Synthesis, chiral high performance liquid chromatographic resolution and enantiospecific activity of a potent new geranylgeranyl transferase inhibitor, 2-hydroxy-3-imidazo[1,2-a]pyridin-3-yl-2-phosphonopropionic Acid

Article abstract of DOI:10.1021/jm900232u

3-(3-Pyridyl)-2-hydroxy-2-phosphonopropanoic acid (3-PEHPC, 1) is a phosphonocarboxylate (PC) analogue of 2-(3-pyridyl)-1- hydroxyethylidenebis(phosphonic acid) (risedronic acid, 2), an osteoporosis drug that decreases bone resorption by inhibiting farnesyl pyrophosphate synthase (FPPS) in osteoclasts, preventing protein prenylation. 1 has lower bone affinity than 2 and weakly inhibits Rab geranylgeranyl transferase (RGGT), selectively preventing prenylation of Rab GTPases. We report here the synthesis and biological studies of 2-hydroxy-3-imidazo[1,2-a]pyridin-3-yl-2- phosphonopropionic acid (3-IPEHPC, 3), the PC analogue of minodronic acid 4. Like 1, 3 selectively inhibited Rab11 vs. Rap 1A prenylation in J774 cells, and decreased cell viability, but was 33-60 × more active in these assays. After resolving 3 by chiral HPLC (>98% ee), we found that (+)-3-E1 was much more potent than (-)-3-E2 in an isolated RGGT inhibition assay, ～17 × more potent (LED 3 μM) than (-)-3-E2 in inhibiting Rab prenylation in J774 cells and >26× more active in the cell viability assay. The enantiomers of 1 exhibited a 4-fold or smaller potency difference in the RGGT and prenylation inhibition assays.

Full text of DOI:10.1021/jm900232u

3454 J. Med. Chem. 2010, 53, 3454–3464
DOI: 10.1021/jm900232u
Synthesis, Chiral High Performance Liquid Chromatographic Resolution and Enantiospecific
Activity of a Potent New Geranylgeranyl Transferase Inhibitor, 2-Hydroxy-3-imidazo[1,2-a]pyridin-
3-yl-2-phosphonopropionic Acid
†
†
†
§
Charles E. McKenna,*,† Boris A. Kashemirov, Katarzyna M. Bzazewska, Isabelle Mallard-Favier, Charlotte A. Stewart,
Javier Rojas,§ Mark W. Lundy,‡z Frank H. Ebetino,‡z Rudi A. Baron, James E. Dunford,^ Marie L. Kirsten,
Miguel C. Seabra, Joy L. Bala,† Mong S. Marma,† Michael J. Rogers,§ and Fraser P. Coxon*,§
_
†Department of Chemistry, University of Southern California, Los Angeles, California 90089-0744, ‡zProcter & Gamble Pharmaceuticals, Mason,
Ohio 45040, §Bone & Musculoskeletal Programme, Institute of Medical Sciences, University of Aberdeen, Aberdeen AB25 2ZD, U.K., Molecular
Medicine, National Heart and Lung Institute, Imperial College London, London SW7 2AZ, U.K., and ^Nuffield Department of Orthopaedic
Surgery, Oxford University, Institute of Musculoskeletal Sciences, Nuffield Orthopaedic Center, Headington, Oxford OX3 7LD, U.K.
Received February 23, 2009
3-(3-Pyridyl)-2-hydroxy-2-phosphonopropanoic acid (3-PEHPC, 1) is a phosphonocarboxylate (PC)
analogue of 2-(3-pyridyl)-1-hydroxyethylidenebis(phosphonic acid) (risedronic acid, 2), an osteoporo-
sis drug that decreases bone resorption by inhibiting farnesyl pyrophosphate synthase (FPPS) in
osteoclasts, preventing protein prenylation. 1 has lower bone affinity than 2 and weakly inhibits Rab
geranylgeranyl transferase (RGGT), selectively preventing prenylation of Rab GTPases. We report here
the synthesis and biological studies of 2-hydroxy-3-imidazo[1,2-a]pyridin-3-yl-2-phosphonopropionic
acid (3-IPEHPC, 3), the PC analogue of minodronic acid 4. Like 1, 3 selectively inhibited Rab11 vs. Rap 1A
prenylation in J774 cells, and decreased cell viability, but was 33-60ꢀ more active in these assays. After
resolving 3 by chiral HPLC (>98% ee), we found that (þ)-3-E1 was much more potent than (-)-3-E2 in
an isolated RGGT inhibition assay, ∼17ꢀ more potent (LED 3 μM) than (-)-3-E2 in inhibiting Rab
prenylation in J774 cells and >26ꢀ more active in the cell viability assay. The enantiomers of 1 exhibited a
4-fold or smaller potency difference in the RGGT and prenylation inhibition assays.
Introduction
important regulators of organelle biogenesis and vesicle trans-
port.7 The SAR for inhibition of RGGT by PCs differs from
that for inhibition of FPPS by N-BPs.5 R-Halo and R-desoxy
derivatives of 1 inhibit RGGT with similar potency to the
parent PC, but R-halo and R-desoxy derivatives are generally
weaker inhibitors of FPPS than 2 itself.8,9
Rab proteins such as Rab11 and Rab25 may contribute to
the aggressiveness and progression of various cancers, includ-
ing those that frequently metastasize to bone.6,10 RGGT is
overexpressed in a subset of human tumors, and there is
evidence that the anticancer effects of some compounds
designed as inhibitors of FPPS might involve suppression of
prenylation mediated by RGGT,11 stimulating interest in the
design of specific inhibitorsfor this enzyme.12 PCanalogues of
N-BP drugs exhibit antitumor properties in vitro, both by
inhibiting cell invasion13 and reducing cell viability,4 and 1 has
exhibited direct antitumor activity in an animal model.14 1 has
only modest potency as an RGGT inhibitor, prompting a
search for a more active analogues.
In this work, we describe the synthesis (6 steps) of 2-hydroxy-
3-imidazo[1,2-a]pyridin-3-yl-2-phosphonopropionic acid
(3-IPEHPC, 3), which is the PC analogue of minodronic acid,
4.15 In contrast to the bisphosphonate 4, the carbon bridging
the two acidic groups 3 is chiral, raising the possibility of
stereospecificity in its biological activity. The same achiral-
chiral relationship exists between 2 and 1. We have developed
a method to resolve the 1 and 3 racemates using chiral
HPLC, permitting individual evaluation of the component
Bisphosphonates are well-established antiresorptive drugs
due to their ability to inhibit osteoclast-mediated bone resorp-
tion. Bisphosphonates also exhibit a range of other biological
properties in vitro, including antitumor, antiparasitic, and
antibacterial activities, raising the possibility of additional
therapeutic uses in vivo.1,2 The most potent bisphosphonate
drugs currently used to treat bone metabolism disorders such
as osteoporosis all contain an amino or other N-containing
group (N-BPs) and have been shown to impede bone resorp-
tion by inhibiting farnesyl diphosphate synthase (FPPS),
thereby blocking the synthesis of isoprenoid lipids required
for the prenylation of small GTPases in osteoclasts.2
Phosphonocarboxylate (PC) analogues of N-BP drugs,
e.g. 3-(3-pyridyl)-2-hydroxy-2-phosphonopropanoic acid (3-
PEHPC, 1), whose structure relates to the bone active drug
2-(3-pyridyl)-1-hydroxyethylidenebis(phosphonic acid) (rised-
ronic acid, 2; formulated as sodium risedronate), exhibit lower
bone affinity than N-BPs but retain some ability to block
protein prenylation.3-6 However, PCs such as 1 selectively
inhibit a different enzyme in the mevalonate pathway, Rab
geranylgeranyl transferase (RGGT), thereby preventing
prenylation of only the Rab family GTPases,3,5 which are
*To whom correspondence should be addressed. For chemistry ques-
tions, contact CE McKenna (mckenna@usc.edu, Tel: þ1-213-740-7007).
For biology questions, contact FP Coxon (f.p.coxon@abdn.ac.uk,
Tel: þ44-1224-558974).
r
pubs.acs.org/jmc
Published on Web 04/15/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3455
enantiomers.16 In a companion study of the kinetics and
mechanism of RGGT-mediated Rab geranylgeranylation,
we found that (þ)-3 had significantly lower IC50 and Ki values
than 1 in inhibiting this enzyme.17 We here compare the
relative inhibitory potencies of the individual 3 and 1 stereo-
isomers in isolated RGGT, J744 cell viability and cellular
prenylation assays, and also examine their selectivity vs.
FPPS. The antiresorptive properties of 1 and 3 are examined
using both in vitro (rabbit osteoclasts) and in vivo (Schenk)
models (Chart 1).
was successfully employed.23 The reduction of 8 was effected,
after some exploration of alternative conditions,24 by hydro-
genation over 10% Pd/C, giving the enamine product
Chart 1. Phosphonocarboxylic Acids 3-PEHPC (1) and3-IPEHPC
(3) and the Parent Bisphosphonic Acids, Risedronic Acid (2) and
Minodronic Acid (4)
Results and Discussion
Synthesis of 3. The point of departure for our preparation of
3 was a constructive synthesis from the patent literature18 in
which condensation of aldehyde 6 with N,N-dimethylglycine
ethyl ester followed by further transformations leads to the
gateway enol 10. However, our preliminary attempts to effect
this initial condensation were unsuccessful, and we ultimately
devised a new route to intermediate 10, outlined in Scheme 1.
