Synthesis and biological activities of potential metabolites of the non-nucleoside reverse transcriptase inhibitor efavirenz

Article abstract of DOI:10.1016/S0960-894X(01)00012-9

Studies on the biotransformation of the clinically important non-nucleoside reverse transcriptase inhibitor efavirenz have shown that oxidation and secondary conjugation are important components of the processing of this molecule in vivo. We have synthesized metabolites of efavirenz to confirm their structure and to evaluate their activity as antivirals.

Full text of DOI:10.1016/S0960-894X(01)00012-9

Bioorganic& Medicinal Chemistry Letters 11 (2001) 619±622
Synthesis and Biological Activities of Potential Metabolites of the
Non-Nucleoside Reverse Transcriptase Inhibitor Efavirenz
Jay A. Markwalder,* David D. Christ Abdul Mutlib, Beverly C. Cordova,
Ronald M. Klabe and Steven P. Seitz
DuPont Pharmaceuticals Company, Experimental Station, PO Box 80500, Wilmington, DE 19880-0500, USA
Received 20 October 2000; accepted 12 December 2000
AbstractÐStudies on the biotransformation of the clinically important non-nucleoside reverse transcriptase inhibitor efavirenz
have shown that oxidation and secondary conjugation are important components of the processing of this molecule in vivo. We
have synthesized metabolites of efavirenz to con®rm their structure and to evaluate their activity as antivirals. # 2001 DuPont
Pharmaceuticals Company. Published by Elsevier Science Ltd. All rights reserved.
The chemotherapy of HIV infection is most e€ective
when mechanistically distinct agents are used in combi-
nation. The ability to suppress viral replication to very
low levels and the necessity to accrue multiple muta-
tions to circumvent agents acting at di€erent sites pre-
sumably contribute to the e€ectiveness of this approach.
There are currently three classes of agents used in the
chemotherapy of HIV infection: nucleoside competitive
reverse transcriptase inhibitors, non-nucleoside reverse
transcriptase inhibitors, and HIV protease inhibitors.
The non-nucleoside reverse transcriptase inhibitor
efavirenz 1 (SustivaTM) was approved by the FDA in
1998 and has become an important component of com-
bination treatment.
some cases subsequently conjugated. The major species
observed in rat urine was a glucuronide. In order to
con®rm structural assignments and obtain sucient
quantities for biological evaluation, we endeavored to
synthesize metabolites of efavirenz.
The syntheticroutes used to prepare the metabolites
parallel the methodology used to prepare efavirenz.2
This chemistry employs directed ortho-metallation of an
anilide derivative to introduce a tri¯uoromethyl ketone.
Addition of acetylides to the tri¯uoromethyl ketone
then gives derivatives suitable for further elaboration
into the benzoxazinone ring system.
The synthesis of the 7-hydroxy compound is shown in
Scheme 1. 2-Chloro-5-nitroanisole 2 was demethylated
and converted to a tri-isopropylsilyl ether. The nitro
group was then reduced to the aniline, which was then
derivatized as the pivalamide. Lithiation of 3 occurred
smoothly, adjacent to the directing amide group. The
organometallicgenerated in this fashion was quenched
with ethyl tri¯uoroacetate to provide the required tri-
¯uoromethyl ketone. The activating pivalamide was
then hydrolyzed and the phenol reprotected to give 4.
The presence of the TIPS protecting group was crucial
for the success of this synthetic sequence. Metallation of
the methoxy version of 3 gave the organolithium
between the amide and methoxy. The bulky silyl ether
e€ectively suppresses this favorable process. The
remainder of the synthesis was completed by addition of
lithium cylopropylacetylide to the tri¯uoromethyl
ketone followed by cyclization with phosgene and
deprotection to give the target molecule 5.
In the course of safety assessment studies on efavirenz, a
number of metabolites were observed in rat urine.1
Mass spectral analysis suggested that many of these
compounds were oxidized species, which had been in
*Corresponding author. Tel.: +1-302-695-1882; e-mail: jay.a.
markwalder@dupontpharma.com
0960-894X/01/$ - see front matter # 2001 DuPont Pharmaceuticals Company. Published by Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00012-9

