Substituted tetrahydroquinolines as potent allosteric inhibitors of reverse transcriptase and its key mutants

Article abstract of DOI:10.1016/j.bmcl.2009.07.031

Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are key elements of multidrug regimens, called HAART (Highly Active Antiretroviral Therapy), that are used to treat HIV-1 infections. Elucidation of the structure-activity relationships of the thiocarbamate moiety of the previous published lead compound 2 provided a series of novel tetrahydroquinoline derivatives as potent inhibitors of HIV-1 RT with nanomolar intrinsic activity on the WT and key mutant enzymes and potent antiviral activity in infected cells. The SAR optimization, mutation profiles, preparation of compounds, and pharmacokinetic profile of compounds are described.

Full text of DOI:10.1016/j.bmcl.2009.07.031

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5119–5123
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Substituted tetrahydroquinolines as potent allosteric inhibitors of reverse
transcriptase and its key mutants
a
a
a
a
a
Dai-Shi Su a, , John J. Lim , Elizabeth Tinney , Bang-Lin Wan , Mary Beth Young , Kenneth D. Anderson ,
Deanne Rudd a, Vandna Munshi b, Carolyn Bahnck b, Peter J. Felock b, Meiqing Lu b, Ming-Tain Lai b,
Sinoeun Touch c, Gregory Moyer c, Daniel J. DiStefano c, Jessica A. Flynn c, Yuexia Liang d, Rosa Sanchez d,
Sridhar Prasad a, Youwei Yan a, Rebecca Perlow-Poehnelt a, Maricel Torrent a, Mike Miller b, Joe P. Vacca a,
Theresa M. Williams a, Neville J. Anthony a
*
a Department of Medicinal Chemistry and Structural Biology, Merck Research Laboratory, WP 14-3, PO Box 4, West Point, PA 19486, United States
b Antiviral Research, Merck Research Laboratory, PO Box 4, West Point, PA 19486, United States
c Vaccines and Biologics Research, Merck Research Laboratory, PO Box 4, West Point, PA 19486, United States
d Drug Metabolism, Merck Research Laboratory, PO Box 4, West Point, PA 19486, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are key elements of multidrug regimens, called
HAART (Highly Active Antiretroviral Therapy), that are used to treat HIV-1 infections. Elucidation of the
structure–activity relationships of the thiocarbamate moiety of the previous published lead compound 2
provided a series of novel tetrahydroquinoline derivatives as potent inhibitors of HIV-1 RT with nanomo-
lar intrinsic activity on the WT and key mutant enzymes and potent antiviral activity in infected cells. The
SAR optimization, mutation profiles, preparation of compounds, and pharmacokinetic profile of com-
pounds are described.
Received 12 May 2009
Revised 26 June 2009
Accepted 2 July 2009
Available online 10 July 2009
Keywords:
Non-nucleoside reverse transcriptase
inhibitor (NNRTI)
SAR
Ó 2009 Elsevier Ltd. All rights reserved.
Molecular modeling
Crystal structure
Mutant virus
Tetrahydroquinoline
Non-nucleoside reverse transcriptase inhibitors (NNRTIs) are
key elements of multidrug regimens, called HAART (Highly Active
Antiretroviral Therapy), that are used to treat HIV-1 infections.
HAART consists of combinations of nucleoside HIV reverse trans-
criptase inhibitors (NRTIs), NNRTIs, and protease inhibitors (PIs).
However, due to the propensity of HIV to rapidly mutate, the effi-
cacy and durability of HAART can be compromised. The most fre-
quent HIV RT mutations observed in patients failing therapy with
first generation NNRTIs are K103N and Y181C. Therefore, new
agents with better activity profiles against both wild type (WT)
HIV-1 and the RT K103N and Y181C strains are needed.
Tetrazole and triazole thioacetanilides, 1, were reported as po-
tent NNRTIs against WT and K103N mutant (Scheme 1).1 It is also
known that the tetrazole and triazole moiety in 1 can be replaced
by tertiary thiocarbamates (2), amides, and carbamates.1c This Let-
ter reports the discovery and SAR of a structurally novel class of
compounds related to compound 2 that exhibit good antiviral effi-
cacy against the WT enzyme and the clinically relevant mutants
K103N and Y181C.
