Inhibition of sphingosine-1-phosphate lyase for the treatment of autoimmune disorders

Article abstract of DOI:10.1021/jm900278w

During nearly a decade of research dedicated to the study of sphingosine signaling pathways, we identified sphingosine-1-phosphate lyase (S1PL) as a drug target for the treatment of autoimmune disorders. S1PL catalyzes the irreversible decomposition of sphingosine-1-phosphate (S1P) by a retro-aldol fragmentation that yields hexadecanaldehyde and phosphoethanolamine. Genetic models demonstrated that mice expressing reduced S1PL activity had decreased numbers of circulating lymphocytes due to altered lymphocyte trafficking, which prevented disease development in multiple models of autoimmune disease. Mechanistic studies of lymphoid tissue following oral administration of 2-acetyl-4(5)-(1(R),2(S),3(R),4-tetrahydroxybutyl)-imidazole (THI) 3 showed a clear relationship between reduced lyase activity, elevated S1P levels, and lower levels of circulating lymphocytes. Our internal medicinal chemistry efforts discovered potent analogues of 3 bearing heterocycles as chemical equivalents of the pendant carbonyl present in the parent structure. Reduction of S1PL activity by oral administration of these analogues recapitulated the phenotype of mice with genetically reduced S1PL expression.

Full text of DOI:10.1021/jm900278w

J. Med. Chem. 2009, 52, 3941–3953
3941
Inhibition of Sphingosine-1-Phosphate Lyase for the Treatment of Autoimmune Disorders
Jeffrey T. Bagdanoff,*,† Michael S. Donoviel,‡ Amr Nouraldeen,‡ James Tarver,† Qinghong Fu,† Marianne Carlsen,†
Theodore C. Jessop,† Haiming Zhang,† Jill Hazelwood,‡ Huy Nguyen,† Simon D. P. Baugh,† Michael Gardyan,†
Kristen M. Terranova,† Joseph Barbosa,† Jack Yan,† Mark Bednarz,† Suman Layek,† Lawrence F. Courtney,† Jerry Taylor,†
Ann Marie Digeorge-Foushee,‡ Suma Gopinathan,‡ Debra Bruce,‡ Traci Smith,‡ Liam Moran,‡ Emily O’Neill,‡ Jeff Kramer,‡
Zhong Lai,† S. David Kimball,† Qingyun Liu,‡ Weimei Sun,‡ Sean Yu,‡ Jonathan Swaffield,‡ Alan Wilson,‡ Alan Main,†
Kenneth G. Carson,† Tamas Oravecz,‡ and David J. Augeri*,†
Lexicon Pharmaceuticals, 350 Carter Road, Princeton, New Jersey, 08540, Lexicon Pharmaceuticals, 8800 Technology Forest Place,
The Woodlands, Texas 77381
ReceiVed March 4, 2009
During nearly a decade of research dedicated to the study of sphingosine signaling pathways, we identified
sphingosine-1-phosphate lyase (S1PL) as a drug target for the treatment of autoimmune disorders. S1PL
catalyzes the irreversible decomposition of sphingosine-1-phosphate (S1P) by a retro-aldol fragmentation
that yields hexadecanaldehyde and phosphoethanolamine. Genetic models demonstrated that mice expressing
reduced S1PL activity had decreased numbers of circulating lymphocytes due to altered lymphocyte trafficking,
which prevented disease development in multiple models of autoimmune disease. Mechanistic studies of
lymphoid tissue following oral administration of 2-acetyl-4(5)-(1(R),2(S),3(R),4-tetrahydroxybutyl)-imidazole
(THI) 3 showed a clear relationship between reduced lyase activity, elevated S1P levels, and lower levels
of circulating lymphocytes. Our internal medicinal chemistry efforts discovered potent analogues of 3 bearing
heterocycles as chemical equivalents of the pendant carbonyl present in the parent structure. Reduction of
S1PL activity by oral administration of these analogues recapitulated the phenotype of mice with genetically
reduced S1PL expression.
Introduction
Sphingolipids represent a wide array of naturally occurring,
amphiphilic modulators of eukaryotic cell-signaling. Containing
polar head groups, multiple hydroxyls, a free amino group, and
lengthy lipid chains, specific members of this family are key
mediators of intra- and extracellular signaling.1 Sphingosine-
1-phosphate (S1Pa)2 (Figure 1), enzymatically derived from
ceramide, is a key signaling lipid. S1P is present in all
mammalian cells and can serve as a secondary messenger in
the signal transduction pathways that regulate the vascular and
immune systems, cell differentiation, and apoptosis.3 It is also
released into the extracellular milieu by a variety of cell types,
making it one of the most abundant biologically active lyso-
phospholipids in circulation.4 S1P functions as an extracellular
ligand for G-protein-coupled receptors (GPCRs) that regulate
Figure 1. Structure of sphingosine-1-phosphate, myriocin, and 1.
T- and B-cell trafficking.5 Five S1P-binding GPCRs (S1P1-S1P5)
have been identified thus far and are expressed in many tissues.6
therapeutic potential of modulating sphingolipid signaling with
sphingolipid-like analogues8 and preceded the development of
Agonism of S1P1 results in lymphopenia, while agonism of
FTY-720 1,9 which is structurally homologous to both myriocin
S1P3 is associated with bradycardia.
Investigation into natural products structurally related to S1P,
such as myriocin,7 implicated control of specific key sphin-
golipid signaling events as a potential treatment for autoimmune
diseases. In vivo studies with myriocin demonstrated the
and sphingosine. A functional consequence of S1P or 1 binding
to S1P1 is receptor internalization, which eventually leads to
inhibition of lymphocyte egress from secondary lymphoid
tissues, resulting in peripheral lymphopenia. These discoveries
led to a closer examination of the importance of S1P gradients
and sphingosine-based signaling in immune regulation.
Endogenous levels of S1P can be influenced by modulating
the activity of the enzymes involved in its synthesis and
degradation. The relationship between S1P and its receptors,
and the array and function of sphingolipid enzymes, has been
recently reviewed.10 Sphingosine is produced by enzymatic
degradation of ceramide.11 Sphingosine kinase 1 (SK1) and
sphingosine kinase 2 (SK2) phosphorylate sphingosine to S1P,12
while S1P phosphatase catalyzes the reverse reaction.13 Alter-
natively, intracellular S1P concentrations are attenuated by S1P
* To whom correspondence should be addressed. For D.A.: phone, (609)
466-5526; fax, (609) 466-6079; E-mail: daugeri@lexpharma.com. For J.B.:
phone, (609) 466-5583; fax, (609) 466-6079; E-mail, jbagdanoff@
lexpharma.com.
†
Lexicon Pharmaceuticals, Princeton, NJ.
Lexicon Pharmaceuticals, The Woodlands, TX.
‡
a Abbreviations : THI, 2-acetyl-4(5)-(1(R),2(S),3(R),4-tetrahydroxybutyl)-
imidazole; S1P, sphingosine-1-phosphate; S1PL, sphingosine-1-phosphate
lyase; KO, knockout; S1P1, sphingosine-1-phosphate receptor 1; SK1,
sphingosine kinase 1; SK2, sphingosine kinase 2; GPCR, G-protein-coupled
receptor; CBC, complete blood count; SAR, structure-activity relationship;
ADME,absorption,distribution,metabolism,excretion;DOP,deoxypyridoxine.
10.1021/jm900278w CCC: $40.75  2009 American Chemical Society
Published on Web 05/21/2009

3942 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Bagdanoff et al.
Figure 3. Structure of THI 3.
activity as well, increasing S1P levels 275% over controls in
HepG2 cells. While 2b registered substantial levels of exposure
when dosed to mice at 10 mg/kg (Cmax ) 451 µM, AUC )
1872 µM·h, %F ) 100, mPPB > 99%), no significant
lymphopenia by was observed by CBC analysis at either 4 or
18 h following a 100 mg/kg oral dose. The high plasma protein
binding may account for the lack of in vivo activity observed.
By contrast, in the positive control leg of this experiment, 1
induced substantial lymphopenia at 10 mg/kg, registering greater
than 90% reduction in lymphocytes from peripheral circulation.
As we were unable to correlate in vitro activity with in vivo
pharmacology, we directed further medicinal chemistry efforts
toward an alternative chemotype.
2-acetyl-4(5)-(1(R),2(S),3(R),4-tetrahydroxybutyl)-imida-
zole (THI 3, Figure 3) is a small molecule present in trace
quantities in caramel color III, a popular food colorant produced
on an industrial scale by high temperature, high pressure reaction
of ammonia with carbohydrates.19 The immunosuppressive
effects of 3 on the immune system have been well docu-
mented.20 Interestingly, oral administration of 3 to mice was
reported21 to result in pharmacology that was strikingly similar
to the phenotype that we observed in our mouse line expressing
reduced S1PL.14 Similar pharmacology was observed with
deoxypyridoxine (DOP).20 Administration of 3 in drinking water
triggered lymphopenia concomitant with decreased S1PL activ-
ity and increased S1P levels in lymphoid tissues. Reduced egress
of mature CD4+ and CD8+ T cells from lymphoid tissues
accounted for lower levels of lymphocytes in peripheral
circulation. While no direct enzymatic evidence was offered,
the observed pharmacology was attributed to S1PL inhibition.
We were intrigued by the striking agreement between the in
vivo pharmacology generated by 3 and the phenotype of our
mouse lines with reduced S1PL expression.
Our work has been prompted by the recognition that agonism
of S1P1 via S1PL inhibition may afford considerable advantages
over direct S1P receptor agonism. In particular, synthetic S1P
receptor agonists must have selectivity against widely distributed
receptor subsets in order to avoid undesired pharmacology. For
example, the synthetic S1P receptor agonist 1 is nonselective
for S1P1, thus agonizing the S1P3 receptor in heart tissue,
causing bradycardia.22 To date, agonists that demonstrate
selectivity for S1P1 over S1P3 have yet to advance significantly
in human clinical trials. By comparison, because S1PL is
expressed primarily in lymphatic tissue, inhibition of S1PL
results in the accumulation of endogenous S1P primarily in these
tissues. As a consequence of localized S1P concentrations, mice
expressing reduced S1PL do not display bradycardia. Advan-
tageously, neutrophil levels are unaffected by S1PL inhibition,
unlike the pharmacology observed with S1P agonists. And,
unlike with 1, severe lymphopenia is not observed with lyase
inhibition. Our genetic models demonstrate that significant
reduction in S1PL expression results only in immunological
phenotypes that show significant promise in mitigating the
outcome of autoimmune disease. While synthetic S1P1 receptor
agonists are currently of high interest in pharmaceutical research
Figure 2. Structure of S1P transition state analogue inhibitor 2a,
oxopyridylpyrimidine 2b.
lyase (S1PL), which, aided by the cofactor pyridoxal phosphate,
catalyzes the irreversible retro-Aldol fragmentation of S1P to
hexadecanal and phosphoethanolamine. Inhibition of S1PL
therefore increases endogenous S1P concentrations in lymphoid
tissues, providing an alternative approach to modulation of S1P
receptors by synthetic ligands.14 Similar to synthetic S1P
receptor agonists, elevated levels of endogenous S1P lead to
lower levels of circulating peripheral lymphocytes and mitigates
the physiologic outcome of aberrant immune function. Thus,
enzymes that mediate the S1P gradient have inspired a new
approach for the treatment of autoimmune diseases.
As a part of a high-throughput gene knockout technology
platform, we conducted a detailed phenotypic analysis of nearly
5000 independent lines of knockout mice by gene targeting and
gene trapping in mouse embryonic stem cells. As a part of this
screen, we generated and analyzed several mouse lines deficient
in the expression of various enzymes and receptors involved in
sphingolipid metabolism and signaling. This work led to the
discovery that homozygous S1PL knockout mice exhibit a
significant decrease in circulating lymphocytes relative to wild-
type cohorts. In mice, only complete knockout of the S1PL gene
resulted in traits other than lymphopenia, leading to reduced
viability. In the S1PL knockout phenotype, high levels of S1P
were observed in the thymus and spleen.15 In addition to the
knockout mice, mouse lines with a wide range of S1PL
expression were also established using a genetic knock-in
strategy. Mice expressing less than 10% wild type S1PL
exhibited lymphopenia without other overt physiological effects.
As expected, these mice had an improved outcome in collagen-
induced arthritis, indicating that pharmacological inhibition of
S1PL represents a novel therapeutic strategy against inflam-
matory disorders.
Background
We began pursuing S1PL as a potential drug target for the
treatment of inflammatory disease through multiple chemical
approaches. Initially, de novo design considerations were applied
to the S1PL reaction substrate (S1P) and the requisite cofactor
(pyridoxal phosphate) to provide precursor 2a (Figure 2), which
we proposed would undergo SK1 or SK2 directed phosphory-
lation in vivo,16 generating a functional transition state analogue
inhibitor. Compound 2a was evaluated in vitro, measuring direct
enzyme inhibition,17 and by an in vivo assay, measuring blood
lymphocyte levels in whole blood by complete blood count
(CBC) four hours following IP administration.18 Compound 2a
displayed no activity in the biochemical assay, possibly due to
inadequate phosphorylation of the two primary alcohols. In
addition, 2a displayed no activity after 100 mg/kg oral
administration to mice (Cmax < 1 µM, AUC < 1 µM·h).
Oxopyridylpyrimidine 2b (Figure 2) was identified by HTS.
It directly inhibited S1PL (IC50 ) 2.1 µM) and exhibited cellular