620
J. A. Markwalder et al. / Bioorg. Med. Chem. Lett. 11 (2001) 619±622
The synthesis of the 8-hydroxy isomer is described in
Scheme 2. 3-Chloro-6-nitroanisole was converted to a
pivalamide derivative 7, which was metallated as before.
In this example the strong amide directing group e€ec-
tively controls the lithiation to give the desired tri-
¯uoromethylketone. The methyl ether was then
converted to the more labile silyl ether. This exchange
was performed prior to the introduction of the acetylene
functionality to avoid exposing advanced intermediates
to strongly Lewis acidic conditions necessary for the
methyl group cleavage. Addition of lithium cyclopro-
pylacetylide to 8 gave the desired carbinol in high yield.
The cyclization of 9 proved to be somewhat more chal-
lenging than anticipated. Treatment of 9 with carbon-
yldiimidazole followed by removal of the silyl ether gave
quinoline 10 as the major product. This reaction is
analogous to results recently published by chemists at
Merck.3 In that work, the reaction was proposed to
proceed through the formation of a highly strained
cyclic allene intermediate. The formation of quinoline
products can be suppressed by modi®cation of the
cyclization conditions. Thus, treatment of 9 with phos-
gene and Hunig's base followed by silyl ether removal
gave the target molecule 11 in good overall yield.
The 5-position is the last site of benzene ring hydrox-
ylation. The synthetic approach to this compound
parallels that used for the 7- and 8-isomers. We thus
prepared the tri¯uoromethyl ketone 12 (Scheme 3).
Upon addition of lithium cyclopropylacetylide, a silyl
migration occurred to give the phenol 13. Attempts to
deprotect 13 with ¯uoride gave the quinoline product
14. The formation of quinoline products has thus far
frustrated e€orts to prepare the 5-hydroxylated analogue.
Scheme 1. Synthesis of 7-hydroxy metabolite. Reagents and condi-
tions: (a) BBr3, CH2Cl2; (b) TIPSOTf, imidazole, DMF; (c) H2, Pd/C,
EtOAc; (d) (CH3)3CCOCl, Et3N, CH2Cl2 (60% for 4 steps); (e) s-
BuLi, ether, 0 ꢀC to rt, CF3CO2Et; (f) HCl, DME, H2O; (g) TIPSOTf,
imidazole, DMF (19% for 3 steps); (h) lithium cyclopropylacetylide,
THF±hexanes, À20 ꢀC; (i) COCl2, i-Pr2NEt, toluene, À20 ꢀC; (j) n-
Bu4NF, THF (56% for 3 steps).
Many of the metabolites observed in vivo have under-
gone secondary processing to form glucuronides or sul-
fates. The synthesis of two of these secondary
metabolites is shown in Scheme 4. Compound 11 was
®rst resolved by chiral phase SCFC.4 The two antipodes
were isolated and characterized ([a]2D5 À29.3ꢀ (c 0.35,
MeOH) and [a]2D5 +30.7ꢀ (c 0.16, MeOH)). The levo-
rotatory isomer was shown to possess the efavirenz
stereochemistry in two ways. First, a small sample of
glucuronide from rat urine was treated with b-glucur-
onidase to give an aglycone that was identical to the
levo-isomer by chiral SCFC. Second, the two isomers
were evaluated for their ability to inhibit the reverse
transcriptase enzyme (vide infra). The residual inhibi-
tory activity resided with the levo-isomer, consistent
Scheme 2. Synthesis of the 8-hydroxy metabolite. Reagents and con-
ditions: (a) SnCl2, EtOH, re¯ux; (b) (CH3)3CCOCl, NEt3, CH2Cl2
(97% for 2 steps); (c) s-BuLi, THF, À20 ꢀC to 0 ꢀC, CF3CO2Et; (d)
HCl, DME, H2O, re¯ux; (e) BBr3, CH2Cl2; (f) TBDMSOTf, DMF,
imidazole, 0 ꢀC (60% for 4 steps); (g) lithium cyclopropylacetylide,
THF±hexanes, À20 ꢀC (76%); (h) carbonyldiimidazole, CH2Cl2, 0 ꢀC
to rt; (i) n-Bu4NF, THF (53% for 2 steps); (j) COCl2, i-Pr2NEt,
toluene, À25 ꢀC to rt; (k) n-Bu4NF, THF (66% for 2 steps).
Scheme 3. Attempted synthesis of the 5-hydroxy metabolite. Reagents
and conditions: (a) lithium cyclopropylacetylide, THF±hexanes,
À20 ꢀC; (b) n-Bu4NF, THF.

