Receptor activity and conformational analysis of 5′-halogenated resiniferatoxin analogs as TRPV1 ligands

Article abstract of DOI:10.1016/j.bmcl.2010.11.012

A series of 5′-halogenated resiniferatoxin analogs have been investigated in order to examine the effect of halogenation in the A-region on their binding and the functional pattern of agonism/antagonism for rat TRPV1 heterologously expressed in Chinese hamster ovary cells. Halogenation at the 5-position in the A-region of RTX and of 4-amino RTX shifted the agonism of parent compounds toward antagonism. The extent of antagonism was greater as the size of the halogen increased (I > Br > Cl > F) while the binding affinities were similar, as previously observed for our potent agonists. In this series, 5-bromo-4-amino RTX (39) showed very potent antagonism with K _i (ant) = 2.81 nM, which was thus 4.5-fold more potent than 5′-iodo RTX, previously reported as a potent TRPV1 antagonist. Molecular modeling analyses with selected agonists and the corresponding halogenated antagonists revealed a striking conformational difference. The 3-methoxy of the A-region in the agonists remained free to interact with the receptor whereas in the case of the antagonists, the compounds assumed a bent conformation, permitting the 3-methoxy to instead form an internal hydrogen bond with the C4-hydroxyl of the diterpene.

Full text of DOI:10.1016/j.bmcl.2010.11.012

Bioorganic & Medicinal Chemistry Letters 21 (2011) 299–302
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Receptor activity and conformational analysis of 50-halogenated
resiniferatoxin analogs as TRPV1 ligands
Kwang Su Lim a, Dong Wook Kang a, Yong Soo Kim a, Myeong Seop Kim a, Seul-Gi Park b, Sun Choi b,
Larry V. Pearce c, Peter M. Blumberg c, Jeewoo Lee a,
⇑
a Research Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National University, Seoul 151-742, Republic of Korea
b College of Pharmacy, Division of Life & Pharmaceutical Sciences, and National Core Research Center for Cell Signaling & Drug Discovery Research,
Ewha Womans University, Seoul 120-750, Republic of Korea
c Laboratory of Cancer Biology and Genetics, Center for Cancer Research, National Cancer Institute, NIH, Bethesda, MD 20892, USA
a r t i c l e i n f o
a b s t r a c t
A series of 50-halogenated resiniferatoxin analogs have been investigated in order to examine the effect of
halogenation in the A-region on their binding and the functional pattern of agonism/antagonism for rat
TRPV1 heterologously expressed in Chinese hamster ovary cells. Halogenation at the 5-position in the
A-region of RTX and of 4-amino RTX shifted the agonism of parent compounds toward antagonism.
The extent of antagonism was greater as the size of the halogen increased (I > Br > Cl > F) while the bind-
ing affinities were similar, as previously observed for our potent agonists. In this series, 5-bromo-4-amino
RTX (39) showed very potent antagonism with Ki (ant) = 2.81 nM, which was thus 4.5-fold more potent
than 50-iodo RTX, previously reported as a potent TRPV1 antagonist. Molecular modeling analyses with
selected agonists and the corresponding halogenated antagonists revealed a striking conformational dif-
ference. The 3-methoxy of the A-region in the agonists remained free to interact with the receptor
whereas in the case of the antagonists, the compounds assumed a bent conformation, permitting the
3-methoxy to instead form an internal hydrogen bond with the C4-hydroxyl of the diterpene.
Ó 2010 Elsevier Ltd. All rights reserved.