Compound 2 exhibited desirable intrinsic potency against the
WT and key mutant K103N and was about 10-fold less potent
against Y181C (IC50: 8 nM (WT), 9 nM (K103N), 80 nM (Y181C)).2
In a cell based antiviral assay (Spread assay),3 compound 2 also
demonstrated good activities and had a minimal shift between
10% fetal bovine serum (FBS) and 50% normal human serum
(NHS) (CIC95: WT, 31 nM (10% FBS), 62 nM (50% NHS)). Conforma-
tional analysis of 2 suggested that the N-methyl group of the thio-
carbamate and the ortho-methyl group of the mesitylene could be
in close proximity. Therefore it appeared reasonable to form a bond
between the two methyl groups, to generate a tetrahydroquinoline
motif which led to compound 3 (Scheme 1 and Table 1). Compound
3 showed improved potency against the WT and K103N enzymes.
However, this modification decreased potency against the Y181C
mutant enzyme compared to the unconstrained analogue 2. Com-
pound 3 also displayed a five-fold decrease in potency in the pres-
ence of 50% NHS in the Spread assay presumably due to its high
* Corresponding author at present address: 1250 S. Collegeville Rd., UP1205,
GlaxoSmithKline, Collegeville, PA 19426, United States. Tel.: +1 610 917 5879.
E-mail address: Dai-Shi.X.Su@gsk.com (D.-S. Su).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.07.031

5120
D.-S. Su et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5119–5123
O
Cl
H
N
X
N
Cl
H
N
N
S
N
N
S
O
SO2NH2
O
SO2NH2
2
1 X = N or CH
O
Cl
H
N
Y
N
X
O
R'
R
Scheme 1.
Table 1
SAR results
O
Cl
H
N
N
6
S
O
8
R'
R
Compounds
R
R0
RT_Pola IC50 (nM)
K103N
Antiviral Activityb CIC95 (nM)
WT
Y181C
10% FBS
50% NHS
3
4
5
6
7
8
9
10
11
12
6-Me
8-Me
6-Cl
SO2NH2
SO2NH2
SO2NH2
SO2NH2
SO2NH2
H
3
6
8
2
1
4
7
2
4
2
3
4
—
4
1
—
14
2
3
2
1300
61
—
31
4
175
95
4
11
7
78
47
141
23
396
219
1250
47
8-Cl
6,8-Cl
6,8-Cl
6,8-Cl
6,8-Cl
6,8-Cl
6,8-Cl
13/18c
125
156
63c
5.6c
97c
40/33c
406
CN
1250
307c
5.7c
CONHBn
SO2CH3
CONH(CH2)3NH2
34c
a
Compounds were evaluated in a standard SPA assay. The values are the geometric mean of at least two determinations. All individual values are within 25% of the mean.
CIC95 (cell culture inhibitory concentration) is defined as the concentration at which the spread of virus is inhibited by >95% in MT-4 human T-lymphoid cells maintained
b
in RPMI 1640 medium containing either 10% fetal bovine serum (FBS) or 50% normal human serum (NHS). The antiviral activity of compound against wild-type (H9IIIB) virus
was measured. The values are the geometric mean of at least two determinations. All individual values are within 25% of the mean.
c
The antiviral activity of compound against wild-type (R8) virus was measured.
protein binding (99% in human plasma). Introducing a methyl
group at the 8-position (4) maintained WT and K103N potency,
and regained activity versus the Y181C mutant. It is known that
the mesitylene group can be replaced with a mono- or di-chloro
substituted phenyl ring.1 Therefore, a chlorine atom was intro-
duced into this series at the 6 and 8 position (5 and 6). A similar
trend was observed in this series of chlorine analogues, where
the 8-chloro compound (6) was significantly more potent than
the 6-chloro isomer (5). The 8-Chloro analogue was also more po-
tent than the corresponding methyl analogue (6 vs 4). Combining
the 6- and 8-chlorine substitution on the aryl ring provided com-
pound 7 with a further increase in potency. Compound 7 exhibited
good potency against WT enzyme, K103N, and in particular Y181C
mutants. These improvements translated into excellent antiviral
activities in cell culture.
group in 7 is orientated differently compared to the one in 3 presum-
ably to minimize its interaction with the 8-chlorine atom.