Our synthesis of 3 begins with a Vilsmeier-Haack for-
mylation of imidazo[1,2-a]pyridine 5 to give the aldehyde 6,
following the approach of Almirante et al.19 as modified by
Gueiffier et al.20 By washing the crude product with water,
we were able to isolate clean 6 without chromatographic
purification. In the next step, ethyl azidoacetate 7, readily
prepared from ethyl bromoacetate and sodium azide,21 was
condensed with 6 using sodium ethoxide dissolved in etha-
nol.22 When following the original procedure,22 the conden-
sation was accompanied by an uncontrolled temperature rise
with release of gas (possibly due to exothermic ethyl azidoa-
cetate rearrangement22). This problem was circumvented
by decreasing the equivalents of azide 7 from 10 to 4,
as suggested by the preparation of the t-butyl esters of
2-azidocinnamate analogues where a smaller azide excess
Figure 1. Enantiomers of 1 have small differences in activity. (A)
Comparative effects of 1 and its enantiomers on Rab prenylation in
J774 cells. Cells were treated for 24 h with 31-500 µM of the stereo-
isomers of 1, or the racemate, then lysed in triton X-114 buffer and
fractionated aqueous phases (containing unprenylated Rab proteins),
electrophoresed and Western blotted for Rab11 or β-actin. (B) Selecti-
vity of 1 enantiomers on GTPase prenylation. J774 cells were meta-
bolically labeled for 18 h with [14C]mevalonate, in the presence or
absence of 1 mM compounds, then radiolabeled; prenylated GTPases
were detected by phosphorimaging after separating proteins in cell
lysates by electrophoresis. (C) Effect of 1 and its enantiomers on activity
of RGGT in isolated enzyme assays, assessed by measuring the incor-
poration of [3H]geranylgeranyl (from [3H]GGPP) into Rab1a.
Scheme 1. Synthesis of 2-Hydroxy-3-imidazo[1,2-a]pyridin-3-yl-2-phosphonopropionic Acid, 3
Reagents and conditions: (a) Vilsmeier reagent, from 2-140 ꢀC, 31%; (b) EtONa/EtOH, from -30 ꢀC to rt, 4 h, 55%; (c) H2/10% Pd-C, MeOH,
2.5 h, rt, 100%; (d) AcOH/H2O (7/1, v/v), 1.5 h, 0 ꢀC, 59%; (e) (EtO)2P(O)H, 70 ꢀC, 21 h; (f ) 6 N HCl, 6 h, reflux, 60%, (e) and (f ) combined.

3456 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
McKenna et al.
Figure 2. Compound 3 is a potent inhibitor of RGGT. (A) Effect of 4, 1, and 3 on J774 cell viability. Cells were treated with compounds for 48 h
and then viable cell number assayed by Alamar blue assay (mean ( SEM; n = 3). (B,C) Effect of 4, 1, and 3 on Rab prenylation in J774 cells.
The cells were treated for 24 h as indicated and then lysed in triton X-114 buffer and fractionated aqueous phases (containing unprenylated Rab
proteins) were Western blotted for Rab11, or β-actin.
9 which was then hydrolyzed by treatment with 87% v/v
acetic acid, providing the key intermediate 10, in an overall
yield of 10% (from 5). A 1H NMR study of this compound in
CDCl3 showed that <1% of the keto tautomer of 10, ethyl
2-oxo-3-(imidazo[1,2-a]pyridin-3-yl)propanoate, was pre-
sent, confirming the expected stabilization of the enol form
by interaction with the extended π orbital system of the
heterocycle. Addition of diethyl phosphite to 10 proceeded
smoothly to provide the phosphonate adduct 11, which
however was found to be of limited stability due to rearran-
gement to its phosphate isomer. Phosphonate-phosphate
rearrangements of R-hydroxy phosphonate diesters are
well documented.25 As a result, crude 11 was not purified
but instead was quickly subjected to hydrolysis in refluxing
concentrated hydrochloric acid, giving the acid 3, which was
isolated as a crystalline solid by precipitation and washing
with ethanol and water. The compound was characterized by
1H and 31P NMR, HRMS, and elemental analysis.
Resolution of 1 and 3 Enantiomers. The role of chirality in
the biological activity of bisphosphonate analogues has received
little attention in the past. R,R-Disubstituted methylenebisphos-
phonate drugs such as 2, zoledronate, alendronate, ibandronate,
and clodronate are prochiral at the P-C-P carbon and have
achiral substituents. We recently reported X-ray crystallographic
evidence that [6,7-dihydro-5H-cyclopenta[c]pyridin-7-yl(hydroxy)-
methylene]bis(phosphonic acid), an analogue of 2 incorporating
a chiral bicyclic substituent, showed stereospecificity in forming
its active site complex with FPPS.26
quinine (ProntoSIL AX QN)-quinidine (ProntoSIL AX
QD; the pseudoenantiomeric quinidine QD column has
opposite chirality to the QN column) weak anion exchange
HPLC columns introduced by Lindner et al.27 Retention of
the analyte by the stationary phase is believed to involve π-π
donor-acceptor, hydrogen bonding, and steric as well as
ion-pairing interactions. Cinchona alkaloid-based columns
have proved effective for the resolution of a number of
different racemic monoacids27 but have not been previously
applied to multifunctional substrates such as 1 or 3, which
contain hydroxy- and nitrogen-containing heterocycle
groups in addition to two different kinds of acid function-
ality with pKas ranging from ∼2 to ∼7.
In preliminary studies using the QN column, we found
that the peaks of the 1 or 3 enantiomers partially overlap and
tail. One-pass separation gave the faster-eluting enantiomer
(E1) with ∼95% ee and the slowly eluting one (E2) with
∼70-80% ee, as determined by reanalysis on QN and QD
columns (the enantiomer with the shorter retention time on
the QN column is dextrorotatory and is therefore referred to
as (þ)-3-E1 or (þ)-1-E1, and the more slowly eluting,
levorotatory enantiomer as (-)-3-E2 or (-)-1-E2; the order
of elution is reversed on the QD column). Attempts to
improve the enantioseparation by use of partial or full ester
derivatives of the desoxy analogue of 1, 13, proved successful
for its P,P-diester analogue 14 (see Supporting Information
for details), but attempts to convert a 1 (or 3) triester to its
P,P-diester 15 under basic conditions led to rearrangement
(to a phosphate ester). We therefore decided to concentrate
on chiral HPLC reprocessing of the first-pass, enantiomer-
enriched fractions obtained from the unmodified compounds.
Because (-)-3-E2 elutes from the QN column within the tail
of the faster-moving (þ)-3-E1 peak, it was advantageous to
carry out its second enrichment pass using the QD column
Unlike the bisphosphonates, all nonequivalently R,R-di-
substituted phosphonocarboxylates have a chiral bridging
R-carbon, but the relationship of this inherent stereoisomer-
ism to their biological activities has remained unexplored.
In search of a general method to resolve racemic PCs,
we investigated separation on the commercially available

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3457
Figure 3. Compound 3 does not affect FPPS or protein farnesylation. (A) Assay for FPPS activity, in which partially purified recombinant
human FPPS was preincubated with 2 or 3 for 10 min prior to commencing the assay by adding [14C]IPP enzyme substrate. The data are the
mean ( SEM (n = 4). (B) Effects of 2 and 3 on prenylation of Ras, Rap1, and Rab6 in J774 cells. The cells were treated for 24 h and then
prenylated and unprenylated proteins separated by triton X-114 fractionation. Resulting aqueous (containing unprenylated proteins; U) and
detergent-enriched (containing prenylated proteins; P) phases were then Western blotted for Ras, Rap1, Rab6, or β-actin. (C) Comparison of
18 and 3 prenylation selectivity in J774 cells. The cells were metabolically labeled for 18 h with [14C]mevalonate, in the presence or absence of
400 μM 3 or 10 μM 18, then radiolabeled; prenylated proteins were detected by phosphorimaging.
(on which it elutes first), whereas the (þ)-3-E1-enriched
fraction was rechromatographed on the original QN col-
umn. The %ee of each enantiomer was estimated to be
g98% after two enrichment passes.
a J774 cell viability assay (IC50 47 μM compared to
2800 μM, respectively), although it is still less than that of
its parent BP 4 (IC50 3 μM; Figure 2A). Rab11 prenylation in
J774 cells was ∼30ꢀ more sensitive to 3 than to 1, the LED of
3 (3 μM) being similar to that of minodronate itself, com-
pared to ∼100 μM for 1 (Figure 2B and C). The similarity
between the potency of 3 and 4 for inhibition of Rab
prenylation highlights the importance of inhibition of pre-
nylation of non-Rab small GTPases in the mechanism of
action of 4, because this compound, which prevents the
modification of all prenylated proteins, reduced J774
viability much more effectively than 3. Moreover, the effect
of 4 on protein prenylation is most likely exclusively the
result of inhibition of FPPS because it has little effect on
RGGT (IC50 > 1 mM; data not shown).
