Hunting the emesis and efficacy targets of PDE4 inhibitors: Identification of the photoaffinity probe 8-(3-azidophenyl)-6-[(4-iodo-1H-1-imidazolyl)methyl]quinoline (APIIMQ) [2]

Full text of DOI:10.1021/jm000065c

3820
J . Med. Chem. 2000, 43, 3820-3823
Ch em istr y. Rolipram (1) is commercially available.
The compounds 2, 3 (KF17625), 4 (KF18280), and 5
shown in Schemes 1 and 2 and Table 1, were prepared
as described in the literature.6 Alkylation of compounds
1 and 3 with NaH and 2 in DMF afforded compounds 6
and 7 (Scheme 1).
Hu n tin g th e Em esis a n d Effica cy Ta r gets
of P DE4 In h ibitor s: Id en tifica tion of th e
P h otoa ffin ity P r obe 8-(3-Azid op h en yl)-6-
[(4-iod o-1H-1-im id a zolyl)m eth yl]qu in olin e
(AP IIMQ)
Reduction of the nitro group in compound 5 resulted
in the aniline 8 (Scheme 2), which was converted to the
aryl azide 9 by diazotization and displacement with
azide. The 3-azido-6-iodo compound 10 was similarly
prepared, except that the aniline 8 was iodinated with
ICl prior to its conversion to the azide.
Compounds 15, 16, and 18 were prepared using
chemistry similar to that described in the literature6b
for compounds in the class of 5, except that the iodo and
azido substituents were introduced by modification of
an aniline, a nitro group, or a carboxylic acid (Schemes
3 and 4).
Compounds 20 and 23 were produced using similar
chemistry except that the imidazole and iodoimidazole
substituents were introduced by displacement of the
bromomethylquinolines 19 and 21 (prepared as in the
literature6b) with imidazole and iodoimidazole, respec-
tively (Scheme 5). The displacement with 5-iodoimid-
azole yielded the 4-iodo 22a as the major product and
the 5-iodo 22b as the minor product. The regiochemistry
of the two iodoimidazoles 22a ,b was determined using
NOE experiments. In all the azide compound sytheses,
care was taken to design the synthesis such that the
light-sensitive azido group was introduced in the last
step.
Conversion of 127I-23 to 125I-23 was accomplished by
palladium-catalyzed tributylstannylation of the iodo-
imidazole to give compound 24 followed by radio-
iodination with Na125I and chloramine-T (Scheme 6).
Recom bin a n t Hu m a n P DE4. PDE4A248 was ex-
pressed as a GST-fusion protein from SF9 cells and
purified to near homogeneity as previously described.5b
P DE Assa y. The compound’s potency on inhibiting
PDE4 catalysis was determined using purified GST-
PDE4A248 as previously described.5d
Resu lts a n d Discu ssion . A number of different
functional groups have been described in the literature
for photolabeling target proteins. These include the aryl
azide, the trifluoromethyl diazerene, and the benzophe-
none moiety.7 One of the most common radioisotopes
with a high specific activity, leading to a high detection
sensitivity, incorporated in photoaffinity probes of this
type is 125I. Because these groups have significant
spatial requirements, it is important to identify posi-
tions on the inhibitor of interest where the groups are
tolerated. To obtain a compound containing these groups
that is also emetic and effective in our functional model,
we chose to work on several different classes of known
emetic PDE4 inhibitors.
The first class of inhibitors we considered was that
based on the prototypical compound rolipram (1), found
to be emetic in both animal models and humans.4
Incorporation of an N-(3-iodo-5-trifluoromethyldiazer-
enobenzyl) substituent on rolipram (Scheme 1) led to a
10-fold loss in potency (see 6 in Table 1). Compound 6
Dwight Macdonald,* He´le`ne Perrier, Susana Liu,
France Laliberte´, Roberta Rasori, Annette Robichaud,
Paul Masson, and Zheng Huang
Merck Frosst Centre for Therapeutic Research, P.O. Box
1005, Pointe Claire-Dorval, Que´bec, Canada, H9R 4P8
Received February 16, 2000
In t r od u ct ion . Type 4 cAMP-specific phosphodi-
esterase (PDE4) is an enzyme responsible for the
hydrolysis of the second messenger c-AMP to AMP in
many cell types.1 Inhibition of this enzyme can signifi-
cantly increase the intracellular c-AMP concentration,
leading to major alterations in cell biochemistry and
function. In particular, some inflammatory processes
can be attenuated with PDE4 inhibitors. For example,
LPS-stimulated TNF-R release in human blood mono-
nuclear cells can be blocked with PDE4 inhibitors.2
Antigen-induced bronchospasm is another pharmaco-
logical event that can be attenuated using PDE4 inhibi-
tors.3 Because both inflammation and bronchoconstric-
tion are major factors in asthma, PDE4 is a promising
therapeutic target for this common and serious disease.
