An efficient and practical synthesis of the HIV protease inhibitor atazanavir via a highly diastereoselective reduction approach

Article abstract of DOI:10.1021/op7001563

An efficient and practical synthesis of the HIV-1 protease inhibitor Atazanavir was developed by employing the diastereoselective reduction of ketomethylene aza-dipeptide isostere 10 as the key and final step. The high diastereoselectivity of the amino ketone reduction by lithium tri-iert-butoxyaluminum hydride in diethyl ether to afford the desired svn-1,2-amino alcohol structure was achieved by Felkin- Anh control as a result of the bulky and chiral N-(methoxycarbonyl)-L-tert-leucinyl moiety as the nitrogen protecting group. The coupling of the two key intermediates, N-(methoxycarbonyl)-L-tert-leucine acylated benzyl hydrazine 7 and chloromethyl ketone 9, via an S_N2 reaction furnished the amino ketone 10 in high yield under our optimized conditions. Our new methodology features the late introduction of the S-hydroxyl group and the early acylation of benzyl hydrazine and chloromethyl ketone with N-(methoxycarbonyl)-L-tert-leucine, respectively, which confers high efficiency and easy purification.

Full text of DOI:10.1021/op7001563

Organic Process Research & Development 2008, 12, 69–75
An Efficient and Practical Synthesis of the HIV Protease Inhibitor Atazanavir via a
Highly Diastereoselective Reduction Approach
Xing Fan, Yan-Li Song, and Ya-Qiu Long*
State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Shanghai Institutes for Biological Sciences,
Chinese Academy of Sciences, 555 Zuchongzhi Road, Shanghai 201203, China
Abstract:
An efficient and practical synthesis of the HIV-1 protease inhibitor
Atazanavir was developed by employing the diastereoselective
reduction of ketomethylene aza-dipeptide isostere 10 as the key
and final step. The high diastereoselectivity of the amino ketone
reduction by lithium tri-tert-butoxyaluminum hydride in diethyl
ether to afford the desired syn-1,2-amino alcohol structure was
achieved by Felkin-Anh control as a result of the bulky and chiral
N-(methoxycarbonyl)-
tecting group. The coupling of the two key intermediates, N-
(methoxycarbonyl)- -tert-leucine acylated benzyl hydrazine 7 and
chloromethyl ketone 9, via an SN2 reaction furnished the amino
ketone 10 in high yield under our optimized conditions. Our new
methodology features the late introduction of the S-hydroxyl group
and the early acylation of benzyl hydrazine and chloromethyl
L-tert-leucinyl moiety as the nitrogen pro-
Figure 1. Chemical structure of Atazanavir.
L
epoxide (2), which was prepared from L-Boc-phenylalanine
1a5–8 or chiral diol 1b.9
The preparation of the key intermediate (2S,3R)-epoxide
usually requires 3 or 4 steps of transformation starting from a
chiral building block,10 which endowed a higher cost and several
limitations such as epimerization and problemic purification.
On the other hand, the double coupling of the free amine of 5
with N-methoxycarbonyl-L-tert-leucine (Scheme 1) in the final
step always was accompanied with low yield and isolation
problems. So we tried to develop a concise and practical process
to prepare Atazanavir. We herein report a more efficient and
practical large-scale synthesis of the title compound by employ-
ing a convergent approach with a highly diastereoselective
reduction as the key and last step.
New Strategy to Synthesize Atazanavir. As illustrated in
Scheme 2, we designed a brand new strategy to synthesize
Atazanavir. The distinct feature of our synthetic route is the
late introduction of the (S)-hydroxyl group on the aza-dipeptide
isostere structure via an asymmetric reduction in the final step.
Correspondingly, the key amino ketone 10 was assembled from
the fragments 7 and 9 via an SN2 reaction. It is noteworthy that
the early introduction of the N-methoxycarbonyl-L-tert-leucinyl
moiety in the two fragments 7 and 9 is another feature of our
ketone with N-(methoxycarbonyl)-L-tert-leucine, respectively, which
confers high efficiency and easy purification.