J. A. Markwalder et al. / Bioorg. Med. Chem. Lett. 11 (2001) 619±622
Table 1. Biological data for efavirenz and its metabolites11
621
Compound
RT IC50 (nM)
RNA IC90 (nM)
1
47(Æ25)
2341
>12,500
2450
9489
1.7(Æ0.5)
63
>1507
422
>985
1014
33
15
16
5
17
18
22
>12,500
1983
hydroxylated analogue. The requisite acetylene was
prepared as shown in Scheme 5. The commercial 1-
hydroxycyclopropane carboxylic acid 19 was converted
in a four-step sequence to the protected aldehyde 20.
The aldehyde was converted to dibromoole®n 21 using a
modi®ed Corey procedure.9 The lithium acetylide was
then generated by exposure of 21 to 2 equiv of n-butyl-
lithium. This species reacted smoothly with 4,6-
dichloro-4-tri¯uoromethyl-1,4-dihydro-2H-3,1-benzox-
azin-2-one10 to provide, after deprotection with tetra-n-
butylammonium ¯uoride, carbinol 22.
Scheme 4. Synthesis of conjugated metabolites. Reagents and condi-
tions: (a) methyl 2,3,4-tri-O-acetyl-d-glucuronate, Ph3P, DEAD,
THF±CH2Cl2; (b) LiOH, THF±H2O; (c) sulfur trioxide±pyridine
complex, pyridine.
The availability of the compounds described in this
paper allowed the unambiguous identi®cation of many
of the metabolites seen in animals dosed with efavirenz.
Compounds 15, 17, and 18 were shown to be identical
to compounds seen in rats, guinea pigs, hamsters and
cynomolgus monkeys. Compound 5 was only observed
in guinea pigs, cynomolgus monkeys and humans. The
carbinol 22 was not detected in any species. Metabolites
were found that had the cyclopropyl methine oxidized;
however, these all had additional oxidation on the ben-
zene ring. Thus, it is likely that methine oxidation
requires prior oxidation of the benzene ring.
We were also interested in determining the activity of
these metabolites as antivirals. As shown in Table 1, all
of the compounds disclosed are signi®cantly weaker
inhibitors of reverse transcriptase and poorer antivirals
than efavirenz. The most potent antiviral 22 has not
been observed in animals. Of the metabolites evaluated
to date, none is of comparable potency to efavirenz. It is
unlikely that a signi®cant portion of the antiviral activ-
ity observed with efavirenz in vivo is attributable to the
metabolites characterized thus far.
Scheme 5. Synthesis of oxidized side-chain analogues. Reagents and
conditions: (a) MeOH, SOCl2; (b) i-Pr3SiOTf, 2,6-lutidine; (c) Dibal,
À78ꢀC; (d) oxalyl chloride, DMSO, NEt3, CH2Cl2; (e) CBr4, Ph3P,
NEt3; (f) n-BuLi, À78ꢀC; (g) 4,6-dichloro-4-tri¯uoromethyl-1,4-dihydro-
2H-3,1-benzoxazin-2-one, À60 ꢀC; (h) n-Bu4NF, THF.
with the earlier results.5 These results suggest that the
levo-isomer 15 has the (S) and the dextro-isomer 16 has
the (R) con®guration.
The carbohydrate was attached as shown in Scheme 4.
Compound 15 was allowed to react with methyl 2,3,4-
tri-O-acetyl-d-glucuronate6 under Mitsunobu condi-
tions7 to give the protected glucuronide as a mixture of
anomers. Saponi®cation of the esters was accomplished
with lithium hydroxide. The mixture of glycosidic link-
age isomers was separated by HPLC and the b-anomer
identi®ed by its characteristic vicinal coupling constant
(J1,2=7.3 Hz). Sulfation was accomplished by exposure
of 15 to sulfur trioxide±pyridine complex8 to give the
pyridine salt 18.
Acknowledgements
The authors would like to thank Al Mical for his expert
assistance on chiral chromatography.
References and Notes
1. Mutlib, A. E.; Chen, H.; Nemeth, G. A.; Markwalder, J. A.;
Seitz, S. P.; Gan, L. A.; Christ, D. D. Drug Met. Disp. 1999,
27, 1319.
2. Pierce, M. E.; Parsons, R. L., Jr.; Radesca, L. A.; Young,
Y. S.; Silverman, S.; Moore, J. R.; Islam, Q.; Choudhury, A.;
The cyclopropylacetylene group is also a site of poten-
tial hydroxylation. As the cyclopropyl methine is the
most activated site, we initiated the synthesis of this