Since the 4-position of the sulfonamide is solvent exposed, we
further explored the SAR of the anilide moiety using a library-based
approach to replace the primary sulfonamide. Selected results are
listed in Table 1. Deletion of the sulfonamide led to compound 8
and reduced potency against the Y181C enzyme and activity in cell
based assays. Replacement of the sulfonamide with a small cyano
group (9) or a large benzyl amide (10) was tolerated. Replacement
of the sulfonamide with methyl sulfone yielded compound 11 with
improved antiviral activities compared to 7 and no shift in potency
in the presence of 50% human serum in the Spread assay. Incorpora-
tion of an amide side chain with a basic amine group (12) was toler-
ated and its antiviral activity was maintained. This result is
consistent with the observation from the co-crystal structures that
the para position of the anilide is solvent exposed and therefore tol-
erates a variety of substituents (Fig. 1).
Co-crystallization4 of compounds 3 and 7 with WT RT revealed
that the 6-substitutent (methyl group in 3 and chlorine atom in 7)
of the phenyl group interacts with Trp-229 and the anilide side chain
is accommodated within the solvent channel lying under Pro-236
(Fig. 1). The anilide carbonyl group in both 3 and 7 is positioned to
interact with Lys-103 backbone NH via a hydrogen bond, and this
interaction is considered to contribute to the excellent potency of
both 3 and 7 against the K103N mutant. The thiocarbamate carbonyl
We then initiated an effort to replace the thiocarbamate and
selected results are summarized in Table 2. Replacement of the
thiocarbamate with carbamate (13), amide (14) and urea (15)
linkers showed modest activities in the enzyme assay, and all
had significant potency shifts in the Spread assay in the presence
of 50% NHS. A cyclopropane ring has been used successfully as a

D.-S. Su et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5119–5123
5121
Figure 2. Modeling study results. Compound
7
is colored in cyan and cis-
Figure 1. Crystal structures of compounds 3 and 7. Compound 3 is colored in green
and 7 in cyan with nitrogen atom in blue, oxygen atom in salmon, sulfur atom in
orange, and chlorine atom in magenta.4
cyclopropyl analogue in yellow with nitrogen atom in blue, oxygen atom in
salmon, sulfur atom in orange, and chlorine atom in magenta. Details are provided
in Ref. 6.
linker in a series of structurally related analogues.5 We specu-
lated that a cyclopropyl ring could be used as a surrogate for
the sulfur-methylene linker in this series. A modeling study sug-
gested that the conformation of a cis-cyclopropyl amide could
mimic the thiocarbamate (Fig. 2).6 Based on these observations,
the sulfur methylene linker was replaced with a cis-cyclopropyl
ring. Reverse phase chiral chromatography of the racemic cis-
cyclopropyl derivative delivered the two enantiomers, the first
eluted enantiomer, 16, and 17, the second. As anticipated, the
antiviral activity and the configuration of the asymmetric centers
were linked, with enzyme inhibition residing predominantly in
16. Compound 16 is about 3–5-fold more potent in the Spread
assay than its enantiomer 17 with a small shift from 10% FBS
to 50% NHS. Based on the molecular modeling, we speculated
that the more active enantiomer 16 has the absolute configura-
tion as shown in Figure 2. Interestingly the related other ring
systems, such as, cis-1-3-cyclopentane, 1,3-cyclohexane, 2-5-tet-
rahydrofuran, 1,3-phenyl, and 2,5-thiophene, showed lower
activity (data not shown).
To assess the antiviral activities of compounds in this series
against a broad panel of clinically relevant mutants, compound 7
and 11 were selected for a PhenoScreen assay performed by Mono-
gram Bioscience.7 Both 7 and 11 have a similar profile of activity
against the panel of clinically isolated mutant viruses (Table 3).
These results further suggest that the cyclopropane function is an
isostere of the thiomethylene region of the thiocarbamate moiety.
