2-Aryl(pyrrolidin-4-yl)acetic acids are potent agonists of sphingosine-1-phosphate (S1P) receptors

Article abstract of DOI:10.1016/j.bmcl.2006.03.090

A series of 2-aryl(pyrrolidin-4-yl)acetic acids were synthesized and their biological activities were evaluated as agonists of S1P receptors. These analogs were able to induce lowering of lymphocyte counts in the peripheral blood of mice and were found to

Full text of DOI:10.1016/j.bmcl.2006.03.090

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3564–3568
2-Aryl(pyrrolidin-4-yl)acetic acids are potent agonists of
sphingosine-1-phosphate (S1P) receptors
Lin Yan,a,* Richard Budhu,a Pei Huo,a Christopher L. Lynch,a,ꢀ Jeffrey J. Hale,a
Sander G. Mills,a Richard Hajdu,b Carol A. Keohane,b Mark J. Rosenbach,b
James A. Milligan,b Gan-Ju Shei,b Gary Chrebet,b James Bergstrom,b
Deborah Cardb and Suzanne M. Mandalab
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Immunology and Rheumatology, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
Received 18 January 2006; revised 23 March 2006; accepted 24 March 2006
Available online 18 April 2006
Abstract—A series of 2-aryl(pyrrolidin-4-yl)acetic acids were synthesized and their biological activities were evaluated as agonists of
S1P receptors. These analogs were able to induce lowering of lymphocyte counts in the peripheral blood of mice and were found to
have good overall pharmacokinetic properties in rat.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Immunosuppressants that dampen the immune response
are used to prevent the rejection of allografts after trans-
plantation and to treat autoimmune disorders. Current
immunotherapies (e.g., calcineurin inhibitors, antime-
tabolites, antiproliferatives, and monoclonal antibodies
R
O
NH2
OH
1, R = -H, FTY720
to T lymphocytes and cytokines) have narrow therapeu-
tic windows; their chronic usage can have limitations
due to the risk of side effects (e.g., infections, nephrotox-
icity, and lymphoproliferative disorders).1 Development
of immunosuppressants with improved safety profiles
based on known modes of drug action and/or new mech-
anisms is highly desired.
2, R = -PO3H2, FTY720 phosphate
OH
OH
P
N
H
O
3
O
N
FTY720 (1, Fig. 1) is a novel immunosuppressant that
has a unique mode of action. FTY720 is a prodrug for
monophosphate 2, which binds to and is an agonist of
four out of the five known sphingosine-1-phosphate
(S1P) receptors (which are members of the superfamily
of seven transmembrane G protein-coupled receptors).2
Agonism of S1P1 alters lymphocyte trafficking by
sequestering lymphocytes into secondary lymphoid
S
F3C
CO2H
4
N
N
CO2H
N
O
5
Figure 1. Structures of FTY720 (1), FTY720-phosphate (2), amino
phosphonic acid analog (3), and two lead azetidine-3-carboxylic acids
4 and 5.
Keywords: Immunosuppressants; S1P agonists; (Pyrrolidin-4-yl)acetic
acids; Peripheral lymphocyte lowering.