Introduction
Atazanavir, trade name Reyataz (formerly known as BMS-
232632), is an antiretroviral drug of the protease inhibitor (PI)
class, approved by the FDA in June of 2003.1,2 Like other
antiretrovirals, it is used to treat infection of human immuno-
deficiency virus (HIV). Unlike most protease inhibitors, Ata-
zanavir appears not to increase cholesterol, triglycerides, or
blood sugar levels, which is a problem to various degrees with
all other PIs.3 Furthermore, the good oral bioavailability and
favorable pharmacokinetic profile enables Atazanavir to be the
first once-a-day protease inhibitor to treat AIDS.4 This can
provide a benefit for people seeking a simplified dosing regimen.
Atazanavir is an aza-peptidomimetic HIV-1 protease inhibi-
tor, bearing an aza-dipeptide isostere structure (Figure 1). The
focus of its synthesis is to construct the aza-dipeptide skeleton
with desired stereochemistry. As outlined in Scheme 1, the
original synthesis of Atazanavir was accomplished through the
nucleophilic attack of the benzylhydrazine (4) on the (2S,3R)-
(5) Giordano, C.; Pozzoli, C.; Benedetti, F. W. O. Patent 012083, 2001.
(6) Fassler, A.; Bold, G.; Capraro, H. G.; Steiner, H. W. O. Patent
9746514, 1997.
(7) Fassler, A.; Bold, G.; Steiner, H. Tetrahedron Lett. 1998, 39, 4925–
4928.
(8) Bold, G.; Faessler, A.; Capraro, H.; Cozens, R.; Klimkait, T.; Lazdins,
J.; Mestan, J.; Poncioni, B.; Roesel, J.; Stover, D.; Tintelnot-Blomley,
M.; Acemoglu, F.; Beck, W.; Boss, E.; Eschbach, M.; Huerlimann,
T.; Masso, E.; Roussel, S.; Ucci-Stoll, K.; Wyss, D.; Lang, M. J. Med.
Chem. 1998, 41, 3387–3401.
* To whom correspondence should be addressed. Phone and Fax: +86-21-
50806876.E-mail: yqlong@mail.shcnc.ac.cn.
(1) Piliero, P. J. Expert. Opin. InVest. Drugs 2002, 11, 1295–1301.
(2) New drugs on the horizon. In Project Inform Wise Words 2003, 13
(Dec), 7-8; http://img.thebody.com/legacyAssets/51/62/ww_0312.pdf.
(3) Sanne, I.; Piliero, P.; Squires, K.; Schnittman, S. J. Acquired Immune
Defic. Syndr. 2003, 32, 18–29.
(9) Xu, Z.; Singh, J.; Schwinden, M. D.; Zheng, B.; Kissick, T. P.; Patel,
B.; Humora, M. J.; Quiroz, F.; Dong, L.; Hsieh, D.-M.; Heikes, J. E.;
Pudipeddi, M.; Lindrud, M. D.; Srivastava, S. K.; Kronenthal, D. R.;
Mueller, R. H. Org. Process Res. DeV. 2002, 6, 323–328.
(10) For an excellent review of HIV-1 protease inhibitor synthesis, see:
Izawa, K.; Onishi, T Chem. ReV. 2006, 106, 2811–2827, and references
therein.
(4) Colombo, S.; Buclin, T.; Cavassini, M.; Décosterd, L. A.; Telenti,
A.; Biollaz, J; Csajka, C. Antimicrob. Agents Chemother. 2006, 50,
3801–3808.
10.1021/op7001563 CCC: $40.75
Published on Web 01/04/2008
2008 American Chemical Society
Vol. 12, No. 1, 2008 / Organic Process Research & Development
•
69

Scheme 1. Literature-reported synthesis of Atazanavir via an epoxide opening approach8,9
Scheme 2. New strategy to synthesize Atazanavir by employing an asymmetric reduction as the key step
strategy, in which the N-methoxycarbonyl-L-tert-leucinyl group
acts as a protective group as well as the functional group, thus
eliminating the need to protect and deprotect the amino
functionality.
