A concise route to valacyclovir hydrochloride

Article abstract of DOI:10.1002/hlca.201000303

An efficient and concise synthesis of valacyclovir hydrochloride (4), which is a prodrug of acyclovir (3) is described. The synthesis was accomplished in two stages by coupling acyclovir with (2S)-2-azido-3-methylbutanoic acid followed by reduction (Schem

Full text of DOI:10.1002/hlca.201000303

592
Helvetica Chimica Acta – Vol. 94 (2011)
A Concise Route to Valacyclovir Hydrochloride
by Prasada Raju V. N. K. V. Vetukuri*, Ravindra Vedantham, Vijayavitthal Thippannachar Mathad,
Pratap Reddy Padi, and Vijaya Anand Ramasamy
Research & Development, Integrated Product Development, Innovation Plaza, Dr. Reddyꢀs
Laboratories Ltd., Bachupally, Ranga Reddy District, Andhra Pradesh-500072, India
(phone: þ 91-4044346412; mobile: þ 91-9989997170; fax: þ 91-4044346840;
e-mail: prasadvvnkv@drreddys.com)
An efficient and concise synthesis of valacyclovir hydrochloride (4), which is a prodrug of acyclovir
(3) is described. The synthesis was accomplished in two stages by coupling acyclovir with (2S)-2-azido-3-
methylbutanoic acid followed by reduction (Scheme 2).
Introduction. – Genital herpes is one of the most unbridled sexually transmitted
diseases (STD), and most commonly caused by infection with herpes simplex virus-
type 2 (HSV-2) [1]. Most sexual transmission of HSV-2 occurs on days without genital
abrasion in the source partner [2]. While lesion in immuno-competent patients may be
benign, those in immuno-compromised patients can be life threatening with high
mortality and morbidity [3]. In the past two decades, the invention of new antiviral
drugs against HSV-2 has been emerging, and a number of antiviral drugs were
developed in recent years. Most of the HSV-2 antiviral drugs target the thymidine
kinase (TK) phosphorylation sites of herpes virus and get activated by viral TK to
become inhibitors of viral DNA polymerases and block viral DNA synthesis [4]. The
currently available and efficient HSV-2 antiviral drugs are penciclovir (1), famciclovir
(2), acyclovir (¼2-amino-1,9-dihydro-9-[(2-hydroxyethoxy)methyl-6H-purin-6-one;
3), and its prodrug valacyclovir hydrochloride (4) (Fig. 1).
Valacyclovir hydrochloride (4), the l-valine ester of acyclovir (3), was designed to
enhance bioavailability of the parent compound. Recent investigations established that
the spectrum of bioavailability of 4 is three times more than that of 3 [5]. Therefore, an
efficient synthetic route to 4 is highly enviable. In this communication, we would like to
divulge our efforts towards the synthesis of 4.
Since 4 is an l-valine ester of 3, most of the hitherto known syntheses involve
coupling of 3 with a protected form of l-valine followed by deprotection (Scheme 1).
Beauchamp et al. reported the synthesis of valacyclovir by condensing acyclovir (3)
with N-[(benzyloxy)carbonyl]-l-valine in the presence of dicyclohexylcarbodiimide
(DCC) followed by removal of the protecting group by hydrogenolysis [6]. Similarly, 4
was prepared from N-[(tert-butoxy)carbonyl]-l-valine or N-formyl-l-valine [7].
Another approach involved condensation of the tosylate of acyclovir with an enamine
derived from l-valine and methyl acetoacetate [8]. Reaction of 3 with the NCA of l-
valine (¼ N-carboxy-l-valine anhydride ¼ (4S)-4-(1-methylethyloxazolidine-2,5-di-
one) was also reported to give 4 in a single step [9] (NCA ¼ N-carboxyanhydride).
ꢀ 2011 Verlag Helvetica Chimica Acta AG, Zꢁrich

Helvetica Chimica Acta – Vol. 94 (2011)
593
Figure. HSV-2 Antiviral drugs
Scheme 1. Previous Synthesis of Valacyclovir Hydrochloride (4)
From the literature, it is evident that protection of the amino group of l-valine has
been the nub of almost all the synthetic routes reported (Scheme 1). The bulky
protecting groups used may cause steric hindrance, which may affect the formation of
the l-valine ester. The protectionꢀdeprotection steps further increase the risk of
racemization.
Results and Discussion. – The use of a small-size versatile synthetic precursor of the
amino group such as an azide group holds great promise to circumvent the problem of
steric hindrance and racemization in the synthesis of the l-valine ester 4. The azido
group being an excellent masked amino equivalent is relatively stable and can be
readily reduced to the desired amino function when required. The protocol, being
highly effective, inspired us to incorporate it in the synthesis of 4 from 3.
The strategy adopted by us was to prepare the a-azido ester 6 of acyclovir followed
by reduction to give the desired a-amino ester. The required starting a-azido acid, (2S)-
2-azido-3-methylbutanoic acid (5), was prepared by following a literature procedure
[10]. Sodium azide was treated with sulfuryl chloride (SO2Cl2) followed by 1H-
imidazole to provide 1H-imidazole-1-sulfonyl azide, which on diazo-transfer reaction
with l-valine in the presence of CuSO4 as a catalyst gave the (2S)-2-azido-3-

