Synthesis of PDE IV inhibitors. First asymmetric synthesis of two of GlaxoSmithKline's highly potent Rolipram analogues

Article abstract of DOI:10.1039/c3ob41646a

Asymmetric syntheses of two of GlaxoSmithKline's highly potent phosphodiesterase IV inhibitors CMPI 1 and CMPO 2 have been accomplished from nitroethane and simple precursors in 8 and 7 steps, respectively. The suggested synthetic strategy involves as a key stage the silylation of enantiopure six-membered cyclic nitronates. In vitro studies of PDE IVB1 inhibition revealed a significant difference in the activity of CMPI 1 and CMPO 2 enantiomers.

Full text of DOI:10.1039/c3ob41646a

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Synthesis of PDE IV inhibitors.† First asymmetric
synthesis of two of GlaxoSmithKline’s highly potent
Rolipram analogues‡
Cite this: Org. Biomol. Chem., 2013, 11,
8082
Petr A. Zhmurov,a Alexey Yu. Sukhorukov,*a Vladimir I. Chupakhin,b
Yulia V. Khomutova,a Sema L. Ioffea and Vladimir A. Tartakovskya
Asymmetric syntheses of two of GlaxoSmithKline’s highly potent phosphodiesterase IV inhibitors CMPI 1
and CMPO 2 have been accomplished from nitroethane and simple precursors in 8 and 7 steps, respecti-
vely. The suggested synthetic strategy involves as a key stage the silylation of enantiopure six-membered
cyclic nitronates. In vitro studies of PDE IVB1 inhibition revealed a significant difference in the activity of
CMPI 1 and CMPO 2 enantiomers.
Received 12th August 2013,
Accepted 10th October 2013
DOI: 10.1039/c3ob41646a
www.rsc.org/obc
Bicyclic pyrrolidine derivatives CMPI 1 (7-[3-(сyclopentyl-
oxy)-4-methoxyphenyl]hexahydro-3H-pyrrolo[1,2-c]imidazol-3-
Introduction
Inhibitors of type IV phosphodiesterase (PDE), an enzyme that one, Fig. 1) and CMPO 2 (7-(3-(cyclopentyloxy)-4-methoxy-
hydrolyzes cyclic adenosine monophosphate in cells, are con- phenyl)tetrahydropyrrolo[1,2-c]oxazol-3(1H)-one, Fig. 1) were
sidered a novel class of highly potent drugs for the complex introduced by GlaxoSmithKline as prospective PDE type IVB
therapy of asthma and chronic obstructive lung disease inhibitors.6 CMPI 1 and CMPO 2 are considerably more
(COPD).2 A prototype of these drugs is the natural alkaloid potent than Rolipram (IC50 = 175 nM) and Cilomilast5f (IC50
=
theophylline, which has been used to treat respiratory diseases 92 nM).7 At the same time biological studies of CMPI 1 and
since the middle of the XXth century.2g,3 However, theophyl- CMPO 2 were conducted for racemates due to the absence of
line has a low therapeutic effect because of the weak and non- an approach to the asymmetric synthesis of their enantio-
selective inhibition of the PDE IV enzyme (IC50 > 10 000 mers.6a It is likely that as in the case of Rolipram8 the binding
nM).2b,3c In the last two decades selective second generation activity of enantiomers is different and one of them could be a
PDE IV inhibitors active at micro- and nanomolar concentrations more potent PDE IV inhibitor than another. In the present
have been developed.2 A gold standard in PDE IV inhibition work, the problem of the asymmetric synthesis of CMPI 1 and
research is the well-known pyrrolidinone Rolipram2b,4 (Fig. 1),
which, however, did not pass clinical trials due to adverse
effects. In 2010–2011 the first oral PDE IV inhibitor Roflumi-
last (Daxas®)2,5a,b was approved by the FDA and European Medi-
cines Agency for clinical use in the USA and European Union
countries. At present, several other catechol-based PDE IV
inhibitors (Apremilast,5c Tetomilast5d and others) are at
different stages of clinical and pre-clinical development.5e
aN.D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991,
Leninsky prospect, 47, Moscow, Russian Federation. E-mail: sukhorukov@ioc.ac.ru;
Fax: +7 499 1355328; Tel: +7 499 1355329
bDepartment of Chemistry, M.V. Lomonosov Moscow State University, 119991,
Leninskie gory, 1, str. 3, Moscow, Russian Federation. E-mail: chupvl@gmail.com;
Fax: +7 495 939 3564; Tel: +7 495 932 8846
†Part 4. For previous article in these series see ref. 1.
‡Electronic supplementary information (ESI) available: Copies of FTIR, NMR 1H,
13C, COSY, HSQC, NOESY spectra of new compounds, details of in vitro assay
and molecular docking. See DOI: 10.1039/c3ob41646a
Fig. 1 PDE IV inhibitors.
8082 | Org. Biomol. Chem., 2013, 11, 8082–8091
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
CMPO 2 was solved and the PDE IV inhibitory activity of their
enantiomers was studied.
According to the suggested approach (Scheme 1) a general
precursor of CMPI 1 and CMPO 2 is an optically pure nitronate
5 bearing a chiral alkoxyl group R* at atom C-6. Stereoselective
asymmetric synthesis of nitronates 5 by [4 + 2] cycloaddition of
nitrolkene 6 to optically pure vinyl ethers of (+)- or (−)-trans-2-
phenylcyclohexanols (Denmark’s approach12) was recently
reported by us.11c Silylation of nitronate 5 to give bromide 711c
followed by nucleophilic substitution of bromine for the azido
or hydroxyl group could furnish the respective functionalized
5,6-dihydro-4H-1,2-oxazines 8 and 9. The reduction of pro-
ducts 8 and 9 involving reductive cleavage of the N–O bond
and recyclization should lead to the necessary optically pure
pyrrolidines 3 and 4, respectively. The latter can be trans-
formed into target products CMPI 1 and CMPO 2 by trivial
procedures.
Results and discussion
Synthesis
Evidently, the target enantiomerically pure products CMPI 1
and CMPO 2 could be formed in a single step from pyrrol-
idines 3 and 4, respectively (Scheme 1). Generally, stereo-
selective synthesis of 2,3-disubstituted pyrrolidines represents
a non-trivial task.9 Recently, we developed an access to trans-
2,3-disubstituted pyrrolidines bearing a –CH2FG fragment at
position C-2 from available six-membered cyclic nitronates.10
The key strategic operation in this synthesis is the silylation of
cyclic nitronates leading to the activation of the methyl group
at C-3.10 This strategy was successfully employed in stereo-
selective synthesis of racemic PDE IV inhibitors CMPI 1 and
CMPO 2,11a as well as racemic11b and asymmetric11c synthesis
of another analogue of Rolipram.
To realize this strategy diastereomerically and enantio-
merically pure cyclic nitronates (+)-(4S,6S,7S,8R)-5 and
(−)-(4R,6R,7R,8S)-5 (d.e. >99%, Scheme 2) were synthe-
sized according to a previously reported procedure from
Scheme 1 Retrosynthetic analysis of CMPI 1 and CMPO 2 molecules.
Scheme 2 Synthesis of 3-bromomethyl-substituted 1,2-oxazines 7.
Org. Biomol. Chem., 2013, 11, 8082–8091 | 8083
This journal is © The Royal Society of Chemistry 2013

View Article Online
Paper
Organic & Biomolecular Chemistry
commercially available nitroethane, isovanillin, cyclopentyl last stereocenter was achieved by stereoselective reduction of
bromide and (+)- or (−)-trans-2-phenylcyclohexanols (>98% the oxime group upon action of sodium cyanoborohydride
ee).11c Silylation of nitronates 5 upon treatment with excess of in
acetic
acid.14
The
resulting
tetrahydrooxazines
trimethylsilyl bromide and triethylamine in dichloromethane (+)-(3R,4S,6S,7S,8R)-10 and (−)-(3S,4R,6R,7R,8S)-10 were
furnished the corresponding 3-bromomethyl-substituted obtained as individual stereoisomers possessing the required
dihydrooxazines 7 accompanied by epimeric 4,6-cis-bromides 3,4-trans-configuration of stereocenters in the 1,2-oxazine ring
7′ as by-products. Diastereomerically pure bromides (coupling constant 3JH3–H4 = 11.0 Hz).