Synthesis of N-(Methoxycarbonyl)-L-tert-leucine Acylated
Benzylhydrazine (7) and Chloromethyl Ketone (9). The
synthesis of the N-(methoxycarbonyl)-L-tert-leucinyl benzyl-
hydrazine (7) is shown in Scheme 3. Nickel(II)-catalyzed
Kumada coupling of the Grignard reagent prepared from
4-bromobenzaldehyde dimethyl acetal (12) to 2-bromopyridine,
followed by acidic hydrolysis, gave the corresponding 4-(py-
ridin-2-yl)-benzaldehyde (3).8 The EDCI/HOBt-mediated cou-
pling of tert-butyl carbazate with N-(methoxycarbonyl)-L-tert-
leucine (8) followed by treatment with 6 N HCl furnished
another building block 6 in 95% yield. The subsequent
condensation of the aldehyde (12) with N-(methoxycarbonyl)-
L-tert-leucinyl hydrazine (6) was carried out in hot ethanol. The
resulting mixture was subjected to the next reaction without
further purification. In general, the hydrazone can be converted
into the corresponding hydrazine by hydrogenation or reduction
with borohydride reagents. However, in our case, the hydro-
genation under the original conditions (Pd/C, MeOH, and 1 atm
H2)5,6 required 3 days to be completed. Replacement of Pd/C
with Raney Ni resulted in a more sluggish reaction. Chemical
reduction using NaBH4 or NaHB(OAc)3 led to product mixtures
with only 30-40% yield of compound 7. We examined the
more active catalyst of Pd(OH)2/C in ethanol, and to our delight,
the hydrogenation proceeded quantitatively under 1 atm H2 in
12 h. Recrystallization from iPrOH-EtOH furnished the desired
Results and Discussion
According to our designed synthetic route (Scheme 2), we
started to synthesize Atazanavir via a more convergent approach.
The two key intermediates, benzyl hydrazine 7 and chlorom-
ethyl ketone 9, were prepared from pyridinyl benzaldehyde 3
and (S)-3-amino-1-chloro-4-phenylbutan-2-one, respectively.
The union of 7 and 9 via an SN2 reaction afforded the
key precursor 10, in which the base and solvent were
optimized, with NaHCO3 in CH3CN being the best choice.
An efficient procedure was developed for the diastereo-
selective reduction of the amino ketone into a syn-amino
alcohol, which served as the key step of the synthesis of
Atazanavir using our new strategy. The asymmetric
reduction parameters such as reducing agent, solvent, and
temperature were optimized for satisfying diastereoselec-
tivty and yield for a large-scale synthesis. The bulky
reductant lithium tri-tert-butoxyaluminum hydride (LTBA)
in diethyl ether proved most effective to predominantly
produce the desired threo amino alcohol via a Felkin-Anh
control.
70
•
Vol. 12, No. 1, 2008 / Organic Process Research & Development

Scheme 3. Synthesis of key intermediates benzylhydrazine (7) and chloromethyl ketone (9)
Scheme 4. Synthesis of Atazanavir
N-(methoxycarbonyl)-L-tert-leucinyl benzylhydrazine 7 in an
overall isolated yield of 78% with ∼100 area % (AP).
Another key intermediate 9 was synthesized conveniently.
As indicated in Scheme 3, the acidic treatment of (S)-tert-butyl
4-chloro-3-oxo-1-phenylbutan-2-ylcarbamate (13) followed by
acylation with N-(methoxycarbonyl)-L-tert-leucine (8) via a
mixed anhydride method afforded the N-(methoxycarbonyl)-
L-tert-leucinyl chloromethyl ketone (9) in 72% overall isolated
yield with 98 AP through recrystallization from petroleum ether
and ethyl acetate. No epimerization was observed under our
reaction conditions. The N-protected R-amino chloromethyl
ketone 13 can be prepared conveniently from L-phenylalanine
in a one-pot three-step sequence through a mixed anhydride
and diazoketone intermediates,11 but for large-scale synthesis,
the starting material 13 was purchased from Desano Company
in Shanghai.
Development of the Diastereoselective Reduction of
Amino Ketone to the syn-Amino Alcohol and Synthesis of
Atazanavir. With the two key intermediates 7 and 9 in hand,
we turned to the synthesis of Atazanavir via a convergent route
as depicted in Scheme 4.
The coupling reaction between 7 and 9 was investigated first.