Article history:
Received 7 October 2010
Accepted 1 November 2010
Available online 5 November 2010
Keywords:
TRPV1 agonist
TRPV1 antagonist
Partial agonist
Halogenation
Resiniferatoxin
Molecular modeling
Resiniferatoxin (RTX, 1),1,2 isolated from Euphorbia resinifera, is
an extremely potent irritant tricyclic diterpene which structurally
falls within the daphnane subclass of phorbol-related diterpenes
but is set apart by its homovanillyl ester group at C-20.3,4 RTX
has proven to function pharmacologically as an ultrapotent agonist
for the transient receptor potential vanilloid 1 (TRPV1) channel,
displaying 103- to 104-fold greater potency than the prototypic
agonist capsaicin.5
RTX or ROPA (resiniferonol orthophenylacetate) based on the three
structural regions including the A-region (4-hydroxy-3-methoxy-
phenyl), B-region (C20 ester), and C-region (diterpene).12–21 In the
SAR of RTX, it should be noted that a divergence in potencies be-
tween ligand binding and agonism for TRPV1 ligands is frequently
observed15 and may reflect the distinction between the properties
of the small fraction of the receptor at the cell surface and the pre-
dominant proportion in internal membranes.22
RTX was found to evoke large inward currents in dorsal root
ganglion (DRG) neurons and TRPV1 transfected cell lines.6 The ac-
tions of RTX are mediated by binding, with picomolar affinity, di-
rectly to the capsaicin-binding site on the TRPV1 receptor.7
Whereas capsaicin under normal conditions produces only short
term desensitization of TRPV1 mediated responses, the apparent
desensitization to RTX can be of very long duration, lasting for
weeks.8,9 RTX is being developed as a potent desensitizing agent
for neurons in the treatment of urinary urge incontinence and
the pain associated with diabetic neuropathy, as well as for cancer
pain.10,11
Previous SAR indicated that although most modifications of RTX
reduced binding affinity and agonism, several have proven to be of
considerable interest (Fig. 1). 5-Iodination in the A-region of RTX
shifted the activity from agonism to antagonism providing a potent
antagonist, 5-iodo RTX (2), for both rat and human TRPV1.16,17
Surprisingly, the carbonate surrogate (3) in the B-region of 5-iodo
RTX showed excellent agonism rather than antagonism and its
R4
OH
OH
OH
R5
H
I
X
1
2
3
4
CH2
CH2
O
O
O
H
O
OCH3
R4
Structure–activity relationships for RTX derivatives have been
investigated mostly through partial modifications starting from
H
I
O
CH2 NHSO2CH3
H
OH
O
O
X
R5
⇑
Corresponding author. Tel.: +82 2 880 7846; fax: +82 2 888 0649.
E-mail address: jeewoo@snu.ac.kr (J. Lee).
Figure 1.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.11.012

300
K. S. Lim et al. / Bioorg. Med. Chem. Lett. 21 (2011) 299–302
CN
F
Br
HO
Br
EtO
Br
b
a
O
O
O
O
H
H
O
O
O
O
a-b
R5
H
H
R4
R3
Br
Br
Br
O
21
20
19
OH
OH
O
O
O
OH
Scheme 4. Reagents and conditions: (a) ethyl cyanoacetate, NaH, DMF, 25%;
(b) NaOH, H2O 100%.
ROPA
31-39
Scheme 1. Reagents and conditions: (a) R-CO2H, EDC, CH2Cl2, 70–90%;
(b) CF3CO2H/CH2Cl2 (1:2) for MOM deprotection, 60–96%.
EtO
F
F
HO
OCH3
NH2
b-d
a
O
O
potency was 4-fold higher than RTX with a one-digit picomolar va-
lue reported for hTRPV1 expressed in HEK cells.21 Substitution of
an isosteric methanesulfonamide on RTX provided a metabolically
stable and potent agonist (4), which was 2.5-fold more potent than
RTX with an EC50 = 0.106 nM for rat TRPV1 heterologously ex-
pressed in CHO cells.20 The results demonstrated that halogenation
and isosteric substitution of the phenolic group in the A-region ap-
pear to be key modifications influencing the functional activity and
potency toward TRPV1.
In this Letter, we have investigated the effect of halogenation in
the A-region of RTX and of the isosteric 4-amino surrogate in a sys-
tematic manner and we describe the synthesis and receptor activ-
ity, as well as conformational analysis for explaining the observed
functional differences.
The halogenated RTX analogs (31–39) were prepared starting
from commercially available ROPA (resiniferonol-9,13,14-orthoph-
enylacetate) by the condensation with the corresponding 2-phenyl
acetic acids (A-region) using EDC (1-[3-(dimethylamino)propyl]-3-
ethylcarbodiimide hydrochloride) coupling reagent, followed by
appropriate O-deprotection if needed (Scheme 1).
The syntheses of the A-region are represented in Schemes 2–6.