(4S,6S,7S,8R)-7 and (4R,6R,7R,8S)-7 were isolated from the reac-
The synthesis of 3-hydroxymethyl-substituted 1,2-oxazines 9
tion mixture by column chromatography. In an analogous (precursors of CMPO 2 enantiomers) proved to be a more chal-
fashion racemic bromide rac-7 was synthesized from racemic lenging task. Thus, direct substitution of bromine for the
trans-2-phenylcyclohexanol (Scheme 2).
hydroxyl group in bromooxazines 7 by hydrolysis with NaOH
Nucleophilic substitution of bromine in enantiomeric or Ag2O in aqueous THF at ambient temperature or reflux con-
(4S,6S,7S,8R)-7 and (4R,6R,7R,8S)-7 for the azido group was ditions failed to provide the desired product. No conversion of
accomplished by the reaction with sodium azide in the pres- the starting material was observed in these experiments.
ence of a catalytic amount (20 mol%) of sodium iodide in However, refluxing of oxazines 7 in aqueous acetone with an
15
aqueous acetone (Scheme 3).13 Corresponding 3-azidomethyl excess of AgNO3 resulted in the substitution of bromine for
substituted 5,6-dihydro-4H-1,2-oxazines (4S,6S,7S,8R)-8 and a nitrate anion.16 The resulting nitrates 11 were identified in
(4R,6R,7R,8S)-8 were isolated in high yields. Installation of the the reaction mixture (1H NMR, HRMS data) and due to their
instability were used in the next stage without special purifi-
cation (Scheme 4). Reduction of nitrates 11 with zinc in acetic
acid allowed a smooth transformation of the –ONO2 fragment
to the hydroxyl group.17 Subsequent addition of sodium cyano-
borohydride to the reaction mixture leads to selective CvN
bond reduction. As a result of these operations tetrahydro-
oxazines (+)-(3R,4S,6S,7S,8R)-12 and (−)-(3S,4R,6R,7R,8S)-12
were obtained in 82% from bromooxazines (4S,6S,7S,8R)-7 and
(4R,6R,7R,8S)-7, respectively. Like products 10 tetrahydro-
oxazines 12 are formed exclusively as 3,4-trans-isomers (3JH3–H4
=
10.9 Hz).
The final step in both syntheses of target PDE IV inhibitors
CMPI 1 and CMPO 2 was the reduction of the 1,2-oxazine cycle
in intermediates 10 and 12 to form a pyrrolidine ring (see
Scheme 3 for CMPI 1 synthesis and Scheme 4 for CMPO 2 syn-
thesis). This multi-step domino process is realized under cata-
lytic hydrogenation conditions, and its result depends on the
nature of the substituent at the C-3 atom of the 1,2-oxazine
ring.10,14a–d Thus, the reduction of tetrahydrooxazine 10
bearing an azidomethyl group at atom C-3 may proceed
ambiguously (Scheme 5). In this process the generation of di-
aminoaldehyde A by a consequent reduction of the azido group
and N–O bond can be expected (Scheme 5).18 Intramolecular
reductive amination of intermediate A may follow two path-
ways, leading to either the necessary pyrrolidine 3 or isomeric
piperidine 13 (Scheme 5).
To prevent the formation of piperidine cycle protection of
the amino group resulting from the azido group reduction is
required. Thus, conducting hydrogenation of tetrahydroox-
azine 10 in the presence of one equivalent of di(tert-butyl)
dicarbonate (Boc2O) at ambient temperature and a hydrogen
pressure of 40 bar resulted in selective reduction of the azido
group, leaving the N–O bond intact (Scheme 6).19 The resulting
1,2-oxazine 14 possessing a Boc-protected primary amino
group was characterized in the reaction mixture.20 Increasing
the temperature of hydrogenation led to the cleavage of the
1,2-oxazine cycle affording pyrrolidine 15 and trans-2-phenyl-
cyclohexanol (1H NMR and HRMS data). The plausible
Scheme 3 Synthesis of CMPI 1 enantiomers.
8084 | Org. Biomol. Chem., 2013, 11, 8082–8091
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 5 Two possible pathways of tetrahydrooxazine 10 reduction.
Catalytic hydrogenation of tetrahydrooxazines (+)-(3R,4S,
6S,7S,8R)-12 and (−)-(3S,4R,6R,7R,8S)-12 proceeds smoothly in
the presence of RANEY® nickel leading to enantiomerically
pure prolinols 4 (see the bottom part of Scheme 4). The initial
trans-2-phenylcyclohexanols were recovered at this stage in
91% yield. Crude prolinols 4 without purification were trans-
formed into target CMPO 2 enantiomers upon treatment with
Im2CO in dichloromethane.22 As a result, (−)-(7S,7aR)-CMPO 2
was obtained from tetrahydrooxazine (+)-(3R,4S,6S,7S,8R)-12,
and (+)-(7R,7aS)-CMPO 2 was obtained from (−)-(3S,4R,6R,
7R,8S)-12 in 77% yield.
NMR data of enantiomerically pure products CMPI 1 and
CMPO 2 are in agreement with literature data for corres-
ponding racemates. Optical rotation angles in pairs (−)-CMPI-1/
(+)-CMPI-1 and (+)-CMPO-2/(−)-CMPO-2 are close to the
absolute value and have opposite signs. The structure of pre-
viously unknown products (+)-8, (−)-8, rac-8, (+)-10, (−)-10, rac-
10, (+)-12 and (−)-12 was confirmed by NMR (1H, 13C, COSY,
HSQC and NOESY), high-resolution mass spectrometry, FTIR-
spectroscopy and elemental analysis.
Scheme 4 Synthesis of CMPO 2 enantiomers.
mechanism of pyrrolidine 15 formation involves the hydro-
genolysis of the N–O bond in 14 generating hemiacetal B, elimin-
ation of trans-2-phenylcyclohexanol to give aldehyde C, its
cyclization to pyrrolidine D, elimination of water forming pyr-
roline E and reduction of the latter (Scheme 6).21 Pyrrolidine
15 is the final product of the hydrogenation and it is not trans-
formed into pyrroloimidazolone CMPI 1 under reaction con-
ditions. The cyclization of crude pyrrolidine 15 into the target
product CMPI 1 could be accomplished only upon reflux in
DMSO (see the bottom part of Scheme 3).11a
In vitro and molecular docking studies
The obtained enantiomers (−)-CMPI-1/(+)-CMPI-1 and
(+)-CMPO-2/(−)-CMPO-2 were tested for their ability to inhibit
FAM-cAMP hydrolysis catalyzed by recombinant human PDE
IVB1 using standard in vitro enzymatic assay. The results are
summarized in Table 1 (column 2).
Employing this procedure, product (−)-(7S,7aR)-CMPI 1 was
obtained from (+)-(3R,4S,6S,7S,8R)-10, and its enantiomer
(+)-(7R,7aS)-CMPI 1 was prepared from (−)-(3S,4R,6R,7R,8S)-10
in 56% average yield (Scheme 3). It should be noted that
enantiopure (+)- and (−)-trans-2-phenylcyclohexanols were
recovered after hydrogenation in 81% average yield.