As indicated in Table 1, the reaction conditions were optimized
in terms of the base, solvent, and temperature. We found that
the SN2 reaction performed in DMF with K2CO3/NaI or in
acetone with NaHCO3/NaI at 50 °C only gave 30% and 50%
yield, respectively. Raising the temperature to 60 °C resulted
in a worse conversion, producing less than 10% yield of the
desired compound 10. Because amino acid derived chloromethyl
ketones readily undergo a general dehydrohalogenation reaction
under mild, basic, non-nucleophilic conditions to afford R,ꢀ-
unsaturated ketones,12 we supposed that the dipeptidyl chlo-
romethyl ketone 9 might undergo a similar transformation to
give unsaturated ketones as the major product under the high
temperature and basic conditions. So, a mild condition for the
SN2 reaction was investigated by lowering the reaction tem-
perature and using the weaker base of NaHCO3 in DMF. As
we expected, the optimized condition of NaHCO3/NaI in DMF
at room temperature readily provided a clean reaction, but the
difficult disposal of DMF solvent renders this method unat-
tractive. We examined MeCN as the solvent and NaHCO3 as
the base with catalytic NaI, and a complete coupling could be
achieved reliably in 12 h at room temperature. Facile workup
to remove unreacted 7 with KHSO4 aqueous solution furnished
compound 10 in yield of 96% with 92 AP.
The final and key step was how to install the S-hydroxyl
group in the aza-dipeptide isostere portion. In general, the
hydroxyl group of the amino alcohol is installed either by the
addition of an organometallic reagent to an amino aldehyde or
(12) Gordon, E. M.; Plušcˇec, J.; Delaney, N. G.; Natarajan, S.; Sundeen,
(11) Albeck, A.; Persky, R. Tetrahedron 1994, 50, 6333–6346.
J. Tetrahedron Lett. 1984, 25, 3277–3280.
Vol. 12, No. 1, 2008 / Organic Process Research & Development
•
71

Table 1. Coupling of N-(methoxycarbonyl)-L-tert-leucinyl-protected benzylhydrazine 7 and chloromethyl ketone 9 under various
conditions
conversion
entry
base
T (°C)
solvent
yield
1
2
3
4
5
K2CO3, NaI
50
50
60
rt
DMF
acetone
DMF
DMF
MeCN
30%
NaHCO3, NaI
NaHCO3, NaI
NaHCO3, NaI
NaHCO3, NaI
50%
<10%
>90%
quant
rt
Table 2. Reduction of N-(methoxycarbonyl)-L-tert-leucinyl protected amino ketone 10 with various reducing agents/conditions
entry
reducing agent
conditions
syn:antia
1
2
3
4
NaBH4
MeOH, 0 °C
THF, 0 °C
THF, 0 °C
Et2O, 0 °C
3:2
3:1
Zn(BH4)2
LiAlH(OtBu)3
LiAlH(OtBu)3
4:1
28:1
a The syn:anti ratio was determined on the crude product with HPLC.
by the reduction of an amino ketone. The reduction of
N-protected amino ketones can be carried out stereoselectively
to produce either the syn- or anti-amino alcohol diastereomer,
depending on the mode of stereocontrol.13 Basically, N-
carbamate protected amino ketone is reduced selectively to anti-
amino alcohols via chelation control, while the N,N-dibenzyl
or N-trityl protecting group renders the amino group sufficiently
bulky to achieve Felkin-Anh control and thus produce syn-
amino alcohols.13,14 It is clear that the nitrogen protecting group
is an important stereocontrol element.
In our case, the nitrogen of the amino ketone 10 is acylated
with (S)-2-(methoxycarbonylamino)-3,3-dimethylbutanoic acid,
which functions as a bulky protecting group. So we assumed
that the reduction of our compound 10 would achieve
Felkin-Anh control to produce the syn-amino alcohol as the
major product.
delight, the reduction of amino ketone 10 with sodium boro-
hydride in methanol or with zinc borohydrides in THF gave
the syn-amino alcohol 11 as major product (Felkin-Anh
control), but only poor stereoselectivities of 3:2 and 3:1 (syn:
anti ratio), respectively, were obtained. The stereochemical
outcome in different solvents here was consistent with published
findings.13–16 Usually, hydroxylic solvents (MeOH and other
alcohols) promote the exchange and/or disproportionation of
ligands attached to boron or aluminum so that the substrate can
become bound to boron or aluminum in the chelated fashion
needed for effective chelation stereocontrol (vide infra).13,16
Consequently borohydride or aluminum hydride reagents in
alcohol solvents give chelation-control reduction, whereas the
nonhydroxylic solvents facilitate a Felkin-Anh controlled
reduction. In our study, the combination of zinc borohydrides
in THF really provides a better syn-selectivity than the reducing
system of sodium borohydride in methanol.