One of the major issues with the development of PDE4
inhibitors, however, has been the side effect of emesis
observed for several prototypical compounds.4 More
recently, it has become clear that some PDE4 inhibitors
are less emetic than others while possessing the same
potency.4,5 For several years, efforts to improve the
therapeutic window of PDE4 inhibitors have involved
the identification of compounds that appeared to be
more potent for inhibiting the enzyme activity and less
potent on the high-affinity rolipram-binding site of the
enzyme.1 Recently, however, it has been shown that the
high-affinity rolipram-binding site of the enzyme is
simply the cofactor-bound active site on the holoenzyme.
The conformational difference between the PDE4 apoen-
zyme and the holoenzyme is responsible for the dif-
ferential binding of inhibitors.5d,e Other recent efforts
have been directed toward the identification of isozyme-
selective compounds.1a
In this communication, we introduce a new approach
to improving the therapeutic window of PDE4 inhibi-
tors, which is aimed at the identification of the specific
targets for emesis and efficacy. To this end we have
prepared an emetic, efficacious, and competitive PDE4
inhibitor (23, APIIMQ) capable of covalently tagging its
biological targets upon photoactivation. This provides
the possibility of identifying the emesis and efficacy
targets of PDE4 inhibitors. To our knowledge, this is
the first reported example of the preparation of a highly
emetic and efficacious PDE4 photoaffinity probe.
* To whom correspondence should be addressed. Tel: 514-428-3089.
Fax: 514-428-4900. E-mail: dwight_macdonald@merk.com.
10.1021/jm000065c CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/30/2000

Communications to the Editor
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21 3821
Sch em e 1a
a
Reagents and conditions: (a) NaH, DMF, rt, 1 h.
Sch em e 2a
Ta ble 1. PDE4A Potency and Emetic Threshold in Ferrets
emetic threshold in ferreta,b
compd PDE4Aa IC50 (nM) dose (mg/kg, po) plasma concn (µM)
1
4 ( 2
35 ( 4
0.1 ( 0.05
>10d
1 ( 0.5
>10d
<0.3c
6
7.0 ( 0.2
4e
7
0.6 ( 0.1
250 ( 20
0.07 ( 0.05
0.2 ( 0.1
3.3 ( 0.5
15 ( 5
0.2 ( 0.1
5
9
10
15
16
18
20
23
0.1 ( 0.05
0.1 ( 0.05
>30d
16.0 ( 0.2
0.8 ( 0.1
<0.3c
>10d
>10d
>10d
28 ( 5
2.7 ( 0.2
9 ( 3
<0.3c
0.1 ( 0.05
0.1 ( 0.05
0.4 ( 0.1
<0.2c
a
b
d
N g 2. See ref 8 for method. c Limit of detection. No emesis
was observed at the dose specified. e Structure of 4:
a
Reagents and conditions: (a) Fe, NH4Cl, EtOH, H2O, reflux;
(b) AcOH, NaNO2, H2O, 0 °C, 15 min; (c) NaN3, 0-25 °C, 2 h; (d)
ICl, CaCO3, MeOH, H2O, rt, 2 h.
was also not emetic at a dose 100-fold higher than the
emetic threshold of rolipram in the ferret.
Another highly potent and emetic PDE4 inhibitor is
compound 4 (KF18280).6a The same substituent as that
used on rolipram to give compound 7 (Scheme 1) also
resulted in a considerable loss in potency (Table 1).
two positions. The resulting compounds 16 and 18
(Schemes 3 and 4) retain considerable potency (28 and
2.7 nM, respectively) but, unfortunately, are not emetic
at 10 mg/kg in the ferret. Introduction of a benzoyl in
place of the azido in 16 resulted in a compound with
modest potency (see 15 in Scheme 3) and still not emetic
at 10 mg/kg in the ferret.