In terms of pharmacokinetics, compounds in this series exhib-
ited extra-hepatic clearance in rats (e.g., for 7, clearance: 109 mL/
min/kg and the half-life: 0.4 h) following iv administration.8 This
may be attributed to the amide hydrolysis of compound. However
compound 7 had reasonable oral bioavailability (30%), distribution
volume (5 L/kg), and half-life (4.9 h) in dogs.8 Replacement of pri-
Table 2
SAR results: replacement of thiocarbamate
O
Y
Cl
H
N
N
X
Cl
O
SO2R
Cl
Compounds
X–Y
R
RT_Pola IC50 (nM)
K103 N
Antiviral activityb CIC95 (nM)
WT
Y181C
10% FBS
50% NHS
13
14
15
OCH2
CH2CH2
NHCH2
NH2
NH2
NH2
19
194
336
21
—
—
116
—
—
187
406
562
2500
5000
>10,000
16
CH3
CH3
18
18
99
176
922
304
922
cis, 1st eluted
cis, 2st eluted
17
210
430
6000
a
Compounds were evaluated in a standard SPA assay. The values are the geometric mean of at least two determinations. All individual values are within 25% of the mean.
CIC95 (cell culture inhibitory concentration) is defined as the concentration at which the spread of virus is inhibited by >95% in MT-4 human T-lymphoid cells maintained
b
in RPMI 1640 medium containing either 10% fetal bovine serum (FBS) or 50% normal human serum (NHS). The antiviral activity of compound against wild-type (H9IIIB) virus
was measured. The values are the geometric mean of at least two determinations. All individual values are within 25% of the mean.

5122
D.-S. Su et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5119–5123
Table 3
Antiviral potency of compounds 7 and 11 against clinically isolated mutant viruses (NL4-3 isolates)a
Mutantion
Compound 7
Compound 11
b
b
IC50 (mM)
IC50 fold shift
IC50 (mM)
IC50 fold shift
WT
L100I
K103N
Y181C
Y188L
G190A
G190S
K103N/V179E
K103N/Y181C
K103N/G190A
Y181C/G190A
K103N/P225H
K101E/G190A
K101E/Y181C/G190A
K103N/Y181C/G190A
V106A/G190A/F227L
0.0012
0.0055
0.0009
0.0168
>1.000
0.0121
0.0619
0.0022
0.0642
0.0082
0.327
0.0032
0.1131
>1.000
>1.000
>1.000
1
4.6
0.8
14
0.0045
0.0505
0.0020
0.0866
>0.25
0.1568
>0.25
0.0075
01493
0.0572
>0.25
0.0093
>0.25
>0.25
>0.25
>0.25
1
11
0.4
19
>833
10.1
51.6
1.8
53.5
6.8
272.5
2.7
94.3
>833
>833
>833
>56
34.8
>56
1.7
33.2
21.3
>56
2.1
>56
>56
>56
>56
a
Phenoscreen assay, Monogram Bioscience, in presence of 40% NHS. For details, see Ref. 7.
The IC50 is defined as the concentration of compound in cell culture required to inhibit 50% of viral replication.
b
mary sulfonamide with methyl sulfone (11) improved the clear-
metabolites were also found in rat liver microsomes, rat hepato-
cytes, rat bile, rat plasma, and human hepatocytes.