Inhibition of Sphingosine-1-Phosphate Lyase
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3943
Scheme 1a
37 provided the C-4 epimer 33, enantiomer 34, C-3 methyl ether
36, and C-6 amide 38, respectively.
Results and Discussion
Our own internal research, coupled with reports published
by other groups,20 strongly supported the notion that the
biological activity of 3 was linked to S1PL inhibition. Published
results showed that the administration of 3 in drinking water
correlated to lymphopenia concomitant with increased concen-
trations of S1P in lymphoid tissue such as spleen and thymus.
While the pharmacology observed following administration of
3 closely resembled the phenotype of mice expressing reduced
levels of S1PL, 3 showed no direct inhibition of S1PL when
tested in our in vitro assay. However, the role of 3 in reduction
of S1PL activity and elevation of S1P was clearly evident in
lysates prepared from mouse spleen tissue following oral
administration of 3 (100 mg/kg). Tissues were harvested near
peak levels of pharmacology (lymphocyte levels by CBC at
18 h, vide infra) and compared directly to vehicle controls.
Results showed a significant reduction in lyase activity that was
concomitant with a 22-fold increase in S1P levels26 in spleen
tissue (Figure 4). This suggested that the established in vivo
efficacy was due to a more complex scenario than a simple
binary ligand-protein interaction. We considered the possibility
that 3 may undergo in vivo processing to an active metabolite.
As is the case with 1, phosphorylation occurs in vivo following
oral administration to provide the biologically active ligand.14
However, detailed analysis of the plasma revealed only the
parent compound. Alternatively, S1PL inhibition may be
dependent on the formation of a higher-order complex (ternary
or higher).27 As a consequence of these findings, we concluded
that 3 exerts biological activity by interaction with another
ligand, peptide, or protein that then goes on to inhibit S1PL.
While further elucidation of the mechanistic nuances remained
a priority, we were motivated to initiate investigation into the
SAR of chemical analogues of 3. In the absence of an in vitro
assay that was predictive of pharmacology, we established an
in vivo assay as our primary screen. Our medicinal chemistry
efforts focused on structural analogues of 3, measuring the in
vivo activity of our investigational compounds by CBC analysis
of lymphocytes in peripheral blood. While in vitro methods are
central to the paradigm of modern drug discovery, direct in vivo
evaluation of investigational agents is a historically validated
approach that accounts for a considerable percentage of
marketed drugs.28
Prior to testing analogues, we conducted a detailed pharma-
cokinetic analysis of 3 to establish a baseline profile (Table 1).
From these studies, it was determined that 3 displayed low
volume of distribution (0.067 L/kg) and low clearance (2.9 mL/
min/kg), providing moderate exposure at a standard oral dose
(35.8 µM·h, 10 mg/kg po).
We established the time-dependent relationship between dose
and onset of pharmacology. A PK/PD study was conducted
whereby 3 was administered at a single 100 mg/kg dose by
oral gavage to nine groups of mice (n ) 5). Each of the nine
groups was evaluated for reduction of circulating lymphocytes
by CBC analysis at various time points between 0 and 48 h
(Figure 5). While circadian rhythm will affect the absolute CBC
data taken from mice during the course of a day, this experiment
determined that the maximum pharmacodynamic effect (41%
reduction) was achieved between 12 and 24 h following a single
oral dose. We concluded that 18 h after oral administration
constituted a reliable time point for measuring peripheral levels
of circulating lymphocytes by CBC analysis.
a Conditions: (a) n-BuLi, THF, then 3,4-O-isopropylidene-D-erythrono-
lactone, -78 °C f rt; (b) NaBH4, i-PrOH; (c) aq HCl, acetone.
and development, small molecule inhibitors of S1PL have
remained relatively unexplored.
Chemistry
An initial approach toward the construction of analogues of
3 involved installation of the polyhydroxylated side chain onto
various heterocyclic cores. Lithiation-halogen exchange of
bromopyridine 4 (Scheme 1) preceded addition of 3,4-O-
isopropylidene-D-erythronolactone, and the intermediate lactol
was subsequently reduced and deprotected under acidic condi-
tions to provide pyridine-based tetraol 5. Thiazole-based tetraol
7
23 was prepared by the same reaction sequence, beginning with
deprotonation of thiazole 6.
Synthetic methods were also applied directly to 3 to provide
other modifications of the core heterocycle. As shown in Scheme
2, N-methylated analogue 8 was made by direct alkylation of 3
with methyl iodide under mildly basic conditions. Substitution
on the open position of the imidazole ring was accomplished
by bis-ketalization of the tetraol side chain of 3 followed by
bromination to provide intermediate 9. Suzuki coupling with a
variety of boronic acids followed by deprotection of the
tetrahydroxy side chain under acidic conditions furnished
analogues 10-12.
Other chemical approaches were applied to 3 to transform
the pendant ketone functionality. Direct reduction of the ketone
provided alcohol 13. Alternatively, condensation with acetyl-
hydrazine provided the acylhydrazone 14. The ketone was
fluorinated to release, after mildly acidic deprotection, difluoride
17. Other changes at the 2-position of 3 were made following
oxidation of the acetyl group. Protection of the tetrahydroxy
side chain and imidazole nitrogen followed by haloform reaction
provided the protected carboxylic acid 15, which was converted
to amide 16 after peptide coupling and deprotection.
The general synthetic approach used to synthesize 3 (Scheme
3) provided access to multiple analogues. Amadori rearrange-
ment24 of D-glucose with dibenzylamine produced D-fruc-
tosamine 18 after hydrogenation of the intermediate N-benzyl
groups. In the Bu¨chi25 reaction, cyclocondensation between 18
and nitrile 19 constructed the central imidazole ring present in
3 after treatment with acetic acid. The polyhydroxylated side
chain stereochemistry derives from the starting sugar. Applica-
tion of the Bu¨chi cyclocondensation reaction to a variety of
nitriles also permitted direct access to a variety of C-2 imidazole
analogues. Imidates were generated in situ by reaction of an
appropriate nitrile with sodium methoxide and reacted with 18
to afford C-2 substituted imidazoles 20-32 after treatment with
acetic acid.
Tandem execution of the Amadori and Bu¨chi reactions also
provided a general method to access side chain analogues,
deoxygenations, and heteroatom substitutions of 3 (Scheme 4).
Amadori rearrangement and Bu¨chi cyclization of D-galactose,
L-glucose, D-glucose C-3 methyl ether 35, and D-glucuronamide

3944 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Bagdanoff et al.
Scheme 2a
a Conditions: (a) K2CO3, MeI, DMF; (b) (MeO)2CMe2, DCE, p-TsOH, 50 °C; (c) Br2, DCM:2.0 M aq Na2CO3 (1:1), 0 °C; (d) Pd(dppf)Cl2, CsF, DMF:
PhMe (1:2), boronic acid, 100 °C; (e) acetone:water:conc HCl (2:2:1); (f) NaBH4, MeOH;(g) H2NNH2Ac, HCl, MeOH; (h) BnBr, K2CO3, DMF; (i) aq
NaOCl, NaOH, dioxane; (j) HATU, Me2NH-HCl, DCM; (k) Pd(OH)2, H2 (65 psi), EtOH; (l) conc HCl:water (2:1), 0 °C f rt; (m) DAST, DCM, 0 °C; (n)
1.0 N aq HCl:THF (1:1), 50 °C.
Scheme 3a
a Conditions: (a) Bn2NH, EtOH, HOAc, 80 °C; (b) H2, Pd-C, EtOH, HOAc; (c) nitrile, MeONa, MeOH, then HOAc.
Scheme 4a
a Conditions: (a) Bn2NH, EtOH, HOAc, 80 °C; (b) H2, Pd-C, EtOH, HOAc; (c) 19, MeONa, MeOH, then HOAc.
Following these studies, we began a chemistry effort to
examine the SAR of analogues of 3 with the intent of
developing an immune modulating agent with enhanced
potency and pharmacokinetics. Investigational agents were
initially evaluated at 100 mg/kg by oral gavage using a 100
mg/kg dose of 3 as a positive control. Blood was drawn at
18 h after dosing and lymphocyte suppression was determined
by CBC analysis. In addition to measuring in vivo activity
by CBC, the ADME and PK profile of each analogue was
established.
droxybutyl side chain manipulations, and (3) core heterocycle
pendant 2-position substitutions. Other design considerations
took into account the physical properties of our chemical lead.
Small polar molecules like 3 often achieve plasma concentra-
tions by paracellular absorption.29 Therefore, we felt it
prudent to not deviate dramatically from the physical
properties of 3, including molecular weight and polar surface
area. We measured Caco-2 values and plasma protein binding
(PPB) as potentially important variables on which the
absorption mechanism may be dependent. With a panel of
pharmacokinetic and in vitro ADME assays in place, we were
poised to interpret the SAR of analogues prepared.
Analogues of 3 may be grouped broadly into three
categories: (1) core heterocycle modifications, (2) tetrahy-

Inhibition of Sphingosine-1-Phosphate Lyase
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3945
Table 2. Effect of Core Modification on PK and PD
Figure 4. Residual lyase activity and S1P levels following oral
administration of 3. (A) Treated group is lysate derived from mouse
spleen tissue harvested 18 h after 100 mg/kg oral administration with
THI; treated + THI group is exogenous addition of THI to lysate from
treated group; control group is spleen lysate from untreated mice; control
+ THI group is exogenous addition of THI to control group lysate.
(B) S1P levels in mouse spleen lysate of control group vs S1P levels
in lysate from mouse spleen tissue harvested 18 h after 100 mg/kg
oral administration with THI. S1P levels measured by radioreceptor-
binding assay.
Table 1. PK/PD Profile, in Vitro and Select Physical Properties of 3
a Reported as % change in circulating lymphocytes relative to vehicle
control 18 h after 100 mg/kg oral dose of HCl salt to mice (n ) 5).
Interanimal variability was less than 20%. b Exposure calculated as t0 f
inf in (µM·h). Obtained from 10 mg/kg po dose to mice (n ) 5). c (nM/s),
obtained at [20 µM]. d Clearance in (mL/min/kg). Calculated from 1 mg/
kg iv dose to mice (n ) 4). f Half-life in h. g Calculated from 1 mg/kg iv
and 10 mg/kg po.
a Maximum exposure in (µM). b Time at maximum exposure (h).
c Half-life in h. d Oral bioavailability calculated from 1 mg/kg iv and 10
mg/kg po. e Exposure calculated as t0 f inf in (µM·h). Obtained from 10
mg/kg po dose to mice (n ) 5). f Clearance in (mL/min/kg). Calculated
from 1 mg/kg iv dose to mice (n ) 4). g (nM/s), obtained at 20 µM.
h Reported as % change in circulating lymphocytes relative to vehicle control
18 h after 100 mg/kg oral dose to mice (n ) 5). Interanimal variability was
less than 20%.
and pyridine 6 produced elevated lymphocyte levels. Substitu-
tion at the 5-position of the core imidazole with a range of
lipophilic substituents provided data on the steric requirements
in that region and allowed us to test the importance of
maintaining the overall polarity of analogues. After dosing to
mice, 5-alkyl analogues 10 and 11 showed diminished activity
while 5-aryl analogue 12 registered elevated lymphocyte levels.
Increased CL and lower exposure, possibly in addition to
reduced intrinsic activity, may have contributed to lower overall
in vivo potency. The incidence of compounds that elevated
lymphocyte counts was low and the significance of these
findings is unclear.
Side Chain Modification SAR and PK. In addition to
studying the effect of core changes to 3, we investigated the
impact of changes to the composition and chirality of the
hydroxylated side chain. Consequently, we were able to exam-
ine the H-bond donor-acceptor properties of the side chain.
The effects of these changes are summarized in Table 3. Oral
dosing of the C-4 epimer 33 revealed a moderately potent
compound with reduced exposure and higher clearance relative
to 3. That the reduced oral potency of 33 correlates with reduced
exposure may indicate that intrinsic potency is not significantly
impacted by stereochemical inversion at C-4. More substantial
side chain alterations, as presented by the enantiomer 34,
completely negated activity. Enantiomer 34 displayed a com-
parable pharmacokinetic profile compared to 3, which under-
scores the importance of the asymmetric configuration derived
from glucose. Manipulation of the side chain functional groups,
as in methyl ether 36, and tertiary amide 38, also resulted in
compounds devoid of activity. That inactive ether 36 exhibited
an improved exposure points to either the importance of the
Figure 5. Time-course study of 3. Single 100 mg/kg dose given to
mice (n ) 5) via oral gavage. Change in circulating peripheral
lymphocytes for treated group (yellow) relative to vehicle control
(green) at the indicated times post gavage. The modest increase in
peripheral circulating lymphocytes at 4 h was deemed an experimental
artifact, which occurred before the onset of pharmacology.
Core Modification SAR and PK. In vivo administration of
8 revealed that methylation at nitrogen rendered 3 inactive, as
the CBC data for this analogue was comparable to vehicle
control (Table 2). That the modestly improved exposure and
bioavailability of imidazole 8 did not translate to an improved
PD effect indicates that an N-H may be essential for intrinsic
activity. Curiously, several compounds produced significantly
higher levels of circulating lymphocytes. For reasons unknown
to us, replacement of the central imidazole ring with thiazole 7