The 5-fluoro-4-hydroxyphenyl A-region was prepared from 2-flu-
oro-6-methoxyphenol employing regioselective formylation
(Scheme 2) and the 5-chloro and 5-bromo-4-hydroxyphenyl A-re-
gions were synthesized starting from vanillin and homovanillic
acid using electrophilic addition of halogens, respectively
(Schemes 2 and 3). The 3,5-dibromophenyl A-region was synthe-
sized from 1,3-dibromo-5-fluorobenzene by nucleophilic substitu-
tion of ethyl cyanoacetate followed by decarboxylation (Scheme 4).
The 5-fluoro-4-amino A-region was prepared from 1,3-difluoro-2-
nitrobenzene employing the reactions of Makosza’s vicarious
NO2
NO2
F
F
F
23
24
22
Scheme 5. Reagents and conditions: (a) t-BuOK, ethyl chloroformate, NMP, 22%;
(b) NaOMe, MeOH, 55%; (c) Pd/C, H2, MeOH, 69%; (d) LiOH, THF–H2O, 82%.
nucleophilic substitution23 and selective nucleophilic substitution
of the methoxy group (Scheme 5). The 4-amino, 5-chloro and 5-
bromo-4-amino A-regions were synthesized from (3-hydroxy-
phenyl)-acetic acid by nitration followed by halogenation of the
4-aminophenyl (Scheme 6).
The halogenated resiniferatoxin analogs were evaluated for
their binding affinities (expressed as Ki values) using a competitive
binding assay with [3H]RTX and for their agonism/antagonism
using a functional 45Ca2+ uptake assay (expressed as EC50 and Ki
(ant) values) for rat TRPV1 heterologously expressed in Chinese
hamster ovary (CHO) cells.24 In this system, resiniferatoxin dis-
played Ki = 0.043 nM and EC50 = 0.27 nM.20 The receptor binding
and functional activities of the halogenated RTX analogs are listed
in Table 1.
Starting with RTX, 5-halogenation on the A-region progressively
shifted the agonism toward antagonism as the size of halogen in-
creased. 5-Fluoronation and 5-chlorination yielded the full agonists
31 and 32, which were 9- and 21-fold less potent, respectively, than
RTX for their functional activity. 5-Bromination caused a partial shift
in function, providing the partial antagonist 33 with Ki (ant) =
7.2 nM and 19% residual agonism. Finally, 5-iodination (I-RTX) affor-
ded the full antagonist 2 with Ki (ant) = 12.2 nM under the same
assay conditions. The 5-halogenated RTX analogs (31–33) bound
OCH3
OH
O
a
b
F
HO
OCH3
OCH3
OCH3
OH
5
HO
e-g
H
c-d
O
O
OMOM
OMOM
OCH3
OH
X
H
X
X
11 X=F
12 X=Cl
9
X=F
7 X=F
8 X=Cl
10 X=Cl
6
Scheme 2. Reagents and conditions: (a) (i) formaldehyde, diethylamine, EtOH, reflux, (ii) CH3I, CH2Cl2, rt, 18 h, (iii) HMTA, HOAc–H2O, 120 °C, 2 h, 37% for 3 steps; (b) NCS,
NaH, THF, 49%; (c) MOMCl, DIPEA, CH3CN, 68% for X@F, 87% for X@Cl; (d) LAH, THF, 98% for X@F and Cl; (e) CBr4, PPh3, benzene, 48% for X@F, 32% for X@Cl; (f) X@F: NaCN,
DMF, 80 °C, 93%, X@Cl: KCN, 18-crown-6, CH3CN, 85%; (g) NaOH, THF–H2O, reflux, 99% for X@F and Cl.
HO
OCH3
OAc
HO
HO
OCH3
OH
OCH3
OH
d
a-c
O
O
O
Br
Br
13
14
15
Scheme 3. Reagents and conditions: (a) cat. H2SO4, EtOH, 99%; (b) oxone, NaBr, acetone–H2O, 56%; (c) NaOH, THF–H2O, 100%; (d) Ac2O, cat. H2SO4, CH2Cl2, 95%.