A first attempt was made with the borohydride reducing
agents in hydroxylic or aprotic solvents (Table 2). To our
(15) Koskinen, A. M. P.; Koskinen, P. M. Tetrahedron Lett. 1993, 34, 6765–
(13) Hoffman, R. V.; Maslouh, N.; Cervantes-Lee, F. J. Org. Chem. 2002,
67, 1045–1056.
6768.
(16) Våbenø, J.; Brisander, M.; Lejon, T.; Luthman, K. J. Org. Chem. 2002,
67, 9186–9191.
(14) Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1991, 30, 1531–1546.
72
•
Vol. 12, No. 1, 2008 / Organic Process Research & Development

Figure 2. Transition state of the reduction.
On the other hand, since the steric hindrance endowed by
the amino ketone or the reducing agents appears to favor
achieving Felkin-Anh control, the use of a more bulky and
more reactive reductant of LiAlH(OtBu)3 in THF at 0 °C
resulted in an obvious increase of the syn:anti ratio (4:1). Similar
stereoselectivity was reported in Hoffman’s work,13 in which
N-trityl-protected amino ketones could be reduced selectively
to syn-amino alcohols with LiAlH(OtBu)3 in THF at -5 °C.
Considering that solvent effect is an important factor to influence
the stereoselectivity,13–16 we examined LiAlH(OtBu)3 in Et2O
with the aim to increase the syn diastereoselectivity further. The
choice of Et2O was inspired by a literature-reported successful
use of LiAlH(OtBu)3 in Et2O to achieve the threo selective
reduction of the halomethyl ketones.10 Also in Hoffman’s work,
it was observed that when changing solvent from THF to ether
in the reduction of N-carbamate-protected amino ketones by
Zn(BH4)2, the stereoselectivity was switched from 2:1 to 1:2
(anti:syn).13 Encouragingly, in our work, clean reaction and
excellent syn-selectivity (28:1) was obtained under very mild
conditions (ice bath, reaction time of 4 h) when employing
LiAlH(OtBu)3 in Et2O as the reagent-solvent combination. This
reduction procedure was scaled up to afford 39 g of Atazanavir
11 with 99.4 AP. The reason why diethyl ether is a preferred
solvent for the Felkin-Anh control reduction of our substrate
over THF is not clear, but this reagent-solvent combination
might be useful for the synthesis of other protease inhibitors
containing a syn-amino alcohol core structure.
the previously reported synthesis of hydroxyethylene dipep-
tide isosteres.17
Mode of Stereocontrol in the Amino Ketone Reduction.
Examination of the literature revealed that high syn-selectivity
was achieved only in the case of N,N-dibenzyl-,14 N-trityl-,13
or N-phthaloyl-protected18 amino ketones. Our study provided
the first example that N-L-tert-leucinyl-protected amino ketone
was reduced by the new reagent-solvent combination of
LiAlH(O-tBu)3 in diethyl ether to exclusively give the syn-1,2-
amino alcohol, which achieved a good Felkin-Anh stereocontrol.
Principally, to achieve Felkin-Anh control in the reduction
of protected amino ketones and thus produce syn-amino
alcohols, the amine group must be modified to make it sterically
bulky. This renders a steric preference for a Felkin-Anh
transition state and minimizes effective chelation involving the
amino nitrogen and the ketone oxygen.13 According to the
principal, we proposed a transition state of the reduction of our
substrate 10, as demonstrated in Figure 2. The N-(methoxy-
carbonyl)-L-tert-leucinyl group installed on the nitrogen of the
amino ketone confers a steric congestion with benzyl group.
In order to stay a low energy conformation, the ketone 10
preferentially adopts a Felkin-Anh transition state, leading to
a high syn-stereoselectivity.
As a comparison study, we examined the stereoselectivity
of the reduction of N-carbamate-protected aza-dipeptide isostere.
