280-fold reduction in activity. Attempts to obtain a
crystal of this ligand with the protein were not successful.

J. I. Trujillo et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3795–3799
3799
10. Blake, T.; Hamper, B.C.; Huang, W.; Kiefer, J.R.; Moon, J.B.; Neal, B.E.; Olson,
K.L.; Pelc, M.J.; Schweitzer, B.A.; Thorarensen, A.; Trujillo, J.I.; Turner, S.R. US
2008/0146569 A1.
11. Carron, C. P.; Trujillo, J. I.; Olson, K. L.; Huang, W.; Hamper, B. C.; Dice, T.; Neal,
B. E.; Pelc, M. J.; Day, J.; Rohrer, D. C.; Kiefer, J. R.; Moon, J. B.; Schweitzer, B. A.;
Blake, T. D.; Turner, S. R.; Woerndle, R.; Case, B. L.; Bono, C. P.; Dilworth, V. M.;
Funckes-Shippy, C. L.; Hood, B. L.; Jerome, G. M.; Kornmeier, C. M.; Radabaugh,
M. R.; Williams, M. L.; Davies, M. S.; Wegner, C. D.; Welsch, D. J.; Abraham, W.
M.; Warren, C. J.; Dowty, M. E.; Hua, F.; Zutshi, A.; Yang, J. Z.; Thorarensen, A.
ACS Med. Chem. Lett. 2010, 1, 59.
12. An oxidized glutathione species believed to be glutathione sulfinic acid is often
observed in H-PGDS inhibitor complexes with amide containing inhibitors.
This occurs predominantly when collection of data occurs at synchrotron
sources but has infrequently been observed from lower energy sources.
13. Lam, P. Y. S.; Jadhav, P. K.; Eyermann, C. J.; Hodge, C. N.; Yu, R.; Bacheler, L. T.;
Meek, J. L.; Otto, M. J.; Rayner, M. M.; Wong, Y. N.; Chang, C.-H.; Weber, P. C.;
Jackson, D. A.; Sharpe, T. R.; Erickson-Viitanen, S. Science 1994, 263, 380.
14. (a) Poornima, C. S.; Dean, P. M. J. Comput. Aided Mol. Des. 1995, 9, 500; (b)
Poornima, C. S.; Dean, P. M. J. Comput. Aided Mol. Des. 1995, 9, 513; (c)
Poornima, C. S.; Dean, P. M. J. Comput. Aided Mol. Des. 1995, 9, 521.
15. Ladbury, J. E. Chem. Biol. 1996, 3, 973–980.
Photon Source and was supported by the companies of the Indus-
trial Macromolecular Crystallography Association through a con-
tract with the Center for Advanced Radiation Sources at the
University of Chicago.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.04.
004.
References and notes
1. Giles, H.; Leff, P. Prostaglandins 1988, 35, 277.
2. (a) Nagai, H. Allergology 2008, 57, 187; (b) Oguma, T.; Asano, K.; Ishizaka, A.
Allergology 2008, 57, 307.
3. Urade, Y.; Eguchi, N. Prostaglandins Other Lipid Mediat. 2002, 68, 375.
4. For a review of hematopoietic prostaglandin D synthase see: Kanaoka, Y.;
Urade, Y. Prostaglandins Leukot. Essent. Fatty Acids 2003, 69, 163. and references
cited therein.
5. (a) Matsuoka, T.; Hirata, M.; Tanaka, H.; Takahashi, Y.; Murata, T.; Kabashima,
K.; Sugimoto, Y.; Kobayashi, T.; Ushikubi, F.; Aze, Y.; Eguchi, N.; Urade, Y.;
Yoshida, N.; Kimura, K.; Mizoguchi, A.; Honda, Y.; Nagai, H.; Narumiya, S.
Science 2000, 287, 2013; (b) Aritake, K.; Kado, Y.; Inoue, T.; Miyano, M.; Urade,
Y. J. Biological Chemistry 2006, 281, 15277.
16. (a) Barillari, C.; Taylor, J.; Viner, R.; Essex, J. W. J. Am. Chem. Soc. 2007, 129,
2577; (b) Raymer, M. L.; Shanschagrin, P. C.; Punch, W. F.; Venkataraman, S.;
Goodman, E. D.; Kuhn, L. A. J. Mol. Biol. 1997, 265, 445; (c) Garcia-Sosa, A. T.;
Mancera, R. L.; Dean, P. M. J. Mol. Model. 2003, 9, 172.
17. Suzuki, A. Mod. Arene Chem. 2002, 53.
18. Fluorescence intensity h-HPGDS TBA enzyme assay-prostaglandin D synthase
(PGDS) converts the substrate prostaglandin H2 (PGH2) to prostaglandin D2.