3946 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Table 3. Effect of Side Chain Modification on PK and PD
Bagdanoff et al.
By contrast, R,R-difluoro analogue 17 displayed remarkable
ability to deplete lymphocyte levels despite its low pharmaco-
kinetic exposure. Initially, the reduced exposure was thought
to be a function of a shorter half-life. However, further study
of 17 in toxicokinetic (TK) experiments revealed significant
toxicity when dosed to mice orally at 100 mg/kg. Concurrent
with PK and toxicology experiments were chemical observations
made during the synthesis of similarly substituted R,R-difluoro
analogues. While our final products were clearly stable during
characterization, hydrolysis products were observed during
purification. We determined that acid-catalyzed hydrolysis of
R,R-difluoro analogue 17 regenerated the parent ketone with
liberation of 2 equiv of HF.30 This observation clarified the short
half-life of 17 and left little doubt regarding the nature of the
toxicity observed in the mouse TK experiments. In fact,
subsequent pathology confirmed findings consistent with HF
poisoning.31
Acylhydrazone 14 gave early indications of enhanced activity
relative to 3 despite its lower exposure. Cognizant of potential
chemical stability issues, we reasoned that the low exposure
may relate to the sensitivity of the acylhydrazone group to the
acidic conditions present in the upper gastrointestinal tract.
Despite the lower concentrations of parent compound measured
in the plasma, we observed an increase in intrinsic potency over
3. Given our SAR at the 2-position, it was postulated that the
H-bond acceptor capacity of the Lewis-basic carbonyl oxygen
was essential to the potency of investigational agents. We
hypothesized that the CdN carbonyl of acylhydrazone 14
functioned as a ketone equivalent and that potency may extend
to other sp2 hybridized nitrogens in this region of the molecule.
We reasoned that heterocycles containing an appropriately
oriented CdN bond within the ring could also function as a
ketone surrogate. Consequently, we prepared and evaluated a
diverse series of analogues bearing heterocycles at the pendant
2-position of the parent imidazole ring.
a Reported as % change in circulating lymphocytes relative to vehicle
control 18 h after 100 mg/kg oral dose of HCl salt to mice (n ) 5).
Interanimal variability was less than 20%. b Exposure calculated as t0 f
inf in (µM·h). Obtained from 10 mg/kg po dose to mice (n ) 5). c (nM/s),
obtained at [20 µM]. d Clearance in (mL/min/kg). Calculated from 1 mg/
kg iv dose to mice (n ) 4). f Half-life in h. g Calculated from 1 mg/kg iv
and 10 mg/kg po.
Table 4. Effect of 2-Substitution on PK and PD
2-Heterocycle SAR and PK. Having identified the 2-position
of the central imidazole as an area tolerant to structural
modification, a series of 2-heterocycles bearing a basic nitrogen
proximal to the imidazole ring were synthesized. Commercial
availability of a wide variety of heterocyclic nitriles, or nitrile
precursors, made a thorough investigation of this SAR possible.
Data within Table 5 confirm that a CdO carbonyl, a CdN
carbonyl, or a structural homologue at the 2-position is a
common feature among active analogues.
As shown in Table 5, thiazoles 22, 23, and 24 exhibited a
range of in vivo activity. The unsubstituted thiazole 24 displayed
significant in vivo potency over the substituted thiazoles 22 and
23. That N-methylimidazole 32 lacks potency suggests a steric
constraint in this region of the molecule. Interestingly, the
opposite trend emerges with triazoles 26 and 27. N-Benzyl
triazole 26 was active while unsubstituted triazole 27 lacked
potency. This trend may reflect the preference for lipophilic
substituents in this region. Alternatively, heteroatoms in this
region may not be tolerated. The latter hypothesis is supported
by the difference in potency between pyrimidine 29 and pyrazine
30. We were pleasantly surprised to discover that imidazole 28
induced particularly significant lymphopenia when administered
orally at 30 mg/kg to mice. While our hypothesis concerning
the importance of a sp2 hybridized nitrogen was generally
consistent with these data, some exceptions exist. For example,
the unsubstituted oxazole 25 proved to be inactive despite the
close structural resemblance to thiazole 24. Oxadiazole 31
displayed only moderate activity in our assay. It is worth noting
that, despite showing lower exposure relative to 3, several
a Reported as % change in circulating lymphocytes relative to vehicle
control 18 h after indicated oral dose of HCl salt to mice (n ) 5). Interanimal
variability was less than 20%. b Exposure calculated as t0 f inf in (µM·h).
Obtained from 10 mg/kg po dose to mice (n ) 5). c (nM/s), obtained at [20
µM]. d Clearance in (mL/min/kg). Calculated from 1 mg/kg iv dose to mice
(n ) 4). f Half-life in h. g Calculated from 1 mg/kg iv and 10 mg/kg po.
C-3 hydroxyl in target binding or steric demands imposed by
the target in this region.
Substitution of the Imidazole 2-Position: SAR and PK
Profile. Substitution at the 2-position of the imidazole also
impacted in vivo activity. But, unlike modification to the side
chain, some changes to the 2-position provided viable im-
munologic agents. The identification of compounds with
enhanced activity prompted us to increase the sensitivity of our
primary screen. Accordingly, we reduced of the oral dose of
analogues and positive control from 100 to 30 mg/kg. (Table
4). Alcohol 13 displayed decreased potency as measured by
CBC. This reduced activity was likely due to a decrease in
intrinsic potency because 13 maintained a favorable PK profile.
Diminished potency was observed with sterically encumbered
aryl ketone 21. The reduced potency correlated with substantially
reduced exposure, potentially compounded by lower intrinsic
potency.

Inhibition of Sphingosine-1-Phosphate Lyase
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3947
Table 5. Effect of 2-Heterocycle Carbonyl Replacements (30 mg/kg) on
analogues of 3, as did several transformations at the pendant
ketone. By contrast, compounds bearing certain heterocycles
at the 2-position of the imidazole core provided superior in vivo
activity. Ongoing efforts targeting S1PL by direct in vivo
evaluation of synthetic analogues of 3 has led to the discovery
of clinically useful compounds. Efficacy studies in animal
models of autoimmune disease and human data from clinical
trials will be the subject of future reports.
PK and PDa
Experimental Procedures
Animal Handling and In Vivo Methods. All procedures were
conducted according to applicable standard operating procedures
and with the approval of the Lexicon Pharmaceuticals Incorporated
institutional animal care and usage committee. The tested analogues
were formulated in sterile water and dosed to 129/SvEvBrd ×
C57BL/6 F1 mice. The pH of intravenous dosing solution was
adjusted to between 5 and 6 and between 3 and 4 for oral
administration. The intravenous dose volume was 3.7 mL/kg body
weight and the oral gavage volume was 5 mL/kg body weight at 1
and 10 mg/kg body weight, respectively. Animals were observed
regularly throughout the day on days in which they were dosed,
and these clinical observations were recorded. For the purpose of
exposure, blood was collected using saphenous vein bleeds in
unanaesthetized animals, except for terminal bleeds, which were
performed using cardiac puncture immediately following euthanasia.
Blood samples were collected at 0.08, 0.25, 0.5, 1, 3, and 6 h post
intravenous dosing and at 0.25, 0.5, 1, 2, 4, 6, and 24 h post oral
administrations. An aliquot of 25 µL of blood in K2-EDTA tubes
was spun following each time point, and 10 µL of the resulting
plasma was transferred to fresh polypropylene tubes for exposure
assessment by liquid chromatography-mass spectrometry.
Measurement of S1PL Activity. Spleens of mice were homog-
enized in ice-cold buffer containing 50 mM KPO4, pH 7.5, 1 mM
EGTA, 1 mM DTT, and complete protease inhibitor cocktail (Roche
Applied Science, Indianapolis, IN). The homogenate was centri-
fuged at 2500g for 10 min, and the supernatant was used for enzyme
activity measurements. Protein concentration was determined using
the Bradford assay. S1PL activity was determined using 33P-labeled
S1P as substrate (American Radiolabeled Chemicals, St. Louis,
MO). The reactions were carried out in a 50 µL volume containing
100 mM KPO4, pH 7.4, 1 mM EDTA, 25 mM NaF, 5 mM DTT,
phosphatase inhibitors I and II (Sigma-Aldrich), 0.1% Triton X-100,
and various amounts of tissue lysate. The reactions were initiated
by addition of pyridoxal phosphate (final concentration 50 µM) and
15 nCi of [33P]-S1P (specific activity 0.66 Ci/mmol, total S1P
concentration 0.5 µM). The reaction mixtures were incubated at
37 °C for 1 h and terminated with the addition of 90 µL of 35%
acetonitrile containing 0.1% TFA. Then 120 µL of the mixtures
were then transferred to a 96-well C18 multi-SPE plate (Millipore
Corporation, Bedford, MA) pre-equilibrated first with 100 µL of
MeOH and then with 200 µL of 20% acetonitrile containing 0.1%
TFA. The plate was briefly centrifuged at 3000 rpm for 1 min, and
the filtrate was collected and assayed for 33P activity by scintillation
counting.
a Reported as % change in circulating lymphocytes relative to vehicle
control 18 h after 30 mg/kg oral dose of HCl salt to mice (n ) 5). Interanimal
variability was less than 20%. b Exposure calculated as t0 f inf in (µM·h).
Obtained from 10 mg/kg po dose to mice (n ) 5). c (nM/s), obtained at [20
µM]. d Clearance in (mL/min/kg). Calculated from 1 mg/kg iv dose to mice
(n ) 4). f Half-life in h. g Calculated from 1 mg/kg iv and 10 mg/kg po.
2-heterocycles showed equivalent and even superior potency.
That this biological activity was achieved with 2-heterocycles
24, 26, 28, and 29 may suggest that superior target affinity
compensates for impaired pharmacokinetics. Alternatively, these
analogues may better address the unresolved mechanistic
subtleties of the interaction between 3 and S1PL (e.g., involve-
ment of a higher-order complex). Another subtlety involves other
mechanism(s) that may also contribute to the antiinflammatory
activity of S1PL inhibition beyond simply limiting lymphocyte
egress. While establishing the precise mechanism remains a
priority, our medicinal chemistry efforts have identified com-
pounds with improved efficacy in models of autoimmune disease
and these results will be the subject of future reports.
Conclusions
Measurement of S1P Concentration. S1P extraction from
mouse sera was performed by mixing 100 µL of serum with 900
µL MeOH containing 0.1% TFA in an Eppendorf tube. After
vortexing for 2 min, 800 µL of water with 0.1% TFA was added
to the samples. Then, 300 µL of the mixture was then transferred
to a 96-well C18 multi-SPE plate pre-equilibrated first with 100
µL of MeOH and then with 200 µL of 20% acetonitrile containing
0.1% TFA. The plates were spun at 3000 rpm for 1 min and the
bound S1P eluted in 100 µL of MeOH. The eluate was dried and
stored at 4 °C overnight before analysis. S1P extraction from single
cell suspensions of mouse splenocytes was performed by using 4-6
× 107 cells. Cells were washed with ice-cold PBS and resuspended
by vortexing for 2 min in 500 µL of MeOH containing 0.1% TFA.
Then 500 µL of water with 0.1% TFA was added to the samples
and S1P was eluted and dried as described for the serum samples,
using 160 µL mixture/well on a C18 multi-SPE plate. S1P levels
Inhibition of S1PL correlates with increased concentrations
of S1P in lymphoid tissue, resulting in decreased levels of
peripherally circulating lymphocytes, and shows therapeutic
potential for the control of autoimmune disease. The pharmacol-
ogy observed after administration of 3 is strikingly similar in
many respects to the phenotype of genetically altered mice
expressing reduced levels of S1PL. While 3 does not show direct
inhibition of S1PL in vitro, oral administration clearly correlates
with reduced lyase activity in lymphoid tissue concomitant with
elevated levels of S1P. Evidently, S1PL inhibition triggerd by
3 involves a more complicated scenario than inhibition by a
discrete binary ligand-protein complex. Further elucidation of
this mechanism of action remains a priority. Alterations to the
side chain and core heterocycle reduced the potency of