Similarly to 1, 3 had almost no effect on FPP synthase
activity at up to 10 μM (Figure 3A; note that the IC50 for the
parent 4 is 3 nM in this assay28) and, unlike 2, had no effect
on the farnesylation of Ras at up to 400 μM (Figure 3B).
Moreover, 3 prevented the incorporation of [14C]mevalonate
into Rabs but, unlike the specific FTase inhibitor FTI-277
(18),29 did not affect farnesylation of nuclear lamins
(Figure 3C). Therefore, like 1, 3 is a selective RGGT
inhibitor both in vitro and in whole cells but is significantly
more potent. Interestingly, the difference in potency of the
parent BPs (4 vs. 2) is only about 4-fold for both inhibition
of FPPS and inhibition of prenylation (data not shown).
Biological Activity of 1 and 3 Stereoisomers
Stereoisomers of 1 Show Small Differences in Activity. The
enantiomers of 1 show a relatively small difference in their
ability to inhibit prenylation of Rab 11 in J774 cells, (þ)-1-E1
being about four times as potent as (-)-1-E2 (LED ∼63 vs.
250 μM, respectively) (Figure 1A). At 1 mM, both stereo-
isomers prevented the incorporation of [14C] mevalonate
into Rab GTPases, but not the Rho and Ras GTPases,
confirming the specificity of these compounds for RGGT
(Figure 1B). Neither (þ)-1-E1 nor (-)-1-E2 affected the
prenylation of Rap1A at concentrations up to 2 mM (data
not shown). Consistent with these results, (þ)-1-E1 was
moderately more potent than isomer (-)-1-E2 in inhibiting
incorporation of tritiated geranylgeranyl into Rab1a by iso-
lated RGGT (IC50 values 39 ( 2.3 μM and 150 ( 54 μM,
respectively; the racemate was of intermediate potency (IC50
51.7 ( 11.5 μM) (Figure 1C).
Compound 3 Is a Potent Inhibitor of Rab Prenylation. (þ)-3
was previously shown to be ∼25ꢀ more potent than 1
in inhibiting geranylgeranylation of Rab1a by isolated
RGGT.17 The greater potency of 3 was confirmed in

3458 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
McKenna et al.
Figure 4. Isomers of 3 have differential activity. (A) The effect of 3 and its enantiomers on the activity of recombinant Rab GGTase in vitro
was assessed by measuring the incorporation of [3H]geranylgeranyl (from [3H]GGPP) into Rap1A (mean ( SEM; n = 2). (B) Effects of 3
enantiomers on J774 cell viability. The cells were treated with the compounds for 48 h and then viable cell number assayed by Alamar blue assay
(mean ( SEM; n = 3). (C,D) Effects of 3 enantiomers compared to 1, 13, 16, and 17 on prenylation in J774 cells. Cells were treated for 24 h with
the indicated concentrations (μM) of the compounds, then lysed in triton X-114 buffer and fractionated into aqueous phases (containing
unprenylated Rab proteins) and Western blotted for unprenylated Rab11, β-actin, or unprenylated Rap1A.
Our finding that the desoxy analogue of 1 has reduced
mineral affinity, but is of similar potency to 1,8 suggested
that a similar modification to 3 could produce an equally
potent Rab prenylation inhibitor with even lower affinity for
bone. However, we have found that the desoxy analogue of 3
is less potent than 3 (data not shown).9

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3459
Chart 2. Structures of Compounds 13 and 16-19
At a high concentration (400 μM), 3 appeared to have an
additive effect with 18 in preventing the prenylation of
∼21 kDa proteins (i.e., those modified by FTase and GGTase
I), further indicating that 3 inhibits prenylation of all proteins
modified with geranylgeranyl groups. In accordance, concen-
trations of g100 μM 3 also inhibited prenylation of proteins
modified by GGTase I, such as Rap1 (Figures 3B and 4D). This
suggests that at concentrations ∼30 times higher than those
necessary for inhibition of RGGT, 3 also weakly inhibits
another enzyme of the mevalonate pathway downstream of
FPPS, i.e., either GGPP synthase or GGTase I (discussed
further in the section below concerning stereoisomers of 3).
Stereoisomers of 3 Differ Markedly in Their Potencies. By
contrast with 1, stereoisomers of 3 show dramatically differ-
ent activities, with (þ)-3-E1 more potent than (-)-3-E2 in
inhibiting RGGT (IC50 values 1.1 μM compared to 67.7 μM,
respectively; Figure 4A). The racemate was of intermediate
potency, although its IC50 appeared somewhat higher than
might be expected from the (þ)-3-E1 value. Large differences
in potency were found with respect to reduction of J774
cell viability (IC50 values 31 μM for isomer (þ)-3-E1 vs. >
800 μM for isomer (-)-3-E2; Figure 4B) and to a lesser
extent, inhibition of Rab11 prenylation (LED ∼3 and
∼50 μM, respectively; Figure 4C). Similar to the racemate,
(þ)-3-E1 inhibited prenylation of Rap1A at concentrations
around 30 times higher than those required for inhibition of
Rab11 prenylation (∼100 μM compared to ∼3 μM). (-)-3-
E2 also inhibited prenylation of Rap1A but only at a
concentration ∼80 times higher than that required for
inhibition of Rab prenylation (Figure 4D). Other PC analo-
gues, both active8 (1, 13, 16) and inactive3 (17) against
RGGT, did not inhibit Rap1A prenylation even at this high
concentration (4 mM) (Figure 4D). This indicates that the
isomers of 3 are both able to inhibit an additional enzyme
target besides RGGT (most likely GGPPS), but their poten-
cies for inhibition of this enzyme are not directly related to
their potencies for inhibition of RGGT (Chart 2).
Figure 5. Compound 3 is a weak inhibitor of GGPPS in intact cells.
(A) The effect of the enantiomers of 3 on activity of GGPPS in
isolated enzyme assays (mean ( SEM; n = 3). (B) The effect of the
enantiomers of 3 on activity of GGTase I in isolated enzyme assays
(mean ( SEM; n = 3). (C) Selectivity of 3 enantiomers for inhibition
of prenylation of GTPases in J774 cells. The cells were metabolically
labeled for 18 h with [3H]GGOH in the absence or presence of
400 μM (þ)-3-E1 or (-)-3-E2 and then lysates separated by
electrophoresis on 12% polyacrylamide-SDS gels and [3H]-labeled,
geranylgeranylated proteins detected by fluorography. (D) Effect of
2, 3, and 19 with/without added GGPP on Rap1A prenylation in
J774 cells. Cells were treated for 24 h with 25 μM 2, 400 μM 3 or
10 μM 19 in the presence or absence of 10 μM GGPP. Proteins in
cell lysates were then separated by electrophoresis on 12% SDS/
PAGE gels and unprenylated Rap1A (uRap1A) or β-actin detected
by Western blotting.
indeed (-)-3-E2) had little effect on the activity of either
GGPPS or GGTase I at concentrations up to 1 mM (Figure 5A
and B). Despite these results, two pieces of evidence indicate
that (þ)-3-E1 achieves its effects on Rap1A prenylation
through weak inhibition of GGPPS. First, (þ)-3-E1 did
not inhibit the incorporation of [3H]GGPP into 21 kDa
proteins (which are exclusive substrates for GGTase I;
Figure 5C), indicating that this compound has no effect on
GGTase I. Second, similar to 2, the inhibition of prenylation
of Rap1A caused by (þ)-3-E1 could be overcome by the
addition of GGPP, whereas the inhibition of Rap1A
prenylation caused by the specific GGTase I inhibitor
GGTI-298 (19)30 was unaffected (Figure 5D). This indicates
that (þ)-3-E1 inhibits an enzyme upstream of GGTase I (i.e.,
GGPPS). It remains unclear why inhibition of GGPPS was
so weak in isolated enzyme assays, but these differences
could reflect differences in the relative concentrations of
The data demonstrate that (þ)-3-E1 is the most potent
selective PC inhibitor of RGGT yet to be identified. Other
selective inhibitors of RGGT are similar in potency toward
this enzyme in vitro (IC50 values 2.1-2.8 μM) but less
effective in whole cell prenylation assays (inhibition of Rab
prenylation at 20-50 μM).12 Although the compound BMS-
1 is a more potent inhibitor than (þ)-3-E1 in enzyme assays,
this compound is also a potent inhibitor of FTase.11
Compound 3 Likely Inhibits GGPP Synthase in Cells.
Surprisingly, in isolated enzyme assays, (þ)-3-E1 (and

3460 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
McKenna et al.
Table 1. Effects of Selected PC and BP Analogues on Enzymes of the Mevalonate Pathway (Isolated Enzyme Assays), J774 Cell Viability, and Rab11
and Rap1A Prenylationa
FPPS
inhibition
reduction
of J774 viability
(IC50, µM)
RGGT
inhibition
(IC50, µM)
inhibition of Rab11
prenylation
inhibition of
Rap 1A prenylation
LED (µM)
inhibition of
GGPPS
inhibition
of GGTase 1
(IC50, µM)
compd
(IC50, µM)
LED (µM)
(IC50, µM)
(()-1
1-E1
1-E2
(()-3
3-E1
3-E2
2
>10003
ND
ND
2800
ND
ND
47
51.7
39
100
63
250
3
no effect at 4 mM
no effect at 2 mM
ND
ND
>2000b
ND
150
no effect at 2 mM
g100
100
ND
ND
ND
ND
>10
ND
12.5c
1.1c
31
3
>1000
>1000
ND
>800
>800
ND
ND
0.0068
0.00328
>800
318
3
67.7c
ND
50
12b
3 b
4000
12b8
3 b
4
>1000 b
0.67
ND
a LED = lowest effective dose. ND = not determined. b Not shown in Figures. c Under the assay conditions of ref 17, IC50 values (Rab1a) of 1.3 and
32 µM were obtained for 3-E1 and 1, respectively. The Ki values (uncompetitive) obtained were 0.211 vs. 33.56 µM. The value for racemic 3 found here might be
expected to be closer to that for 3-E1. However, conservatively, the enantiomer IC50 ratio is at least ∼20:1.