*
Corresponding author. Tel.: +1 732 594 5419; fax: +1 732 594
2210; e-mail: lin_yan@merck.com
organs,3 while agonism of S1P3 has been linked to bra-
dycardia in rodents.4 FTY720 has shown immunosup-
pressive efficacy in animal models5 and is currently in
ꢀ
Present address: Abbott Laboratories, Dept. R4N6, Bldg. AP-10,
100 Abbott Park Road, Abbott Park, IL 60064, USA.
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.03.090

L. Yan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3564–3568
3565
Phase III clinical trials for the prevention of rejection of
kidney after transplantation and in Phase II for the
treatment of multiple sclerosis.6
The synthesis of pyrrolidine-4-carboxylic acid and (pyrr-
olidin-4-yl)acetic acid analogs that have a pendant
substituted thiophene group similar to that found in 4
is shown in Scheme 1. Decarboxylative arylation of ani-
sole with pyroglutamic acid using Eaton’s reagent11 fol-
lowed by demethylation furnished phenol 8. Alkylation
with bromide 10 (prepared from alcohol 99) gave lactam
11. N-Boc protection of 11 followed by acylation with
ethyl chloroformate provided 12 as a 1:1 cis:trans mix-
ture.12 Carbonyl reduction was effected by removal of
the N-Boc group, treatment with Meerwein’s salt, and
reduction with NaCNBH3 to give pyrrolidine 13. Sapon-
ification of 13 gave pyrrolidine-4-carboxylic acid 14 as a
1:1 cis:trans mixture of diastereomers. Hydrolysis of
N-Boc protected 13 led to a carboxylic acid, which
was reduced to an alcohol and subsequently converted
into iodide 15. Cyanide displacement, acid hydrolysis,
and removal of the N-Boc furnished 5-aryl(pyrrolidin-
4-yl)acetic acid 16 as a 1:1 cis:trans mixture of diastereo-
mers. S1P receptor binding studies2a showed that
(pyrrolidin-3-yl)acetic acid 16 was a more potent S1P1
receptor agonist than was pyrrolidine-4-carboxylic acid
14 (Table 1). Based on these data, (pyrrolidin-4-yl)acetic
acid analogs were targeted for further investigation.
Studies from these laboratories have previously identi-
fied a series of 3-(N-alkylamino)propyl-phosphonic
acids7 as well as conformationally constrained analogs
of those compounds (e.g., 3, S1P1 IC50 = 0.1 nM),8 as
potent agonists of S1P1 that induce peripheral lym-
phocyte lowering in mice after iv administration.
Replacement of the phosphonic acid in these earlier
compounds with other acid groups, such as carboxyl-
ate, was investigated with the hope of increasing oral
bioavailability, but these structural changes were
found to decrease binding to S1P1 in instances where
their lipophilic tails were kept constant.7 The rational
combination of leads identified from high-throughput
screening of the Merck sample collection and scaffolds
selected from the conformationally constrained 3-
aminopropyl-phosphonic acids led to the discovery
of a series of azetidine-3-carboxylic acids, such as 49
and 5,10 as potent, selective, and orally bioavailable
S1P1 receptor agonists. This paper describes the design
and synthesis of a novel series of S1P1 receptor ago-
nists that utilize (pyrrolidin-4-yl)acetic acid as a zwit-
terionic pharmacophore and that have properties
comparable to azetidine-3-carboxylic acid analogs 4
and 5.
The synthesis of (pyrrolidin-4-yl)acetic acid analogs
based on 5 is shown in Scheme 2. Conversion of phenol
8 to the corresponding triflate followed by palladium
a, b
c
+
O
+
N
O
CO2H
N
H
Br
OH
CH3O
H
HO
F3C
F3C
S
S
8
10
9
6
7
d
CO2C2H5
O
O
N
Boc
N
H
e, f
O
O
S
S
F3C
F3C
12
g, h, i
11
CO2C2H5
CO2H
j
N
H
N
H
O
O
S
S
F3C
F3C
13
e, j, k, l
14
I
CO2H
m, n, o
N
N
H
O
O
Boc
S
S
F3C
F3C
15
16
Scheme 1. Reagents and conditions: (a) P2O5, CH3SO3H, 100 ꢁC (54%); (b) BBr3, CH2Cl2, ꢀ78 ꢁC to 0 ꢁC (80%); (c) CBr4, PPh3, CH2Cl2, rt (100%);
(d) K2CO3, CH3CN, 60 ꢁC (96%); (e) (Boc)2O, DMAP, CH2Cl2, rt (90%); (f) 1—LiN(TMS)2, THF, ꢀ78 ꢁC; 2—ClCO2Et, THF, ꢀ78 ꢁC (83%); (g)
HCl, EtOH, rt (81%); (h) (CH3)3OÆBF4, CH2Cl2, rt (94%); (i) NaCNBH3, CH3OH, pH 3–4, rt (97%); (j) NaOH, EtOH (100%); (k) 1—ClCO2Et,
Et3N, THF, 0 ꢁC to rt; 2—NaBH4, THF, H2O, 0 ꢁC to rt (93%); (l) I2, PPh3, imidazole, CH2Cl2, rt (56%); (m) NaCN, DMSO, 60 ꢁC (68%); (n) 40%
KOH, CH3OH, H2O, 100 ꢁC (100%); (o) 20% CF3CO2H, CH2Cl2, rt (16%).