As can be seen from the data in Table 1 (−)-(7S,7aR)-CMPI 1
is ten times more potent than its enantiomer (+)-(7R,7aS)-
CMPI 1. For CMPO 2 the (−)-enantiomer is also more potent
than the (+)-enantiomer, although the difference is smaller
compared to CMPI 1. Interestingly, IC50 values for (+)-(7R,7aS)-
CMPI 1 and (+)-(7R,7aS)-CMPO 2 are essentially the same
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 8082–8091 | 8085

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 PDEIVB1 inhibition data and predicted affinities for enantiomers and
racemates of CMPI-1 and CMPO-2
Affinity (kcal mol−1
(AutoDock Vina)
)
Compound
IC50 (nM)
(−)-(7S,7aR)-CMPI 1
(+)-(7R,7aS)-CMPI 1
rac-CMPI 1
(−)-(7S,7aR)-CMPO 2
(+)-(7R,7aS)-CMPO 2
rac-CMPO 211a
15
150
27a
−10.2
−10.5
—
−10.5
−10.4
—
35
150
70, 75a
a Data from ref. 6a.
differences in the experimentally detected IC50 (Table 1,
column 3). An accurate prediction of the ligand’s affinity to the
studied protein is a very complicated task, and most of the
docking software is not well suited for it.26
On the other hand, qualitative analysis of interactions of
the ligands with the protein reveals some differences in their
binding which may influence the inhibition activity (Fig. 2). All
of the compounds share identical lipophilic interactions with
pockets formed by Phe414, Phe446, Phe506, Met411, Met503,
Ile410 and partly by Tyr233 and Thr407. The main difference
is the interactions with Zn2+ ions and with amino acids that
bind Mg2+ and Zn2+ ions: His234, His238, His278, Asp275 and
Asp392. Thus, (−)-(7S,7aR)-CMPI 1 and (−)-(7S,7aR)-CMPO 2
form strong interactions with Zn2+, that does not occur for
(+)-enantiomers (Fig. 2). The NH-group of the (−)-(7S,7aR)-
CMPI 1 also forms hydrogen bonds with Asp392 and, probably,
His234 (Fig. 2). This may explain the higher activity of
(−)-(7S,7aR)-CMPI 1 compared to (−)-(7S,7aR)-CMPO 2, which
cannot serve as
a hydrogen bond donor. In general,
(−)-(7S,7aR)-CMPI 1 and (−)-(7S,7aR)-CMPO 2 enantiomers fit
better into the protein pocket as compared to (+)-(7R,7aS)-
CMPI 1 and (+)-(7R,7aS)-CMPO 2. Though this model is plaus-
ible, we suppose that these interactions account for the higher
inhibition activity (around 5–10 times) of (−)-CMPI 1 and
(−)-CMPO 2 over their (+)-enantiomers. These results can be
used in further design and development of new highly potent
PDE IV inhibitors.
Experimental
Scheme 6 Plausible mechanism of hydrogenation of tetrahydrooxazine 10 to
pyrrolidine 15.
All reactions were performed in oven-dried (150 °C) glassware.
Catalytic hydrogenations were carried out in a steel autoclave
with external stirring and heating. Column chromatography
was performed using Kieselgel 40–60 μm 60A silica gel. 1D and
2D NMR spectra were recorded at room temperature in CDCl3.
The chemical shifts (1H, 13C) are given ppm (δ) in relation to
the solvent signal. Multiplicities are indicated by s (singlet), d
(doublet), t (triplet), q (quartet), m (multiplet), br (broad).
Numeration of atoms is given in Schemes 3 and 4. Peaks in IR-
spectra data are reported in cm−1 with the following relative
intensities: s (strong), m (medium), w (weak), br (broad), sh
(shoulder). Elemental analysis was performed by the Analytical
(within the accuracy of the experiment), indicating that the
nature of heteroatom in position 2 does not influence the inhi-
bition activity.
To reveal the nature of the difference in PDE IVB inhibitory
activity of these enantiomers molecular docking studies were
performed using the AutoDock Vina23 software. 3KKT24 struc-
tures were selected taking into account the quality of the struc-
ture and the similarity of the co-crystallized compounds with
studied ones.25 The predicted affinities of CMPI 1 and
CMPO 2 enantiomers to PDE IVB are very similar despite the
8086 | Org. Biomol. Chem., 2013, 11, 8082–8091
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
bromide, DMSO, RANEY® nickel (50% slurry in water), Im2CO
were purchased from commercial sources and used as
received. Diastereomerically pure (d.e. >98%) cyclic nitronates
(+)-5, (−)-5 and rac-5 were synthesized according to a previously
described procedure.11c
PDE inhibition studies were conducted at BPS Bioscience
(San Diego, USA). PDE IVB1 assay kit was used (BPS catalogue
number 60342) for this purpose.
For docking studies with AutoDock Vina23 human PDE IVB
(PDB ID: 3KKT24) was selected. The protein structure was pre-
pared according to a classical AutoDock scenario: ligand and
water molecules were removed, atoms of Zn2+ and Mg2+ were
retained in the protein structure, nonpolar hydrogens, Gastei-
ger–Huckel charges and studied ligands CMPI 1 and CMPO 2
were added to proteins. Spatial structures of the ligands were
calculated with OpenBabel.27 Top scoring poses of docked
molecules according to predicted free energy of binding were
selected for further discussion of protein–ligand interactions
(Table 1). Visualization of interactions of molecules CMPI 1
and CMPO 2 with PDE IVB was done using Pymol v. 0.99.28
trans-3-(Azidomethyl)-4-[3-(cyclopentyloxy)-4-methoxyphenyl]-
6-{[trans-2-phenylcyclohexyl]oxy}-5,6-dihydro-4H-1,2-
oxazine (8)
Me3SiBr (1.37 mL, 10.4 mmol) was added to a solution of
racemic or enantiopure nitronate 5 (1.0 g, 2.09 mmol) and
Et3N (0.43 mL, 3.1 mmol) in 6.0 mL of MeCN at −25 °C under
argon. The mixture was kept at −25 °C for 18 h with occasional
stirring, and then it was diluted with EtOAc (20 mL) and
poured into a mixture of EtOAc (200 mL) and a saturated solu-
tion of K2CO3 (200 mL). The aqueous layer was back-extracted
with EtOAc (2 × 75 mL). The combined organic layers were
washed with a saturated solution of K2CO3 (100 mL), water
(100 mL), and brine (100 mL), dried (Na2SO4), and evaporated
under vacuum. The residue was preadsorbed on silica gel and
subjected to a column chromatography (eluent: heptane–
EtOAc = 20 : 1 → 10 : 1 → 5 : 1 → 1 : 1) to give 0.081 g (7%) of
bromide 4,6-cis-7′,11c 0.29 g (26%) of labile bromide 4,6-trans-
711c (containing ca. 5% of trans-2-phenylcyclohexanol) and
0.42 g (42%) of initial nitronate 5. The recovered nitronate 5
can be silylated again by the same procedure to give additional
0.135 g of bromide 4,6-trans-7 and 0.046 g of 4,6-cis-7′.
Fig. 2 Prediction of interactions of CMPI 1 (A) and CMPO 2 (B) enantiomers
with PDE IVB (PDB ID: 3KKT) by AutoDock Vina. Green − (−)-(7S,7aR)-CMPI 1
and (−)-(7S,7aR)-CMPO 2, blue − (+)-(7R,7aS)-CMPI 1 and (+)-(7R,7aS)-CMPO 2.
Important amino acids are shown in bulk, main chain of amino acids is hidden,
protein structure is visualized as a thin tube, Zn2+ ion is shown in green, Mg2+
ion is shown in grey.
Laboratory of the Institute of Organic Chemistry. Concen- The overall yield of racemic or enantiopure bromide 4,6-trans-7
trations c in optical rotation angles are given in g per 100 mL. ca. 38%.
