Synthesis of "neoprofen", a rigidified analogue of ibuprofen, exemplifying synthetic methodology for altering the 3-D topology of pharmaceutical substances

Article abstract of DOI:10.1039/c6ob01351a

3,3-Dimethylcyclopentanes (neopentylenes) are ubiquitous in Nature but largely absent from synthetic pharmaceutical libraries. Neopentylenes define a hydrophobic and rigid 3-D topology with distinct molecular pharmacology, as exemplified here with two neo

Full text of DOI:10.1039/c6ob01351a

Organic &
Biomolecular Chemistry
View Article Online
View Journal
COMMUNICATION
Synthesis of “neoprofen”, a rigidified analogue of
ibuprofen, exemplifying synthetic methodology
for altering the 3-D topology of pharmaceutical
substances†
Ron R. Ramsubhag,a Chelsea L. Massaro,a Christina M. Dadich,a
Andrew J. Janeczek,a Tung T. Hoang,a Elizabeth A. Mazzio,b Suresh Eyunni,b
Karam F. A. Solimanb and Gregory B. Dudley*a,c
Cite this: DOI: 10.1039/c6ob01351a
Received 22nd June 2016,
Accepted 1st August 2016
DOI: 10.1039/c6ob01351a
www.rsc.org/obc
3,3-Dimethylcyclopentanes (neopentylenes) are ubiquitous in
Nature but largely absent from synthetic pharmaceutical libraries.
Neopentylenes define a hydrophobic and rigid 3-D topology with
distinct molecular pharmacology, as exemplified here with two
neopentylene-fused analogues of the synthetic anti-inflammatory
drug, ibuprofen.
Rigid structural features lend themselves to specific bio-
molecular interactions. This Communication features pre-
liminary results from synthetic efforts focused on the
installation of the neopentylene ring fusion (i.e., in polycyclic
3,3-dimethyl-cyclopentanes), a rigid and hydrophobic struc-
tural motif that, in principle, can enhance the three-
dimensional topology of known and novel pharmacophores.
Neopentylene ring-fused compounds are ubiquitous in
Nature (cf. Fig. 1, top), but synthetic variants are largely absent
from pharmaceutical screening libraries and drug discovery
programs. Illudol1 and pentalenene,2 for example, have long
stood as challenging targets for total synthesis, in part due to
the difficulty of crafting the hydrophobic neopentylene ring
fusion. Syntheses of tremulenolide A,3 illudalic acid,4 and
alcyopterosin A5 pose similar problems. Medicinal chemistry
efforts associated with illudalic acid and alcyopterosin A
focused on simplified synthetic analogues in which the neo-
pentylene ring fusion was deleted6 or truncated,7 respectively,
for synthetic convenience. The structurally simplified ana-
logues were universally less potent than their neopentylene-
fused congeners.
aDepartment of Chemistry and Biochemistry, Florida State University, Tallahassee,
Florida 32306-4390, USA. E-mail: gdudley@chem.fsu.edu
bCollege of Pharmacy and Pharmaceutical Sciences, Florida Agricultural and
Mechanical University, Tallahassee, Florida 32307, USA
Fig. 1 Top: Natural products featuring a neopentylene ring fusion.
Middle: Ibuprofen (IBU) and “neoprofen” (1), the current focus. Bottom:
Naproxen and flurbiprofen.
cC. Eugene Bennett Department of Chemistry, West Virginia University, Morgantown,
West Virginia 26506, USA
One rare example of a synthetic neopentylene is the ibupro-
fen analog 1 (Fig. 1, middle), in which the 4-isobutyl sub-
stituent has been replaced with a 3,4-neopentylene ring.
†Electronic supplementary information (ESI) available: Experimental pro-
cedures, characterization and molecular docking data, copies of NMR spectra.
See DOI: 10.1039/c6ob01351a
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Communication
Organic & Biomolecular Chemistry
Ibuprofen analog 1, which we refer to herein as “neoprofen”, relevant interactions relative to ibuprofen. Computational
has been prepared in 9 steps from benzaldehyde as described modelling provided insight into the potential validity of this
in a 1979 patent.8 The pharmacology of synthetic neopentyl- hypothesis and guided the design of a second ibuprofen
enes is largely unexplored; our central hypothesis is that the analog. Computationally overlaying ibuprofen (IBU) and 1 in
neopentylene provides a rigid, topologically unique, and the cyclooxygenase II (COX-2) enzyme binding pocket illus-
hydrophobic anchor that perturbs biomolecular interactions trated how the compounds may bind (Fig. 2). The carboxylic
relative to truncated and/or more conformationally flexible acid of IBU forms a salt bridge with arginine 121 of the COX-2
analogues. Here we offer some initial validation of this central enzyme, locking the molecule in place. The isobutyl tail of IBU
hypothesis in the context of neoprofen (1) and “homo- extends into a hydrophobic pocket formed by phenylalanine
neoprofen” (2), including preliminary pharmacological activity and tyrosine residues. Analog 1 can form a similar salt bridge
reminiscent of but not identical to ibuprofen (IBU).
with arginine 121. However, the neopentylene ring of 1 is more
The preliminary pharmacological assessments described rigid and compact than the isobutyl group of IBU, such that
herein are understood in the context of known structure– the neopentylene ring cannot extend as deeply into the hydro-
activity relationship (SAR) observations regarding the IBU core. phobic pocket as the isobutyl group.