594
Helvetica Chimica Acta – Vol. 94 (2011)
methylbutanoic acid (5) (Scheme 2). DCC-Mediated coupling of acyclovir (3) with 5
provided the a-azido ester 6 in good yield and purity. Then, 6 was reduced to
valacyclovir hydrochloride (4) under catalytic-hydrogenation conditions with Raney
nickel as a catalyst as described in the Exper. Part.
Scheme 2. The Azide Route to Valacyclovir Hydrochloride (4)
It was envisaged that N2-substituted acyclovir would give better results in the DCC
condensation step due to enhanced solubility as compared to acyclovir (3). In this
context, two different N2-substituted acyclovirs, namely, N2-[(dimethylamino)methy-
lene] and N2-trityl derivatives 7 and 8 (trityl ¼ triphenylmethyl) were prepared. The
above choices were made to account for different types of N2-substituents, i.e., of the
imine and alkyl type. N2-[(Dimethylamino)methylene]acyclovir (7) was prepared by
modifying the literature procedure [11], and N2-tritylacyclovir (8) by tritylation of 3.
To compare the reactivity of the N2-substituted acyclovir derivatives 7 and 8 with
that of acyclovir, the former were coupled with 5 under standard DCC conditions
(Scheme 3). As expected, both 7 and 8 gave better results in terms of yield within 1 h
(92% of 9 and 95% of 10, resp.). This outcome clearly endorses the enhanced solubility
of 7 and 8 in DMF under normal reaction conditions1). The N2-[(dimethylamino)-
methylene] group in 9 was deprotected with aqueous HCl solution at 508 to provide 6 in
95% yield. Similarly, the trityl deprotection of 10 was carried out with AcOH at 508 to
give 6 in 85% yield.
Conclusion. – We have developed a succinct synthesis of valacyclovir hydrochloride
(4) by coupling of acyclovir (3) with azido acid 5 derived from l-valine, followed by
reduction with Raney nickel.
We greatly appreciate the supportive environment encouraged at Dr. Reddyꢀs Laboratories Ltd. We
are thankful to the Intellectual Property Management (IPM), the NMR group of Dr. Reddyꢀs Research
Foundation (DRF), and the analytical group of the Integrated Product Development Organization
(IPDO) for their generous support (IPDO communication number: IPDO-IPM-00190).
N2-[(Dimethylamino)methylene]acyclovir (7) is ca. two times more soluble than acyclovir (3), and
N2-tritylacyclovir (8) is highly soluble in DMF.
1
)