[α]D values are given in 10−1 deg cm2 g−1. Analytical thin-layer
To a stirred solution of enantiopure or racemic bromide 7
chromatography was performed on silica gel plates with (0.367 g, 0.677 mmol) in acetone (12 mL) was added a solution
QF-254. Visualization was accomplished with UV light, solu- of NaN3 (0.205 g, 3.15 mmol) and NaI (0.022 g, 0.15 mmol) in
tion of FeCl3 in HCl/H2O–Et2O or the solution of anisalde- water (2.5 mL). The mixture was stirred at r.t. for 24 h, then
hyde/H2SO4 in ethanol. Glacial acetic acid was recrystallized evaporated under vacuum. The residue was preadsorbed on
two times. CH2Cl2 (technical grade), MeCN (technical grade), silica gel and subjected to a column chromatography (eluent:
Et3N, and Me3SiBr were redistilled from CaH2, acetone was hexane–EtOAc = 20 : 1 → 10 : 1 → 5 : 1 → 3 : 1) to give 0.303 g
redistilled from Na2SO4. Hexane, EtOAc and methanol were (89%) of enantiopure (oil) or racemic (crystalline) azide 8. For
technical grade and distilled without drying agents. analytical purposes racemic azide 8 was recrystallized from
(+)-(1S,2R)-2-Phenylcyclohexanol (>98% ee), (−)-(1R,2S)-2- MeOH (mp 100–101 °C). Rf 0.72 (hexane–EtOAc = 1 : 1). Found:
phenylcyclohexanol (99% ee), ethyl vinyl ether, SnCl4, NaN3, C, 69.01; H, 7.54; N, 11.12. Calc. for C29H36N4O4: C, 69.02; H,
NaI, NaBH3CN, NH4OAc, nitroethane, isovanillin, cyclopentyl 7.19; N, 11.10. FTIR (KBr, characteristic bands): vmax/cm−1
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 8082–8091 | 8087

View Article Online
Paper
Organic & Biomolecular Chemistry
2955s, 2930s, br, 2855m, 2855m, 2839w, 2106s (N3), 1603w (75.47 MHz, HSQC, DEPT): 24.0 (9-C), 24.7 (23-C), 26.0 (10-C),
(CvN), 1591m, 1514s, 1443m, sh, 1334m, sh, 1276m, 1245s, 31.2 (12-C), 32.7 and 32.8 (22-C), 33.4 (5-C), 36.6 (11-C), 37.3
1212m, 1139m, sh, 1108s, 1037s, 1007m, 913m, 872m, sh, (4-C), 50.2 (13-C), 50.9 (8-C), 56.1 (20-C), 61.4 (3-C), 77.5 (7-C),
1
759m, sh, 705m. H NMR (300 MHz, COSY, HSQC): 1.20–1.32 80.5 (21-C), 93.4 (6-C), 112.3 (18-C), 114.3 (15-C), 119.4 (19-C),
(m, 1 H, HC(12)), 1.34–1.49 (m, 1 H, HC(11)), 1.51–1.57 (m, 126.7 (p-Ph), 128.0 and 128.8 (o- and m-Ph), 133.8 (14-C), 144.5
1 H, HC(9)), 1.58–1.69 (m, 2 H, HC(23)), 1.74–1.96 (m, 10 H, (i-Ph), 147.8 and 149.0 (16-C and 17-C). HRMS: m/z 507.2962
+
HC(9), H2C(10), HC(11), H2C(22) and HC(23)), 1.97–2.05 (m, (positive ions); calcd for [C29H39N4O4 ]: 507.2966.
2 H, H2C(5)), 2.34–2.45 (m, 1 H, HC(12)), 2.59 (ddd, J = 11.9,
(+)-(3R,4S,6S,7S,8R)-10 (obtained from (4S,6S,7S,8R)-8): col-
11.0, 3.2 Hz, 1 H, HaxC(8)), 2.80 (dd, J = 11.9, 7.7 Hz, 1 H, orless oil, [α]2D4 +174.8 (c 0.87 in MeOH).
HaxC(4)), 3.20 (s, 2 H, H2C(13)), 3.81 (s, 3 H, H3C(20)), 3.94
(−)-(3S,4R,6R,7R,8S)-10 (obtained from (4R,6R,7R,8S)-8): col-
(ddd, J = 11.0, 10.2, 4.0 Hz, 1 H, HaxC(7)), 4.70 (m, 1 H, HC(21)), orless oil, [α]2D7 −172.8 (c 0.87 in MeOH).
5.39 (dd, J = 2.3, 1.8 Hz, 1 H, HeqC(6)), 6.48 (d, J = 1.7 Hz, 1 H,
Rac-10 (obtained from rac-8): white solid, mp 109–112 °C
HC(15)), 6.52 (dd, J = 8.1, 1.7 Hz, 1 H, HC(19)), 6.77 (d, J = (crystallized from MeOH).
8.1 Hz, 1 H, HC(18)), 7.18–7.35 (m, 5 H, Ph). 13C NMR
3,4-trans-3,6-trans-3-(Hydroxymethyl)-4-[3-(cyclopentyloxy)-4-
methoxyphenyl]-6-{[trans-2-phenylcyclohexyl]oxy}-1,2-
oxazinane (12)
(75.47 MHz, HSQC, DEPT): 24.0 (9-C), 24.6 (23-C), 26.1 (10-C),
30.4 (12-C), 32.1 (13-C), 32.7 (22-C), 33.8 (11-C), 34.2 (4-C), 50.8
(8-C), 51.4 (13-C), 56.1 (20-C), 76.1 (7-C), 80.5 (21-C), 90.7 (6-C),
112.3 (18-C), 114.8 (15-C), 120.7 (19-C), 125.8 (p-Ph), 127.9 and To a stirred solution of enantiopure bromide (4S,6S,7S,8R)-7 or
128.0 (o- and m-Ph), 130.7 (14-C), 144.6 (i-Ph), 148.0 and 149.6 (4R,6R,7R,8S)-7 (0.125 g, 0.23 mmol) in a mixture of water
(16-C and 17-C), 155.8 (3-C). HRMS: m/z 505.2809 (positive (3 mL) and acetone (5 mL) was added AgNO3 (0.196 g,
+
ions); calcd for [C29H37N4O4 ]: 505.2809.
1.15 mmol). The mixture was refluxed for 3 h, then cooled to
r.t. and acetone was removed under vacuum. The residue was
diluted with AcOEt (100 mL) and water (50 mL) and trans-
ferred into a separating funnel. The aqueous layer was back-
extracted with EtOAc (2 × 25 mL). The combined organic layers
3,4-trans-3,6-trans-3-(Azidomethyl)-4-[3-(cyclopentyloxy)-4-
methoxyphenyl]-6-{[trans-2-phenylcyclohexyl]oxy}-1,2-
oxazinane (10)
To a stirred solution of enantiopure or racemic 1,2-oxazine 8 were washed with water (50 mL) and brine (50 mL), dried with
(0.300 g, 0.6 mmol) in AcOH (3 mL) was added NaBH3CN Na2SO4 and evaporated under vacuum to give crude nitrates 11
(0.134 g, 2.13 mmol) under argon atmosphere. After 40 min a [1H NMR (300 MHz): 1.23–1.45, 1.55–1.96 and 1.96–2.04 (3 m,
second portion of NaBH3CN (0.045 g, 0.71 mmol) was added 17 H, H2C(5), H2C(9), H2C(10), H2C(11), HC(12), H2C(22) and
and the mixture was stirred for an additional 1 h. The resulting H2C(23)), 2.36 (m, 1 H, HC(12)), 2.58 (ddd, J = 12.2, 10.5,
solution was poured into a mixture of EtOAc (100 mL)/satu- 3.4 Hz, 1 H, HaxC(8)), 2.90 (dd, J = 12.2, 7.3 Hz, 1 H, HaxC(4)),
rated solution of K2CO3 (100 mL). The aqueous layer was back- 3.81 (s, 3 H, H3C(20)), 3.96 (ddd, J = 10.5, 10.1, 3.7 Hz, 1 H,
extracted with EtOAc (2 × 50 mL). The combined organic layers HaxC(7)), 4.22 (d, J = 13.2 Hz, 1 H, HC(13)), 4.36 (d, J = 13.2 Hz,
were washed with a saturated solution of K2CO3 (50 mL), water 1 H, HC(13)), 4.70 (m, 1 H, HC(21)), 5.37 (dd, J = 2.0, 1.5 Hz,
(50 mL), brine (50 mL), dried with Na2SO4 and evaporated 1 H, HeqC(6)), 6.52 (d, J = 1.6 Hz, 1 H, HC(15)), 6.55 (dd, J =
under vacuum. The residue was subjected to a column chromato- 8.2, 1.6 Hz, 1 H, HC(19)), 6.77 (d, J = 8.2 Hz, 1 H, HC(18)),
graphy on silica gel (eluent: EtOAc–hexane = 1 : 10 → 1 : 5 → 7.14–7.36 (m, 5 H, Ph). HRMS: m/z 525.2588 (positive ions);
+
1 : 3) to yield 0.286 g (94%) of tetrahydro-1,2-oxazine 10. For calcd for [C29H37N2O7 ]: 525.2601]. The crude product was dis-
analytical purposes racemic product rac-10 was recrystallized solved in AcOH (6 mL) and Zn powder (0.15 g) was added.