3948 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Bagdanoff et al.
in the eluates were measured with a whole cell receptor binding
assay using HEK293 cells expressing the S1P1 (Edg1) receptor.
Five × 104 HEK293-S1P1 cells/well were seeded into clear bottom
96-well plates coated with poly-D-lysine (BD BioSciences) and
incubated overnight at 37 °C. Cells were placed on ice for 15 min
then washed twice with ice-cold binding buffer (25 mM Tris-HCl,
pH 7.5, 100 mM NaCl, 50 mM NaF, 0.4% fatty acid free BSA,
fraction 5). Dried S1P eluates or standards were resuspended in 50
µL of ice-cold binding buffer containing 25000 cpm [33P]-S1P
(specific activity 0.6 Ci/mmol) and incubated on ice for 60 min.
After washing twice with 100 µL of ice-cold binding buffer, 50
µL of lysis buffer (0.1% SDS, 0.4% NaOH, 2% Na2CO3) was added
and the plates were shaken for 5 min. Then 40 µL of cell extract
was then transferred to a white opaque microplate (Costar,
Cambridge, MA, cat. no. 3912) and counted after the addition of
120 µL of MS 40 microscint. S1P content of extracts was estimated
from standard curves (between 42 pM and 2.5 µM), generated with
cold S1P.
(2.7 L), dibenzylamine (539.14 g, 2.733 mol), and glacial acetic
acid (160 mL, 2.795 mol). The suspension was set stirring under
nitrogen and heated to reflux. After 4 h, the reaction was allowed
to gradually cool to room temperature and stirred under nitrogen
overnight. The resulting suspension was filtered, and the solids were
washed with ethanol (2 × 750 mL) and heptane (500 mL). The
filter cake was dried under vacuum at 50 °C overnight to afford a
white solid (923 g). A 12 L three-neck round-bottom flask with
mechanical stirrer, subsurface gas inlet bubbler, and gas outlet tube
was charged with a portion of this solid (219.50 g, 610.71 mmol),
ethanol (3.0 L), and acetic acid (1.5 L). The resulting suspension
was sparged with nitrogen for 30 min and then treated with 10%
Pd/C DeGussa type, 50 wt % water (44.47 g, 2.22 g active Pd).
Hydrogen was bubbled into the stirring mixture for 4 h then sparged
with nitrogen for 30 min, and the product solids were dissolved
upon addition of water (800 mL). The resulting solution was filtered
through to remove insoluble catalyst solids, and the yellow filtrate
was concentrated under vacuum to give a tan solid, which was
dissolved in water (150 mL) warmed to 55 °C. Ethanol (1.0 L)
was added over 5 min to the warm solution to provide a turbid
solution, which formed a suspension upon standing. The thickening
suspension was let stand at 55 °C for 15 min, then gradually cooled
to room temperature while stirring. The suspended solids were
filtered and filter cake was washed with 2-propanol (2 × 250 mL)
and heptane (1 × 250 mL) and then were dried in a vacuum oven
to afford fructosamine-HOAc 18 (124.16 g, 85% yield) as a fine
off-white chalky powder. Characterization of the obtained material
was in good agreement with literature values.32
General Procedure B: 1-(4-((1R,2S,3R)-1,2,3,4-Tetrahydroxy-
butyl)-11H-imidazol-2-yl)ethanone (THI, 3). To a three-neck, 12
L round-bottom flask equipped with a mechanical stirrer and a
temperature controller was added ethoxyacrylonitrile (10.0 g, 103.0
mmol) and methanol (200 mL). The solution was treated with 25
wt % sodium methoxide in methanol (14.1 mL, 60.0 mmol) and
maintained at room temperature for 1 h. Then, fructosamine-HOAc
(23.9 g, 100.0 mmol) was added in one portion, and the slurry was
stirred at room temperature for 6 h. Then another portion of 25 wt
% sodium methoxide in methanol (11.8 mL, 50.0 mmol) was added
and the reaction was maintained at room temperature for 16 h. The
mixture was then diluted with water (200 mL) and treated with
acetic acid (11.8 mL, 200.0 mmol). After stirring at 60 °C for 1 h,
the solution was concentrated to 100 mL total volume. The slurry
was cooled to 0 °C and stirred for 1 h then filtered and the solids
were washed with water (2 × 10 mL). The solids were collected
and dried under vacuum at 50 °C to afford a pale-yellow solid,
which was slurried in water (100 mL) and heated to boiling for 15
min before cooling to 0 °C. The mixture was stirred for 1 h then
filtered, and the solids were washed with water (2 × 10 mL) and
dried to constant weight in a vacuum oven at 50 °C to provide 3
(19.5 g, 82% yield) as a pale-yellow solid. The product was
analytically pure but may be purified further by recrystallization
from MeOH/Et2O or by precipitating from water upon treatment
with 2.0 M aq NaOH solution. 1H NMR (400 MHz, 1.0 M HCl in
D2O) δ ppm 7.48 (d, J ) 2.0 Hz, 1 H), 5.09 (s, 1 H), 3.40-3.70
(m, 4H), 2.53 (d, J ) 2.4 Hz, 3H). 13C NMR (400 MHz, 1.0 M
HCl in D2O) δ ppm 185.0, 139.4, 138.0, 119.5, 73.0, 70.9, 65.0,
63.2, 26.7. MS (EI) m/z: 231 [M + H]+. HRMS calcd for
C22H15N2O5 [M + H]+ 231.0981, found 231.0976. HPLC (Zorbax
C8 4.6 mm × 150 mm, 4% MeCN:10 mM aq NH4OAc, 4 min
isocratic) tR ) 1.11 min, 100% integrated area.
Chemical Methods. All reactions were conducted under a static
atmosphere of argon or nitrogen and stirred magnetically unless
otherwise noted. Reagents, starting materials, and solvents were
purchased from commercial suppliers and used as received. Flash
column chromatography was carried out using prepacked silica gel
columns from Biotage or ISCO or by slurry preparation using EMD
1
silica gel 60 (particle size 0.040-0.063 mm). H and 13C NMR
spectra were collected on Bruker ARX300, DRX400, or DPX400,
or Varian Mercury 400 MHz NMR spectrometers. Chemical shifts
are reported in parts per million (ppm) relative to tetramethylsilane
in δ units, and coupling constants (J values) are given in hertz (Hz).
Data are reported in the following format: chemical shift, multiplic-
ity, coupling constants, and assignment. Reactions were monitored
by TLC using 0.25 mm E. Merck silica gel plates (60 F254) and
were visualized with UV light. Analytical HPLC spectra were
collected on Shimadzu HPLC systems equipped with a UV detector
measuring absorbance at 220 and 254 nm. Mass spectra were
obtained on Waters ZQ or ZMD LCMS systems equipped with an
autosampler, an ELSD detector, a UV detector measuring absor-
bance at 220 and 254 nm, and a mass detector. High-resolution
mass spectra were obtained on a Waters LCT Premier XE
Micromass MS Technologies instrument equipped with an au-
tosampler. Elemental analysis was conducted by Robertson Microlit
Laboratory, Madison, NJ.
2-Ethoxyacrylonitrile (19). A 250 mL jacketed three-neck
round-bottom flask was equipped with a magnetic stir bar, rubber
septum with temperature probe, plastic stopper, and pressure-
equalized addition funnel with gas bubbler. The vessel was charged
with bromoacetaldehyde diethylacetal (35.33 g, 179.28 mmol) and
tin(II) chloride (167 mg, 0.88 mmol). The suspension was cooled
to 3 °C before the addition of trimethylsilyl cyanide (17.69 g, 178.31
mmol) over 15 min, while maintaining the reaction temperature
below 15 °C. The mixture was held at about 10 °C for 30 min,
then warmed to 20 °C for 2 h or until GC analysis showed
consumption of bromoacetaldehyde diethylacetal. The resulting
yellow solution was diluted with MTBE (85 mL). Diethylamine
(19.61 g, 268.11 mmol) was added over 10 min while the
temperature was maintained below 35 °C. The resulting thick slurry
was diluted with additional portion of MTBE (25 mL). The reaction
mixture was stirred at about 25 °C for 2 h or until reaction was
complete as determined by GC analysis. The mixture was filtered
and the collected solids were washed MTBE (2 × 30 mL). The
combined filtrate was concentrated under vacuum. The resulting
crude ethoxyacrylonitrile was polish-filtered and vacuum distilled
to give 19 as a clear liquid (11.2 g, 67% yield); bp: 46-49 °C/31
mmHg. 1H NMR (400 MHz, CDCl3) δ ppm 4.85 (d, J ) 3.54 Hz,
1H), 4.75 (d, J ) 3.28 Hz, 1H), 3.73 (q, J ) 6.99 Hz, 2H), 1.24 (t,
J ) 6.95 Hz, 3H). 13C NMR (100 MHz, CDCl3) δ ppm 136.58,
115.23, 101.11, 65.54, 14.40.
1-(6-((1R,2S,3R)-1,2,3,4-Tetrahydroxybutyl)pyridin-2-yl)etha-
none (5). 2-Bromo-6-(2-methyl-1,3-dioxolan-2-yl)pyridine 433 (3.97
g, 16.30 mmol) was suspended in THF (120 mL) and cooled to
-78 °C. n-Butyl lithium (1.6 M in hexanes, 12.2 mL) was added
dropwise, and the reaction was warmed to -40 °C for 45 min.
The reaction was recooled to -78 °C, and a solution of 3,4-O-
isopropylidene-D-erythronolactone (4.64 g, 29.3 mmol) in THF (40
mL) was added dropwise over 30 min. After stirring for 2 h, the
reaction was quenched with sat. aq NH4Cl (50 mL). The reaction
was warmed to room temperature and then extracted with EtOAc
(3 × 50 mL). The combined organic portions were washed with
General Procedure A: Fructosamine-HOAc (18). A 12 L three-
neck round-bottom flask with mechanical stirrer, rubber septum with
temperature probe, and reflux condenser with gas bubbler was
charged with D-(+)-glucose (504.08 g, 2.798 mol), absolute ethanol

Inhibition of Sphingosine-1-Phosphate Lyase
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3949
water (50 mL) and brine (30 mL) and then dried over MgO4,
filtered, and concentrated to provide a yellow solid, which was used
without further purification. To a -60 °C solution of the intermedi-
ate lactol (2.77 g, 8.57 mmol) in isopropyl alcohol (248 mL) was
added solid sodium borohydride (3.24 g, 85.7 mmol) portionwise
over 20 min. The reaction was removed from the cold bath and
stirred overnight. The reaction was then concentrated and the residue
diluted with brine (100 mL) and extracted with EtOAc (3 × 100
mL). The combined organics were dried over MgO4, then filtered
and concentrated to provide a crude mixture of diastereomers, which
were separated by preparative HPLC. The desired diastereomer
(0.53 g, 1.64 mmol) was taken up in 1:1 acetone:water (50 mL),
treated with conc HCl (14 mL), and the reaction was heated to 80
°C for 2 h. At completion, the acetone was removed under vacuum,
and the remaining aqueous portion washed with ether (3 × 50 mL).
The aqueous portion was concentrated and then purified by HPLC
to provide 5 (0.18 g, 8% yield, 3 steps) as a white solid. The
absolute configuration at the benzylic center was confirmed by
MHz, chloroform-d) δ ppm 11.03 (br s, 1 H), 5.03 (d, J ) 8.1 Hz,
1 H), 4.11-4.24 (m, 2 H), 3.99-4.07 (m, 1 H), 3.95 (dd, J ) 8.6,
4.5 Hz, 1 H), 2.64 (s, 3 H), 1.52 (s, 3 H), 1.48 (s, 3 H), 1.33 (s, 3
H), 1.22 (s, 3 H). 13C NMR (100 MHz, chloroform-d) δ ppm
189.08, 144.12, 129.77, 117.03, 111.09, 110.21, 81.11, 76.47, 73.64,
67.42, 26.98, 26.92, 26.45, 25.48, 25.26. MS (EI) m/z: 389, 391
[M + H]+. HRMS calcd for C15H21BrN2O5 [M + H]+ 389.0712,
found 389.0753.
General Procedure C: 1-(5-Methyl-4-((1R,2S,3R)-1,2,3,4-tetra-
hydroxybutyl)-1H-imidazol-2-yl)ethanone (10). To a solution of the
bromide 9 (1.12 g, 2.88 mmol) in toluene (12 mL) and DMF (6
mL) was added cesium fluoride (875 mg, 5.76 mmol), Pd(dppf)Cl2
(118 mg, 0.144 mmol), and methyl boronic acid (346 mg, 5.76
mmol). The resulting slurry was iteratively evacuated and then
backfilled with nitrogen while sonicating for 15 min and then heated
to 100 °C for 16 h. At completion, the reaction was filtered over a
pad of celite. The filter pad was washed with toluene (2 × 10 mL),
and the combined organics were concentrated and then dried under
vacuum. The crude material was purified by flash chromatography
over silica gel (30-50% EtOAc:hexanes eluent) to provide a plastic
mass that was dissolved acetone (6 mL) and water (6 mL) and
treated with concentrated HCl (2 mL). The reaction was maintained
at room temperature for 2 h. At completion, the reaction was
concentrated and then dissolved in water (5 mL) and lyophilized
1
single crystal X-ray analysis. H NMR (400 MHz, 1.0 M HCl in
D2O) δ ppm 7.85-8.09 (m, 2 H), 7.74 (d, J ) 7.6 Hz, 1 H), 5.10
(d, J ) 2.3 Hz, 1 H), 3.94 (dd, J ) 8.2, 2.4 Hz, 1 H), 3.76-3.87
(m, 2 H), 3.63 (dd, J ) 11.6, 6.3 Hz, 1 H), 2.70 (s, 3 H). 13C NMR
(100 MHz, 1.0 M HCl in D2O) δ ppm 203.94, 160.89, 152.13,
139.05, 125.95, 122.29, 74.24, 72.66, 71.61, 63.39, 26.15. MS (EI)
m/z: 242 [M + H]+; mp: 142-144 °C. HRMS calcd for C11H15NO5
[M + H]+ 242.1028, found 242.1019. HPLC (Zorbax C8 4.6 mm
× 150 mm, 10-90% MeCN:10 mM aq NH4OAc, 3 min gradient)
tR ) 1.72 min, 100% integrated area.
1-(1-Methyl-4-((1R,2S,3R)-1,2,3,4-tetrahydroxybutyl)-1H-imida-
zol-2-yl)ethanone (8). To a room temperature slurry of 3 (190 mg,
0.83 mmol) and K2CO3 (229 mg, 1.66 mmol) in DMF (5 mL) was
added MeI (56 µL, 0.91 mmol). After 4 h, the reaction was filtered
and concentrated and then purified by HPLC. Pure fractions were
combined and concentrated and then suspended in water (5 mL)
and lyophilized to afford 8 (121 mg, 60% yield) as a white powder.
1H NMR (400 MHz, MeOD) δ ppm 7.19 (s, 1 H), 4.86 (s, 1 H),
3.86 (s, 3 H), 3.61-3.75 (m, 3 H), 3.55 (dd, J ) 13.6, 5.3 Hz, 1
H), 2.44-2.49 (m, 3 H). 13C NMR (100 MHz, MeOD) δ ppm
191.61, 145.23, 143.42, 126.99, 75.52, 73.14, 69.06, 65.09, 36.70,
27.31. Regiochemistry of N-methylation determined by NOE
between N-CH3 and Ar-H. MS (EI) m/z: 245 [M + H]+. HRMS
calcd for C10H16N2O5 [M + H]+ 245.1137, found 245.1120. HPLC
(C30 Develosil RP Aqueous 4.6 mm × 50 mm, 5-50% [methanol/
water/TFA (95/5/0.1)]:water, 4 min gradient) tR ) 1.52 min, 99%
integrated area.
1-(5-Bromo-4-((4S,4′R,5R)-2,2,2′,2′-tetramethyl-4,4′-bi(1,3-dioxo-
lan)-5-yl)-1H-imidazol-2-yl)ethanone (9). To a round-bottom flask
charged with THI (11.00 g, 47.61 mmol) suspended in dichloro-
ethane (100 mL) was added 2,2-dimethoxypropane (100 mL) and
p-toluenesulfonic acid (1.84 g, 9.60 mmol). The mixture was heated
to 50 °C for 24 h. At completion, the resulting solution was
neutralized with triethylamine (5 mL), washed with water (2 ×
100 mL) and brine (100 mL), and then dried over MgO4, filtered,
and concentrated to provide the intermediate THI-bisketal (12.01
g, 81% yield) as a clear yellow plastic mass. 1H NMR (400 MHz,
chloroform-d) δ ppm 10.54-11.31 (m, 1 H), 7.24 (s, 1 H), 5.00
(d, J ) 7.5 Hz, 1 H), 4.08-4.30 (m, 3 H), 3.94-4.03 (m, 1 H),
2.59-2.66 (m, 3 H), 1.43-1.51 (m, 6 H), 1.26-1.42 (m, 6 H).
MS (EI) m/z: 311 [M + H]+. HRMS calcd for C22H15N2O5 [M +
H]+ 311.1607, found 311.1579.
1
to provide 10 (84 mg, 27% yield, 2 steps) as a white powder. H
NMR (300 MHz, 1.0 M DCl in D2O) δ ppm 4.83 (d, J ) 1.5 Hz,
1 H), 3.49-3.64 (m, 3 H), 3.43 (dd, J ) 11.3, 6.7 Hz, 1 H), 2.43
(s, 3 H), 2.19 (s, 3 H). MS (EI) m/z: 245 [M + H]+. HRMS calcd
for C10H16N2O5 [M + H]+ 245.1137, found 245.1131. HPLC (C30
Develosil RP Aqueous 4.6 mm × 50 mm, 10-95% [methanol/
water/TFA (95/5/0.1)]:water, 4 min gradient) tR ) 1.35 min, 97%
integrated area.
1-(5-Ethyl-4-((1R,2S,3R)-1,2,3,4-tetrahydroxybutyl)-1H-imida-
zol-2-yl)ethanone (11). General procedure C was applied to bro-
moimidazole 9 (1.26 g, 3.25 mmol) with vinylboronic acid pinacol
ester (1.00 g, 6.49 mmol, 2.00 equiv), Pd(dppf)Cl2-DCM (133 mg,
0.163 mmol), and cesium fluoride (1.14 g, 7.50 mmol). After
purification by flash chromatography (20-40% EtOAc:hexanes
eluent), the intermediate was dissolved in ethanol (10 mL), treated
with 10 wt % Pd-C (100 mg), and maintained under hydrogen (1
atm) for 1 h. At completion, the reaction was sparged with nitrogen,
filtered, and concentrated. Subsequent ketal deprotection by general
method C provided 11 (454 mg, 54% yield, 3 steps) as a white
solid. 1H NMR (400 MHz, 1.0 M DCl in D2O) δ ppm 5.09 (d, J )
2.7 Hz, 1 H), 3.57-3.72 (m, 2 H), 3.42-3.54 (m, 2 H), 2.63 (q, J
) 7.7 Hz, 2 H), 2.44-2.55 (m, 3 H), 1.02-1.14 (m, 3 H). MS
(EI) m/z: 259 [M + H]+. HRMS calcd for C11H18N2O5 [M + H]+
259.1294, found 259.1284. HPLC (Zorbax C8 4.6 mm × 150 mm,
5-20% [methanol/water/TFA (95/5/0.1)]:water with 0.1% TFA, 6
min gradient) tR ) 3.34 min, 99% integrated area.
1-(5-Phenyl-4-((1R,2S,3R)-1,2,3,4-tetrahydroxybutyl)-1H-imida-
zol-2-yl)ethanone (12). General procedure C was applied to bro-
moimidazole 9 (778 mg, 2.00 mmol) using phenylboronic acid (488
mg, 4.00 mmol), Pd(dppf)Cl2-DCM (82 mg, 0.10 mmol, 0.05
equiv), and cesium fluoride (603 mg, 4.00 mmol). After purification
by flash chromatography (10-30% EtOAc:hexane), ketal depro-
tection provided 12 (288 mg, 47% yield, 2 steps) as a white powder.
1H NMR (400 MHz, 1.0 M DCl in D2O) δ ppm 7.36-7.50 (m, 5
H), 5.16 (d, J ) 2.0 Hz, 1 H), 3.46-3.62 (m, 3 H), 3.37 (dd, J )
11.7, 5.9 Hz, 1 H), 2.56 (s, 3 H). MS (EI) m/z: 307 [M + H]+.
HRMS calcd for C15H18N2O5 [M + H]+ 307.1294, found 307.1282.
HPLC (Zorbax C8 4.6 mm × 150 mm, 10-90% MeCN:10 mM
aq NH4OAc, 3 min gradient) tR ) 2.42 min, 97% integrated area.
(1R,2S,3R)-1-(2-(1-Hydroxyethyl)-1H-imidazol-4-yl)butane-1,2,3,4-
tetraol (13). To a room temperature slurry of 3 (230 mg, 1.00 mmol)
in MeOH (10 mL) was added granular sodium borohydride (189
mg, 5.00 mmol). The reaction was maintained at room temperature
overnight. The reaction then was cooled to 0 °C, treated with 1 N
aqueous HCl solution (20 mL), and allowed to warm to room
temperature for 1 h before concentration. The resulting paste was
diluted with water (5 mL) and then adjusted to pH ∼9 with 1 N aq
To solution of the intermediate THI-bisketal (6.23 g, 20.00 mmol)
in DCM (100 mL) was added a 2.0 M solution of aqueous Na2CO3
(100 mL). The rapidly stirring biphasic mixture was cooled to 0
°C and then treated with bromine (3.36 g, 21.00 mmol) dropwise.
The cold bath was removed, and the mixture was maintained for
10 min. At completion, the layers were separated and the aqueous
phase was washed with DCM (100 mL). The combined organics
were washed with brine (100 mL), dried over MgO4, and then
filtered and concentrated. The crude material was purified by flash
chromatography over silica gel (5% MeOH:DCM eluent) to provide
1
bromide 9 (7.24 g, 93% yield) as a white solid. H NMR (400