Figure 7. Mevalonate pathway, showing the prenylated proteins
analyzed in this study and sites of action of inhibitors, including 1
(3-PEHPC) and 3 (3-IPEHPC). GGPPS is suggested as an addi-
tional site of action for 3.
Conclusions
The novel PC analogue of minodronic acid, 3, can be
Figure 6. Compound 3 inhibits bone resorption by rabbit osteo-
synthesized in six steps from imidazo[1,2-a]pyridine. As pre-
clasts in vitro. (A) Rabbit osteoclasts were treated for 24 h, then
viously observed for a series of phosphonocarboxylate analo-
lysed in triton X-114 buffer and aqueous phases (containing un-
prenylated Rab proteins) were Western blotted for Rab11 or
gues of risedronate,3,5,8 the structural modification of
replacing one phosphonic acid group by a carboxylic acid
group in the drug “backbone” of the parent N-BP drug
redirects the active target from FPPS to RGGT, a more
downstream enzyme of the mevalonate pathway, causing a
selective inhibition of prenylation of Rab proteins (Figure 7).
Such compounds have potential therapeutic value, for
example as antitumor agents, because Rab GTPases have
been implicated in the pathogenesis of certain cancers that
metastasize to bone.6 In support of this, 1 has been shown
to possess antitumor effects both in vitro4,13 and in vivo.14,32
The lower affinity of PCs for bone mineral8 indicates that they
are likely to be targeted to the bone microenvironment5 but
not as tightly associated with the bone surface as higher
affinity BPs.
We have now demonstrated that 3 is a potent RGGT
inhibitor compared to previously studied PCs and exhibits
high selectivity for this enzyme. By separating chiral PC
inhibitors into their component enantiomers for the first time,
we have demonstrated that inhibition of RGGT can be highly
dependent on analogue stereochemistry at the R-carbon and
that almost all the activity of racemic 3 stems from one
enantiomer [(þ)-3-E1], which is the most potent PC-type
selective RGGT inhibitor discovered to date. This compound
illustrates the potential of bisphosphonate PC analogues as a
new platform for drug design,6 in addition to providing a new
tool for investigating Rab-dependent biological processes.3,5,33
β-actin. (B) Rabbit osteoclasts seeded onto dentine discs were
treated with (þ)-3-E1 for 48 h, then fixed, stained with TRITC-
phalloidin, and the number of actin rings counted. Resorption pit
area per disk was then determined by reflected light microscopy.
Results are expressed as percentage of control and are the mean (
SEM of six independent experiments.
enzyme and substrates in the cells compared to the enzyme
assays.
A summary of the biological activities of compounds 1-4,
1-E1-E2, and 3-E1-E2 is presented in Table 1.
Compound 3 Inhibits Rab Prenylation in Osteoclasts but is a
Weak Inhibitor of Bone Resorption. Compound 3 was much
more potent (∼15ꢀ) than 1 in inhibiting Rab prenylation in
rabbit osteoclasts (effective at 12.5 μM compared to 200 μM
for 1 (Figure 6A)). 3 was also an inhibitor of bone resorption
(Figure 6B), but was only around 3-fold more potent than 1
in vitro (IC50 ∼330 μM compared to 970 μM for 1), and in
vivo in the Schenk growing rat model (LED = 2.8 mg/kg
compared to 12 mg/kg for 1),31 which may be somewhat less
than anticipated from its potency for inhibiting Rab prenyla-
tion in osteoclasts. The in vivo data are for the racemates,
thus the LED of the E-1 isomer of 3 should be lower, e.g.
∼ 1.4 mg/kg. Unlike 1, 3 affected the osteoclast cytoskeleton
at concentrations that also inhibited resorption. The reason
for the discrepancy between the effects of these two PCs on
osteoclasts remains unclear.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3461
Itwill be of interest tococrystallize RGGT12 withthe inhibitor
3 in complex, which we predict to show preferential binding
of (þ)-3-E1.
9.53 (m, CH, 1H), 9.98 (s, CHO, 1H) (for a discussion of the 1H
NMR of imidazo[1,2-a]pyridine analogues, see Paudler and
Blewitt34).
Ethyl Azidoacetate (7).21 To a cooled (0-5 ꢀC) solution of
ethyl bromoacetate (40 g, 0.24 mol) in 250 mL of acetone, a
solution of sodium azide (38.9 g, 0.60 mol) in 200 mL of water
was added dropwise. The temperature was increased to 63 ꢀC
and maintained at that temperature for 4 h. The reaction
mixture was then cooled to room temperature and extracted
with CH2Cl2 (7 ꢀ 70 mL). The combined extracts were washed
with saturated NaHCO3 (3 ꢀ 40 mL) and water (3 ꢀ 20 mL) and
dried over MgSO4. Solvents were evaporated under reduced
pressure, and the residue was dried under vacuum. The product
was obtained as a colorless oil in 96% yield (29.9 g). 1H NMR,
CDCl3: 1.36 (t, 3JHH = 6.80 Hz, CH3, 3H), 3.92 (s, CH2N3, 2H),
4.31 (q, 3JHH = 6.80 Hz, CH2, 2H).
Experimental Section
General Methods, Reagents and Materials. Compounds 1 and
17 were gifts from Procter & Gamble Pharmaceuticals, Inc.
Compounds 12, 13, and 16 were synthesized by our published
method.8 Synthesis of the partial esters of 13 (used for HPLC
studies), 14, and 15 is discussed in the Supporting Informa-
tion. All solvents and reagents were reagent grade, purchased
commercially and used without further purification, except as
mentioned below. Phosphorus oxychloride and triethylamine
were freshly distilled before use. HPLC were carried out
using a Dynamax Rainin model SD-200 pump equipped with
a Shimadzu SPD-10A VP UV-vis detector on ProntoSIL
AX QN 8 mm ꢀ 150 mm and ProntoSIL AX QD 4 mm ꢀ
150 mm chiral columns (Bischoff Chromatography Leonberg,
Germany). Columns were eluted isocratically with 0.7 M
TEAAc containing 75% MeOH at pH 5.8 for enantioseparation
of 3, or with 0.25 M TEAAc, 75% MeOH at pH 6.9 for
enantioseparation of 1. Optical rotations for both enantiomers
of 3 dissolved in water at pH 6.3 were determined on a Jasco
P-2000 polarimeter, cell path length 10 cm. NMR spectra were
measured on a Varian Mercury 400 spectrometer. Chemical
shifts (δ) are reported in parts per million (ppm) relative to
internal residual CHCl3 in CDCl3 (δ 7.29), or internal residual
HDO in D2O (pH ∼12, δ 4.8). Elemental analysis of racemic 3
was performed by Gallbraith Laboratories Inc., Knoxville,
TN. HRMS analysis was performed at the UC Riverside, MS
Laboratory, using a VG-ZAB MS spectrometer in FAB
mode. UV spectra were measured on a Beckmann Coulter
DU 800 spectrophotometer. Concentrations of 3 enantiomers
in aqueous solution (pH 6.3) were assigned by UV, using ε=
6450 at 280 nm.
For biological assays, PC compounds were dissolved in PBS,
the pH was adjusted to 7.4 with 1 M NaOH, and the solution
filter-sterilized before use. Antibodies used for immunoblotting
were: antiunprenylated Rap1A (sc-1482) from Santa Cruz
Biotechnology, anti-Rab11 (05-853) from Upstate Serologicals
(Lake, Placid, NY), and β-actin from Sigma (Poole, UK). 23c6
antivitronectin receptor (VNR) antibody was a generous gift
of Prof. Mike Horton (University College London, UK). [3H]-
GGOH was obtained from American Radiochemicals Ltd. (St
Louis, MO). [14C]Mevalonic acid lactone, Enhance reagent, and
[1-3H]trans-geranylgeranyl pyrophosphate (GGPP) were pur-
chased from NEN (Hounslow, UK). Solvent was removed from
[3H]GGOH and [14C]mevalonic acid lactone by evaporating in
nitrogen. All other reagents were from Sigma Chemical Co
(Poole, UK) unless stated otherwise.
(Z )-2-Azido-3-imidazo[1,2-a]pyridin-3-yl-acrylic Acid Ethyl
Ester (8).22 To ethanolic sodium ethoxide, generated by the
addition of metallic sodium (4.5 equiv) to ethanol (140 mL) and
cooled to -30 ꢀC, a solution of aldehyde 6 (4 g, 0.0273 mol) and
ethyl azidoacetate 7 (14.1 g, 0.109 mol) in ethanol (120 mL) was
added dropwise. The temperature was slowly increased to room
temperature, and the mixture allowed to stand for 4 h. The
reaction was quenched with a saturated aqueous solution of
NH4Cl (360 mL) and extracted with diethyl ether (1200 mL).