K. S. Lim et al. / Bioorg. Med. Chem. Lett. 21 (2011) 299–302
301
EtO
OCH3
NH2
HO
OH
EtO
OH
c-d
a-b
O
O
O
O
NO2
27
f
25
26
e-f
HO
OCH3
NH2
HO
OCH3
NH2
O
X
30
28 X=Cl
29 X=Br
Scheme 6. Reagents and conditions: (a) cat. H2SO4, EtOH, 98%; (b) HNO3, AcOH, 29%; (c) K2CO3, CH3I, acetone, 71%; (d) 10% Pd/C, H2, THF–EtOH, 97%; (e) oxone, NaCl or NaBr,
acetone–H2O, 30%; (f) LiOH, THF–H2O, 60–90%.
Table 1
Potencies of RTX analogs for binding to rat TRPV1 and for inducing calcium influx in CHO/TRPV1 cells
O
O
H
O
R5
H
R4
R3
O
OH
O
O
R3
R4
R5
RTX bindinga (Ki = nM)
Agoa,b (EC50 = nM)
Anta,c (Ki = nM)
1 (RTX)
2 (I-RTX)
31
32
33
34
35
36
37
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
Br
OCH3
OCH3
OCH3
OCH3
OH
OH
OH
OH
OH
OAc
H
NH2
NH2
NH2
NH2
H
I
F
Cl
Br
Br
Br
H
F
0.043
0.27
NE
1.88 0.24
5.7 1.7
(19%)
NE
12.2 4.0
NE
0.61 0.08
0.26 0.04
1.03 0.14
0.17 0.05
0.30 0.02
110 36
0.13 0.03
0.70 0.16
0.85 0.02
1.99 0.27
NE
7.2 1.6
(30%)
NE
NE
NE
(37%)
1070 110
0.46 0.10
1.14 0.23
(34%)
38
39
Cl
Br
(56%)
2.81 0.26
NE
NE, no effect.
a
Values represent the mean SEM of three or more independent experiments for Ki or EC50
.
b
c
Values in parentheses represent the mean % of agonism relative to capsaicin.
Values in parentheses represent the mean % of antagonism of the capsaicin stimulated 45Ca2+ uptake response.
to TRPV1 with lower affinity than did RTX (1) but showed little
difference as the size of the halogen increased, with a range of
Ki = 0.17–1.03 nM. Compound 34, the O-acetyl analog of 33, dis-
played modestly weaker binding affinity and antagonism compared
to 33. The 3,5-dibromo analog 35 was a full agonist, albeit with
weaker binding affinity and agonist potency.
difference was examined as the size of the halogen increased, with
a range of Ki = 0.7–1.99 nM. Unfortunately, the 5-iodo 4-amino
analog could not be investigated due to synthetic difficulty.
In order to investigate the structural basis for the difference in
the functional activities of the TRPV1 agonists and their haloge-
nated analogs as antagonists, we performed conformational analy-
sis of RTX (1) and 36 as agonists and I-RTX (2) and 39 as
antagonists using the SYBYL Grid Search method.25 The resulting
lowest energy conformers are as shown in Figure 2 in which over-
lays of RTX vs I-RTX and 36 vs 39 are represented in Figure 2A and
B, respectively.
In their lowest energy conformers, the antagonists 2 and 39
showed bent conformations in which the methoxy group appeared
to form an intramolecular hydrogen bond with the C-4 hydroxyl
group of the diterpene core. In contrast, the agonists did not show
this intramolecular hydrogen-bond in their lowest conformers and
instead allowed the A-region to remain free to interact with resi-
dues in the binding site. The analysis indicates that the 5-halogen
promoted bending of the B-region to permit this internal hydro-
gen-bond formation between 3-methoxy of the A-region and
C4-OH of the diterpene and the preference for this conformation
was higher as the size of the halogen increased. A prediction is thus
that the binding of this bent conformation, unlike the more
Next, the halogenated 4-amino RTX analogs were explored as
bioisosteres of the corresponding 4-phenolic agonists. The parent
4-amino RTX (36) showed high affinity binding and potent full
agonism. The values of Ki = 0.13 nM and EC50 = 0.46 nM were only
2–3 fold less potent than RTX. As was the case with the 5-haloge-
nated RTX derivatives, 5-halogenation on the A-region in 4-amino
RTX also shifted the agonism toward antagonism progressively as
the size of the halogen increased. 5-Fluoronation produced full
agonist 37, which was 2.5-fold less potent than the parent 36, as
found in the RTX series. However, 5-chlorination already started
to shift the agonism of 36 toward antagonism to afford the partial
antagonist 38 and 5-bromination yielded the full antagonist 39. Of
particular note, compound 39 displayed high potency as an antag-
onist with Ki (ant) = 2.81 nM, which was 4-fold better than I-RTX
under the same assay conditions. As found in the RTX series, the
binding affinities of the 5-halogenated 4-amino RTX analogs
(37–39) were weaker than that of 4-amino RTX (36) and little

302
K. S. Lim et al. / Bioorg. Med. Chem. Lett. 21 (2011) 299–302
Figure 2. The lowest energy conformers of (A) RTX and I-RTX, and (B) 36 and 39. Carbon atoms are shown in green for RTX, white for I-RTX, purple for 36 and pink for 39.