from MeOH. Rf 0.56 (hexane–EtOAc = 1 : 1). Found: C, 68.80; After stirring the mixture for 3 h an additional portion of Zn
H, 7.63; N, 11.20. Calc. for C29H38N4O4: C, 68.75; H, 7.56; N, (0.05 g) was added and the mixture was left overnight. Then
11.06. FTIR (KBr, characteristic bands) vmax/cm−1 3217m, sh NaBH3CN (0.087 g, 1.38 mmol) was added and the mixture
(N–H), 2933s, br, 2859m, 2095s (N3), 1593m, 151s, sh, 1445m, was left stirring for 0.5 h. Then the second portion of
1336m, sh, 1273s, 1261s, 1164m, 1141m, 1096m, 1081w, NaBH3CN (43 mg, 0.68 mmol) was added and after stirring for
901m, 806m, 759m, 702m, 535m. 1H NMR (300 MHz, COSY, 0.5 h the resulting mixture was poured into EtOAc (100 mL)/
HSQC): 1.25–1.49 (m, 2 H, HC(10) and HC(12)), 1.56–1.73 (m, saturated solution of K2CO3 (100 mL). The aqueous layer was
3 H, HC(9) and HC(23)), 1.77–2.02 (m, 12 H, H2C(5), HC(9), back-extracted with EtOAc (50 mL). Combined organic layers
HC(10), H2C(11), H2C(22) and HC(23)), 2.18–2.26 (m, 1 H, were washed with a saturated solution of K2CO3 (50 mL), water
HC(12)), 2.36 (ddd, J = 11.7, 11.0, 4.3 Hz, 1 H, HaxC(4)), 2.45 (50 mL), brine (50 mL), dried with Na2SO4 and evaporated under
(dd, J = 12.5, 7.3 Hz, 1 H, HC(13)), 2.68 (ddd, J = 11.8, 11.8, vacuum. The residue was subjected to a column chromato-
3.6 Hz, 1 H, HaxC(8)), 2.79 (dd, J = 12.5, 2.0 Hz, 1 H, HC(13)), graphy on silica gel (eluent: EtOAc–hexane = 1 : 10 → 1 : 5 →
3.00 (dddd, J = 12.5, 11.0, 7.3, 2.0 Hz, 1 H, HaxC(3)), 3.75–3.82 1 : 3 → 1 : 1) to yield 0.091 g (82% from bromide 7) of tetra-
(s and m, 4 H, HC(7) and H3C(20)), 4.12 (d, J = 12.5 Hz, 1 H, hydro-1,2-oxazine (+)-12 or (−)-12 as oil. Rf 0.19 (hexane–EtOAc
HN(2)), 4.74 (m, 1 H, HC(21)), 4.98 (d, J = 1.8 Hz, 1 H, HeqC(6)), = 1 : 1). Found: C, 72.08; H, 8.04; N, 2.85. Calc. for C29H39NO5:
1
6.57 (s, 1 H, HC(15)), 6.59 (d, J = 8.1 Hz, 1 H, HC(19)), 6.77 (d, C, 72.32; H, 8.16; N, 2.91. H NMR (300 MHz, COSY, HSQC):
J = 8.1 Hz, 1 H, HC(18)), 7.29–7.47 (m, 5 H, Ph). 13C NMR 1.20–1.45 (m, 2 H, HC(10) and HC(12)), 1.56–1.74 (m, 3 H,
8088 | Org. Biomol. Chem., 2013, 11, 8082–8091
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
HC(9) and HC(23)), 1.75–2.00 (m, 12 H, H2C(5), HC(9), HC(10), (m, 6 H, H2C(16) and HC(17)), 2.05 (dddd, J = 12.8, 11.9, 10.1,
H2C(11), H2C(22) and HC(23)), 2.18–2.27 (m, 1 H, HC(12)), 9.2 Hz, 1 H, H′C(6)), 2.38 (dddd, J = 12.8, 10.1, 8.2, 4.6 Hz, 1 H,
2.30 (ddd, J = 12.2, 10.9, 3.8 Hz, 1 H, HaxC(4)), 2.57 (dd, J = H″C(6)), 2.74 (ddd, J = 11.0, 10.1, 9.2 Hz, 1 H, HC(7)), 3.26–3.31
11.2, 9.0 Hz, 1 H, HC(13)), 2.68 (ddd, J = 12.5, 10.3, 3.3 Hz, (m, 1 H, H′C(5)), 3.33 (dd, J = 9.9, 9.9 Hz, 1 H, H″C(1)), 3.51
1 H, HaxC(8)), 3.01 (ddd, J = 10.9, 9.0, 1.0 Hz, 1 H, HaxC(3)), (dd, J = 9.9, 7.7 Hz, 1 H, H′C(1)), 3.64–3.75 (m, 2 H, HC(7a) and
3.10 (dd, J = 11.2, 1.0 Hz, 1 H, HC(13)), 3.74 (ddd, J = 10.5, H′C(5)), 3.82 (s, 3 H, H3C(14)), 4.76 (m, 1 H, HC(15)), 5.55 (br,
10.2, 3.6 Hz, 1 H, HaxC(7)), 3.81 (s, 3 H, H3C(20)), 3.92 (br., 1 H, HN(2)), 6.72 (s, 1 H, HC(9)), 6.73 (d, J = 7.3 Hz, 1 H,
2 H, NH(2) and OH), 4.74 (m, 1 H, HC(21)), 4.99 (s, 1 H, HC(13)), 6.83 (d, J = 7.3 Hz, 1 H, HC(12)). 13C NMR
H
eqC(6)), 6.58 (s, 1 H, HC(15)), 6.61 (d, 1 H, J = 8.2 Hz, (75.47 MHz, HSQC, DEPT): 24.0 (17-C), 32.7 (16-C), 34.4 (6-C),
HC(19)), 6.76 (d, J = 8.2 Hz, 1 H, HC(18)), 7.29 (m, 1 H, p-Ph), 41.3 (1-C), 45.2 (5-C), 48.5 (7-C), 56.2 (14-C), 66.1 (7a-C), 80.6
7.40–7.46 (m, 4 H, o- and m-Ph). 13C NMR (75.47 MHz, HSQC, (15-C), 112.4 (12-C), 114.7 (9-C), 119.6 (13-C), 131.7 (8-C), 147.9
DEPT): 24.0 (9-C), 24.7 (23-C), 25.9 (10-C), 31.2 (12-C), 32.7 and and 149.3 (10-C and 11-C), 165.8 (3-C). Characteristic
32.8 (22-C), 32.9 (5-C), 36.5 (11-C), 36.8 (4-C), 50.8 (8-C), 56.0 2D-NOESY correlations: HC(7a)/HC(9) and HC(13), H3C(14)/
(20-C), 61.6 (13-C), 63.1 (3-C), 77.6 (7-C), 80.5 (21-C), 93.7 (6-C), HC(12), HC(7)/H″C(6), H″C(6)/H″C(5), H′C(6)/H′C(5), HC(7a)/
112.3 (18-C), 114.4 (15-C), 119.4 (19-C), 126.6 (p-Ph), 128.2 and H′C(1). NMR spectra are in accordance with literature data for
128.7 (o- and m-Ph), 134.0 (14-C), 144.7 (i-Ph), 147.7 and 148.8 racemic CMPI 1.6a,11a
(16-C and 17-C). HRMS: m/z 482.2901 (positive ions); calcd for
[C29H40NO5 ]: 482.2901.