The combined organic extracts were dried over Na2SO4 and
concentrated in vacuo. The crude product was chromato-
graphed on silica gel, eluted with EtOAc:hexane (3:2), giving a
yellowish solid (3.87 g, 55%). Stereochemical assignment fol-
3
lows Chezal et al.22 1H NMR CDCl3: 1.45 (t, JHH = 7.2 Hz,
CH3, 3H), 4.44 (q, 3JHH = 7.2 Hz, CH2, 2H), 7.00 (dt, 3JHH
7.0, 4JHH = 0.8, CH, 1H), 7.12 (s, CHd, 1H), 7.35 (ddd, 3JHH
=
=
8.8, 7.0, 4JHH = 1.2, CH, 1H), 7.73 (bd, 3JHH = 8.8, CH, 1H),
8.24 (d, 3JHH = 7.0, CH, 1H), 8.54 (s, CH, 1H).
(Z )-2-Amino-3-imidazo[1,2-a]pyridin-3-yl-acrylic Acid Ethyl
Ester (9). A mixture of azide 8 (4.63 g, 0.018 mol) and 10%
palladium on charcoal (1.05 g) in methanol (130 mL) was stirred
under H2 for 2.5 h at rt. The reaction progress was controlled by
TLC (CH2Cl2/acetone 9:1). The reaction mixture was filtered
and the filtrate concentrated under reduced pressure. The
product was obtained quantitatively (4.15 g). 1H NMR, CDCl3:
1.42 (t, 3JHH = 6.80 Hz, CH3, 3H), 4.37 (q, 3JHH = 6.80 Hz, CH2,
3
2
2H), 6.61 (s, CHdC, 1H), 6.92 (dt, JHH = 6.80 Hz, JHH
1.2 Hz, CH, 1H), 7.26 (ddd, 3JHH = 6.80 Hz, 10.0 Hz, 2JHH
=
=
1.2 Hz, CH, 1H), 7.66 (m, CH, 1H), 7.87 (s, CH, 1H), 8.15
(m, CH, 1H).
(Z )-2-Hydroxy-3-imidazo[1,2-a]pyridin-3-yl-acrylic Acid Ethyl
Ester (10). A solution of enamine 9 (3.5 g, 0.015 mol) in acetic
acid (5 mL) and water (35 mL) was stirred for 1.5 h at 0 ꢀC. A
solid precipitated. The mixture was concentrated in vacuo, and
the semisolid residue was washed with water and dried in
vacuo, leaving the product as an off-white solid (2.08 g, 59%).
Imidazo[1,2-a]pyridin-3-carbaldehyde (6).19,20 To DMF (120
mL, 1.55 mol) cooled to 2 ꢀC, freshly distilled phosphorus
oxychloride (61 mL, 0.65 mol) was slowly added. The tempera-
ture was allowed to gradually rise to room temperature. After
recooling to 2 ꢀC and a solution of imidazo[1,2-a]-pyridine 5
(10 g, 0.085 mol) in DMF (60 mL) was added dropwise. The
mixture was warmed to 105 ꢀC, and the temperature rose to
140 ꢀC. The oil bath was removed until the temperature stabi-
lized at 120 ꢀC. Heating was then continuted for 45 min at 120 ꢀC
and 2.5 h at 85 ꢀC. After cooling, the reaction mixture was
poured into 5% HCl (600 mL) cooled by a water bath and then
brought to pH 9 by adding 20% aq NaOH. The resulting
solution was extracted with CH2Cl2 (1200 mL) overnight. The
organic layer was separated and dried over MgSO4, the solvent
removed under reduced pressure, and the crude product washed
with water (5 ꢀ 15 mL) and dried: 3.49 g (31%) of a white solid
1H NMR: 1.43 (t, 3JHH = 7.20 Hz, CH3, 3H), 4.42 (q, 3JHH
=
7.20 Hz, CH2, 2H), 6.78 (s, CHdC, 1H), 6.96 (dt, 3JHH = 6.80
=
Hz, 2JHH = 1.2 Hz, CH, H), 7.26 (m, CH, 1H), 7.72 (d, 3JHH
3
9.20 Hz, CH, 1H), 8.22 (d, JHH = 6.80 Hz, CH, 1H), 8.32
(s, CH, 1H).
2-(Diethoxyphosphoryl)-2-hydroxy-3-imidazo[1,2-a]pyridin-
3-yl-propionic Acid Ethyl Ester (11). A mixture of enol 10 (1.45 g,
0.006 mol) in diethyl phosphite (25 mL) was stirred for 21 h at
70 ꢀC. After cooling, it was concentrated in vacuo and subjected
to hydrolysis without further isolation or purification due to its
instability. 1H NMR, CDCl3: 1.40, 1.42, 1.43 (3t, 3JHH = 6.80
Hz, 3CH3, 9H), 3.62 (dd, 2JHH = 16.00, 3JHP = 6.80 Hz, CH,
2
3
1H), 3.84 (dd, JHH = 16.00, JHP = 4.40 Hz, CH, 1H),
3
4.18-4.34 (m, 3CH2O, 6H), 6.87 (t, JHH = 6.80 Hz, CH,
=
1H), 7.22-7.26 (m, CH, 1H), 7.48 (s, CH, 1H), 7.68 (d, 3JHH
1
3
was obtained. H NMR (CDCl3): 7.17 (dt, JHH = 6.80 Hz,
8.80 Hz, CH, 1H), 8.38 (d, 3JHH = 7.20 Hz, CH, 1H).
4JHH = 0.8 Hz, CH, 1H), 7.60 (ddd, JHH = 8.40, 6.80 Hz,
2-Hydroxy-3-imidazo[1,2-a]pyridin-3-yl-2-phosphonopropionic
Acid (3). A solution of the crude ester 11 in concentrated
3
4JHH = 1.6 Hz, CH, 1H), 7.83 (m, CH, 1H), 8.35 (s, CH, 1H),

3462 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
McKenna et al.
hydrochloric acid (45 mL) was refluxed for 5 h. After evapora-
tion under reduced pressure, the residue was dissolved partially
in 100 μL of water and precipitated by addition of ethanol (50
mL). The precipitate was filtered, washed with water (500 μL)
and then ethanol (40 mL), filtered, washed again with ethanol,
and dried, giving an off-white solid (1.03 g, 60% from 10). Two
overlapping peaks of approximately equal area on an AX QN
HPLC column (see below). 1H NMR (D2O, pH 12.8): 3.35 (dd,
2JHH = 15.6, 3JHP = 7.2, CHHCP, 1H), 3.75 (dd, 2JHH = 15.6,
3JHP = 1.2, CHHCP, 1H), 6.90 (t, 3JHH = 6.4, CH, 1H), 7.26
(bdd, 3JHH = 9.2, 6.4, CH, 1H), 7.35 (s, CH, 1H), 7.45 (d, 3JHH
= 9.2, CH, 1H), 8.41 (d, 3JHH = 6.4, CH, 1H). 31P NMR (D2O,
pH 12.8): 15.06. 13C NMR (D2O, pH 12.8): 29.30 (CH2CP),
FPPS and GGPPS Assays. FPP synthase and GGPP synthase
were purified and assayed as described.37 Briefly, compounds
under investigation were preincubated with 1 pmol FPPS or
6 pmol GGPPS for 10 min and then the reaction initiated by
adding 20 μL of substrate containing 50 μM [1-14C]IPP
(4 μCi/mmol) and either 50 μM GPP for FPPS or 50 μM FPP
for GGPPS. After 5 min, the assay was terminated by adding
200 μL of 1:4 conc HCl/methanol. Reaction products were then
extracted with 0.5 mL of scintillation fluid (Microscint E,
Perkin-Elmer) and radioactivity in the upper phase determined
by scintillation counting. Data was analyzed using Graphpad
Prism.
Assessment of Rab Prenylation by Western Blotting. The
effects of the compounds on protein prenylation were studied
using triton X-114 fractionation in which prenylated proteins
partition into the detergent-rich phase, whereas unprenylated
proteins remain in the aqueous phase.5 Briefly, cells were treated
for 24 h, then lysed in 20 mM Tris, 150 mM NaCl, pH 7.5, 1%
Triton X-114, a sample taken for determination of protein
concentration, and then remaining lysate incubated at 37 ꢀC
for 10 min. Following centrifugation at 13000g for 2 min, the
aqueous and detergent-rich phases were separated and then
Triton X-114 added back to the aqueous phase to 1% v/v and
the extraction process repeated. Aqueous phases, equivalent to
20 mg of unfractionated lysate, were electrophoresed on 12%
gels and Western blotted for Rab11 (an abundant, ubiquitous
Rab), β-actin, or unprenylated Rap1A. Effects are expressed as
the lowest effective dose (from 3 independent experiments), i.e.,
the lowest dose at which unprenylated Rap1A or Rab11 was
detected.