3950 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Bagdanoff et al.
NaOH and allowed to stand overnight in a +4 °C refrigerator. The
resulting precipitant was filtered and washed with cold water (5
mL). The solid was treated with 1 N aqueous HCl solution (1 mL)
and water (2 mL), then lyophilized to provide the alcohol 13 (115
tography (20-40% EtOAc:hexane eluent) to give a oil, which was
dissolved in ethanol (8 mL). The solution was sparged with nitrogen
for 10 min before the addition of palladium hydroxide on carbon
(10 wt %, 100 mg). The reaction vessel was charged with H2 (65
psi) and stirred vigorously for 20 h. At completion, the reaction
was filtered, and the filtrate evaporated to provide an oil, which
was dissolved in acetone and cooled to 0 °C in an ice bath. The
solution was treated with 2:1 conc HCl:water (6 mL) and the
reaction was allowed to warm to room temperature. After 18 h,
the reaction was concentrated to provide 16 (23 mg, 7% yield, 2
1
mg, 50% yield) as a white solid. H NMR (400 MHz, 1.0 M DCl
in D2O) δ ppm 7.21 (s, 1 H), 5.07 (q, J ) 6.8 Hz, 1 H), 5.02 (s, 1
H), 3.48-3.74 (m, 4 H), 1.47 (s, 3 H). MS (EI) m/z: 233 [M +
H]+. HRMS calcd for C9H16N2O5 [M + H]+ 233.1137, found
233.1137. HPLC (Zorbax C8 4.6 mm × 150 mm, 5-40% methanol
with 0.1% TFA:water with 0.1% TFA, 5 min) tR ) 2.76 min, 100%
integrated area.
1
steps) as a white solid. H NMR (400 MHz, MeOD) δ ppm 7.54
(s, 1 H), 5.20 (s, 1 H), 3.75-3.86 (m, 2 H), 3.63-3.73 (m, 2 H),
3.16 (s, 6 H). 13C NMR (100 MHz, 1.0 M DCl in D2O) δ ppm
156.09, 136.47, 135.67, 117.52, 72.79, 70.77, 64.59, 62.98, 38.58,
36.09. MS (EI) m/z: 260 [M + H]+. HRMS calcd for C10H17N3O5
[M + H]+ 260.1246, found 260.1231. HPLC (C30 Develosil RP
Aqueous 4.6 mm × 50 mm, 5-80% [methanol/water/TFA (95/5/
0.1)]:water with 0.1% TFA, 4 min gradient) tR ) 0.35 min, 100%
integrated area.
(E)-N′-(1-(4-((1R,2S,3R)-1,2,3,4-Tetrahydroxybutyl)-1H-imida-
zol-2-yl)ethylidene)acetohydrazide (14). To a room temperature
suspension of 3 (160 mg, 0.70 mmol) in methanol (3 mL) and water
(1 mL) was added acetic hydrazide (56 mg, 0.75 mmol) and
concentrated aqueous hydrochloric acid (50 µL). The suspension
was stirred at 50 °C for 24 h, and the resulting solution was cooled
to room temperature and then diluted with tetrahydrofuran (10 mL).
A white precipitate was collected and washed with tetrahydrofuran
1
to yield the 14 (129 mg, 65% yield) as a white solid. H NMR
(1R,2S,3R)-1-(2-(1,1-Difluoroethyl)-1H-imidazol-4-yl)butane-
1,2,3,4-tetraol (17). THI-bisketal (see preparation of 9) (2.0 g, 6.45
mmol) was dissolved in DCM (20 mL) and cooled to 0 °C in an
ice bath. A solution of diethylaminosulfur trifluride (3.12 g, 19.35
mmol) in DCM (10 mL) was added in dropwise via syringe to the
cooled solution. Then the reaction was allowed to warm to room
temperature and maintained for 3 h and then cooled to 0 °C before
quenching by slow addition of sat. aq NaHCO3 (20 mL). The ice
bath was removed, and the reaction mixture was stirred at room
temperature for 10 min and then transferred to a separatory funnel.
The layers were separated, and the aqueous layer was washed with
DCM (10 mL). The combined organics were concentrated, and the
resulting crude material was purified by preparatory HPLC. The
resulting intermediate was lyophilized to provide a white solid,
which was dissolved in THF (10 mL) and treated with 1 N aq HCl
(10 mL). The reaction was then heated to 50 °C for 3 h. At
completion, the reaction was concentrated and then redissolved in
water (40 mL) and lyophilized to provide 17 (0.84 g, 53% yield, 2
steps) as a white powder. 1H NMR (400 MHz, MeOD) δ ppm 7.57
(s, 1 H), 5.19 (d, J ) 1.3 Hz, 1 H), 3.73-3.85 (m, 2 H), 3.61-3.73
(m, 2 H), 2.20 (t, J ) 13.2 Hz, 3 H). 19F NMR (376 MHz, MeOD)
δ ppm -91.65, -88.21. MS (EI) m/z: 253 [M + H]+. HRMS calcd
for C9H14F2N2O4 [M + H]+ 253.1000, found 253.0988. HPLC
(Sunfire C18 4.6 mm × 50 mm, 0% MeCN:10 mM aq NH4OAc,
4 min isocratic) tR ) 3.10 min, 98% integrated area.
(1R,2S,3R)-1-(2-(Perfluoroethyl)-1H-imidazol-4-yl)butane-1,2,3,4-
tetraol (20). To a solution of benzyl 2,2,3,3,3-pentafluoropro-
panoate34 (0.253 g, 1.00 mmol) in methanol (2.0 mL) was added
fructosamine-HOAc 18 (0.24 g, 1.00 mmol). The reaction was
maintained at room temperature overnight. The reaction was then
treated with a 25 wt % solution of sodium methoxide (0.06 g, 1.10
mmol) in MeOH. The reaction was stirred at room temperature for
4 h before the reaction was adjusted to pH ∼4 with acetic acid.
The reaction was diluted with water (3 mL) and purified directly
by HPLC. The product was lyophilized to provide 20 (6 mg, 2%)
as a white powder. 1H NMR (400 MHz, MeOD) δ ppm 7.16 (s, 1
H) 4.94 (d, J ) 1.8 Hz, 1 H) 3.71 (dd, J ) 11.1, 3.0 Hz, 1 H)
3.58-3.67 (m, 2 H) 3.54 (dd, J ) 10.9, 5.1 Hz, 1 H). MS (EI)
m/z: 307 [M + H]+. HRMS calcd for C9H11F5N2O4 [M + H]+
307.0717, found 307.0708. HPLC (Sunfire C18 4.6 mm × 50 mm,
5-50% MeCN:10 mM aq NH4OAc, 6 min gradient) tR ) 3.13
min, 100% integrated area.
(400 MHz, DMSO-d6) δ ppm 2.2 (s, 3H), 2.5 (s, 3H) 3.4-3.7 (m,
4H), 4.3 (m, 2H), 4.6-5.0 (m, 4H), 7.0 (m, 1H), 10.30 (s, 1H),
10.37 (s, 1H). 13C NMR (100 MHz, DMSO-d6) δ ppm 172.78,
165.91, 144.49, 143.77, 140.50, 74.00, 73.89, 71.29, 63.45, 20.71,
12.50. MS (EI) m/z: 287 [M + H]+. HRMS calcd for C11H18N4O5
[M + H]+ 287.1355, found 287.1360. HPLC (C30 Develosil RP
Aqueous 4.6 mm × 50 mm, 6-40% [methanol/water/TFA (95/5/
0.1)]:water with 0.1% TFA, 5 min gradient) tR ) 0.23 min, 100%
integrated area.
1-Benzyl-4-((4S,4′R,5R)-2,2,2′,2′-tetramethyl-4,4′-bi(1,3-dioxolan)-
5-yl)-1H-imidazole-2-carboxylic Acid (15). To a room temperature
solution of THI-bisketal (see preparation of 9) (20.0 g, 64.5 mmol)
in DMF (200 mL) was added K2CO3 (12.5 g, 90.3 mmol) and
benzyl bromide (10.7 mL, 90.3 mmol). The reaction was heated at
50 °C for 18 h. Then, an additional portion of benzyl bromide (5
mL, 42.0 mmol) was added and the temperature was increased to
60 °C. After 3 h, the reaction was quenched with water (200 mL),
cooled to 0 °C, and extracted with EtOAc (3 × 200 mL). The
organic extracts were washed with water (200 mL) and then brine
(100 mL) and dried over sodium sulfate. After concentration, flash
chromatography (20-40% EtOAc:hexane eluent) provided the fully
protected intermediate (16.10 g, 62% yield) as an plastic mass that
was used without further purification. 1H NMR (400 MHz, chloroform-
d) δ ppm 7.25-7.38 (m, 4 H), 7.18 (dd, J ) 7.6, 1.8 Hz, 2 H), 5.58
(d, J ) 5.8 Hz, 2 H), 4.92 (d, J ) 7.8 Hz, 1 H), 4.35 (dd, J ) 7.6, 6.8
Hz, 1 H), 4.25 (dd, J ) 17.7, 6.3 Hz, 1 H), 4.10 (dd, J ) 8.6, 6.3 Hz,
1 H), 4.01 (dd, J ) 8.6, 4.8 Hz, 1 H), 2.65 (s, 3 H), 1.50 (d, J ) 16.7
Hz, 6 H), 1.27 (m, 6 H). 13C NMR (100 MHz, chloroform-d) δ ppm
142.64, 140.44, 136.48, 129.01, 128.29, 127.88, 127.10, 124.52,
110.16, 109.71, 80.69, 75.52, 66.88, 52.00, 27.58, 27.22, 27.12, 26.54,
25.42.
A portion of the product obtained above (13.00 g, 32.50 mmol)
was dissolved in dioxane (120 mL) and treated with NaOH (13.20
g, 33.00 mmol) dissolved in commercial bleach (200 mL, 6%
NaOCl). After 2 h of vigorous stirring, the layers were separated
and the aqueous layer was washed with EtOAc (3 × 150 mL).
The combined organic extracts were washed with brine (2 × 100
mL) and then dried over celite. Filtration and evaporation afforded
a solid that was further dried in vacuo to afford the carboxylic acid
1
15 (13.0 g, 100% yield) as a white solid. H NMR (400 MHz,
MeOD) δ ppm 7.10-7.45 (m, 1 H) 5.05 (br s, 1 H) 3.72-3.93 (m,
2 H) 3.66 (dd, J ) 10.7, 4.9 Hz, 2 H) 3.06 (br s, 1 H) 1.11-1.17
(m, 2 H) 1.05-1.11 (m, 2 H). MS (EI) m/z: 403 [M + H]+.
N,N-Dimethyl-4-((1R,2S,3R)-1,2,3,4-tetrahydroxybutyl)-1H-imi-
dazole-2-carboxamide (16). To a solution of acid 15 (500 mg, 1.24
mmol) in DCM (4 mL) was added HATU (520 mg, 1.37 mmol),
dimethylamine hydrochloride (202 mg, 2.48 mmol), and triethyla-
mine (517 µL, 3.72 mmol). After 20 h, the reaction was treated
with an additional portion of HATU (520 mg, 1.37 mmol) and
dimethylamine hydrochloride (202 mg, 2.48 mmol). After another
18 h, the reaction was concentrated and purified by flash chroma-
Phenyl(4-((1R,2S,3R)-1,2,3,4-tetrahydroxybutyl)-1H-imidazol-2-
yl)methanone (21). Prepared according to the general procedure B
from fructosamine-HOAc 18 (0.782 g, 3.27 mmol) and 2,2-difluoro-
2-phenylacetonitrile (500 mg, 3.27 mmol) to provide the spontane-
ous hydrolysis product, ketone 21 (350 mg, 34% yield) as a yellow
solid. 1H NMR (400 MHz, 1.0 M DCl in D2O) δ ppm 8.03 (d, J )
7.3 Hz, 1 H) 7.68 (t, J ) 7.5 Hz, 1 H) 7.54 (t, J ) 7.8 Hz, 1 H)
7.43 (s, 1 H) 5.02 (d, J ) 3.3 Hz, 1 H) 3.82 (td, J ) 7.8, 3.0 Hz,
1 H) 3.70-3.78 (m, 1 H) 3.60 (dd, J ) 11.5, 6.3 Hz, 1 H). 13C
NMR (100 MHz, 1.0 M DCl in D2O) δ ppm 180.45, 139.24, 138.53,