(−)-(7S,7aR)-CMPI 1 (obtained from (+)-10): colorless oil;
[α]2D6 −33.6 (c 1.0 in MeOH).
+
(+)-(3R,4S,6S,7S,8R)-12 (obtained from (4S,6S,7S,8R)-7): col-
orless oil, [α]2D5 +193.9 (c 1.0 in MeOH).
(+)-(7R,7aS)-CMPI 1 (obtained from (−)-10): colorless oil;
[α]2D7 +32.3 (c 1.0 in MeOH).
(−)-(3S,4R,6R,7R,8S)-12 (obtained from (4R,6R,7R,8S)-7): col-
Rac-1 (obtained from rac-10): mp 134–139 °C (lit. mp
139–141 °C,11a 118–120 °C6a).
orless oil, [α]2D5 −191.2 (c 1.0 in MeOH).
(−)-trans-2-Phenylcyclohexanol (obtained from (−)-10): mp
61–64 °C (lit. 63–66 °C, Aldrich), [α]2D7 = −53.0 (c 1.83 in MeOH)
[lit.29 [α]D −56.8 (c 1.42 in MeOH)]. 1H NMR spectra are in
trans-7-[3-(Cyclopentyloxy)-4-methoxyphenyl]hexahydro-3H-
pyrrolo[1,2-c]imidazol-3-one (CMPI 1)
RANEY® nickel (ca. 50 mg, washed with MeOH) was placed in accordance with literature data.29
a test tube equipped with a magnetic stirrer and charged with
(+)-trans-2-Phenylcyclohexanol (obtained from (+)-10): mp
a solution of enantiopure or racemic tetrahydro-1,2-oxazine 10 64–65 °C (lit. 64–66 °C, Aldrich), [α]2D4 +55.2 (c 1.83 in MeOH)
(0.119 g, 0.24 mmol) and Boc2O (0.05 mL, 0.24 mmol) in [lit.30 [α]D20 +57.3 (c 5.0 in MeOH)]. 1H NMR spectra are in
MeOH (5 mL). The test tube was placed in a steel autoclave accordance with literature data.29
which was then flushed and filled with H2 to a pressure of 40
trans-7-[3-(Cyclopentyloxy)-4-methoxyphenyl]hexahydro-1H-
bar. The autoclave was slowly (1 h) heated to 70–75 °C and the
pyrrolo[1,2-c][1,3]oxazol-3-one (CMPO 2)
mixture was stirred at this temperature for 4.5 h. Then the
autoclave was cooled to r.t., slowly depressurized and the cata- RANEY® nickel (ca. 0.03 g, washed with MeOH) was placed in
lyst was filtered off. The solvent was evaporated, and the a test tube equipped with a magnetic stirrer and charged
residual mixture of trans-2-phenylcyclohexanol and pyrrolidine with
a
solution of enantiopure tetrahydro-1,2-oxazine
1
15 was dried under vacuum [data for 15: H NMR (300 MHz): (+)-(3R,4S,6S,7S,8R)-12 or (−)-(3S,4R,6R,7R,8S)-12 (0.081 g,
1.42 (s, 9 H, Boc), 1.55–1.68, 1.76–1.99 and 2.18–2.30 (3 m, 11 0.168 mmol) in MeOH (2 mL). The test tube was placed in a
H, H2C(4), H2C(15), H2C(16), NH), 2.75 (ddd, J = 8.3, 8.1, steel autoclave which was then flushed and filled with H2 to a
8.1 Hz, 1 H, HC(3)), 3.02–3.21 (m, 4 H, HC(2), H2C(5), HC(6)), pressure of 25 bar. After the mixture was stirred at r.t. for
3.26 (dd, J = 5.7, 9.3 Hz, 1 H, HC(6)), 3.83 (s, 3 H, H3C(13)), 3.5 h, the autoclave was slowly depressurized and the catalyst
4.78 (m, 1 H, HC(14)), 4.92 (br, 1 H, NHBoc), 6.75 (d, J = was filtered off. The solvent was evaporated, and the residue
8.0 Hz, 1 H, HC(12)), 6.77 (s, 1 H, HC(8)), 6.81 (d, J = 8.0 Hz, was dried under vacuum. The product was dissolved in CH2Cl2
1 H, HC(11)). HRMS: m/z 391.2586 (positive ions); calcd for (1 mL) and 1,1′-carbonyldiimidazole (0.054 g, 0.336 mmol) was
+
[C22H35N2O4 ]: 391.2591].
added. After stirring at r.t. for 2 h the mixture was concen-
The residue was dissolved in DMSO (5.4 mL) and the solu- trated under vacuum and the residue was subjected to column
tion was gently refluxed for 30 min under argon. Then the chromatography on silica gel (eluent: EtOAc–hexane = 1 : 10 →
solvent was evaporated under vacuum (ca. 100 °C/10 Torr) and 1 : 5 → 1 : 1). The first fraction contained (+)- or (−)-trans-2-phenyl-
the residue was subjected to column chromatography on silica cyclohexanol (0.027 g, 91%) and the second fraction contained
gel (eluent: EtOAc–hexane = 1 : 10 → 1 : 5 → 1 : 1 → MeOH– target (−)- or (+)-pyrrolooxazolidinone CMPO 2 (0.041 g, 77%).
EtOAc = 0 : 1 → 1 : 10). Two fractions were collected: the Rf 0.33 (EtOAc–hexane = 1 : 1). 1H NMR (300 MHz, COSY,
EtOAc–hexane fraction contained trans-2-phenylcyclohexanol HSQC): 1.54–1.72 (m, 2 H) and 1.80–1.98 (m, 6 H) (H2C(16)
(34 mg, 81%) and the EtOAc–MeOH fraction contained target and H2C(17)), 2.15 (dddd, J = 12.7, 11.7, 9.5, 8.8 Hz, 1 H, H′C(6)),
enantiopure or racemic pyrroloimidazolone CMPI 1 (42 mg, 2.49 (dddd, J = 12.7, 10.3, 8.1, 2.2 Hz, 1 H, H″C(6)), 2.79 (ddd,
56% from 10). Rf 0.71 (MeOH–EtOAc = 1 : 3). 1H NMR J = 10.3, 9.2, 7.3 Hz, 1 H, HC(7)), 3.48 (ddd, J = 11.0, 9.5,
(300 MHz, COSY, HSQC): 1.56–1.71 (m, 2 H, HC(17)), 1.79–1.94 2.2 Hz, 1 H, H′C(5)), 3.74 (ddd, J = 11.0, 8.8, 8.1 Hz, 1 H,
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 8082–8091 | 8089

View Article Online
Paper
Organic & Biomolecular Chemistry
H″C(5)), 3.82–3.88 (m, 1 H, HC(7a)), 3.84 (s, 3 H, H3C(14)),
4.22 (dd, J = 9.2, 2.1 Hz, 1 H, H″C(1)), 4.43 (dd, J = 9.2, 8.4 Hz,
1 H, H′C(1)), 4.73–4.81 (m, 1 H, HC(15)), 6.73 (s, 1 H, HC(9)),
6.75 (d, J = 7.9 Hz, 1 H, HC(13)), 6.86 (d, J = 7.9 Hz, 1 H,
HC(12)). 13C NMR (75.47 MHz, HSQC, DEPT): 24.0 (17-C), 32.8
(16-C), 34.5 (6-C), 45.8 (5-C), 49.2 (7-C), 56.2 (14-C), 65.8 (7a-C),
66.2 (1-C), 80.7 (15-C), 112.5 (12-C), 114.5 (9-C), 119.4 (13-C),
130.4 (8-C), 148.1 and 149.6 (10-C and 11-C), 161.8 (3-C).