Helvetica Chimica Acta – Vol. 94 (2011)
595
Scheme 3. Coupling of 7 and 8 with Azido Acid 5
a) 5, DCC, Et3N, DMF, 158, 1 h. b) HCl, 508. c) AcOH, 508.
Experimental Part
General. Solvents and reagents were used without further purification. Solvent removal was
accomplished by a Bꢁchi rotary evaporator at 30 – 40 Torr. IR Spectra: Perkin-Elmer-1650 FT-IR
1
spectrometer; in KBr dispersion; ˜n in cmꢀ1. H-NMR Spectra: Mercury-Plus/Varian Gemini-2000 FT-
NMR spectrometer; in (D6)DMSO and CDCl3 at 200 or 400 MHz; d in ppm rel. to Me4Si as internal
standard, J in Hz. MS: HP-5989A LC/MS spectrometer at 70 eV; in m/z (rel. %). HR-MS: Micromass-
LCT-Premier-XE mass spectrometer equipped with an ESI lack spray source for accurate mass values
(Water Corporation, Milford, MA, USA); in m/z.
(2S)-2-Azido-3-methylbutanoic Acid 2-[(2-Amino-1,6-dihydro-6-oxo-9H-purin-9-yl)methoxy]ethyl
Ester (6). (2S)-2-Azido-3-methylbutanoic acid (5; 90% in AcOEt; 21.3 g, 0.13 mol) was taken in DMF
(400 ml) under N2 and cooled to 10 – 158. DCC (27.5 g, 0.13 mol) was charged to it, and the mixture was
agitated for 30 min. Acyclovir (3; 20 g, 0.089 mol) was added to the reaction mass at 10 – 158 followed by
N,N-dimethylpyridin-4-amine (DMAP; 1.62 g, 0.013 mol), and the resulting mixture was stirred for
45 min at 10 – 158. The separated solid was filtered, and the filtrate was cooled to 0 – 58. H2O (1.35 l) was
added to the filtrate in a dropwise manner to precipitate the solid. Filtration of the solid followed by
drying at 608 gave the desired 6 (25.5 g, 82%). White solid. IR: 3470, 3293, 3191, 2931, 2099, 1732, 1655,
1
1629, 1104. H-NMR (400 MHz, (D6)DMSO): 10.65 (s, 1 H); 7.81 (s, 1 H); 6.50 (s, 2 H); 5.36 (s, 2 H);
4.28 – 4.34 (m, 1 H); 4.18 – 4.23 (m, 1 H); 4.10 (d, J ¼ 5.6, 1 H); 3.67 – 3.74 (m, 1 H); 1.96 – 2.05 (m, 1 H);
0.89 (d, J ¼ 6.8, 3 H); 0.80 (d, J ¼ 6.8, 3 H). HR-MS: 351.1537 ([M þ H]þ, C13H19N8O4þ ; calc. 351.1529).
l-Valine 2-[(2-Amino-1,6-dihydro-6-oxo-9H-purin-9-yl)methoxy]ethyl Ester Hydrochloride (1:1)
(4). Raney nickel (4.0 g) was added to a soln. of 6 (20 g, 0.057 mol) in EtOH (300 ml), and the resulting
suspension was vigorously stirred at 50 – 608. After completion of the reaction, the mixture was cooled to
r.t., and the unwanted solid was separated by filtration. The pH of the filtrate was adjusted to 2.5 – 3.0
with 10% aq. HCl soln. (35 ml). The resulting homogeneous mass was treated with charcoal at 50 – 608.
The mixture was filtered, and the filtrate was concentrated at 508. The residue was washed with acetone
(100 ml), and the product was separated by filtration and finally dried at 508: 4 (16.9 g, 82%). IR: 3472,
3327, 3110, 2925, 1732, 1631, 1607, 1100. 1H-NMR (400 MHz, (D6)DMSO): 10.81 (s, 1 H); 8.46 (s, 2 H);
7.84 (s, 1 H); 6.66 (s, 2 H); 5.37 (s, 2 H); 4.35 – 4.41 (m, 1 H); 4.18 – 4.23 (m, 1 H); 3.83 (d, J ¼ 4.0, 1 H);
3.71 – 3.74 (m, 2 H); 2.07 – 2.12 (m, 1 H); 0.90 (d, J ¼ 7.2, 3 H); 0.87 (d, J ¼ 6.8, 3 H); 0.80 (d, J ¼ 6.8, 3 H).

596
Helvetica Chimica Acta – Vol. 94 (2011)
REFERENCES
[1] J. S. Smith, N. J. Robinson, J. Infect. Dis. 2002, 186, 3; L. Corey, H. H. Handsfield, J. Am. Med. Assoc.
2000, 283, 791; F. Xu, M. R. Sternberg, B. J. Kottiri, J. Am. Med. Assoc. 2006, 8, 964.
[2] G. J. Mertz, O. Schmidt, J. L. Jourden, Sex. Transm. Dis. 1985, 12, 33; A. Langenberg, L. Corey, R.
Ashley, W. Leong, S. A. Straus, N. Eng. J. Med. 1999, 341, 1432.
[3] M. Vajpayee, N. Malhotra, Indian J. Pharmacol. 2000, 32, 330.
[4] S. K. Tyring, D. Baker, W. Snowden, J. Infect. Dis. 2002, 186, 40.
[5] R. Gupta, A. Wald, E. Krantz, S. Selke, T. Warren, M. Vargas-Cortes, G. Miller, L. Corey, J. Infect.
Dis. 2004, 190, 1374; L. M. Beauchamp, T. A. Krenitsky, Drugs Future 1993, 18, 619.
[6] L. M. Beauchamp, G. F. Orr, P. DeMiranda, T. Burnette, T. A. Krenitsky, Antiviral Chem.
Chemother. 1992, 3, 157.
[7] M. Y. Etinger, L. M. Yudovich, M. Yuzefovich, G. A. Nisnevich, B. Z. Dolitzki, B. Pertsikov, B.
Tishin, D. Blasberger, PCT Int. Appl. 041647, 2003.
[8] M. Montoro, S. Pirovano, PCT Int. Appl. 003553, 1998.
[9] W. P. Jackson, PCT Int. Appl. 029311, 1994.
[10] E. D. Goddard-Borger, R. V. Stick, Org. Lett. 2007, 9, 3797.
[11] C. McGuigan, M. Derudas, J. J. Bugert, G. Anderi, R. Snoeck, J. Balzarini, Bioorg. Med. Chem. Lett.
2008, 18, 4364.
Received August 12, 2010