3566
L. Yan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3564–3568
Table 1. Inhibition (IC50, nM) of 33P-S1P binding to S1P receptorsa,b
accomplished using the procedure analogous to that of
19. The configuration of methyl acetates on C-3 of
pyrrolidone relative to the aryl group on C-5 was
determined using 1D NOE experiments (Scheme 2,
inset). Direct NOE effects between H3 and H5 on the
cis analog were not observed. The cis and the trans
configurations were, therefore, assigned based on their
distinct NOE patterns generated by irradiating H4 and
CO2H
R
N
H
N
O
N
Compound
R
Stereo
S1P1 S1P3 S1P4 S1P5
4
1.2 530 1600 23
0.6 12,000 70 1.0
0
H4 , respectively.
5
14
16
22a
25
2.2
4800 2700 130
2100 390 390
The S1P receptor binding affinities (IC50) for new com-
pounds were determined in competitive binding assays
in transfected Chinese hamster ovary (CHO) cell mem-
branes expressing S1P receptors with 33P-labeled S1P
as the ligand.2a The functional activities (EC50) were
determined by measuring the binding of 35S-GTPcS to
S1P receptors expressed in CHO cell membranes.2a All
of the new pyrrolidine acids were found to be full ago-
nists (in a range of 70–120% of max) of S1P1,3,5 recep-
tors and inverse agonists of S1P4 receptor; none of the
new compounds were found binding to S1P2 receptor
(IC50 > 10 lM).15 Compound-induced peripheral blood
lymphocyte (PBL) lowering in mice was measured by
the reduction percentage of the absolute PBL counts
determined at a 3-h time point after the oral administra-
tion of the test compound in comparison to those from
vehicle controls.2a The murine PBL lowering has been
previously shown to correlate with immunosuppressive
efficacy in rodents.16
trans
cis
5.7 >1000 150 300
22b
22c
22d
22e
22f
22g
22h
7.9
1500 160 140
trans
trans
trans
1.0 >1000 120 68
1.4
3100 110 57
5.5 >1000 230 50
F3C
F
F
trans chiral 1 1.2
trans chiral 2 0.8
690
32
5.1
F
F
1100 800 11
460 110 21
The S1P receptor binding data for pyrrolidine-4-carbox-
ylic acid 14 and (pyrrolidin-4-yl)acetic acid 16, both of
which have substituted thiophene ether side chain, are
shown in Table 1. The binding affinity of pyrrolidine-
4-carboxylic acid 14 for S1P1 is about one order of mag-
nitude less than that of (pyrrolidin-4-yl)acetic acid 16,
indicating that (pyrrolidin-4-yl)acetic acid provides the
more optimal spacial orientation between the amino
group and the carboxylic acid. Compound 16 is about
1000-fold selective for S1P1 over S1P3 and has modest
affinity for both S1P4 and S1P5. Compound 16 was also
found to lower peripheral lymphocyte count in mice
after a relative high oral dose (39% vs control after a
10 mpg/kg po dose).
F
F
trans
2.0
a Displacement of 33P-labeled sphingosine-1-phosphate (S1P) by test
compounds from human S1P receptors expressed on CHO cell
membranes. Data are reported as means for n = 3 measurements. SD
were generally within 20% of the average.
b S1P2 IC50 values for new compounds were all greater than 10 lM.
catalyzed cyanation13 and N-Boc protection gave lactam
17. Allylation of 17 furnished a 76:3:1 mixture of readily
separable 3,5-trans, 3,5-cis and bis-allyl adducts. The
higher trans stereoselectivity presumably resulted from
the favorable approaching of allyl iodide to the less-
hindered face of the enolate intermediate of 17. The
double bond of 18 was oxidized under Sharpless’
conditions14 and subsequent methylation provided pyr-
rolidinone 19. Conversion of 19 to pyrrolidine 20 was
achieved using the procedures analogous to those
described in Scheme 1. Treatment of nitrile 20 with
hydroxylamine furnished N-hydroxyamidine 21, which
was coupled to various benzoic acids using EDC
followed by dehydration at elevated temperature and
removal of the N-Boc to provide the amino acids exem-
plified in Table 2. Most of the derivatized benzoic acids
used to prepare the new analogs were either commercial-
ly available or synthesized as described previously.8
The receptor binding data of new (pyrrolidin-4-yl)acetic
acids having derivatized oxadiazole ring are also shown
in Table 1. The racemic 2,4-trans analog 22a was deter-
mined to be slightly more S1P1 potent than the cis ana-
log 22b; this was found to generally hold for this series
of compounds. For other trans analogs, sterically
more-hindered groups, such as cyclopentyl and cyclo-
hexyl groups, enhanced S1P1 binding affinities by about
half a log unit. Fluorination of the alkyl groups slightly
enhanced receptor binding affinities but decreased the
selectivity for S1P1 over S1P3. The former, but not the
latter, was observed of analogs of 5.10
Several of the new (pyrrolidin-4-yl)acetic acid analogs
were selected for further evaluation in the mouse PBL
lowering assay. Analogs 22b, 22c, 22d, 22g, and 22h
were all found to induce a PBL response after oral
administration, with 22f, 22g, and 22h being able to
The cis ester 23 was prepared by racemizing the trans
allyl adduct 18 using LiHMDS at ꢀ78 ꢁC and then
converting the resulting allyl group to methyl ester 23.