The fact that compounds 16 and 18 were not detect-
able in the plasma when dosed in the ferret suggested
that the high lipophilicity and poor aqueous solubility
(leading to poor pharmacokinetics) may be responsible
for the lack of emesis from these compounds. To
introduce polarity into the molecule, we took note of the
fact that the highly polar imidazole-substituted com-
pound 20 (Scheme 5) has a very low emetic threshold.
Substitution of this compound with an iodo on the
imidazole ring and an azido on the phenyl ring afforded
compound 23 (Scheme 5), which is quite potent on
PDE4A (0.4 nM) and is also emetic in the ferret at very
low doses (0.1 mg/kg). Compound 23 is also potent in
our human whole blood assay for the inhibition of LPS-
The most potent class of inhibitors we have considered
is the arylquinoline, as typified by compound 5.6b
Substitution of the nitro group for an azido gave
compound 9 (Scheme 2). This compound not only retains
high potency (0.2 nM) but also has a very low emetic
threshold. The next requirement was to identify a
suitable position to locate the iodine substituent. Simple
substitution on the ring para to the azido by iodinating
the aniline intermediate 8 (Scheme 2) resulted in a loss
in potency (compound 10). Because the potency of these
compounds is very sensitive to substitution on the
8-phenyl, we next chose to place a relatively small group
only in the meta position, which tolerates such groups.
Another position that tolerates a variety of substituents
is the 6-position on the quinoline. We therefore intro-
duced the azido and iodo substituents at either of these

3822 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21
Communications to the Editor
Sch em e 3a
Sch em e 5a
a
Reagents and conditions: (a) NBS, DMF, rt, 18 h; (b) glycerol,
H2SO4, H2O, sodium 3-nitrobenzenesulfonate, 150 °C, 30 min; (c)
3-aminobenzeneboronic acid, Pd(PPd3)4, NaCO3, DME, reflux, 5
h; (d) HI, NaNO2, KI, PEG-200, CH2Cl2, 0-25 °C, 18 h; (e)
diphenylphosphoryl azide, Et3N, t-BuOH, reflux, 18 h; (f) TFA,
CH2Cl2, rt, 3 h; (g) AcOH, NaNO2, H2O, 0 °C, 15 min; (h) NaN3,
0-25 °C, 2 h.
Sch em e 4a
Reagents and conditions: (a) imidazole, THF, reflux, 3 h; (b)
phenylboronic acid, Pd(PPh3)4, NaCO3, DME, 75 °C, 18 h; (c)
5-iodoimidazole, Cs2CO3, DMF, rt, 18 h; (d) Fe, NH4Cl, THF,
EtOH, H2O, 70 °C, 40 min; (e) AcOH, NaNO2, H2O, 0 °C, 25 min;
(f) NaN3, 0-25 °C, 30 min.
a
Sch em e 6a
a
Reagents and conditions: (a) 3-nitrobenzeneboronic acid,
Pd(PPd3)4, NaCO3, DME, reflux, 6 h; (b) diphenylphosphoryl azide,
Et3N, t-BuOH, reflux, 10 h; (c) TFA, CH2Cl2, rt, 30 min; (d) HI,
NaNO2, KI, PEG-200, CH2Cl2, 0-25 °C, 2 h; (e) Fe, NH4Cl, EtOH,
H2O, reflux, 1 h; (f) AcOH, NaNO2, H2O, 0 °C, 30 min; (g) NaN3,
rt, 3 h.
a
Reagents and conditions: (a) Bu3SnSnBu3, Pd(dba)2, Ph3As,
induced TNF-R (IC50 ) 2 ( 1 µM)2 and in our guinea
pig ovalbumin-induced bronchoconstriction model (ED50
) 0.3 ( 0.2 mg/kg).3b To evaluate the protein labeling
ability of compound 23, we converted it to the 125I
version (Scheme 6).
DMF, 135 °C, 1 h; (b) Na125I, chloramine-T, DMF, H2O, rt, 1 h.