Characteristic 2D-NOESY correlations: HC(7a)/HC(9) and
HC(13), HC(7)/H″C(6), HC(7)/H″C(1), HC(9)/H′C(6), H″C(6)/
H″C(5), H′C(6)/H′C(5). NMR spectra are in accordance with
literature data for racemic CMPO 2.6a,11a
Notes and references
1 A. Yu. Sukhorukov, Y. D. Boyko, Yu. V. Nelyubina,
S. Gerard, S. L. Ioffe, V. A. Tartakovsky, L. Malleret and
A. Belaaouaj, J. Org. Chem., 2012, 77, 5465.
2 For recent reviews see: (a) J. M. Michalski, G. Golden,
J. Ikari and S. I. Renard, Clin. Pharmacol. Ther., 2012, 91,
134; (b) D. Wang and X. Cui, Int. J. COPD, 2006, 4, 373;
(c) D. Spina, Br. J. Pharmacol., 2008, 155, 308; (d) C. Kroegel
and M. Foerster, Expert Opin. Invest. Drugs, 2007, 16, 109;
(e) B. J. Lipworth, Lancet, 2005, 365, 167; (f) K. Fan Chung,
Eur. J. Pharmacol., 2006, 533, 110; (g) W. M. Brown,
Int. J. COPD, 2007, 2, 517; (h) G. P. Currie, C. A. Butler,
W. J. Anderson and C. Skinner, Br. J. Clin. Pharmacol.,
2008, 65, 803; (i) S. P. Page and D. Spina, Curr. Opin.
Pharmacol., 2012, 12, 275; ( j) M. Yeadon, N. Clarke and
J. Ward, in New Drugs and Targets for Asthma and COPD, ed.
T. T. Hansel and P. J. Barnes, Karger, Basel, 2010, vol. 39,
pp. 269–278.
3 (a) K. Ito, S. Lim, G. Caramori, B. Cosio, K. F. Chung,
I. M. Adcock and P. J. Barnes, Proc. Natl. Acad. Sci. U. S. A.,
2002, 99, 8921; (b) J.-P. Wu, Q. Wu, X. Sun and H. Sun,
Chin. Med. J., 2013, 126, 965; (c) L. Hendeles,
M. Weinberger, G. Milavetz, M. Hill and L. Vaughan, Chest,
1985, 87, 758.
(−)-(7S,7aR)-CMPO
2
(obtained from (+)-12): mp
137–139 °C; [α]2D6 −69.1 (c 0.83 in MeOH).
(+)-(7R,7aS)-CMPO
2
(obtained from (−)-12): mp
138–139 °C; [α]2D6 +71.4 (c 0.83 in MeOH).
(−)-trans-2-Phenylcyclohexanol (obtained from (−)-12): mp
61–63 °C (lit. 63–66 °C, Aldrich), [α]2D4 = −54.9 (c 0.5 in MeOH)
[lit.29 [α]D −56.8 (c 1.42 in MeOH)]. 1H NMR spectra are in
accordance with literature data.29
(+)-trans-2-Phenylcyclohexanol (obtained from (+)-12): mp
59–60 °C (lit. 64–66 °C, Aldrich), [α]2D6 +54.8 (c 1.0 in MeOH)
[lit.30 [α]2D0 +57.3 (c 5.0 in MeOH)]. 1H NMR spectra are in
accordance with literature data.29
4 (a) J.-W. Park, S. W. Ryter, S. Y. Kyung, S. P. Lee and
S. H. Jeong, Eur. J. Pharmacol., 2013, 706, 76;
(b) R. Horowski and M. Sastre-y-Hernandez, Curr. Ther. Res.
Clin. Exp., 1985, 38, 23.
Conclusions
In conclusion, the first asymmetric syntheses of two highly potent
PDE IV inhibitors CMPI 1 and CMPO 2 were developed. These
syntheses provide both enantiomers of each product CMPI 1
and CMPO 2 in 8 and 7 steps (11% and 15% overall yields,
respectively) starting from available isovanillin, nitroethane
and (+)- or (−)-trans-2-phenylcyclohexanols. The suggested
strategy is based on the original process of silylation of easily
available six-membered cyclic nitronates. This strategy is flex-
ible, providing a wide variation of substituents in the pyrrol-
idine unit, and, thus, enabling the synthesis of large libraries
of CMPI 1 and CMPO 2 analogs. (−)-(7S,7aR) enantiomers of
CMPI 1 and CMPO 2 were shown to be considerably more
potent inhibitors of the PDE IVB1 enzyme compared to
their (+)-(7R,7aS) enantiomers. A rational model based on
molecular docking studies was suggested to explain the differ-
ence in the PDE inhibitory activity of CMPI 1 and CMPO 2
enantiomers.
5 (a) D. J. Reid and N. T. Pham, Ann. Pharmacother., 2012, 46,
521; (b) N. A. Pinner, L. A. Hamilton and A. Hughes, Clin.
Ther., 2012, 34, 56; (c) P. H. Schafer, A. Parton,
A. K. Gandhi, L. Capone, M. Adams, L. Wu, J. B. Bartlett,
M. A. Loveland, A. Gilhar, Y. F. Cheung, G. S. Baillie,
M. D. Houslay, H. W. Man, G. W. Muller and D. I. Stirling,
Br. J. Clin. Pharmacol., 2010, 159, 842; (d) S. J. Bickston,
K. R. Snider and M. R. Kappus, Exp. Opin. Investig. Drugs,
2012, 21, 1845; (e) A. Gavalda and R. S. Roberts, Exp. Opin.
Ther. Patents, 2013, 23, 997; (f) M. Giembycz, Br. J. Clin.
Pharmacol., 2006, 62, 138.
6 (a) M. F. Brackeen, D. J. Cowan, J. A. Stafford,
F. J. Schoenen, J. M. Veal, P. L. Domanico, D. Rose,
A. B. Strickland, M. Verghese and P. L. Feldman, J. Med.
Chem., 1995, 38, 4848; (b) E. K. Jackson and J. A. Carcillo,
US Pat., 5 849 774, 1998; (c) J.-A. Karlson and D. Aldous,
Exp. Opin. Ther. Patents, 1997, 7, 989.
7 E. E. Polymeropoulos and N. Hofgen, Quant. Struct.-Act.
Relat., 1997, 16, 231.
8 J. Demnitz, L. LaVecchia, E. Bacher, T. H. Keller, T. Müller,
F. Schürch, H.-P. Weber and E. Pombo-Villar, Molecules,
1998, 3, 107.
Acknowledgements
The financial support from the Russian Foundation for Basic
Research (grant 11-03-00737a) and the Russian President’s
Council for Grants (grant MK-3918.2013.3) is greatly acknowl-
9 P.-O. Delaye, T. K. Pradhan, E. Lambert, J.-L. Vasse and
J. Szymoniak, Eur. J. Org. Chem., 2010, 3395 and references
cited therein.
edged. The authors thank Dr Henry Zhu from BPS Bioscience 10 (a) A. Yu. Sukhorukov and S. L. Ioffe, Chem. Rev., 2011, 111,
for conducting PDE in vitro assay.