Agonists RTX and 36 are depicted as ball-and-stick. Antagonists I-RTX and 39 are depicted as capped stick. Hydrogen bonds are displayed in black dashed lines, and non-polar
hydrogens are undisplayed for clarity.
12. Acs, G.; Lee, J.; Marquez, V. E.; Wang, S.; Milne, G. W. A.; Lewin, N. E.; Blumberg,
P. M. J. Neurochem. 1995, 65, 301.
13. Walpole, C. S. J.; Bevan, S.; Bloomfield, G.; Breckenridge, R.; James, I. F.; Ritchie,
extended conformation of RTX agonists, should fail to induce the
shift in the conformation of the tetrameric TRPV1 channel associ-
ated with channel opening. This model thus suggests novel strate-
gies for antagonist design.
In conclusion, we have systematically modified the aromatic A-
region of RTX and its 4-amino surrogate by halogenation at the 5-
position in order to explore the role of halogens in the reversal of
activity from agonism to antagonism. 5-Halogenation converted
the agonists to partial or full antagonists, and the extent of antag-
onism reflected the order of I > Br > Cl > F. Antagonism was further
favored in derivatives of the 4-amino RTX surrogate compared to
derivatives of RTX itself. Of particular note, the 5-bromo 4-amino
T.; Szallasi, A.; Winter, J.; Wrigglesworth, R. J. Med. Chem. 1996, 39, 2939.
14. Appendino, G.; Cravotto, G.; Palmisano, G.; Annunziata, R.; Szallasi, A. J. Med.
Chem. 1996, 39, 3123.
15. Acs, G.; Lee, J.; Marquez, V. E.; Blumberg, P. M. Mol. Brain Res. 1996, 35, 173.
16. Wahl, P.; Foged, C.; Tullin, S.; Thomsen, C. Mol. Pharmacol. 2001, 59, 9. Iodo-
RTX.
17. Seabrook, G. R.; Sutton, K. G.; Jarolimek, W.; Hollingworth, G. J.; Teague, S.;
Webb, J.; Clark, N.; Boyce, S.; Kerby, J.; Ali, Z.; Chou, M.; Middleton, R.;
Kaczorowski, G.; Jones, A. B. J. Pharmacol. Exp. Ther. 2002, 303, 1052.
18. McDonnell, M. E.; Zhang, S.-P.; Dubin, A. E.; Dax, S. L. Bioorg. Med. Chem. Lett.
2002, 12, 1189.
19. Appendino, G.; Ech-Chahad, A.; Minassi, A.; Bacchiega, S.; De Petrocellis, L.; Di
Marzo, V. Bioorg. Med. Chem. Lett. 2007, 17, 132.
20. Choi, H.-K.; Choi, S.; Lee, Y.; Kang, D. W.; Ryu, H.; Maeng, H.-J.; Chung, S.-J.;
Pavlyukovets, V. A.; Pearce, L. V.; Toth, A.; Tran, R.; Wang, Y.; Morgan, M. A.;
Blumberg, P. M.; Lee, J. Bioorg. Med. Chem. 2009, 17, 690.
21. Appendino, G.; Ech-Chahad, A.; Minassi, A.; De Petrocellis, L.; Di Marzo, V.
Bioorg. Med. Chem. Lett. 2010, 20, 97.
22. Toth, A.; Blumberg, P. M.; Chen, Z.; Kozikowski, A. P. Mol. Pharmacol. 2004, 65,
282.