As shown in Scheme 5, N-Boc-protected amino ketone (14)
was reduced by NaBH4in MeOH or Zn(BH4)2in THF to give
anti-1,2-amino alcohols (15) as the major product, with an anti:
syn ratio of 3:2 and >20:1, respectively. The reduction
proceeded via a chelation control predominantly. This result is
in agreement with the reported general amino ketone reduction
protocol.
It is worthwhile to note that the construction of the
syn-vicinal amino alcohol via amino ketone reduction in
our approach was achieved directly on the aza-dipeptide
isostere structure, devoid of any protecting group ex-
changing. The N-(methoxycarbonyl)-L-tert-leucinyl group
played a dual role in our route: on the one hand, it is the
structural element of the molecule; on the other hand, it
serves as the protecting group that is bulky enough to enforce
the Felkin-Anh control. This is the distinct advantage over
Conclusions
In conclusion, we have successfully developed an efficient
and convenient route toward the synthesis of HIV protease
(17) Tao, J. H.; Hoffman, R. V. J. Org. Chem. 1997, 62, 6240–6244.
(18) Sengupta, S.; SenSarma, D. S. Tetrahedron Lett. 2001, 42, 485–487.
Vol. 12, No. 1, 2008 / Organic Process Research & Development
•
73

Scheme 5. Reduction of the N-carbamate-protected aza-dipeptide isostere
inhibitor Atazanavir. Distinct from literature-reported proce-
dures, our new approach employed the amino ketone reduction
of aza-dipeptide isostere with a new reagent-solvent combina-
tion of LiAlH(OtBu)3 in Et2O as the key and final step, which
achieved Felkin-Anh stereocontrol to offer the desired syn-
1,2-amino alcohol exclusively. Furthermore, the N-(methoxy-
carbonyl)-L-tert-leucinyl group played a dual role in our
convergent route: on the one hand, it is the structural element
of the molecule, and on the other hand, it serves as the protecting
group of the two fragments of benzylhydrazine and chlorom-
ethyl ketone. The new strategy offers an alternative and efficient
approach to prepare hydroxyethylene aza-dipeptide isostere with
high diastereoselectivity and good yield.
MHz, CDCl3): δ 8.73 (d, 1H, J ) 4.5 Hz), 7.99 (d, 2H, J
) 8.4 Hz), 7.85–7.75 (m, 2H), 7.48 (d, 2H, J ) 8.4 Hz),
7.30 (m, 1H), 5.38 (d, 1H, J ) 9.6 Hz), 4.05 (qAB, 2H, J
) 12.6 Hz, 16.8 Hz), 3.78 (d, 1H, J ) 9.6 Hz), 3.67 (s,
3H), 0.97 (s, 9H). 13C NMR (100 MHz, CDCl3): δ 170.3,
157.1, 157.0, 149.6, 138.6, 138.2, 136.7, 129.2 × 2, 126.9
× 2, 122.1, 120.4, 61.1, 55.5, 52.3, 34.5, 26.4 × 3. MS
(EI): m/z 370 (M)+. HRMS (EI): calcd for C20H26O3N4
(M+) 370.2005, found 370.2011.
Methyl (S)-1-((S)-4-Chloro-3-oxo-1-phenylbutan-2-ylcar-
bamoyl)-2,2-dimethylpropyl Carbamate (9). To a solution
of tert-butyl (S)-4-chloro-3-oxo-1-phenylbutan-2-ylcar-
bamate (38.2 g, 128.3 mmol) in dry THF (100 mL) was
added 4 N HCl (100 mL). The mixture was stirred for
4 h at 50 °C and concentrated to remove the THF, and
the residue was washed with dichloromethane (40 mL)
three times. The aqueous layer was concentrated to give
a white solid. To a cold solution of 8 (26.4 g, 140 mmol)
in dry THF (350 mL) was added NMM (28 mL, 256
mmol) followed by careful addition of isobutyl chloro-
formate (18.44 mL, 128 mmol). After stirring for 20 min
at –25 °C, the above white solid was added to the solution.
The reaction mixture was stirred for 1 h at -20 °C and
for 1–3 additional hours at room temperature. The solvent
was removed in vacuum, and the residue was taken up in
ethyl acetate. The usual workup to give an oily residue,
crystallized from petroleum ether/ ethyl acetate to afford
Experimental Section
All materials were purchased from commercial suppliers.