5004; (b) A. Yu. Sukhorukov, A. D. Dilman and S. L. Ioffe,
8090 | Org. Biomol. Chem., 2013, 11, 8082–8091
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Chem. Heterocycl. Compd., 2012, 48, 49; (c) A. Yu. 18 Under catalytic hydrogenation conditions the azido group
Sukhorukov, A. V. Lesiv, O. L. Eliseev, Yu. A. Khomutova,
S. L. Ioffe and A. O. Borissova, Eur. J. Org. Chem., 2008,
4025; (d) A. Yu. Sukhorukov, A. V. Lesiv, Yu. A. Khomutova
is reduced faster than the N–O bond: A. Yu. Sukhorukov,
A. N. Semakin, A. V. Lesiv, Yu. A. Khomutova and
S. L. Ioffe, Russ. J. Org. Chem., 2007, 43, 1106.
and S. L. Ioffe, Synthesis, 2009, 741; (e) A. Yu. Sukhorukov, 19 The study was conducted on racemic tetrahydro-1,2-
A. V. Lesiv, O. L. Eliseev, Yu. A. Khomutova and S. L. Ioffe, oxazine rac-10 obtained from racemic rac-7 (Scheme 3).
Synthesis, 2009, 2570; (f) A. A. Tabolin, A. V. Lesiv and 20 Tetrahydro-1,2-oxazine 14 was characterized in the reaction
S. L. Ioffe, Synthesis, 2009, 3099.
mixture: [1H NMR (300 MHz): 1.4 (s, 9 H, Boc), 1.27–1.56
and 1.72–2.04 (2 m, 15 H, H2C(9), H2C(10), H2C(11), HC(12),
H2C(22) and H2C(23)), 2.10–2.24 (m, 3 H, HC(12) and
H2C(5)), 2.45 (ddd, J = 10.8, 8.5, 3.0 Hz, 1 H, HaxC(8)), 2.90
(dd, J = 10.0, 9.8, 3.3 Hz, 1 H, HaxC(4)), 2.93 (m, 2 H,
H2C(13)), 3.36 (m, 1 H, HC(3)), 3.62–3.74 (m, 2 H, HC(7)
and NH), 3.79 (s, 3 H, H3C(20)), 4.56 (br, 1 H, NHBoc), 4.74
(m, 1 H, HC(21)), 4.94 (d, J = 1.0 Hz, 1 H, HeqC(6)), 6.56 (s,
1 H, HC(15)), 6.58 (d, J = 7.5 Hz, 1 H, HC(19)), 6.77 (d, J =
7.5 Hz, 1 H, HC(18)), 7.21–7.46 (m, 5 H, Ph). HRMS: m/z =
581.3569 (positive ions); calcd for [C34H49N2O6+]:
581.3585].
11 (a) P. A. Zhmurov, A. A. Tabolin, A. Yu. Sukhorukov,
M. S. Klenov, A. V. Lesiv, S. L. Ioffe and V. A. Tartakovsky,
Russ. Chem. Bull., 2011, 60, 2390; (b) A. Yu. Sukhorukov,
A. V. Lesiv, Yu. A. Khomutova, S. L. Ioffe and
V.
A.
Tartakovsky,
Synthesis,
2009,
1999;
(c) A. Yu. Sukhorukov, Y. D. Boyko, Yu. A. Khomutova, Yu.
V. Nelyubina, S. L. Ioffe and V. A. Tartakovsky, J. Org.
Chem., 2011, 76, 7893.
12 On the [4 + 2] cycloaddition of nitroalkenes to chiral vinyl
ethers see: (a) S. E. Denmark and A. Thorarensen, Chem.
Rev., 1996, 96, 137; (b) S. E. Denmark, M. E. Schnute and
C. B. W. Senanayake, J. Org. Chem., 1993, 58, 1859; 21 For an indepth discussion on the mechanism of 1,2-
(c) S. E. Denmark and E. A. Martinborough, J. Am. Chem.
Soc., 1999, 121, 3046.
13 A. S. Cantrell, P. Engelhardt, M. Högberg, S. R. Jaskunas,
oxazine hydrogenation see review10a and ref. 10c.
22 D. Dunkan and T. Livinghouse, J. Org. Chem., 2001, 66,
5237.
N. G. Johansson, C. L. Jordan, J. Kangasmetsä, 23 O. Trott and A. J. Olson, J. Comput. Chem., 2010, 31,
M. D. Kinnick, P. Lind, J. M. Morin Jr., M. A. Muesing, 455.
R. Noreen, B. Oberg, P. Pranc, C. Sahlberg, R. J. Ternansky, 24 F. C. Bernstein, T. F. Koetzle, G. J. Williams, E. E. Meyer Jr.,
R. T. Vasileff, L. Vrang, S. J. West and H. Zhang, J. Med.
Chem., 1996, 39, 4261.
M. D. Brice, J. R. Rodgers, O. Kennard, T. Shimanouchi
and M. Tasumi, J. Mol. Biol., 1977, 112, 535.
14 For stereoselective reduction of cyclic oxime ethers see ref. 25 Despite the structural diversity of the PDE IVB structures
10, 11 and (a) R. Zimmer, T. Arnold, K. Homann and
H.-U. Reissig, Synthesis, 1994, 1050; (b) E. Schmidt,
H.-U. Reissig and R. Zimmer, Synthesis, 2006, 2074;
(c) H.-U. Reissig, K. Homann, F. Hiller and R. Zimmer, Syn-
thesis, 2007, 2681; (d) J. K. Gallos, V. C. Sarli, A. C. Varvogli,
available to the moment (∼35) and their co-crystallized
ligands the binding site of the PDE IVB is rather rigid. For
example, overlapping of the 3KKT and 1XLX results in the
almost identical binding site except the C-capped terminus
(see Fig. 1 in ESI‡), suggesting that the interactions of
ligands with these binding sites will be very similar. Thus,
only one PDE IVB structure (3KKT) was used to study
binding of CMPI 1 and CMPO 2 enantiomers.
C.
N. G. Argyropoulos, Tetrahedron Lett., 2003, 44, 3905.
15 For hydrolysis of benzyl bromides to alcohols with AgNO3/
Z.
Papadoyanni,
S.
D.
Papaspyrou
and
H2O see: T. Hayashi, K. Hayashizaki, T. Kiyoi and Y. Ito, 26 T. Cheng, X. Li, Y. Li, Z. Liu and R. Wang, J. Chem. Inf.
J. Am. Chem. Soc., 1988, 110, 8153. Model., 2009, 49, 1079.
16 Formation of nitrates in reaction of halides with AgNO3 27 N. M. O’Boyle, M. Banck, C. A. James, C. Morley,
was observed before: (a) C. Wessler, A. Homann, U. Fricke
and J. Lehmann, Eur. J. Med. Chem., 2003, 38, 581;
T. Vandermeersch and G. R. Hutchison, J. Cheminformatics,
2011, 3, 33.
(b) M. J. Haddadin, A. M. A. Kattan and C. H. Issidorides, 28 The PyMOL Molecular Graphics System, Version 0.99, DeLano
J. Org. Chem., 1985, 50, 129; (c) H. Kwart and
M. W. Brechbiel, J. Org. Chem., 1982, 47, 461.
Scientific LLC.
29 S. B. King and K. B. Sharpless, Tetrahedron Lett., 1994, 35,
17 For reduction of nitrates to alcohols with zinc see:
5611.
E. Baciocchi, T. Del Giacco, S. M. Murgia and 30 G. Cravotto, G. B. Giovenzana, T. Pilati, M. Sisti and
G. V. Sebastiani, Tetrahedron, 1988, 44, 6651.
G. Palmisano, J. Org. Chem., 2001, 66, 8447.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 8082–8091 | 8091