Inhibition of Sphingosine-1-Phosphate Lyase
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3951
135.84, 134.08, 130.18, 129.67, 119.88, 73.11, 70.99, 65.16, 63.22.
MS (EI) m/z: 293 [M + H]+. HRMS calcd for C14H16N2O5 [M +
H]+ 293.1137, found 293.1162. HPLC (Sunfire C18 4.6 mm × 50
General Procedure D: (1R,2S,3R)-1-(2-(1H-1,2,4-Triazol-3-yl)-
1H-imidazol-4-yl)butane-1,2,3,4-tetraol (27). A solution of 26 (275
mg, 0.80 mmol) in methanol (10 mL) and 0.5 M aqueous HCl
(14 mL) was sparged with nitrogen for 10 min. Then, palladium
hydroxide on carbon (20% w/w, 100 mg) was added, and the
reaction vessel was charged with H2 (65 psi). After 18 h, the
reaction was sparged with nitrogen and then filtered over a pad
of celite and washed with methanol (10 mL). The solvent was
removed to give an off white solid which was then triturated
with ether:THF (1:1) to give the deprotected triazole 27 (180
mg, 100% yield) as an off white solid. 1H NMR (400 MHz,
MeOD) δ ppm 8.72 (s, 1 H), 7.54 (s, 1 H), 5.24 (d, J ) 1.3 Hz,
1 H), 3.78-3.88 (m, 2 H), 3.66-3.76 (m, 2 H). 13C NMR (100
MHz, 1.0 M DCl in D2O) δ ppm 148.04, 145.55, 135.49, 135.09,
117.08, 72.39, 70.40, 64.25, 62.57. MS (EI) m/z: 256 [M + H]+.
HRMS calcd for C16H20N5O4 [M + H]+ 256.1048, found
256.1033. HPLC (Zorbax C8 4.6 mm × 150 mm, 5-90%
MeCN:10 mM aq NH4OAc, 4 min gradient) tR ) 0.80 min, 99%
integrated area.
mm, 5-90% MeCN:10 mM aq NH4OAc, 3 min gradient) tR
)
0.80 min, 100% integrated area.
(1R,2S,3R)-1-(2-(2-Methylthiazol-4-yl)-1H-imidazol-4-yl)butane-
1,2,3,4-tetraol (22). Prepared according to the general procedure B
from fructosamine-HOAc 18 (1.00 g, 4.13 mmol) and 2-methylthi-
azole-4-carbonitrile (0.50 g, 4.13 mmol) to provide 22 (0.68 g, 58%
1
yield). H NMR (400 MHz, MeOD) δ ppm 7.40 (s, 1 H) 7.32 (d,
J ) 3.3 Hz, 1 H) 6.39 (dd, J ) 3.6, 0.9 Hz, 1 H) 3.76-3.89 (m,
2 H) 3.67-3.75 (m, 2 H) 2.46 (s, 3 H). 13C NMR (100 MHz, 1.0
M DCl in D2O) δ ppm 171.51, 138.56, 136.39, 134.69, 116.74,
108.40, 72.85, 70.85, 64.68, 62.98, 18.16. MS (EI) m/z: 286 [M +
H]+. HRMS calcd for C11H15N3O4S [M + H]+ 286.0862, found
286.0842. HPLC (Zorbax C8 4.6 mm × 150 mm, 5-90% MeCN:
10 mM aq NH4OAc, 3 min gradient) tR ) 1.76 min, 99% integrated
area.
(1R,2S,3R)-1-(2-(5-methylthiazol-4-yl)-1H-imidazol-4-yl)butane-
1,2,3,4-tetraol (23). Prepared according to the General procedure
B from fructosamine-HOAc 18 (2.00 g, 8.25 mmol) and 5-meth-
ylthiazole-4-carbonitrile35 (1.02 g, 8.25 mmol) to provide 23 (1.70
(1R,2S,3R)-1-(1H,1′H-2,2′-biimidazol-4-yl)butane-1,2,3,4-tetraol
(28). Prepared according to the general procedure B from
fructosamine-HOAc 18 (1.94 g, 8.37 mmol) and 1-benzyl-1H-
imidazole-2-carbonitrile (0.78 g, 8.37 mmol) to provide, after
hydrogenation by general procedure D, compound 28 (60 mg,
1
g, 72% yield). H NMR (400 MHz, MeOD) δ ppm 9.01 (s, 1 H)
1
7.55 (s, 1 H) 5.24 (d, J ) 1.3 Hz, 1 H) 3.78-3.89 (m, 2 H)
3.67-3.77 (m, 2 H) 2.77 (s, 3 H). 13C NMR (100 MHz, 1.0 M
DCl in D2O) δ ppm 155.22, 140.38, 137.84, 134.89, 133.71, 117.14,
72.84, 70.84, 64.63, 62.98, 11.52. MS (EI) m/z: 286 [M + H]+.
HRMS calcd for C11H15N3O4S [M + H]+ 286.0862, found
286.0864. HPLC (Luna Phenyl-hexyl 4.6 mm × 50 mm, 5-90%
MeCN:10 mM aq NH4OAc, 3 min gradient) tR ) 0.93 min, 100%
integrated area.
3% yield) as a white solid. H NMR (400 MHz, MeOD) δ ppm
7.14 (d, J ) 4.3 Hz, 3 H), 5.02 (s, 1 H), 3.73-3.89 (m, 3 H),
3.66 (br s, 1 H). 13C NMR (100 MHz, 1.0 M DCl in D2O) δ
ppm 138.20, 127.82, 122.98, 119.80, 72.87, 70.78, 64.77, 63.02.
MS (EI) m/z: 255 [M + H]+. HRMS calcd for C10H15N4O4 [M
+ H]+ 255.1092, found 255.1085. HPLC (HILIC Silica 4.6 mm
× 50 mm, 100-80% MeCN:10 mM aq NH4OAc, 3 min gradient)
tR ) 2.23 min, 99% integrated area.
(1R,2S,3R)-1-(2-(pyrimidin-2-yl)-1H-imidazol-4-yl)butane-1,2,3,4-
tetraol (29). Prepared according to the general procedure B from
fructosamine-HOAc 18 (2.39 g, 10.00 mmol) and pyrimidine-
2-carbonitrile (1.05 g, 10.00 mmol) to provide 29 (0.68 g, 26%
(1R,2S,3R)-1-(2-(Thiazol-4-yl)-1H-imidazol-4-yl)butane-1,2,3,4-
tetraol (24). Prepared according to the general procedure B from
fructosamine-HOAc 18 (0.93 g, 3.88 mmol) and 5-methylthi-
azole-4-carbonitrile (0.43 g, 3.88 mmol) to provide 24 (95 mg,
1
1
yield). H NMR (400 MHz, DMSO-d6) δ ppm 8.82 (d, J ) 4.8
9% yield). H NMR (400 MHz, 1.0 M DCl in D2O) δ ppm 9.01
Hz, 2 H), 7.39 (t, J ) 4.9 Hz, 1 H), 7.14 (s, 1 H), 4.80 (d, J )
1.5 Hz, 1 H), 3.61 (dd, J ) 10.7, 3.1 Hz, 2 H), 3.40 (dd, J )
10.8, 5.8 Hz, 2 H). 13C NMR (100 MHz, 1.0 M DCl in D2O) δ
ppm 158.25, 151.04, 140.91, 136.18, 122.70, 118.04, 72.43,
70.45, 64.41, 62.62. MS (EI) m/z: 267 [M + H]+. HRMS calcd
for C11H14N4O4 [M + H]+ 267.1093, found 267.1079. HPLC
(Zorbax C8 4.6 mm × 150 mm, 5% MeCN:10 mM aq NH4OAc,
4 min isocratic) tR ) 1.01 min, 100% integrated area.
(1R,2S,3R)-1-(2-(Pyrazin-2-yl)-1H-imidazol-4-yl)butane-1,2,3,4-
tetraol (30). Prepared according to the general procedure B from
fructosamine-HOAc 18 (1.62 g, 7.00 mmol) and pyrazine-2-
carbonitrile (0.74 g, 7.00 mmol) to provide 30 (0.72 g, 39% yield).
1H NMR (300 MHz, deuterium oxide) δ ppm 8.96 (d, J ) 1.5 Hz,
1 H), 8.52 (dd, J ) 2.7, 1.5 Hz, 1 H), 8.42 (d, J ) 2.7 Hz, 1 H),
7.22 (br s, 1 H), 4.91 (br s, 1 H), 3.78 (dd, J ) 7.1, 3.6 Hz, 1 H),
3.59-3.75 (m, 2 H), 3.47-3.58 (m, 1 H). 13C NMR (100 MHz,
1.0 M DCl in D2O) δ ppm 146.28, 145.98, 141.42, 140.38, 138.82,
136.66, 118.52, 72.84, 70.85, 64.82, 63.02. MS (EI) m/z: 267 [M
+ H]+. HRMS calcd for C11H14N4O4 [M + H]+ 267.1093, found
267.1091. HPLC (HILIC Silica 4.6 mm × 50 mm, 100-40%
MeCN:10 mM aq NH4OAc, 3 min gradient) tR ) 1.47 min, 99%
integrated area.
(d, J ) 1.8 Hz, 1 H), 7.94 (d, J ) 1.8 Hz, 1 H), 7.19 (s, 1 H),
4.95 (d, J ) 3.8 Hz, 1 H), 3.84 (dd, J ) 7.0, 3.8 Hz, 1 H),
3.67-3.80 (m, 2 H), 3.59 (dd, J ) 11.5, 6.5 Hz, 1 H). 13C NMR
(100 MHz, 1.0 M DCl in D2O) δ ppm 157.72, 139.06, 138.58,
134.67, 123.70, 116.73, 72.85, 70.85, 64.69, 62.98. MS (EI) m/z:
272 [M + H]+. HRMS calcd for C10H13N3O4S [M + H]+
271.0627, found 272.0693. HPLC (HILIC Silica 4.6 mm × 50
mm, 97-40% MeCN:10 mM aq NH4OAc, 3 min gradient) tR )
1.21, 100% integrated area.
(1R,2S,3R)-1-(2-(Oxazol-4-yl)-1H-imidazol-4-yl)butane-1,2,3,4-
tetraol (25). Prepared according to the general procedure B from
fructosamine-HOAc 18 (510 mg, 2.13 mmol) and oxazole-4-
carbonitrile (200 mg, 2.13 mmol) to provide 29 (64 mg, 12% yield).
1H NMR (400 MHz, MeOD) δ ppm 8.29 (s, 1 H), 8.22 (s, 1 H),
7.12 (br s, 1 H), 5.01 (br s, 1 H), 3.71-3.88 (m, 3 H), 3.56-3.71
(m, 1 H). 13C NMR (100 MHz, 1.0 M DCl in D2O) δ ppm 154.05,
140.93, 136.68, 134.86, 125.27, 116.84, 72.85, 70.83, 64.66, 62.98.
MS (EI) m/z: 256 [M + H]+. HRMS calcd for C10H13N3O5 [M +
H]+ 256.0933, found 256.0936. HPLC (Sunfire C18 4.6 mm × 50
mm, 10-90% MeCN:10 mM aq NH4OAc, 2 min gradient) tR
0.30 min, 100% integrated area.
)
(1R,2S,3R)-1-(2-(1-Benzyl-1H-1,2,4-triazol-3-yl)-1H-imidazol-4-
yl)butane-1,2,3,4-tetraol (26). Prepared according to the general
procedure B from fructosamine-HOAc (2.73 g, 11.40 mmol) and
1-benzyl-1H-1,2,4-triazole-3-carbonitrile36 (2.1 g, 11.40 mmol)
(1R,2S,3R)-1-(2-(1,2,4-Oxadiazol-3-yl)-1H-imidazol-4-yl)butane-
1,2,3,4-tetraol (31). Prepared according to the general procedure B
from fructosamine-HOAc 18 (478 mg, 2.00 mmol) and 1,2,4-
oxadiazole-3-carbonitrile (512 mg, 2.00 mmol) to provide 31 (205
1
1
to provide 26 (3.21 g, 81% yield). H NMR (400 MHz, MeOD)
mg, 40% yield). H NMR (400 MHz, 1.0 M DCl in D2O) δ ppm
δ ppm 8.80 (s, 1 H), 7.51 (s, 1 H), 7.33-7.45 (m, 5 H), 5.58 (s,
2 H), 5.21 (d, J ) 1.3 Hz, 1 H), 3.76-3.86 (m, 2 H), 3.66-3.73
(m, 2 H). MS (EI) m/z: 346 [M + H]+. HRMS calcd for
C16H20N5O4 [M + H]+ 346.1515, found 346.1501. HPLC (Sunfire
C18 4.6 mm × 50 mm, 5-90% [methanol/water/TFA (95/5/
0.1)]:water with 0.1% TFA, 3 min gradient) tR ) 1.89 min, 99%
integrated area.
9.43 (s, 1 H), 7.64 (s, 1 H), 5.24 (s, 1 H), 3.72-3.84 (m, 3 H),
3.61-3.68 (m, J ) 13.1, 6.8 Hz, 1 H).13C NMR (100 MHz, 1.0 M
DCl in D2O) δ ppm 169.3, 157.4, 138.6, 132.8, 119.9, 73.8, 71.6,
65.8, 63.8. MS (EI) m/z: 257 [M + H]+. HRMS calcd for
C9H12N4O5 [M + H]+ 257.0886, found 257.0870. HPLC (Luna
Phenyl-hexyl 4.6 mm × 50 mm, 5% [methanol/water/TFA (95/5/
0.1)]:water, isocratic) tR ) 2.41 min, 100% integrated area.