ance (84 mL/min/kg) and the half-life (1.7 h) in rats slightly. Intro-
duction of cis-cyclopropane (16) lowered clearance in both rats and
dogs (22 mL/min/kg and 14 mL/min/kg, respectively), but no
improvement in the half-life (0.5 h and 1.3 h in rats and dogs,
respectively). Further study of 7 revealed that it underwent oxida-
tion of tetrahydroquinoline ring in human liver microsomes in the
presence of NADPH. The oxidative metabolism was mediated by
CYP3A4 and could be inhibited significantly by ketoconazole, a
known inhibitor specific for human CYP3A4. These oxidative
Compounds described in this Letter were prepared according to
the route depicted in Scheme 2. Chlorination of commercially avail-
able tetrahydroquinoline with NCS provided a mixture of mono (Ia
and Ib) and bis (Ic) chlorinated derivatives (in a ratio of ꢀ3:1:1)
which can be easily separated by silica gel chromatography. Treat-
ment of the 6,8-dichloro tetrahydroquinoline (Ic) with phosgene,
coupling with methyl thioglycolate, followed by hydrolysis of the
methyl ester afforded the thiocarbamate acid (II). The final com-
NH
Cl
NH
NH
NH
a
Cl
O
+
+
ratio: ~3:1:1
Cl
Cl
Ic
Ib
Ia
O
Cl
H
N
OH
NH
Cl
c
N
S
N
S
Cl
b
Cl
O
O
Cl
SO2NH2
Cl
H2N
Cl
7
Ic
II
Cl
SO2NH2
e
O
O
O
O
OH
NH
d
N
OH
N
O
O
Cl
O
cis-III
cis-IV
Cl
SO2Me
O
O
f
N
N
H
Cl
Cl
Cl
H2N
cis-16/17
SO2Me
Cl
Scheme 2. Reagents and conditions: (a) NCS, DCM, (b) (i) phosgene, DCM, pyridine, 0 °C, 40%; (ii) methyl thioglycolate, pyridine, microwave, 90 °C, 30 min, (iii) 5 N NaOH,
73%; (c) oxalyl chloride, DMF (cat.), THF, 0 °C, 23%; (d) tBuOK, THF, 72%; (e) NCS, microwave, 60 °C, 2 h, 13–51%; (f) (COCl)2, DMF (cat), DCM, 0 °C, 30 min, then aniline,
pyridine, microwave, 60 °C, 1 h, 35%.

D.-S. Su et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5119–5123
5123
pound could then be prepared by converting the acid to the acid
References and notes
chloride with oxalyl chloride and reaction with the requisite aniline.
The carbamate, urea, and amide analogues were synthesized in an
analogous fashion. The cyclopropane derivatives 16 and 17 were
synthesized from commercially available 3-oxabicycl-[3.1.0]-hex-
ane-2,4-dione. The coupling of the chlorinated tetrahydroquinoline
Ic with 3-oxabicycl-[3.1.0]-hexane-2,4-dione was not successful un-
der numerous conditions presumably due to the steric hindrance
caused by the 8-chlorine atom. Therefore, the parent tetrahydro-
quinoline was deprotonated with KOtBu and acylated with 3-oxabi-
cycl-[3.1.0]-hexane-2,4-dione to form the cis-cyclopropane car-
boxylic acid (III) and was then chlorinated. The chlorination with
NCS under thermal conditions was unsuccessful; however, micro-
wave heating condition yielded dichlorinated tetrahydroquinoline
acid (IV) in modest yield. The anilide formation via the intermediate
acid chloride afforded the racemate which was resolved into 16 and
17 by chiral HPLC.
1. (a) Muraglia, E.; Kinzel, O. D.; Laufer, R.; Miller, M. D.; Moyer, G.; Munshi, V.;
Orvieto, F.; Cecilla, P.; Pescatore, G.; Rowley, M.; Williams, P. D.; Summa, V.
Bioorg. Med. Org. Lett. 2006, 16, 2748; (b) De La Rosa, M.; Kim, H. W.; Gunic, E.;
Jenket, C.; Boyle, U.; Koh, Y.-H.; Korboukh, I.; Allan, M.; Zhang, W.; Chen, H.; Xu,
W.; Nilar, S.; Yao, N.; Hamatake, R.; Lang, S.; Hong, Z.; Zhang, Z.; Girardet, J.-L.
Bioorg. Med. Org. Lett. 2006, 16, 4444; (c) O’Meara, J. A.; Jakalian, A.; LaPlante, S.;
Bonneau, P. R.; Coulombe, R.; Faucher, A.-M.; Guse, I.; Landry, S.; Racine, J.;
Simoneau, B.; Thavonekham, B.; Yoakim, C. Bioorg. Med. Org. Lett. 2007, 17, 3362;
(d) Gagnon, A.; Amad, M. H.; Bonneau, P. R.; Coulombe, R.; DeRoy, P. L.; Doyon,
L.; Duan, J.; Garneau, M.; Guse, I.; Jakalian, A.; Jolicoeur, E.; Landry, S.; Malenfant,
E.; Simoneau, B.; Yoakim, C. Bioorg. Med. Org. Lett. 2007, 17, 4437.
2. Experimental protocols for the HIV-1 RT polymerase SPA assay are described in
the following patent application: Saggar, S. A.; Sisko, J. T.; Tucker, T. J.; Tynebor, R.