622
J. A. Markwalder et al. / Bioorg. Med. Chem. Lett. 11 (2001) 619±622
Fortunak, J. M. D.; Nguyen, D.; Luo, C.; Morgan, S. J.;
Davis, W. P.; Confalone, P. N. J. Org. Chem. 1998, 63, 8536.
3. Frey, L. F.; Tillyer, R. D.; Ouellet, S. G.; Reamer, R. A.;
Grabowski, E. J. J.; Reider, P. J. J. Am. Chem. Soc. 2000, 122,
1215.
6. Herzig, J.; Nudelman, A.; Gottlieb, H. E. Carbohydr. Res.
1988, 177, 21.
7. Budman, G. T.; Green, D. V. S.; Voyle, M. J. Organomet.
Chem. 1990, 338, 117.
8. Bernauer, K.; Borgulya, J. Synthesis 1983, 29.
9. Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 36, 3769.
10. Christ, D. D.; Cocuzza, A. J.; Ko, S. S.; Markwalder, J.
A.; Mutlib, A. E.; Parsons, R. L., Jr.; Patel, M.; Seitz, S. P.;
U.S. Patent 5 874 430, 1999; Chem. Abstr. 1999, 130, 196659.
11. Assays were performed as described in: Corbett, J. W.;
Ko, S. S.; Rodgers, J. D.; Je€rey, S.; Bacheler, L. T.; Klabe,
R. M.; Diamond, S.; Lai, C.-M.; Rabel, S. R.; Saye, J. A.;
Adams, S. P.; Trainor, G. L.; Anderson, P. S.; Erickson-
Viitanen, S. K. Antimicrob. Agents Chemother. 1999, 43, 2893.
4. Column: Chiralpak AD1 (Chiral Technologies Inc.) Sol-
vent: 25% methanol, 75% CO2. Pressure: 150 atm. The (S)-
isomer (15) elutes prior to (R)-isomer (16).
5. Young, S. D.; Britcher, S. F.; Tran, L. O.; Payne, L. S.;
Lumma, W. C.; Lyle, T. A.; Hu€, J. R.; Anderson, P. S.;
Olsen, D. B.; Carroll, S. S.; Pettibone, D. J.; O'Brien, J. A.;
Ball, R. G.; Balani, S. K.; Lin, J. H.; Chen, I-W.; Schleif,
W. A.; Sardana, V. V.; Long, W. J.; Byrnes, V. W.; Emini,
E. A. Antimicrob. Agents Chemother. 1995, 39, 2602.