1
80.51 (d, JCP = 138.3, CP), 112.10, 115.40 (2s, 2CH), 122.93
(d, JCP = 16.8, C), 125.26 (s, 2CH), 130.07 (s, CH), 145.12
3
(s, C), 179.55 (s, CdO). Anal. Calcd for C10H11N2O6P(H2O)2-
(C2H5OH)0.16: C 37.61%, H 4.88%, N 8.50%. Found: C
37.44%, H 5.00%, N 8.63%. HRMS for C10H10N2O6P- (m/z):
found 285.0283; calcd: 285.028.
Separation of 3 Enantiomers by Chiral HPLC. 3 was eluted
isocratically on a ProntoSIL AX QN column (8 mm ꢀ 150 mm)
with 0.7 M triethylamine/acetic acid containing 75% MeOH at
pH 5.8. Each injection was done using 5-7 mg of 3, which was
dissolved in 100 μL buffer and TEA at pH 6.7. The two fractions
corresponding to partially overlapping peaks were separated
and evaporated under reduced pressure. During the evaporation
process in glass vessels, but not in plastic (PP) vessels, an adduct
sometimes forms in variable amounts (2-10%), detectable by
chiral HPLC and 31P NMR analysis, that reverts (without
racemization) to 3 (E1 or E2) on treatment with water at pH
7-8 (Supporting Information). The origin and structure of this
compound will be discussed elsewhere.35 On the basis of reana-
lysis using QN and QD columns, the %ee of the enantiomer
fraction with the shorter retention time, (þ)-3-E1, was about
95%. The enantiomer with the longer retention time, (-)-3-E2
was obtained in about 70-80% ee. The fraction enriched with
(þ)-3-E1 was further purified using the same ProntoSIL AX QN
column, whereas the fraction enriched in (-)-3-E2 was repur-
ified using a ProntoSIL 4 mm ꢀ 150 mm AX QD column, which
reversed the elution order, eliminating tailing of (þ)-3-E1 into
(-)-3-E2. The enantiomeric purity of both enantiomers was
g98%, defined by chiral HPLC using the ProntoSIL AX QN
and AX QD columns to define the %ee of (-)-3-E2 and (þ)-3-
E1, respectively. The optical rotations of the two enantiomers
were opposite, as expected:
In some cases in which prenylation of Rap1A was assessed,
cells were lysed simply in RIPA buffer38 because the antibody
recognizes only the unprenylated form of Rap1A and therefore
fractionation is not required.
Cell Viability Assay. The number of viable J774 cells follow-
ing treatment with the compounds for 48 h was assessed as
previously described.3 Experiments were carried out in repli-
cates of 6 and the data expressed are the mean ( SEM of at least
four independent experiments.
Incorporation of [14C]Mevalonate and [3H]GGOH into Pre-
nylated Proteins in Intact Cells. Detection of prenylated proteins
in J774 macrophages was carried out as described previously.3
Briefly, cells were depleted of mevalonate by incubation with
5 μM mevastatin for 4 h and then transferred into fresh
medium containing 5 μM mevastatin and either 7.5 μCi/mL
[14C]mevalonic acid lactone or 30 mCi/mL [3H]GGOH plus
compounds under investigation. After 18 h, the cells were lysed
in RIPA buffer and then 50 μg of cell lysate from each treatment
were electrophoresed on 12% polyacrylamide-SDS gels under
reducing conditions. After electrophoresis, the gels were fixed in
10% (v/v) acetic acid, 40% (v/v) methanol, 50% (v/v) distilled
water, and then [14C]-labeled gels were dried and labeled pro-
teins visualized on a BioRad personal FX imager after exposure
to a Kodak phosphorimaging screen. [3H]-Labeled gels were
incubated in Enhance for 30 min prior to drying. [3H]-Labeled
proteins were then visualized by exposing the gel to pre-
flashed Hyperfilm-MP (Amersham, Aylesbury, UK) for 6 days
at -70 ꢀC.
23
ð þ Þ-3-E1 : ½RꢁD ¼ þ 112:4 ( 0:2 ðc ¼ 0:68g=dL; H2O, pH 6:3Þ
23
ð-Þ-3-E2 : ½RꢁD ¼ -110:9 ( 0:2 ðc ¼ 0:64g=dL; H2O, pH 6:3Þ
Determination of the Extinction Coefficient of 3. First, UV
spectra of racemic 3 in aqueous solutions at pH 9, 7, 5, and 3
were acquired, which displayed an isosbestic point at 296 nm
between pH 9 and pH 7. The extinction coefficient determined
for pure racemic 3 (a sample which passed elemental analysis
1
and was pure by 31P, H, 13C NMR; MW = 286.2) in 0.1 N
phosphate buffer, at pH 6.3 was 6450 ( 3 at 280 nm (λmax
)
(n = 3).
RGGT and GGTase I Assays. RGGT activity was measured
by determining the incorporation of [3H]GGPP into His-tagged
canine Rab1a protein as previously described.36 The final con-
centrations in the prenylation reaction were 50 mM sodium
HEPES, pH 7.2, 5 mM MgCl2, 1 mM DTE, 1 mM NP-40, 4 μM
Rab1a, 5 μM GGPP (800 dpm/pmol), 2 μM REP-1, and 50 nM
RGGT. All reactions were allowed for 30 min at 37 ꢀC in a 25 μL
volume in glass tubes (Fisher). The assay was conducted in
duplicate and repeated in three independent reactions. The
experimental data were fitted using the enzyme kinetic module
from Systat Software and are expressed as mean ( SEM of three
independent experiments.
Rabbit Osteoclast Cultures. Mature osteoclasts were isolated
from 2-day old New Zealand White rabbits as previously
described39 and partially purified through an FCS gradient.
Cells were then resuspended in fresh a-MEM (4 mL/rabbit) and
100 mL seeded on to dentine discs in 96-well plates. Discs were
rinsed in PBS 90 min after seeding the cells and then cultured in
fresh a-MEM for 48 h. Cells were then fixed in 4% formalde-
hyde, and the number of osteoclasts, number of F-actin rings,
and resorption pit area determined as previously described.39 In
experiments for assessment of effects on prenylation, cultures of
pure rabbit osteoclasts in 12-well plates were also generated by
incubating marrow cells isolated from long bones of neonatal

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3463
rabbits with 1,25 (OH)2D3, as previously described,40 and
Western blotted as described above.
(7) (a) Pereira-Leal, J. B.; Seabra, M. C. The mammalian Rab family
of small GTPases: definition of family and subfamily sequence
motifs suggests a mechanism for functional specificity in the Ras
superfamily. J. Mol. Biol. 2000, 301, 1077–1087. (b) Leung, K. F.;
Baron, R.; Seabra, M. C. Geranylgeranylation of Rab GTPases. J. Lipid
Res. 2006, 47, 467–475.
(8) Marma, M. S.; Xia, Z. D.; Stewart, C.; Coxon, F.; Dunford, J. E.;
Baron, R.; Kashemirov, B. A.; Ebetino, F. H.; Triffitt, J. T.;
Russell, R. G. G.; McKenna, C. E. Synthesis and biological
evaluation of alpha-halogenated bisphosphonate and phosphono-
carboxylate analogues of risedronate. J. Med. Chem. 2007, 50,
5967–5975.
(9) Blazewska, K. M.; Kashemirov, B. A.; Stewart, C. A.; Coxon,
F. P.; Baron, R.; Rogers, M. J.; Seabra, M. C.; Ebetino, F. H.;
McKenna, C. E. 42nd Western Regional Meeting of the American
Chemical Society, Las Vegas, NV, Sept 23-27, 2008; Abstract
WRM-070.
(10) (a) Cheng, K. W.; Lahad, J. P.; Kuo, W. L.; Lapuk, A.; Yamada,
K.; Auersperg, N.; Liu, J. S.; Smith-McCune, K.; Lu, K. H.;
Fishman, D.; Gray, J. W.; Mills, G. B. The RAB25 small GTPase
determines aggressiveness of ovarian and breast cancers. Nat. Med.
2004, 10, 1251–1256. (b) Gebhardt, C.; Breitenbach, U.; Richter, K. H.;
Furstenberger, G.; Mauch, C.; Angel, P.; Hess, J. C-fos-dependent
induction of the small ras-related GTPase Rab11a in skin carcinogen-
esis. Am. J. Pathol. 2005, 167, 243–253.
(11) Lackner, M. R.; Kindt, R. M.; Carroll, P. M.; Brown, K.; Cancilla,
M. R.; Chen, C. Y.; de Silva, H.; Franke, Y.; Guan, B.; Heuer, T.;
Hung, T.; Keegan, K.; Lee, J. M.; Manne, V.; O’Brien, C.; Parry,
D.; Perez-Villar, J. J.; Reddy, R. K.; Xiao, H. J.; Zhan, H. J.;
Cockett, M.; Plowman, G.; Fitzgerald, K.; Costa, M.;
Ross-Macdonald, P. Chemical genetics identifies Rab geranylger-
anyl transferase as an apoptotic target of farnesyl transferase
inhibitors. Cancer Cell 2005, 7, 325–336.
(12) (a) Watanabe, M.; Fiji, H. D. G.; Guo, L.; Chan, L.; Kinderman,
S. S.; Slamon, D. J.; Kwon, O.; Tamanoi, F. Inhibitors of protein
geranylgeranyltransferase I and Rab geranylgeranyltransferase
identified from a library of allenoate-derived compounds. J. Biol.