125I-23 specifically labels recombinant PDE4 in the
presence of other cellular proteins under photolysis
conditions. Shown in Figure 1 is the SDS gel radioactiv-
ity image of 100 µg of SF9 cell lysate proteins spiked
with 5 ng of GST-PDE4A248 after photolysis with 1 nM
125I-23. A band with a mobility of about 110 kDa that
comigrated with pure GST-PDE4A248 on Commassie
stain (data not shown) was specifically labeled after
photolysis (inside circle in lane 3). This band was absent
in the null cell lysate. No labeling was observed in
absence of light (data not shown) indicating that the
formation of the covalent adduct was photochemically
specific. An active-site-directed potent PDE4 inhibitor,
CDP-8405b (10 µM, Ki ∼ 4 nM), specifically competed
with the photolabeling of 125I-23 (lane 4), demonstrating
that 23 is an active-site-directed-specific photoprobe.
The probe is also highly sensitive and specific because
5 ng of GST-PDE4A248 was readily identified in the
presence of 100 µg of total cellular proteins. The probe’s
high detection sensitivity and its high emetic potency
in vivo suggest that 23 will be a powerful tool in
F igu r e 1. Radioactivity image of the 100 µg of cell lysate
spiked with 5 ng of GST-PDE4A248 photolabeled with 1 nM
125I-23: lane 1, 14C-labeled molecular weight marker; lane 2,
null cell lysate; lane 3, cell lysate spiked with 5 ng of GST-
PDE4A248; lane 4, cell lysate spiked with 5 ng of GST-PDE4A248
in the presence of 10 µM CDP-840. The two weaker bands
immediately below the intensely labeled GST-PDE4A248 are
protelytic fragments derived from GST-PDE4A248 as judging
from their positive immunoreactivity with a PDE4-specific
antibody.
studying the distribution of its high-affinity targets in
specific tissues and in associating these targets with its
biological responses such as efficacy and emesis. Further

Communications to the Editor
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 21 3823
(5) (a) Hughes, B.; Howat, D.; Lisle, H.; Holbrook, M.; J ames, T.;
Gozzard, N.; Blease, K.; Hughes, P.; Kingaby, R.; Warrellow, G.;
Alexander, R.; Head, J .; Boyd, E.; Eaton, M.; Perry, M.; Wales,
M.; Smith, B.; Owens, R.; Catterall, C.; Lumb, S.; Russell, A.;
Allen, R.; Merriman, M.; Bloxham, D.; Higgs, G. The inhibition
of antigen-induced eosinophilia and bronchoconstriction by CDP
840, a novel stereoselective inhibitor of phosphodiesterase type
4. Br. J . Pharmacol. 1996, 118, 1183-1191. (b) Perry, M. J .;
O’Connell, J .; Walker, C.; Crabbe, T.; Baldock, D.; Russell, A.;
Lumb, S.; Huang, Z.; Howat, D.; Allen, R.; Merriman, M.; Walls,
J .; Daniel, T.; Hughes, B.; Laliberte, F.; Higgs, G. A.; Owens, R.
J . CDP840: A Novel Inhibitor of PDE-4. Cell Biochem. Biophys.
1998, 29, 113-132. (c) Christensen, S. B.; Guider, A.; Forster,
C. J .; Gleason, J . G.; Bender, P. E.; Karpinski, J . M.; DeWolf,
W. E., J r.; Barnette, M. S.; Underwood, D. C.; Griswold, D. E.;
Cieslinski, L. B.; Burman, M.; Bochnowicz, S.; Osborn, R. R.;
Manning, C. D.; Grous, M.; Hillegas, L. M.; Bartus, J . O.; Ryan,
M. D.; Eggleston, D. S.; Haltiwanger, R. C.; Torphy, T. J . 1,4-
Cyclohexanecarboxylates: Potent and Selective Inhibitors of
Phosphodiesterase 4 for the Treatment of Asthma. J . Med.
Chem. 1998, 41, 821-835. (d) Laliberte´, F.; Han, Y.; Govindara-
jan, A.; Giroux, A.; Liu, S.; Bobechko, B.; Lario, P.; Bartlett, A.;
Gorseth, E.; Gresser, M.; Huang, Z. Conformational Difference
between PDE4 Apoenzyme and Holoenzyme. Biochemistry 2000,
39, 6449-6458. (e) Xu, R. X.; Hassell, A. M.; Vanderwall, D.;
Lambert, M. H.; Holmes, W. D.; Luther, M. A.; Rocque, W. J .;
Milburn, M. V.; Zhao, Y.; Ke, H.; Nolte, R. T. Atomic Structure
of PDE4: Insights into Phosphodiesterase Mechanism and
Specificity. Science 2000, 288, 1822-1825.
characterization of the probe in addition to its use in
the identification of biological targets in various tissue
types is underway.