Transformation of 23 to the final amino acid was

L. Yan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3564–3568
3567
a, b, c
NC
d
O
N
H
O
O
N
N
trans-3,5 configuration
HO
NC
Boc
Boc
H4'
H4
8
H4
H5
H4'
17
18
CO2CH3
H3
CO2CH3
H3
H5
e
O
O
N
N
NC
NC
Boc
Boc
CO2CH3
CO2CH3
19
f, g, h, c
NC
O
N
N
NC
cis-3,5 configuration
Boc
Boc
19
20
H4'
H4
H4'
CO2CH3
H4
H5
CO2CH3
H3
H3
O
H5
i
O
N
N
CO2CH3
j, f , k
CO2H
NC
NC
Boc
Boc
R
23
N
N
N
H
N
HO
Boc
ON
H2N
21
22
Scheme 2. Reagents and conditions: (a) PhN(Tf)2, EtN-i-Pr2, DMF, rt (88–100%); (b) Pd(PPh3)4, Zn(CN)2, DMF, 80 ꢁC; (c) (Boc)2O, DMAP,
CH2Cl2, rt (over two steps, 61–88%); (d) 1—LDA, THF, ꢀ78 ꢁC; 2—allyl iodide, THF, ꢀ78 ꢁC (76–86%); (e) 1—RuCl3ÆxH2O, NaIO4,
CCl4:CH3CN:H2O, rt; 2—TMSCHN2, benzene, CH3OH, rt (over two steps, 94%); (f) 20% CF3CO2H, CH2Cl2, rt (89%); (g) (CH3)3OÆBF4, CH2Cl2,
rt; (h) NaCNBH3, CH3OH, pH 3–4, rt (85–100%); (i) NH2OHÆHCl, NaHCO3, CH3OH, reflux (74–93%); (j) derivatized benzoic acid, EDC, CH3CN,
1 h at rt and 16 h at 120 ꢁC (40–60%); (k) NaOH, EtOH, rt (100%). The relative configurations of substitutes on the pyrrolidinone ring were assigned
based on 1D NOE 1H NMR experiments (inset). Curved arrows point away from the irradiated proton to the proton that exhibits NOE effect.
Table 2. Rat pharmacokinetic data (2.0 mpk po, 1.0 mpk iv) for selected S1P receptor agonistsa
Compound
PK parameters
4
Clp = 12.3 mL/min/kg, Vdss = 10.0 L/kg, t1/2 = 6.7 h, %F = 79
Clp = 4.1 mL/min/kg, Vdss = 2.8 L/kg, t1/2 = 8.5 h, %F = 67
Clp = 7.4 mL/min/kg, Vdss = 6.6 L/kg, t1/2 = 10.7 h, %F = 79
Clp = 3.4 mL/min/kg, Vdss = 2.9 L/kg, t1/2 = 10.9 h, %F = 27
Clp = 2.1 mL/min/kg, Vdss = 1.3 L/kg, t1/2 = 5.6 h, %F = 36
Clp = 3.5 mL/min/kg, Vdss = 2.2 L/kg, t1/2 = 4.4 h, %F = 23
5
22a
22b
22g
22h
a Plasma compound concentrations used to calculate pharmacokinetic parameters were obtained after iv administration (1.0 mpk) and po admin-
istration (2.0 mpk) of test compounds to male Sprague–Dawley rats (n = 2), respectively.
do so after doses of less than 1.0 mpk po.17 The pharma-
codynamic ED50 value for 22g in this assay was deter-
mined to be approximately 0.3 mpk po, making it
comparable to 4 (ED50 = 0.4 mpk po). Rat pharmacoki-
netics for selected compounds (Table 2) appeared to
parallel those for analogs of 5, with isobutyl analogs
22a and 22b as well as fluoroalkyl analogs 22g and 22h
all being low clearance compounds with good oral
bioavailability.