Chem. 2008, 283, 9571–9579. (b) Guo, Z.; Wu, Y. W.; Tan, K. T.; Bon,
R. S.; Guiu-Rozas, E.; Delon, C.; Nguyen, U. T.; Wetzel, S.; Arndt, S.;
Goody, R. S.; Blankenfeldt, W.; Alexandrov, K.; Waldmann, H.
Development of selective RabGGTase inhibitors and crystal structure
of a RabGGTase-inhibitor complex. Angew. Chem., Int. Ed. 2008, 47,
3747–3750.
(13) Boissier, S.; Ferreras, M.; Peyruchaud, O.; Magnetto, S.; Ebetino,
F. H.; Colombel, M.; Delmas, P.; Delaisse, J.-M.; Clezardin, P.
Bisphosphonates inhibit breast and prostate carcinoma cell inva-
sion, an early event in the formation of bone metastases. Cancer
Res. 2000, 60, 2949–2954.
(14) Fournier, P. G. J.; Dauhine, F.; Lundy, M. W.; Rogers, M. J.;
Ebetino, F. H.; Clezardin, P. Lowering Bone Mineral Affinity of
Bisphosphonates as a Therapeutic Strategy to Optimize Skeletal
Tumor Growth Inhibition In Vivo. Cancer Res. 2008, 68, 8945–
8953.
(15) Takeuchi, M.; Sakamoto, S.; Kawamuki, K.; Kurihara, H.;
Nakahara, H.; Isomura, Y. Studies on novel bone resorption
inhibitors. II. Synthesis and pharmacological activities of fused
aza-heteroarylbisphosphonate derivatives. Chem. Pharm. Bull.
1998, 46, 1703–1709.
(16) (a) Kashemirov, B. A.; Marma, M. S.; Mallard, I.; Ebetino, F. H.;
Coxon, F. P.; Rojas, J.; Rogers, M. J.; McKenna, C. E. 16th
International Conference on Phosphorus Chemistry, Birmingham,
UK, July 4-9, 2004; Abstract PS2-041. (b) Bala, J.; Mallard, I.;
Kashemirov, B. A.; Coxon, F. P.; Rogers, M J.; Ebetino, F. H.;
McKenna, C. E. 40th Western Regional Meeting of the American
Chemical Society, Anaheim, CA USA, Jan. 22-25 2006; Abstract
WRM-129.
(17) Baron, R. A.; Tavare, R.; Figueiredo, A. C.; Blazewska, K. M.;
Kashemirov, B. A.; McKenna, C. E.; Ebetino, F. H.; Taylor, A.;
Rogers, M. J.; Coxon, F. P.; Seabra, M. C. Phosphonocarboxylates
Inhibit the Second Geranylgeranyl Addition by Rab Geranylger-
anyl Transferase. J. Biol. Chem. 2009, 284, 6861–6868.
(18) Ebetino, F. H.; Bayless, A. V.; Dansereau, S. M. Phosphonocar-
boxylate compounds pharmaceutical compositions, and methods
for treating abnormal calcium and phosphate metabolism. U.S.
Patent 5,760,021, 1998.
Acknowledgment. This research was supported by Procter
& Gamble Pharmaceuticals, Inc. C. E. McKenna was a
consultant for Procter & Gamble Pharmaceuticals, Inc. to
12/2008, during which period some of this work was done.
_
K. M. Bzazewska was a 2007-9 WiSE Fellow.
Supporting Information Available: NMR spectra of 1, 3, and
6-15; HPLC traces for the 1 and 3 enantiomer separations;
pH dependence of 3 UV spectrum; synthetic and chiral HPLC
separation details for 13-15; stability of 13 to racemization;
RGGT inhibition curves for isolated 3-E1 and E2 adducts.
This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) (a) Ling, Y.; Sahota, G.; Odeh, S.; Chan, J. M. W.; Araujo, F. G.;
Moreno, S. N. J.; Oldfield, E. Bisphosphonate inhibitors of Tox-
oplasma gondi growth: In vitro, QSAR, and in vivo investigations.
J. Med. Chem. 2005, 48, 3130–3140. (b) Papapoulos, S. E. Bispho-
sphonate actions: physical chemistry revisited. Bone 2006, 38, 613–
616. (c) Ling, Y.; Li, Z. H.; Miranda, K.; Oldfield, E.; Moreno, S. N. J.
The farnesyl-diphosphate/geranylgeranyl-diphosphate synthase of
Toxoplasma gondii is a bifunctional enzyme and a molecular target
of bisphosphonates. J. Biol. Chem. 2007, 282, 30804–30816. (d)
Kotsikorou, E.; Song, Y. C.; Chan, J. M. W.; Faelens, S.; Tovian, Z.;
Broderick, E.; Bakalara, N.; Docampo, R.; Oldfield, E. Bisphosphonate
inhibition of the exopolyphosphatase activity of the Trypanosoma
brucei soluble vacuolar pyrophosphatase. J. Med. Chem. 2005, 48,
6128–6139. (e) Zhang, Y.; Leon, A.; Song, Y.; Studer, D.; Haase, C.;
Koscielski, L. A.; Oldfield, E. Activity of nitrogen-containing and non-
nitrogen-containing bisphosphonates on tumor cell lines. J. Med.
Chem. 2006, 49, 5804–5814. (f ) Roelofs, A. J.; Thompson, K.; Gordon,
S.; Rogers, M. J. Molecular mechanisms of action of bisphosphonates:
current status. Clin. Cancer Res. 2006, 12, 6222S–6230S. (g) Russell,
R. G. G.; Watts, N. B.; Ebetino, F. H.; Rogers, M. J. Mechanisms of
action of bisphosphonates: similarities and differences and their poten-
tial influence on clinical efficacy. Osteoporosis Int. 2008, 19, 733–759.
(h) Dunford, J. E.; Kwaasi, A. A.; Rogers, M. J.; Barnett, B. L.; Ebetino,
F. H.; Russell, R. G. G.; Oppermann, U.; Kavanagh, K. L. Structure-
activity relationships among the nitrogen containing bisphosphonates in
clinical use and other analogues: time-dependent inhibition of human
farnesyl pyrophosphate synthase. J. Med. Chem. 2008, 51, 2187–2195.
(i) Nancollas, G. H.; Tang, R.; Phipps, R. J.; Henneman, Z.; Gulde, S.;
Wu, W.; Mangood, A.; Russell, R. G. G.; Ebetino, F. H. Novel insights
into actions of bisphosphonates on bone: differences in interactions with
hydroxyapatite. Bone 2006, 38, 617–627. (j) Peng, Z.-Y.; Mansour,
J. M.; Araujo, F.; Ju, J.-Y.; McKenna, C. E.; Mansour, T. E. Some
phosphonic acid analogs as inhibitors of pyrophosphate-dependent
phosphofructokinase, a novel target in Toxoplasma gondii. Biochem.
Pharm. 1995, 49, 105–113.
(2) Russell, R. G. G.; Xia, Z.; Dunford, J. E.; Oppermann, U.; Kwaasi,
A.; Hulley, P. A.; Kavanagh, K. L.; Triffitt, J. T.; Lundy, M. W.;
Phipps, R. J.; Barnett, B. L.; Coxon, F. P.; Rogers, M. J.; Watts,
N. B.; Ebetino, F. H. Bisphosphonates-an update on mechanisms
of action and how these relate to clinical efficacy. Skeletal Biol.
Med., Part B 2007, 1117, 209–257.
(3) Coxon, F. P.; Helfrich, M. H.; Larijani, B.; Muzylak, M.; Dunford,
J. E.; Marshall, D.; McKinnon, A. D.; Nesbitt, S. A.; Horton,
M. A.; Seabra, M. C.; Ebetino, F. H.; Rogers, M. J. Identification
of a novel phosphonocarboxylate inhibitor of Rab geranylgeranyl
transferase that specifically prevents Rab prenylation in osteoclasts
and macrophages. J. Biol. Chem. 2001, 276, 48213–48222.
(4) Roelofs, A. J.; Hulley, P. A.; Meijer, A.; Ebetino, F. H.; Graham,
R.; Russell, G.; Shipman, C. M. Selective inhibition of Rab
prenylation by a phosphonocarboxylate analogue of risedronate
induces apoptosis, but not S-phase arrest, in human myeloma cells.
Int. J. Cancer 2006, 119, 1254–1261.
(5) Coxon, F. P.; Ebetino, F. H.; Mules, E. H.; Seabra, M. C.;
McKenna, C. E.; Rogers, M. J. Phosphonocarboxylate inhibitors
of Rab geranylgeranyl transferase disrupt the prenylation and
membrane localization of Rab proteins in osteoclasts in vitro and
in vivo. Bone 2005, 37, 349–358.
(19) Almirante, L; Mugnaini, A.; Detoma, N.; Gamba, A.; Murmann,
W.; Hidalgo, J. Imidazole Derivatives. 4. Synthesis and Pharma-
cologic Activity of Oxygenated Derivatives of Imidazo [1,2-a]-
Pyridine. J. Med. Chem. 1970, 13, 1048–1051.
(6) Cheng, K. W.; Lahad, J. P.; Gray, J. W.; Mills, G. B. Emerging role
of RAB GTPases in cancer and human disease. Cancer Res. 2005,
65, 2516–2519.