23. Stahly, G. P.; Stahly, B. C.; Lilje, K. C. J. Org. Chem. 1984, 49, 579.
24. Wang, Y.; Szabo, T.; Welter, J. D.; Toth, A.; Tran, R.; Lee, J.; Kang, S. U.; Lee, Y.-S.;
Min, K. H.; Suh, Y.-G.; Park, M.-K.; Park, H.-G.; Park, Y.-H.; Kim, H.-D.; Oh, U.;
Blumberg, P. M.; Lee, J. Mol. Pharm. 2002, 62, 947 [published erratum appears
in Mol. Pharmacol. 2003, 63, 958].
25. The 3D structures of the tested compounds were generated with Concord and
energy minimized using an MMFF94s force field and MMFF94 charge until the
rms of the Powell gradient was 0.05 kcal molꢀ1 Aꢀ1 in SYBYL 8.1.1 (Tripos Int.,
St. Louis, MO, USA).
RTX analog (39) was
a potent, full antagonist with Ki
(ant) = 2.81 nM, which was 4.5-fold more potent than I-RTX (2) un-
der our conditions. Molecular modeling of selected agonists and
antagonists demonstrated that the 3-methoxy of the A-region in
agonists remained free to interact with the receptor for agonism,
whereas a 5-halogen in the antagonists favored a bent B-region,
allowing the 3-methoxy to form an intramolecular hydrogen bond
with the C4-hydroxyl of the diterpene.
Acknowledgments
This research was supported by Grant R11-2007-107-02001-0
from the National Research Foundation of Korea (NRF), the Na-
tional Core Research Center (NCRC) program (R15-2006-020) of
MEST and NRF through the Center for Cell Signaling & Drug Discov-
ery Research at Ewha Womans University (to S. Choi), and by the
Intramural Research Program of the National Institutes of Health,
Center for Cancer Research, National Cancer Institute. We thank
numerous research fellows for some of the biological analyses.
The conformational analysis of each compound was performed using SYBYL
Grid Search method (force field: MMFF94s; charges: MMFF94; minimization
method: powell; termination: gradient 0.05 kcal molꢀ1 Aꢀ1
; and max
iterations: 10,000). The torsion angles defined for the Grid Search were C29–
C14–O10–C7, C14–O10–C7–C8, O10–C7–C8–C1 and O1–C16–C33–C34 and
they were altered in 30° increments. 20,736 unique conformers were produced
for each compound. Among them, the lowest energy conformers were selected
and the resulting conformers were superimposed using the Fit Atoms tool in
SYBYL. All computational calculations were undertaken on an IntelÒ XeonTM
Quad-core workstation with Linux Cent OS release 4.6.
References and notes
1. Appendino, G.; Szallasi, A. Life Sci. 1997, 60, 681.
2. Blumberg, P. M.; Szallasi, A.; Acs, G. In Capsaicin in the study of pain; Wood, J. N.,
Ed.; Academic Press: London, 1993; pp 45–62. Chapter 3.
16 17
33
34
3. Hergenhahn, M.; Adolf, W.; Hecker, E. Tetrahedron Lett. 1975, 19, 1595.
4. Adolf, W.; Sorg, B.; Hergenhahn, M.; Hecker, E. J. Nat. Prod. 1982, 45, 347.
5. Szallasi, A.; Blumberg, P. M. Neuroscience 1989, 30, 515.
6. Szallasi, A.; Szabo, T.; Biro, T.; Modarres, S.; Blumberg, P. M.; Krause, J. E.;
Cortright, D. N.; Appendino, G. Br. J. Pharmacol. 1999, 128, 428.
7. Biro, T.; Acs, G.; Acs, P.; Modarres, S.; Blumberg, P. M. J. Investig. Dermatol. Symp.
Proc. 1997, 2, 56.
O
O
O
H
H
8
O
10
7
1
O
29
14
8. Liu, L.; Simon, S. A. Brain Res. 1998, 809, 246.
OH
9. Szallasi, A.; Joo, F.; Blumberg, P. M. Brain Res. 1989, 503, 68.
10. Bley, K. R. Expert Opin. Investig. Drugs 2004, 13, 1445.
11. Brown, D. C.; Iadarola, M. J.; Perkowski, S. Z.; Erin, H.; Shofer, F.; Laszlo, K. J.;
Olah, Z.; Mannes, A. J. Anesthesiology 2005, 103, 1052.
O
O
OH
RTX