For example, naproxen9 and flurbiprofen10 (Fig. 1, bottom) are
As noted above, our central hypothesis is that the neo-
both better inhibitors of the COX-2 enzyme than is ibuprofen. pentylene ring of 1 will perturb pharmacological interactions
Based on X-ray crystal data, molecular docking, and muta- relative to the more conformationally flexible (and more
genesis studies, the latter substances (naproxen and flurbiprofen) linear) isobutyl group of IBU. We therefore expected there to
bind analogously to IBU in the COX-2 binding site, with the be observable differences in the molecular pharmacology of
differences in affinity associated with variable extension of IBU and 1. As a logical extension of these hypotheses, we
hydrophobic aromatic regions of the molecules into a hydro-
phobic enzyme pocket.11 Neoprofen is more massive but
compact and conformationally restricted relative to IBU, so it
is not expected to extend as deeply into the hydrophobic
pocket of the COX-2 enzyme.12
Our synthesis of neoprofen (1, Scheme 1) draws on tandem
fragmentation/olefination methodology13 for the synthesis of
neopentyl-tethered 1,6-enynes, which are suitable for prepar-
ing diverse synthetic neopentylenes by strategic application of
myriad enyne cycloisomerization and annulation methods.
Here we expand this methodology to tandem fragmentation/
vinylogous olefination to prepare neopentyl-tethered dienyne 5
in three steps from commercially available dimedone. Addition
of methyllithium to 5 is followed by a rhodium-catalyzed intra-
molecular [4 + 2] formal Diels–Alder reaction,14 with sub-
sequent DDQ oxidation to give rise to indane 7. We employed
a similar sequence of Rh-catalyzed cycloisomerization and oxi-
dation in our recent synthesis of alcyopterosin A.15 Here, ter-
tiary alcohol 7 was formally rearranged to primary alcohol 9 by
dehydration followed by hydroboration/oxidation. Finally, two-
stage oxidation provided neoprofen (1).
Our synthesis of neoprofen was guided by the hypothesis
Fig. 2 Ibuprofen (IBU, blue) vs. neoprofen (1, green) overlaid in the
that the neopentylene ring would perturb pharmacologically COX-2 enzyme pocket.
Scheme 1 Synthesis of neoprofen (1) (see ESI† for details).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Communication
reasoned that if we could lengthen and restore some confor-
mational flexibility to 1, then we might observe less of a differ-
ence in molecular pharmacology as compared to IBU.
Therefore, we designed and synthesized “homoneoprofen” 2
(cf. Fig. 1), which we reasoned would penetrate deeper into the
hydrophobic pocket (by analogy to naproxen or flurbiprofen)
than 1 and may offset perturbations in binding associated
with the switch from the isobutyl chain of IBU to the neo-
pentylene ring of 1. Additional molecular docking studies on
murine cyclooxygenase II (COX-2) suggest that analogues
1 and 2 should bind COX-2 in the same pocket as IBU (see
ESI† for details).
To obtain “homoneoprofen” (2, Scheme 2), we started from
dienyne 5 (cf. Scheme 1). DIBAL-H reduction gave primary
allylic alcohol 10. Oxidative cycloisomerization was again
accomplished in a two-stage process using Wilkinson’s catalyst
in TFE, followed by addition of DDQ. In this case, however,
DDQ treatment resulted in a productive secondary oxidation of
the expected primary alcohol to aldehyde 10a. Aldehyde 10a
was isolated but not purified before being converted to unsatu-
rated ester 11 using Horner–Wadsworth–Emmons conditions.
Hydrogenation of 11 gave saturated ester 12, and finally sapo-
nification provided homoneoprofen 2.
Fig. 3 Human COX-2 activity inhibition assay; activity of homoneopro-
fen (2, FSU2, red bar) was intermediate between that of ibuprofen (IBU,
yellow bar) and neoprofen (1, FSU1, blue bar). The data represent COX-2
activity (% control) and are presented as the mean
st. dev, n = 2.
Statistical difference from the controls were determined by a students
t-test. *P < 0.05 (see ESI† for details).