(20) Gueiffier, A.; Lhassani, M.; Elhakmaoui, A.; Snoeck, R.; Andrei,
G.; Chavignon, O.; Teulade, J. C.; Kerbal, A.; Essassi, E. M.;
Debouzy, J. C.; Witvrouw, M.; Blache, Y.; Balzarini, J.; DeClercq,

3464 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
McKenna et al.
E.; Chapat, J. P. Synthesis of acyclo-C-nucleosides in the imidazo-
[1,2-a]pyridine and pyrimidine series as antiviral agents. J. Med.
Chem. 1996, 39, 2856–2859.
C-8/C-9 stereochemistry on chiral recognition. Chirality 1999, 11,
522–528.
(28) Dunford, J. E.; Thompson, K.; Coxon, F. P.; Luckman, S. P.;
Hahn, F. M.; Poulter, C. D.; Ebetino, F. H.; Rogers, M. J.
Structure-activity relationships for inhibition of farnesyl dipho-
sphate synthase in vitro and inhibition of bone resorptionin vivo by
nitrogen-containing bisphosphonates. J. Pharmacol. Exp. Ther.
2001, 296, 235–242.
(29) Lerner, E. C.; Qian, Y. M.; Hamilton, A. D.; Sebit, S. M. Disrup-
tion of Oncogenic K-RAS4B Processing and Signaling by a Potent
Geranylgeranyltransferase-I Inhibitor. J. Biol. Chem. 1995, 270,
26770–26773.
(30) McGuire, T. F.; Qian, Y. M.; Vogt, A.; Hamilton, A. D.; Sebti,
S. M. Platelet-Derived Growth Factor Receptor Tyrosine Phos-
phorylation Requires Protein Geranylgeranylation But Not Far-
nesylation. J. Biol. Chem. 1996, 271, 27402–27407.
(31) Ebetino, F. H.; Bayless, A. V.; Amburgey, J.; Ibbotson, K. J.;
Dansereau, S.; Ebrahimpour, A. Elucidation of a pharmacophore
for the bisphosphonate mechanism of bone antiresorptive activity.
Phosphorus, Sulfur Silicon Relat. Elem. 1996, 110, 217–220.
(32) Lawson, M. A.; Coulton, L.; Ebetino, F. H.; Vanderkerken, K.;
Croucher, P. I. Geranylgeranyl transferase type II inhibition pre-
vents myeloma bone disease. Biochem. Biophys. Res. Commun.
2008, 377, 453–457.
(21) Hooper, N.; Beeching, L. J.; Dyke, J. M.; Morris, A.; Ogden, J. S.;
Dias, A. A.; Costa, M. L.; Barros, M. T.; Cabral, M. H.; Moutinho,
A. M. C. A study of the thermal decomposition of 2-azidoethanol
and 2-azidoethyl acetate by ultraviolet photoelectron spectroscopy
and matrix isolation infrared spectroscopy. J. Phys. Chem. A 2002,
106, 9968–9975.
(22) Chezal, J. M.; Delmas, G.; Mavel, S.; Elakmaoui, H.; Metin, J.;
Diez, A.; Blache, Y.; Gueiffier, A.; Rubiralta, M.; Teulade, J. C.;
Chavignon, O. Solid-supported heterocumulenes: preparation and
crystal structure of azaaplysinopsins. J. Org. Chem. 1997, 62, 4085–
4087.
(23) Kondo, K.; Morohoshi, S.; Mitsuhashi, M.; Murakami, Y.
Synthetic utility of tert-butyl azidoacetate on the Hemetsberger-
Knittel reaction (synthetic studies of indoles and related com-
pounds part 47). Chem. Pharm. Bull. 1999, 47, 1227–1231.
(24) (a) Bleicher, K. H.; Gerber, F.; Wuthrich, Y.; Alanine, A.; Capret-
ta, A. Parallel synthesis of substituted imidazoles from 1,2-ami-
noalcohols. Tetrahedron Lett. 2002, 43, 7687–7690. (b) Shin, C.;
Yonezawa, Y.; Obara, T.; Nishio, H. Dehydrooligopeptides. 8. Con-
venient Syntheses of Various Dehydrotyrosine Derivatives Protected
with Useful N,O-Protecting Groups Via N-Carboxy Dehydrotyrosine
Anhydrides. Bull. Chem. Soc. Jpn. 1988, 61, 885–891.
(25) (a) Pudovik, A. N.; Guryanov, I. V.; Banderov, Lv.; Romanov,
G. V. Phosphonate-Phosphate Rearrangement of Esters of Al-
pha-Hydroxyalkylphosphonic Acids. J. Gen. Chem. U.S.S.R. 1968,
38, 142–149. (b) Hammerschmidt, F.; Zbiral, E. Novel Synthetic
Aspects of the Phosphonate-Phosphate Rearrangement. 2. Synthesis
of Enolphosphates from 1-Oxoalkanphosphonates and Sulfur-Ylides.
Monatsh. Chem. 1980, 111, 1015–1023. (c) Hammerschmidt, F.
Phosphonate-Phosphate Rearrangement and Phosphate-Phosphonate
Rearrangement and Its Application. 2. Stereochemistry of Phosphona-
te-Phosphate Rearrangement at Carbon as Exemplified by the Isomer-
ization of (R)-(þ)-(1-Hydroxy-1-phenylethyl)phosphonic and (S)-(-)-
(1-Hydroxy-1-phenylethyl)phosphonic acid diethyl ester. Monatsh.
Chem. 1993, 124, 1063–1069. (d) McKenna, C. E.; Kashemirov,
B. A. Recent progress in carbonylphosphonate chemistry. In New Aspects
in Phosphorus Chem. I; Springer: Heidelberg, Germany, 2002; Vol. 220,
pp 201-238. (e) Sanchez, G. V., Jr.; Kashemirov, B. A.; McKenna, C. E.
40th Western Regional Meeting of the American Chemical Society,
Anaheim, CA, Jan 22-25, 2006; Abstract WRM-130.
(26) Deprele, S.; Kashemirov, B. A.; Hogan, J. M.; Ebetino, F. H.;
Barnett, B. L.; Evdokimov, A.; McKenna, C. E. Farnesyl pyropho-
sphate synthase enantiospecificity with a chiral risedronate analog,
[6,7-dihydro-5H-cyclopenta[c]pyridin-7yl(hydroxy)methylene]bis-
(phosphonic acid) (NE-10501): synthetic, structural, and modeling
studies. Bioorg. Med. Chem. Lett. 2008, 18, 2878–2882.
(27) (a) Lammerhofer, M.; Lindner, W. Quinine and quinidine deriva-
tives as chiral selectors. 1. Brush type chiral stationary phases for
high-performance liquid chromatography based on cinchonan
carbamates and their application as chiral anion exchangers.
J. Chromatogr., A 1996, 741, 33–48. (b) Maier, N. M.; Nicoletti, L.;
Lammerhofer, M.; Lindner, W. Enantioselective anion exchangers
based on cinchona alkaloid-derived carbamates: influence of
(33) (a) Kotti, T.; Head, D. D.; McKenna, C. E.; Russell, D. W.
Biphasic requirement for geranylgeraniol in hippocampal long-
term potentiation. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 11394–
11399. (b) Tian, X.; Jin, R. U.; Bredemeyer, A. J.; Oates, E. J.;
·
Blazewska, K. M.; McKenna, C. E.; Mills, J. C. RAB26 and RAB3D
e
Are Direct Transcriptional Targets of MIST1 That Regulate Exocrine
Granule Maturation. Mol. Cell Biol. 2010, 30, 1269–1284.
(34) Paudler, W. W.; Blewitt, H. L. NMR Spectra and Pi-Electron
Densities of Some Imidazo(1,2-a) Pyridines. Tetrahedron 1965, 21,
353–361.
·
(35) Blazewska, K. M.; Haiges, R.; Kashemirov, B. A.; Ebetino, F. H.;
e
McKenna, C. E. 18th International Conference on Phosphorus
Chemistry, Wroclaw, Poland, July 12-15, 2010; idem, manuscript
e
in preparation.
(36) Seabra, M. C.; James, G. L. Prenylation assays for small GTPases.
Transmembr. Signal Protoc. 1998, 84, 251–260.
(37) Kavanagh, K. L.; Dunford, J. E.; Bunkoczi, G.; Russell, R. G. G.;
Oppermann, U. The crystal structure of human geranylgeranyl
pyrophosphate synthase reveals a novel hexameric arrangement
and inhibitory product binding. J. Biol. Chem. 2006, 281, 22004–
22012.
(38) Frith, J. C.; Rogers, M. J. Antagonistic effects of different classes of
bisphosphonates in osteoclasts and macrophages in vitro. J. Bone
Miner. Res. 2003, 18, 204–212.
(39) Coxon, F. P.; Helfrich, M. H.; Van’t Hof, R.; Sebti, S.; Ralston,
S. H.; Hamilton, A.; Rogers, M. J. Protein geranylgeranylation is
required for osteoclast formation, function, and survival: inhibi-
tion by bisphosphonates and GGTI-298. J. Bone Miner. Res. 2000,
15, 1467–1476.
(40) Coxon, F. P.; Frith, J. C.; Benford, H. L.; Rogers, M. J. Isolation
and purification of rabbit osteoclasts. Bone Res. Protoc. 2003, 80,
89–99.