Acknowledgments
We thank Ms. Susan Li and Dr. Larry Colwell for coor-
dinating the rat pharmacokinetic studies.
References and notes
1. Kovarik, J. M.; Burtin, P. Expert Opin. Emerg. Drugs
2003, 8, 47.
In summary, a series of 2-aryl(pyrrolidin-4-yl)acetic
acids have been identified as potent and selective S1P1
receptor agonists. Unlike the previously described ami-
no phosphonic acids,7,8 these amino acids exhibit good
overall pharmacokinetic properties in rat and effectively
lower peripheral lymphocytes in mice after oral admin-
istration. This work demonstrates that a (pyrrolidin-4-
yl)acetic acid scaffold can be utilized to afford potent
and selective S1P1 receptor agonists. Further, this also
supports the continued investigation of this structurally
novel scaffold as a potential replacement for the zwitter-
ionic azetidine-3-carboxylic acid pharmacophore of S1P
receptor agonists such as 4 and 5.
2. (a) Mandala, S.; Hajdu, R.; Bergstrom, J.; Quackenbush,
E.; Xie, J.; Milligan, J.; Thornton, R.; Shei, G. J.; Card,
D.; Keohane, C.; Rosenbach, M.; Hale, J.; Lynch, C. L.;
Rupprecht, K.; Parson, W.; Rosen, H. Science 2002, 296,
346; (b) Brinkmann, V.; Davis, M. D.; Heise, C. E.;
Albert, R.; Cottens, S.; Hof, R.; Bruns, C.; Prieschl, E.;
Baumruker, T.; Hiestand, P.; Foster, C. A.; Zollinger, M.;
Lynch, K. R. J. Biol. Chem. 2002, 277, 21453; (c) Hale, J.
J.; Yan, L.; Neway, W. E.; Hajdu, R.; Bergstrom, J. D.;
Milligan, J. A.; Shei, G.-J.; Chrebet, G. L.; Thornton, R.
A.; Card, D.; Rosenbach, M.; Rosen, H.; Mandala, S.
Bioorg. Med. Chem. 2004, 12, 4803; (d) Albert, R.;
Hinterding, K.; Brinkmann, V.; Guerini, D.; Muller-
¨
Hartwieg, C.; Knecht, H.; Simeon, C.; Streiff, M.; Wagner,

3568
L. Yan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3564–3568
´
T.; Welzenbach, K.; Zecri, F.; Zollinger, M.; Cooke, N.;
Francotte, E. J. Med. Chem. 2005, 48, 5373.
Card, D.; Ferrer, M.; Hodder, P.; Strulovici, B.; Rosen,
H.; Mandala, S. J. Med. Chem. 2004, 47, 6662.
10. Li, Z.; Chen, W.; Hale, J. J.; Lynch, C. L.; Mills, S.
G.; Hajdu, R.; Keohane, C. A.; Rosenbach, M. J.;
Milligan, J. A.; Shei, G.-J.; Parent, S. A.; Chrebet, G.;
Parent, S. A.; Bergstrom, J.; Card, D.; Forrest, M.;
Quackenbush, E. J.; Wickman, L. A.; Vargas, H.;
Evans, R. M.; Rosen, H.; Mandala, S. J. Med. Chem.
2005, 48, 6169.
3. Cyster, J. G. Annu. Rev. Immunol. 2005, 23, 127.
4. (a) Forrest, M.; Sun, S. -Y.; Hajdu, R.; Bergstrom, J.;
Card, D.; Doherty, G.; Hale, J.; Keohane, C.; Meyers, C.;
Milligan, J.; Mills, S.; Nomura, N.; Rosen, H.; Rosen-
bach, M.; Shei, G.-J.; Singer, I. I.; Tian, M.; West, S.;
White, V.; Xie, J.; Proia, R. L.; Mandala, S. J. Pharmacol.