With 1 and 2 in hand, we proceeded to sample their activity
in the human COX-2 enzyme and relative to IBU (Fig. 3). Com-
pared to ibuprofen (IC50 0.02 µg mL−1 or 1 µM in our assays),
neoprofen 1 showed relatively poor activity: IC50 4 µg mL−1 long-term goal of understanding the unique hydrophobic 3-D
(20 µM). This observation is consistent with our central topology of the neopentylene ring fusion and its potential role
hypothesis: replacing the isobutyl side chain with a neopentyl- in drug discovery. Neopentylene ring-fused polycyclic struc-
ene ring perturbs molecular pharmacological interactions. In tures are ubiquitous in Nature but largely absent from modern
this case, activity was reduced as a consequence of the rigid, synthetic pharmaceutical screening libraries. The omission of
compact neopentylene ring. The intermediate activity of homo- neopentylene-fused molecular substances is perhaps related to
neoprofen 2 (IC50 0.4 µg mL−1 or 2 µM; i.e., between that of historical limitations in the synthetic chemistry of neopentyl-
IBU and 1) is consistent with our design criteria for 2 of restor- tethered bifunctional building blocks; current and on-going
ing two-dimensional length and conformational flexibility so methodology in these laboratories is aimed at addressing and
as to extend more deeply into the COX-2 binding pocket than overcoming these limitations. The synthesis of so-called neo-
does neoprofen (1). These observations are qualitatively con- profen (1) and homoneoprofen (2) reflect the difficulties
sistent with the SAR trends associated with naproxen9,11 and associated with preparing neopentylene-fused compounds—
flurbiprofen.10
more work is certainly needed in this area—but these com-
In conclusion, we have designed, prepared, and analysed pounds also highlight the potential pharmacological signifi-
two neopentylene-fused analogues of ibuprofen, toward the cance of this compact, rigid, hydrophobic, and topologically
unique structural feature. Although neoprofen (1) is higher
molecular weight than ibuprofen by virtue of its one extra
carbon atom, neoprofen (1) seemingly cannot occupy as much
conformational space as ibuprofen in the relevant COX-2
binding pocket due. Preliminary support for this tentative
interpretation of the assay data comes from the broader SAR
associated with naproxen and flurbiprofen as well as the
restored binding affinity of homoneoprofen (2), which adds to
the deviation from ibuprofen in terms of molecular weight but
restores the 2-D length and conformational flexibility that was
found to be advantageous in molecular docking simulations
(see ESI† for details). On the other hand, the rigidity and
compact nature of 1 (and other neopentylenes) should be gen-
erally advantageous for ligation in compact and rigidly defined
Scheme 2 Synthesis of homoneoprofen (2) (see ESI† for details).
binding pockets. Therefore, we anticipate that neopentylene
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Communication
Organic & Biomolecular Chemistry
ring-fused structures should have strategic value in molecular
pharmacology.
7 L. M. Finkielsztein, A. M. Bruno, S. G. Renou and
G. Y. M. Iglesias, Bioorg. Med. Chem., 2006, 14, 1863–1870.
8 T. G. Payne, Indaneacetic acid derivatives, U.S. Patent
4166131, issued August 28, 1979.
9 O. Laneuville, D. K. Breuer, D. L. Dewitt, T. Hla, C. D. Funk
and L. W. Smith, J. Pharmacol. Exp. Ther., 1994, 271, 927–
934.
This research was supported by grants from the National
Science Foundation (NSF-CHE 1300722) and the National Insti-
tute of Minority Health and Health Disparities of the National
Institutes of Health through Grant Number 8 G12MD007582-
28 and Grant Number 1P20 MD006738-01. We thank Prof. Wei
Yang (FSU) for helpful discussions and assistance with mole- 10 R. G. Kurumbail, A. M. Stevens, J. K. Gierse, J. J. McDonald,
cular modelling.
A. R. Stegeman, J. Y. Pak, D. Gildehaus, J. M. Miyashiro,
T. D. Penning, K. Seibert, P. C. Isakson and W. C. Stallings,
Nature, 1996, 384, 644–648.
11 K. C. Duggan, M. J. Walters, J. Musse, J. M. Harp,
J. R. Kiefer, J. A. Oates and L. J. Marnett, J. Biol. Chem.,
2010, 285, 34950–34959.
12 There may also be important metabolic difference between
the neopentylene ring fusion of neoprofen compared to the
isobutyl group of IBU, but such differences would not be
apparent in our preliminary assays.
13 T. T. Hoang and G. B. Dudley, Org. Lett., 2013, 15, 4026–
4029.
14 R. S. Jolly, G. Luedtke, D. Sheehan and T. Livinghouse,
J. Am. Chem. Soc., 1990, 112, 4965–4966.
Notes and references
1 T. C. McMorris, M. S. R. Nair and M. Anchel, J. Am. Chem.
Soc., 1967, 89, 4562–4563.
2 B. K. Koe and A. B. Sobin, Antibiot. Annu., 1957, 672.
3 W. A. Ayer and E. R. Cruz, J. Org. Chem., 1993, 58, 7529.
4 J. F. Ammirati, Poisonous Mushrooms of the Northern United
States and Canada, E. P. Dutton, New York, N.Y., 1972, p. 84.
5 J. A. Palermo, M. F. Rodriguez Brasco, C. Spagnuolo and
A. M. Seldes, J. Org. Chem., 2000, 65, 4482–4486.
6 Q. Ling, Y. Huang, Y. Zhou, Z. Cai, B. Xiong, Y. Zhang,
L. Ma, X. Wang, X. Li and J. Li, Bioorg. Med. Chem., 2008,
16, 7399–7409.
15 T. T. Hoang, M. Birepinte, N. M. Kramer and G. B. Dudley,
Org. Lett., 2016, 18, 3470–3473.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016