The depletion of PGH2 was measured via an Fe(II) reduction of the remaining
PGH2 to malondialdehyde (MDA) and 12-HHT. The enzyme assay is based on
the quantitative formation of a fluorescent complex from non fluorescent
compounds MDA and 2-thiobarbituric acid (TBA), essentially as described in
US patent application US 2004/152148 by Lambalot (see Supplementary data
for details).
6. Howhy, M.; Spadola, L.; Lundquist, B.; Hawtin, P.; Dahmen, J.; Groth-Clausen, I.;
Nilsson, E.; Persdotter, S.; von Wachenfeldt, K.; Folmer, R. H. A.; Edman, K. J.
Med. Chem. 2008, 51, 2178.
7. (a) Li, Z.; Xiong, J. Sabol, J.S.; WO 2004/016223 A2.; (b) Cao, Y.; Sabol, J.S.; Ayers,
S; WO 2006/015195 A1.; (c) Aldous, S.C.; Jiang, J.Z.; Lu, J.; Ma, L.; Mu, L.;
Munson, H.R.; Sabol, J.S.; Thurairatnam, S.; Vandeusen, C.L. WO 2007/041634
A1.; (d) Aldous, S.C.; Fennie, M.W.; Jiang, J.Z.; John, S.; Mu, L.; Pedgrift, B.;
Pribish, J.R.; Rauckman, B.; Sabol, J.S.; Stoklosa, G.T.; Thurairatnam, S.;
Vandeusen, C.L. WO 2008/121670 A1.; (e) Aldous, S.C.; Fennie, M.W.;Jiang,
J.Z.; John, S.; Pedgrift, B.; Pribish, J.R.; Rauckman, B.; Sabol, J.S.; Stoklosa, G.T.;
Thurairatnam, S.; Vandeusen, C.L. WO 2008/121670 A1.
8. (a) Hesterkamp, T.; Barker, J.; Davenport, A.; Whittaker, M. Curr. Top. Med.
Chem. 2007, 7, 1582; (b) Aicher, B.; Kelter, A.R.; Coulter, T.S.; Taylor, S.;
Davenport, A.J.; Hesterkamp, T.; Kirchhoff, C. WO 2008/122787 A1.
9. (a) Urade, Y.; Tanaka, Y.; Yamane, K.; Togawa, M. WO 2007/007778 A1.; (b)
Urade, Y.; Shigeno, K.; Tanaka, Y.; Kuze, J.; Tsuchikawa, M.; Hosoya, T. JP
2007051121 A.; (c) Yamane, K.; Tanaka, Y.; Shigeno, K.; Hosoya, T.; Inoue, S.;
Kitade, M.; Harada, T.; Aoyagi, H.; Miyoshi, N.; Mutoh, T.; Togawa, M.; Kiniwa,
M.; Yamasaki, Y. Abstracts of Papers, 235th National Meeting of the American
Chemical Society, New Orleans, LA, April 6–10, 2008; American Chemical
Society: Washington, DC, 2008; MEDI 26.
19. Bissantz, C.; Kuhn, B.; Stahl, M. J. Med. Chem. 2010, 53, 5061. For a recent
perspective on molecular interactions see.
20. (a) For recent publications regarding inhibitors of hHPGDS see: Smythe, M.L.;
Flanagan, J.U. WO 2011/150457.; (b) Urade, Y., Kitade, M.; Shigeno, K. Yamane,
K.; Tanaka, K. WO 2011/090062.; (c) Takada, Ch. Saeki, K.; Kawaski, H.; Tagaya,
M.; Komai, M.; Ikemura, T.; Yamaguchi, S.; Iida, K. WO 2011/043359.; (d)
Urade, Y.; Kitade, M.; Shigeno, K.; Yamane, K.; Tanaka, K. WO 2010/0104024.;
(e) Christ, A.; Labzin, L.; Bourne, G. T.; Fukunishi, H.; Weber, J. E.; Sweet, M. J.;
Smythe, M. L.; Flanagan, J. U. J. Med. Chem. 2010, 53, 5536; (f) Weber, J. E.;
Oakley, A. J.; Christ, A. N.; Clark, A. G.; Hayes, J. D.; Hall, R.; Hume, D. A.; Board,
P. G.; Smythe, M. L.; Flanagan, J. U. Eur. J. Med. Chem. 2010, 45, 447; (g) Endres,
G.W.; Lee, P.H.; Olson, K.L.; Kramer, J.B.; Ciske, F.L.; Barret, S.D. WO 2010/
033977 A2.; (h) Huang, W.; Mantell, S.J.; Strang, R.S.; Thorarensen, A.; Trujillo,
J.I.; Turner, S.R.; Yeap, S.K. WO 2009/153720 A1.; (i) Trujillo, J.I. et al.
manuscript in preparation.