Exp. Ther. 2004, 309, 758; (b) Sanna, M. G.; Liao, J.; Jo,
E.; Alfonso, C.; Ahn, M.-Y.; Peterson, M. S.; Webb, B.;
Lefebvre, S.; Chun, J.; Gray, N.; Rosen, H. J. Biol. Chem.
2004, 279, 13839.
5. (a) Matloubian, M.; Lo, C. G.; Cinamon, G.; Lesneski, M.
J.; Xu, Y.; Brinkamnn, V.; Allende, M. L.; Proia, R. L.;
Cyster, J. G. Nature 2004, 427, 355; (b) Allende, M. L.;
Dreier, J. L.; Mandala, S.; Proia, R. L. J. Biol. Chem.
2004, 279, 15396.
6. (a) Budde, K.; Schmouder, R. L.; Brunkhorst, R.;
Nashan, B.; Lucker, P. W.; Mayer, T.; Choudhury, S.;
Skerjanec, A.; Kraus, G.; Neumayer, H. H. J. Am. Soc.
Nephrol. 2002, 13, 1073; (b) Budde, K.; Schmouder, R. L.;
Nashan, B.; Brunkhorst, R.; Lucker, P. W.; Mayer, T.;
Brookman, L.; Nedelman, J.; Skerjanec, A.; Bohler, T.;
Neumayer, H. H. Am. J. Transplant. 2003, 3, 846.
7. Hale, J. J.; Doherty, G.; Toth, L.; Zhen, L.; Mills, S. G.;
Hajdu, R.; Keohane, C. A.; Rosenbach, M.; Milligan, J.;
Shei, G.-J.; Chrebet, G.; Bergstrom, J.; Card, D.; Rosen,
H.; Mandala, S. Bioorg. Med. Chem. Lett. 2004, 14, 3495.
8. Yan, L.; Hale, J. J.; Lynch, C. L.; Budhu, R.; Gentry, A.;
Mills, S. G.; Hajdu, R.; Keohane, C. A.; Rosenbach, M.
J.; Milligan, J. A.; Shei, G.-J.; Chrebet, G.; Bergstrom, J.;
Card, D.; Rosen, H.; Mandala, S. Bioorg. Med. Chem.
Lett. 2004, 14, 4861.
11. Rigo, B.; Fasseur, D.; Cherepy, N.; Couturier, D. Tetra-
hedron Lett. 1989, 30, 7057.
12. The acylation reaction required using 2 equiv of LiHMDS
to complete probably due to the generation of more acidic
a-H of 12 than that of N-Boc 11. The addition of excess
base was responsible for the racemization of the resulting
ester.
13. Kubota, H.; Rice, K. C. Tetrahedron Lett. 1998, 39, 2907.
14. Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless,
K. B. J. Org. Chem. 1981, 46, 3936.
15. The values of IC50 and EC50 generally fall within a range
of 10-fold; therefore, only the values of IC50 of S1P1,3,4,5
receptors will be compared and discussed. Inverse agonist
activity was defined as the inhibition of both basal and
agonist stimulated 35S-GTPcS binding in the 35S-GTPcS
functional assay (see Ref. 2a). Compound 2 (30 nM) was
used as the agonist.
16. Hale, J. J.; Neway, W.; Mills, S. G.; Hajdu, R.; Keohane,
C. A.; Rosenbach, M.; Milligan, J.; Shei, G.-J.; Chrebet,
G.; Bergstrom, J.; Card, D.; Koo, G. C.; Koprak, S. L.;
Jackson, J. J.; Rosen, H.; Mandala, S. Bioorg. Med. Chem.
Lett. 2004, 14, 3351.
17. Compounds 22f and 22g were prepared by chiral separa-
tion of their corresponding racemic methyl ester on
Chiralcel OD 2 · 25 cm column (eluted with EtOH/
heptane (v:v = 15:85) at rate of 8.0 mL/min), followed by
saponification.
9. Hale, J. J.; Lynch, C. L.; Neway, W.; Mills, S. G.; Hajdu,
R.; Keohane, C. A.; Rosenbach, M. J.; Milligan, J. A.;
Shei, G.-J.; Parent, S. A.; Chrebet, G.; Bergstrom, J.;