Unless specified otherwise, all reagents and solvents were used
as supplied by manufacturers. Melting points are uncorrected.
1H NMR spectra (300 MHz) and 13C NMR spectra (100 MHz)
were recorded in CDCl3. Specific rotations (uncorrected) were
determined in a Perkin-Elmer 241 polarimeter. Low and high
resolution mass spectra were determined on a MAT-95 mass
spectrometer. HPLC analysis results are described as area %
(AP).
N-1-[N-(Methoxycarbonyl)-L-tert-leucinyl]-N-2-[4-(pyri-
dine-2-yl)-benzyl] Hydrazine (7). To the solution of 6(31.65
g, 132.1 mmol) in EtOH (250 mL) was added triethyl-
amine (21.9 mL, 158.5 mmol) at room temperature under
N2. After 1 h of stirring at room temperature, a solution
of 3 (24.18 g, 132.1 mmol) in EtOH (100 mL) was added.
The mixture was heated to reflux for 4 h, then cooled to
room temperature, treated with Pd(OH)2 (5 g), and stirred
under 1 atm H2 at room temperature for 12 h. The catalyst
was filtered off, and the filtrate was concentrated under
reduced pressure to give an oily residue that was
recrystallized from isopropyl ether/ethanol to yield 7 as a
9 as a white solid (33.8 g, 71.9% yield, 97.9 AP). Mp:
1
115–116 °C. [R]20 ) -3.9°(c ) 0.92, CHCl3). H NMR
D
(300 MHz, CDCl3): δ 7.27–7.35 (m, 3H), 7.16 (d, 2H, J
) 6.6 Hz), 6.23 (d, 1H, J ) 6.0 Hz), 5.27 (d, 1H, J ) 9.3
Hz), 4.93–5.01 (m, 1H), 4.14 (d, 1H, J ) 15.6 Hz),
3.85–3.94 (m, 2H), 3.68 (s, 3H), 3.01–3.02 (m, 2H), 0.95
(s, 9H). 13C NMR (100 MHz, CDCl3): δ 200.6, 170.8,
157.0, 135.1, 129.0 × 2, 128.9 × 2, 127.5, 62.5, 57.1,
52.4, 47.7, 37.5, 34.4, 26.4 × 3. MS (EI): m/z 368 (M+).
HRMS (EI): calcd for C18H25ClN2O4 (M+) 368.1503,
found 368.1517.
white solid (37 g, 75.7% yield, 100.0 AP). Mp: 134–135
1
°C; [R]20 ) -22.5° (c ) 0.91, CHCl3). H NMR (300
D
74
•
Vol. 12, No. 1, 2008 / Organic Process Research & Development

1-[4-(Pyridin-2-yl)phenyl]-5(S)-2,5-bis{[N-(methoxycar-
bonyl)-L-tert-leucinyl]amino}-4-oxo-6-phenyl-2-azahexane (10).
To a solution of 9 (28 g, 76 mmol) in MeCN (250 mL) was
added NaI (12.4 g, 83 mmol). The mixture was stirred for 15
min at room temperature, and a solution of 7 (30.7 g, 83 mmol)
in MeCN (350 mL) was added. The reaction mixture was stirred
for another 15 min and treated with NaHCO3 (12.7 g, 152
mmol). The mixture was stirred for 12 h and concentrated under
reduced pressure to remove the MeCN. The residue was diluted
with ethyl acetate (500 mL), and the organic phase was washed
with KHSO4 and brine, dried over Na2SO4, and concentrated
in vacuum to give crude product 10 (51 g, 96.2% yield, 91.7
AP) as a yellow solid that was directly used for the next step
without further purification. [R]22D ) +3° (c ) 0.15, CHCl3).