3952 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13
Bagdanoff et al.
(1R,2S,3R)-1-(1′-Methyl-1H,1′H-2,4′-biimidazol-4-yl)butane-
1,2,3,4-tetraol (32). Prepared according to the general procedure
B from fructosamine-HOAc 18 (2.00 g, 8.25 mmol) and
1-methyl-1H-imidazole-4-carbonitrile (0.88 g, 8.25 mmol) to
Supporting Information Available: HPLC conditions for
determining purity of all final products, select elemental analyses,
and characterization data of final products. This material is available
free of charge via the Internet at http://pubs.acs.org.
1
provide 32 (0.90 g, 41% yield). H NMR (400 MHz, MeOD) δ
ppm 9.22 (s, 1 H), 8.33 (d, J ) 1.5 Hz, 1 H), 7.65 (s, 1 H), 5.26
(d, J ) 1.3 Hz, 1 H), 4.10 (s, 3 H), 3.78-3.88 (m, 2 H),
3.68-3.77 (m, 2 H). 13C NMR (100 MHz, 1.0 M DCl in D2O)
δ ppm 138.59, 136.30, 132.08, 125.65, 118.05, 117.42, 72.93,
70.83, 64.67, 63.06, 36.77. MS (EI) m/z: 269 [M + H]+. HRMS
calcd for C11H16N4O4 [M + H]+ 269.1250, found 261.1242.
HPLC (Zorbax C8 4.6 mm × 150 mm, 5-90% [methanol/water/
TFA (95/5/0.1)]:water, 3 min gradient) tR ) 0.32 min, 100%
integrated area.
References
(1) (a) Kitatani, K.; Idkowiak-Baldys, J.; Hannun, Y. A. The sphingolipid
salvage pathway in ceramide metabolism and signaling. Cell. Signal-
ling 2008, 20, 1010–1018. (b) Gomez-Munoz, A.; Steinbrecher, U. P.
Control of cell signaling by ceramides, sphingosine-1-phosphate and
ceramide-1-phosphate. Recent Res. DeV. Lipids 2004, 7, 65–86. (c)
Smith, W. L.; Merrill, A. H. Sphingolipid metabolism and signaling.
J. Biol. Chem. 2002, 277, 25841–25842. (d) Lahiri, S.; Futerman, A. H.
The metabolism and function of sphingolipids and glycosphingolipids.
Cell. Mol. Life Sci. 2007, 64, 2270–2284.
1-(4-((1R,2R,3R)-1,2,3,4-Tetrahydroxybutyl)-1H-imidazol-2-yl)-
ethanone (33). The Amadori and Bu¨chi rearrangements detailed in
general procedure A and general procedure B, respectively, were
applied to D-galactose (5.00 g, 27.78 mmol) to generate 33 (60
mg, 1% yield, 2 steps). 1H NMR (400 MHz, 1.0 M DCl in D2O) δ
ppm 7.24 (s, 1 H), 4.67 (br s, 1 H), 3.80-3.88 (m, 1 H), 3.76 (dd,
J ) 8.1, 2.5 Hz, 1 H), 3.49-3.60 (m, 2 H), 2.44 (m, 2 H), 2.44 (s,
3 H). MS (EI) m/z: 231 [M + H]+. HRMS calcd for C9H15N2O5
[M + H]+ 231.0981, found 231.0968. HPLC (Sunfire C18 4.6 mm
× 50 mm, 10-90% MeCN:10 mM aq NH4OAc, 2 min gradient)
tR ) 0.22 min, 100% integrated area.
1-(4-((1S,2R,3S)-1,2,3,4-Tetrahydroxybutyl)-1H-imidazol-2-yl)-
ethanone (34). The Amadori and Bu¨chi rearrangements detailed in
general procedure A and general procedure B, respectively, were
applied to L-(-)-glucose (5.00 g, 27.78 mmol) to generate 34 (2.76
g, 44% yield, 2 steps). 1H NMR (300 MHz, 1.0 M DCl in D2O) δ
ppm 7.25 (s, 1 H), 4.86 (d, J ) 1.7 Hz, 1 H), 3.32-3.48 (m, 3 H),
3.23-3.32 (m, 1 H), 2.31 (s, 3 H). MS (EI) m/z: 231 [M + H]+.
HRMS calcd for C9H15N2O5 [M + H]+ 239.0981, found 231.0970.
HPLC (C30 Develosil RP Aqueous 4.6 mm × 50 mm, 2-40%
[methanol/water/TFA (95/5/0.1)]:water with 0.1% TFA, 5 min
gradient) tR ) 2.14 min, 100% integrated area.
(2) (a) Rivera, J.; Proia, R. L.; Olivera, A. The alliance of sphingosine-
1-phosphate and its receptors in immunity. Nat. ReV. 2008, 8, 753–
763. (b) Gomez-Munoz, A.; Steinbrecher, U. P. Control of cell
signaling by ceramides, sphingosine-1-phosphate and ceramide-1-
phosphate. Recent Res. DeV. Lipids 2004, 7, 65–86.
(3) (a) Reiss, U.; Oskouian, B.; Zhou, J.; Gupta, V.; Sooriyakumaran, P.;
Kelly, S.; Wang, E.; Merrill, A. H.; Saba, J. J. Sphingosine-phosphate
lyase enhances stress-induced ceramide generation apoptosis. J. Biol.
Chem. 2004, 279, 1281–1290. (b) Moore, A. N.; Kampfl, A. W.; Zhao,
X.; Hayes, R. L.; Dask, P. K. Sphingosine 1-phosphate induces
apoptosis of cultured hippocampal neurons that requires protein
phosphatases and activator protein-1 complexes. Neuroscience 1999,
94, 405–415.
(4) (a) Hait, N. C.; Oskeritzian, C. A.; Paugh, S. W.; Milstein, S.; Spiegel,
S. Sphingosine kinases, sphingosine 1-phosphate, apoptosis and
disease. Biochim. Biophys. Acta 2006, 1758, 2016–2026. (b) Saba,
J. D.; Hla, T. Point-counterpoint of sphingosine 1-phosphate metabo-
lism. Circ. Res. 2004, 94, 724–734. (c) Pyne, S.; Pyne, N. J.
Sphingosine 1-phosphate signaling in mammalian cells. Biochem. J.
2000, 349, 385–402.
(5) (a) Wang, W.; Huang, M.-C.; Goetzl, E. J. Type 1 sphingosine
1-phosphate G protein-coupled receptor (S1P1) mediation of enhanced
IL-4 generation by CD4 T cells from S1P1 transgenic mice. J. Immun.
2007, 178, 4885–4890. (b) Chi, H.; Flavell, R. A. Cutting edge:
regulation of T cell trafficking and primary immune responses by
sphingosine 1-phosphate receptor 1. J. Immun. 2005, 174, 2485–2488.
(c) Goetzl, E. J.; Liao, J.-J.; Huang, M.-C. Regulation of the roles of
sphingosine 1-phosphate and its type 1 G protein-coupled receptor in
T cell immunity and autoimmunity. Mol. Cell Biol. Lipids 2008, 1781,
503–507.
(6) For a general review, see: (a) Goetzl, E. J.; Wang, W.; McGiffert, C.;
Huang, M.-C.; Graeler, M. H. Sphingosine 1-phosphate and its G
protein-coupled receptors constitute a multifunctional immunoregu-
latory system. J. Cell. Biochem. 2004, 92, 1104–1114.
(7) Yoshikawa, M.; Yokokawa, Y.; Okuno, Y.; Yagi, N.; Murakami, N.
Synthesis, Immunosuppressive Activity, and Structure-Activity Re-
lationships of Myriocin Analogs, 2-epi-Myriocin, 14-Deoxymyriocin,
Z-14-Deoxymyriocin, and Nor-deoxymyriocin. Chem. Pharm. Bull.
1995, 43, 1647–1653.
(8) (a) Miyake, Y.; Kozutsumi, Y.; Nakamura, S.; Fujita, T.; Kawasaki,
T. Serine palmitoyltransferase is the primary target of a sphingosine-
like immunosuppressant, ISP-1/ myriocin. Biochem. Biophys. Res.
Commun. 1995, 211, 396–403. (b) Chen, J. K.; Lane, W. S.; Schreiber,
S. L. The Identification of Myriocin-Binding proteins. Chem. Biol.
1999, 6, 221–235.
(9) (a) Yoshikawa, M.; Yokokawa, Y.; Okuno, Y.; Yagi, N.; Murakami,
N. Syntheses, immunosuppressive activity, and structure-activity
relationships of myriocin analogs, 2-epi-myriocin 14-deoxomyriocin,
Z-14-deoxomyriocin, and nor-deoxomyriocins. Chem. Pharm. Bull.
1995, 43, 1647–1653. (b) Fujita, T.; Hirose, R.; Hamamichi, N.; Kitao,
Y.; Sasaki, S.; Yoneta, M.; Chiba, K. 2-Substituted 2-aminoethanol:
minimum essential structure for immunosuppressive activity of ISP-I
(myriocin). Bioorg. Med. Chem. Lett. 1995, 5, 1857–1860.
(10) (a) Rivera, J.; Proia, R. L.; Olivera, A. The alliance of sphingosine-
1-phosphate and its receptors in immunity. Nat. ReV. Immunol. 2008,
8, 753–763. (b) Billich, A.; Baumruker, T. Sphingolipid Metabolizing
Enzymes as Novel Theraputic Targets. In Lipids in Health and Disease;
Quinn, P. J.; Wang, X., Eds.; Springer Science Business Media: New
York, 2008, 487-522.
1-(4-((1R,2R,3R)-2,3,4-Trihydroxy-1-methoxybutyl)-1H-imida-
zol-2-yl)ethanone (36). The Amadori and Bu¨chi rearrangements
detailed in general procedure A and general procedure B, respec-
tively, were applied to D-(+)-glucose-3-hydroxmethyl ether 35 (1.90
g, 9.60 mmol) to generate 36 (22 mg, 1.0% yield) after HPLC
1
purification. H NMR (400 MHz, MeOD) δ ppm 7.30 (s, 1 H),
4.65 (d, J ) 3.0 Hz, 1 H), 3.79 (dd, J ) 11.0, 3.0 Hz, 1 H),
3.70-3.76 (m, 1 H), 3.59-3.68 (m, 2 H), 3.35 (d, J ) 8.0 Hz, 3
H), 1.92 (s, 3 H). MS (EI) m/z: 245 [M + H]+. HRMS calcd for
C10H17N2O5 [M + H]+ 245.1137, found 245.1115. HPLC (Sunfire
C18 4.6 mm × 50 mm, 6-30% [methanol/water/TFA (95/5/0.1)]:
water with 0.1% TFA, 5 min gradient) tR ) 1.07 min, 100%
integrated area.
(2S,3S,4R)-4-(2-Acetyl-1H-imidazol-4-yl)-2,3,4-trihydroxy-N,N-
dimethylbutanamide (38). The Amadori and Bu¨chi rearrangements
detailed in general procedure A and general procedure B, respec-
tively, were applied to D-(+)-glucose-6-dimethyamide 37 (2.12 g,
9.60 mmol) to generate 38 (60 mg, 2.0% yield) after purification
by preparatory HPLC. 1H NMR (400 MHz, 1.0 M DCl in D2O) δ
ppm 7.31 (s, 1 H), 5.00 (d, J ) 3.3 Hz, 1 H), 4.75 (d, J ) 7.8 Hz,
1 H), 3.95 (dd, J ) 7.8, 3.3 Hz, 1 H), 3.07 (s, 3 H), 2.84 (s, 3 H),
2.53 (s, 3 H). MS (EI) m/z: 272 [M + H]+. HRMS calcd for
C11H17N3O5 [M + H]+ 272.1246, found 272.1200. HPLC (Sunfire
C18 4.6 mm × 50 mm, 10-90% [methanol/water/TFA (95/5/0.1)]:
water, 4 min gradient) tR ) 0.37 min, 96% integrated area.
Acknowledgment. We thank the analytical group at Lexicon
Pharmaceuticals in Princeton, NJ, for their assistance in the
characterization of compounds and the animal handling and
vivarium operation groups in The Woodlands, TX for their
maintenance of animal resources.
(11) (a) Igarashi, Y. Functional roles of sphingosine, sphingosine 1-phos-
phate, and methylsphingosines: in regard to membrane sphingolipid
signaling pathways. J. Biochem. 1997, 122, 1080–1087. (b) Elrick,
M. J.; Fluss, S.; Colombini, M. Sphingosine, a product of ceramide
hydrolysis, influences the formation of ceramide channels. Biophys.
J. 2006, 91, 1749–1756.
Note Added after ASAP Publication. This manuscript was
released ASAP on May 21, 2009, with an incomplete author listing.
The correct version was posted on July 2, 2009.
(12) SK I is believed to play the more dominant role. See: (a) Kharel, Y.;
Lee, S.; Snyder, A. H.; Sheasley-O’Neill, S. L.; Morris, M. A.; Setiady,