Con clu sion . In conclusion, we have identified com-
pound 23 (APIIMQ) as a potent, efficacious, and com-
petitive radiolabeled inhibitor of PDE4, which covalently
labels the enzyme upon light activation. Compound 23
also has a very low emetic threshold (0.1 mg/kg, po, in
ferrets). This probe should be very useful for the
identification of the respective targets through which
PDE4 inhibitors cause emesis and also produce their
efficacy.
Ack n ow led gm en t. We are grateful to Dr. Michel
Gallant and Dr. Cameron Black for kindly providing
samples of compounds 2-4.
Su p p or tin g In for m a tion Ava ila ble: Synthetic chemistry
methods, biological procedures, and chemical characterization
data for compounds 22-24 is available free of charge via the
Internet at http://pubs.acs.org.
Refer en ces
(6) (a) Suzuki, F.; Kuroda, T.; Tamura, T.; Sato, S.; Ohmori, K.;
Ichikawa, S. New Antiinflammatory Agents. 2. 5-Phenyl-3H-
imidazo[4,5-c][1,8]naphthyridin-4(5H)-ones: A New Class of
Nonsteroidal Antiinflammatory Agents with Potent Activity Like
Glucocorticoids. J . Med. Chem. 1992, 35, 2863-2870. (b) Wil-
helm, R. S.; Fatheree, P. R.; Chin R. L. Quinolines as Type IV
Phosphodiesterase Inhibitors. Patent WO9422852, 1994. (c)
Gallant, M.; Sawyer, N.; Metters, K. M.; Zamboni, R. J . A novel
photoaffinity probe for the LTD4 receptor. Bioorg. Med. Chem.
1998, 6, 63-72.
(7) (a) Prestwich, G. D.; Dorman, G.; Elliott, J . T.; Marecak, D. M.;
Chaudhary, A. Benzophenone Photoprobes for Phosphoinositi-
des, Peptides and Drugs. Photochem. Photobiol. 1997, 65, 222-
34. (b) Brunner, J . New Photolabeling and Cross-linking Meth-
ods. Annu. Rev. Biochem. 1993, 62, 483-514.
(1) (a) Houslay, M. D.; Sullivan, M.; Bolger, G. B. The Multienzyme
PDE4 Cyclic Adenosine Monophosphate-Specific Phosphodi-
esterase Family: Intracellular Targeting, Regulation, and Selec-
tive Inhibition by Compounds Exerting Antiinflammatory and
Antidepressant Actions. Adv. Pharmacol. 1998, 44, 225-342. (b)
Spina, D.; Landells, L. J .; Page, C. P. The Role of Phosphodi-
esterase Enzymes in Allergy and Asthma. Adv. Pharmacol. 1998,
44, 33-89.
(2) Brideau, C.; Van Staden, C.; Styhler, A.; Rodger, I. W.; Chan,
C.-C. The effects of phosphodiesterase type 4 inhibitors on tumor
necrosis factor-R and leukotriene B4 in a novel human whole
blood assay. Br. J . Pharmacol. 1999, 126, 979-988.
(3) (a) J ones, T. R.; McAuliffe, M.; McFarlane, C. S.; Piechuta, H.;
Macdonald, D.; Rodger, I. W. Effects of a selective phosphodi-
esterase IV inhibitor (CDP-840) in
a leukotriene-dependent
nonhuman primate model of allergic asthma. Can. J . Physiol.
Pharmacol. 1998, 76, 210-217. (b) Danahay, H.; Broadley, K.
J . PDE4 inhibition and a corticosteroid in chronically antigen
exposed conscious guinea-pigs. Clin. Exp. Allergy 1998, 28, 513-
22.
(8) (a) King G. L. Animal models in the study of vomiting. Can. J .
Physiol. Pharmacol. 1990, 68, 260-268. (b) Robichaud, A.;
Tattersall, F. D.; Choudhury, I.; Rodger, I. W. Emesis induced
by inhibitors of type IV cyclic nucleotide phosphodiesterase (PDE
IV) in the ferret. Neuropharmacology 1999, 38, 289-297.
(4) Burnouf, C.; Pruniaux, M.-P.; Szilagyi, C. M. Phosphodiesterases
4 Inhibitors. Annu. Rep. Med. Chem. 1998, 33, 91-109.
J M000065C