M.; Su, D.-S.; Anthony, N. J. Preparation of Heterocyclic Phenoxy Compounds as
HIV Reverse Transcriptase Inhibitor. US Patent Appl. US 2007/021442A1, 2007.
3. Vacca, J. P.; Dorsey, B. D.;Schleif, W. A.;Levin, R. B.; McDaniel, S. L.;Darke, P. L.; Zygay,
J.; Quintero, J. C.; Blahy, O. M.; Roth, E.; Sardana, V. V.; Schlabach, A. J.; Graham, P. I.;
Condra, J. H.; Gotlib, L.; Holloway, M. K.; Lin, J.; Chen, I.; Vastag, K.; Ostovic, D.;
Anderson, P. S.; Emini, E. A.; Huff, J. R. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 4096.
4. Compounds
3 and 7 were soaked into a pre-existing crystal of HXB2 RT
In summary, we have designed and synthesized a series of novel
tetrahydroquinoline derivatives as potent inhibitors of HIV-1 RT
with nanomolar intrinsic activity on the WT and key mutant en-
zymes and potent antiviral activity in infected cells. Based on the
modeling study results, tetrahydroquinoline amides have been
identified as an effective isostere for the N-methyl mesitylene
amide moiety, and the cis-cyclopropyl ring was used as a surrogate
for the sulfur-methylene linker. Compounds in this series tended
towards high plasma clearance in rats after iv administration.
CYP3A4 mediated oxidation of the tetrahydroquinoline ring was
identified as a significant metabolic pathway in vitro and in vivo.
These compounds represent a novel and potent NNRTIs with activ-
ity against a number of clinically significant mutant viruses.
containing a weakly bound efavirenz analog. Electron density was mapped at a
resolution of 3.0 Å (3) and 2.7 Å (7) and the structure solved by Fourier
synthesis. The distances between carbonyl O and Lys103-N are 2.88 Å and 2.74 Å
for compound 3 and 7, respectively. The structures have been deposited in the
PDB; PDB codes are 310R and 310S for compounds 3 and 7, respectively.
5. Sahlberg, C.; Noreen, R.; Engelhardt, P.; Hogberg, M.; Kangasmetsa, J.; Vrang, L.;
Zhang, H. Bioorg. Med. Org. Lett. 1998, 8, 1511.
6. Approximately 100 conformations of each compound were generated using two
algorithms: a distance geometry (see, (a) Crippen, C. M.; Havel, T. F. In Distance
Geometry and Molecular Conformation; Bawden, E., Ed.; Research Studies Press:
Wiley, New York, 1988; (b) Kuszewski, J.; Nilges, M.; Brunger, A. T. J. Biomol.
NMR 1992, 2, 33) and a knowledge based method (see, Feuston, B. P.; Miller, M.
D.; Culberson, J. C.; Nachbar, R. B.; Kearsley, S. L. J. Chem. Inf. Comput. Sci. 2001,
41, 754.). All conformations were then subjected to energy minimization using a
distance dependent dielectric constant of compound and applying the MMFFs
forcefield (Halgren, T. A. J. Comput. Chem. 1996, 17, 490, 520, 553, 616; Halgren,
T. A.; Nachbar, R. B. J. Comput. Chem. 1996, 17, 587.).
7. Assays performed by Monogram Bioscience, San Francisco, CA. Values are the
average of two determinations. Detailed assay protocols are available at: http://
www.monogramhiv.com/assays/hcp/phenolHIVTechnology.aspx.
8. For rat singles, Sprague-Dawley rats (n = 3), po: 10 mg/kg, iv: 2 mg/kg. For dog
singles, Beagle dogs (n = 2), po: 1 mg/kg, iv: 0.5 mg/kg. For rat iv cassette (co-
administration with other analogues): 0.25 mg/kg, For dog iv cassette (co-
administration with other analogues): 0.125 mg/kg. Interanimal variability was
less than 20%.
Acknowledgments
We are pleased to acknowledge the efforts of Drs. C. W. Ross,
S. M. Pitzenberger, Mr. S. L. Varga, and Mrs. J. S. Murphy. We are
grateful to Mr. Thomas J. Tucker for helpful discussions.