1H NMR (300 MHz, CDCl3): δ 8.65 (d, 1H, J ) 5.1 Hz),
7.98–7.88 (m, 3H), 7.74–7.64 (m, 2H,), 7.42 (d, 2H, J ) 8.1
Hz), 7.22–7.16 (m, 3H), 7.07 (d, 2H, J ) 7.2 Hz), 5.69 (d, 1H,
J ) 9.0 Hz), 5.51 (d, 1H, J ) 10.2 Hz), 4.80 (d, 1H, J ) 6.0
Hz), 4.20–4.07 (m, 3H), 3.82 (d, 1H, J ) 17.4 Hz), 3.75 (d,
1H, J ) 9.6 Hz), 3.64 (s, 3H), 3.59 (s, 3H), 3.04–2.86 (m, 2H),
0.94 (s, 9H), 0.80 (s, 9H). 13C NMR (100 MHz, CDCl3): δ
207.6, 173.1, 170.8, 169.8 × 2, 157.0, 149.6, 138.8, 137.3,
136.7, 135.6, 129.5 × 2, 129.0 × 2, 128.8 × 2, 127.2 × 2,
126.9, 122.1, 120.4, 62.4, 61.2, 60.2, 59.9, 57.7, 52.3, 36.9,
34.5, 34.3, 26.4 × 3, 26.3 × 3. MS (ESI): m/z 703.6 (M +
1)+, 725.6 (M + Na)+. HRMS (ESI): calcd for C38H50N6O7Na
(M + Na)+ 725.3639, found 725.3615.
reflux was observed. The slurry was cooled to room tempera-
ture, and the solid was collected by filtration, washed with 100
mL of cool isopropyl ether-EtOH solution, and dried at 50 °C
under vacuum to give 18 g of crude free base. The combined
crude product (46 g) was recrystallized from the mixture solvent
of ethanol–water twice to afford Atazanavir (39 g, 62%, AP
99.4) as a white solid. Mp: 198–200 °C, [R]20D ) -44.3° (c )
1
0.9, EtOH) [lit.8 [R]20 ) -47° (c ) 1.0, EtOH)]. H NMR
D
(300 MHz, CDCl3): δ 8.59 (d, 1H, J ) 0.8 Hz), 7.89–7.80 (m,
4H), 7.51 (d, 1H, J ) 8.4 Hz), 7.37–7.32 (m, 1H), 7.23–7.17
(m, 4H), 7.15–7.10 (m, 1H), 4.14 (t, 1H, J ) 7.2 Hz), 3.99 (s,
2H), 3.84 (s, 1H), 3.75 (d, 1H, J ) 9.6 Hz), 3.68 (s, 1H), 3.63
(s, 3H), 3.59 (s, 3H), 2.96–2.82 (m, 3H), 2.69 (d, 1H, J ) 12.9
Hz), 0.82 (s, 9H), 0.71 (s, 9H). MS (EI): m/z 704 (M)+. Anal.
Calcd for C38H52N6O7: C, 64.75; H, 7.44; N, 11.92. Found: C,
64.41; H, 7.33; N, 11.69.
Acknowledgment
We thank Desano Company in Shanghai for providing the
reagent of (S)-tert-butyl 4-chloro-3-oxo-1-phenylbutan-2-ylcar-
bamate. We also thank Shaoxu Huang for running analytic
HPLC to identify the purity. Financial support from Chinese
Academy of Sciences (KSCX2-YW-R-20) and Science and
Technology Commission of Shanghai Municipality (07QH14018)
is greatly appreciated.
Supporting Information Available
1-[4-(Pyridin-2-yl)phenyl]-5(S)-2,5-bis{[N-(methoxycar-
bonyl)-L-tert-leucinyl]amino}-4(S)-hydroxy-6-phenyl-2-aza-
hexane (11, Atazanavir). To the solution of 10 (24.57 g, 35.0
mmol) in diethyl ether (400 mL) was added LTBA (20.83 g,
87.5 mmol) at 0 °C. The mixture was stirred for 4 h at 0 °C
and then quenched with water (2 mL). The solvent was removed
in vacuum, and the residue was taken up in dichloromethane
(200 mL) and washed with brine. The organic layer was
concentrated to provide a yellow solid that was treated with
370 mL of isopropyl ether-EtOH solution and heated until
The synthesis for the known compounds 3, 6, 8, and 12;
1
analytical data for new compounds; H NMR spectra for
compounds 3, N-Boc-6, and 7–11 and 13C NMR spectra for
compounds 7, 9, and 10; HPLC chart for the purity identification
of compounds 7, 9, 10, and 11. This material is available free
of charge via the Internet at http://pubs.acs.org.
Received for review July 10, 2007.
OP7001563
Vol. 12, No. 1, 2008 / Organic Process Research & Development
•
75