Inhibition of Sphingosine-1-Phosphate Lyase
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 13 3953
Y.; Zhu, R.; Zigler, M. A.; Burcin, T. L.; Ley, K.; Tung, K. S. K.;
Engelhard, V. H.; Macdonald, T. L.; Pearson-White, S.; Lynch, K. R.
Sphingosine Kinase 2 Is Required for Modulation of Lymphocyte
Traffic by FTY720. J. Biol. Chem. 2005, 280, 36865–36872. (b)
Baumruker, T.; Bornancin, F.; Billich, A. The role of sphingosine and
ceramide kinases in inflammatory responses. Immunol. Lett. 2005, 96,
175–185.
(22) Sanna, M. G.; Liao, Jiayu; Jo, E.; Alfonso, C.; Ahn, M.-Y.; Peterson,
M. S.; Webb, B.; Lefebvre, S.; Chun, J.; Gray, N.; Rosen, H.
Sphingosine 1-phosphate (S1P) receptor subtypes S1P1 and S1P3,
respectively, regulate lymphocyte recirculation and heart rate. J. Biol.
Chem. 2004, 279, 13839–13848.
(23) (a) Pyne, S. G.; Ung, A. T. Diastereoselective Synthesis of (1S,2S,3R)-
and (1R,2R,3R)-2-Acetyl-5-(1,2,3,4-tetrahydroxybutyl)thiazole. Synlett
1998, 3, 280–282. (b) Ung, A. T.; Pyne, S. G.; Skelton, B. W.; White,
A. H. Asymmetric synthesis of (1R,2S,3R)-2-acetyl-5-(1,2,3,4-tetrahy-
droxybutyl)thiazole. Tetrahedron 1996, 52, 14069–14078.
(24) (a) Hodge, John, E.; Rist, Carl, E. Amadori rearrangement under new
conditions and its significance for nonenzymatic browning reactions.
J. Am. Chem. Soc. 1953, 75, 316–322. (b) Hodge, J. E.; Fisher, B. E.
Amadori Rearrangement Products. In Methods in Carbohydrate
Chemistry, Volume II: Reactions of Carbohydrates Whistler, R. L.,
Wolfrom, M. L., Bemiller, J. N., Eds.; Academic Press: New York,
1963; Vol. II, pp 99-103.
(25) Buchi, G.; Halweg, K. M. A Convenient Synthesis of 2-Acetyl-4(5)-
(1(R),2(S),3(R),4-tetrahydroxybutyl)-imidazole. J. Am. Chem. Soc.
1984, 50, 1134–1136.
(26) Murata, N.; Sato, K.; Kon, J.; Tomura, H.; Okajima, F. Quantitative
Measurement of Sphingosine-1-Phosphate by Radioreceptor-Binding
Assay. Anal. Biochem. 2000, 282, 115–120.
(27) Mukhopadhyay, D.; Howell, K. S.; Riezman, H.; Capitani, G. J. Biol.
Chem. 2008, 283, 20159–20169.
(28) (a) Drews, J. Drug Discovery: A Historical Perspective. Science 2000,
287, 1960–1964. (b) Sams-Dodd, F. Target-based drug discovery: is
something wrong? Drug DiscoVery Today 2005, 10, 139–147. (c)
Sams-Dodd, F. Drug discovery: selecting the optimal approach. Drug
DiscoVery Today 2006, 11, 465–472.
(13) (a) Johnson, K. R.; Johnson, K. Y.; Becker, K. P.; Bielawski, J.; Mao,
C.; Obeid, L. M. Role of human sphingosine-1-phosphate phosphatase
1 in the regulation of intra- and extracellular sphingosine-1-phosphate
levels and cell viability. J. Biol. Chem. 2003, 278, 34541–34547. (b)
Mechtcheriakova, D.; Wlachos, A.; Sobanov, J.; Kopp, T.; Reuschel,
R.; Bornancin, F.; Cai, R.; Zemann, B.; Urtz, N.; Stingl, G.; Zlabinger,
G.; Woisetschlaeger, M.; Baumruker, T.; Billich, A. Sphingosine
1-phosphate phosphatase 2 is induced during inflammatory responses.
Cell. Signalling 2007, 19, 748–760.
(14) (a) Bandhuvula, P.; Fyrst, H.; Saba, J. D. A rapid fluorescence assay
for sphingosine-1-phosphate lyase enzyme activity. J. Lipid Res. 2007,
48, 2769–2778. (b) Bandhuvula, P.; Saba, J. D. Sphingosine-1-
phosphate lyase in immunity and cancer: silencing the siren. Trends
Mol. Med. 2007, 13, 210–217.
(15) Vogel, P.; Donoviel, M. S.; Read, R.; Hansen, G. M.; Hazlewood, J.;
Anderson, S.; Sun, W.; Swaffield, J.; Oravecz, T. Incomplete Inhibition
of Sphingosine 1-Phosphate Lyase Modulates Immune System Func-
tion yet Prevents Early Lethality and Non-Lymphoid Lesions. PLoS
One 2008, 4, 4112.
(16) It has been established that FTY720 is phosphorylated by sphingosine
kinases in vivo and that phospho-FTY720 is the active S1P agonist.
See: (a) Billich, A.; Bornancin, F.; Devay, P.; Mechtcheriakova, D.;
Urtz, N.; Baumruker, T. Phosphorylation of the Immunomodulatory
Drug FTY720 by Sphingosine Kinases. J. Biol. Chem. 2003, 278,
47408–47415. (b) Zemann, B.; Kinzel, B.; Mueller, M.; Reuschel, R.;
Mechtcheriakova, D.; Urtz, N.; Bornancin, F.; Baumruker, T.; Billich,
A. Sphingosine kinase type 2 is essential for lymphopenia induced
by the immunomodulatory drug FTY720. Blood 2006, 107, 1454–
1458.
(29) Kondoh, M.; Yagi, K. Progress in absorption enhancers based on tight
junction. Expert Opin. Drug DeliVery 2007, 4, 275–286.
(30) To our knowledge, acid catalyzed hydrolysis of R-polyfluorinated
imidazoles has not previously been reported. For references involving
alkaline catalysis, see: (a) Fujii, S.; Maki, Y.; Kimoto, H.; Cohen,
L. A. Synthesis and alkaline hydrolysis of (pentafluoroethyl)imidazoles.
J. Fluorine Chem. 1987, 35, 437–454. (b) Hayahawa, Y.; Kimoto,
H.; Cohen, L. A.; Kirk, K. L. Synthesis of (Trifluoromethyl)imidazoles
with Additional Electronegative Substituents. An Approach to Recep-
tor-Activated Affinity Labels. J. Org. Chem. 1998, 63, 9448–9454.
(31) (a) MSDS for hydrogen fluoride. (b) For a discussion on the effects
of HF poisoning, see: Heard, K.; Delgado, J. Oral Decontamination
with Calcium or Magnesium Salts Does Not Improve Survival
Following Hydrofluoric Acid Ingestion. J. Toxicol. Clin. Toxicol. 2003,
41, 789–792.
(32) Hodge, J. E.; Fisher, B. E. Amadori Rearrangement Products. In
Methods in Carbohydrate Chemistry, Volume II: Reactions of Car-
bohydrates Whistler, R. L., Wolfrom, M. L., Bemiller, J. N., Eds.;
Academic Press: New York, 1963; Vol. II, pp 99-103.
(33) Hatanaka, M.; Takahashi, K.; Nakamura, S.; Mashino, T. Preparation
and antioxidant activity of R-pyridoin and its derivatives. Bioorg. Med.
Chem. Lett. 2005, 13, 6763–6770.
(34) Nakajima, N.; Saito, M.; Ubukata, M. Preparation and reaction of
4-methoxybenzyl (MPM) and 3,4-dimethoxybenzyl (DMPM) perflu-
oroimidates. Tetrahedron Lett. 1998, 39, 5565–5568.
(35) Abarbri, M.; Thibonnet, J.; Berillon, L.; Dehmel, F.; Rottlander, M.;
Knochel, P. Preparation of New Polyfunctional Magnesiated Hetero-
cycles Using a Chlorine-, Bromine-, or Iodine-Magnesium Exchange.
J. Org. Chem. 2000, 65, 4618–4634.
(36) Ohta, S.; Kawasaki, I.; Fukuno, A.; Yamashita, M.; Tada, T.;
Kawabata, T. Synthesis and application of triazole derivatives.
Synthesis of 3- and 5-acyl-1,2,4-triazoles via lithiation of 1-alkyl-1H-
1,2,4-triazoles. Chem. Pharm. Bull. 1993, 41, 1226–1231.
(17) Assay conducted with [33P]-S1P pyridoxal phosphate and a lysate of
S1P lyase. Production of [33P]-phosphoethanolamine was monitored
to determine lyase activity.
(18) FTY720 was used as a positive control in parallel with vehicle control
group (n ) 5 mice/group).
(19) Kroeplien, U.; Rosdorfer, J.; Van der Greef, J.; Long, R. C., Jr.;
Goldstein, J. H. 2-Acetyl-4(5)-(1,2,3,4-tetrahydroxybutyl)imidazole:
detection in commercial caramel color III and preparation by a
model browning reaction. J. Org. Chem. 1985, 50, 1131–1133.
(20) (a) Spector, R.; Huntoon, S. Effects of Caramel Color (Ammonia
Process) on Mamalian Vitamin B6 Metabolism. Toxicol. Appl.
Pharmacol. 1982, 62, 172–178. (b) Houben, G. F.; Kuijpers, M. H. M.;
Van Loveren, H.; Penninks, A. H.; Sinkeldam, E. J.; Seinen, W. Effects
of ammonia caramel and tetrahydroxybutylimidazole on the immune
system of rats. Arch. Toxicol. 1989, 13, 183–187. (c) Gobin, S. J. P.;
Phillips, J. A. Immunosuppressive effects of 2-acetyl-4-tetrahydroxy-
butyl imidazole (THI) in the rat. Clin. Arch. Toxicol. Exp. Immunol.
1991, 85, 335–340. (d) Bradbury, M. G.; Doherty, K. V.; Parish,
C. R.; Lyons, A. B. The immunosuppressive compound 2-acetyl-
4-tetrahydroxybutyl imidazole inhibits the allogeneic mixed lym-
phocyte reaction by sequestration of a recirculating subpopulation
of T cells. Immunology 1996, 87, 80–85. (e) Gugsyan, R.; Coward,
A.; O’Connor, L.; Shortman, K.; Scollay, R. Emigration of mature T
cells from the thymus is inhibited by the imidazole-based compound
2-acetyl-4-tetrahydroxybutylimidazole. Immunology 1998, 93, 398–
404.
(21) Schwab, S. R.; Pereira, J. P.; Matloubian, M.; Xu, Y.; Huang, Y.;
Cyster, J. G. Lymphocyte Sequestration Through S1P Lyase
Inhibition and Disruption of S1P Gradients. Science 2005, 309,
1735–1739.
JM900278W