d’

‘ .

ah

I
.7 3

‘.
,3:
Iii
2"?»
‘1'.
24-.
5‘}
g;

'!

-n
.\
um

,.

M— .
. “a.

~£§W

 

 

0/, \\

(ow/~39 (or;

LIBRARIES
MICHIGAN STATE UNIVERSITY
EAST LANSING, MICI-I 48824-1048

This is to certify that the
thesis entitled

PREPARATION OF THIOALKYL SUBSTITUTED SMALL RING
NITROGEN HETEROCYCLES AND 5-DISUBSTITUTED-
ET HYL-2—OXAZOLIDINONES FOR ANTIBACTERIAL AND
ANTIVIRAL CANDIDACY

presented by
Zhen Xu

has been accepted towards fulfillment
of the requirements for the

MS. degree in Chemistry

 

 

’ PM U 449:“ 0"”

\ Major Profisor’sySignature
July 30, 2004

 

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/C|RClDateDuo.p85-p.15

 

PREPARATION OF THIOALKYL SUBSTITUTED SMALL RING NITROGEN
HETEROCYCLES AND s-DISUBSTITUTED-ETHYL-z-OXAZOLIDINONES
FOR ANTIBACTERIAL AND ANTIVIRAL CANDIDACY
By

Zhen Xu

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCES
Department of Chemistry

2004

ABSTRACT
PREPARATION OF THIOALKYL SUBSTITUTED SMALL RING NITROGEN
HETEROCYCLES AND S-DISUBSTITUTED-ETHYL-Z-OXAZOLIDINONES FOR
ANTIBACTERIAL AND ANTIVIRAL CANDIDACY
By

Zhen Xu

Oxazolidinones are a new type of synthetic antibacterial drugs with unique mechanism
of action. General and concise syntheses towards novel antibacterial oxazolidinone
intermediates from naturally occurring carbohydrates were developed. An efficient
four-step sequence was achieved in the synthesis of homochiral
S-D-erythro-Z-oxazolidinones using 2-deoxy-D—ribose as chiral synthon.
Furthermore, a methodology of in situ deprotection and rearrangement of
N-p-methoxybenzyl-B-hydroxyamide to form 5-substituted-2-oxazolidinones was
succeeded. An application of this methodology in the solid-state synthesis was

suggested.

A template family of novel homochiral thioalkyl substituted small ring nitrogen
heterocycles was established from carbohydrate precursors. The target molecules were
designed as the transition-state analogs of glycosidase inhibitors. Preliminary bio-test

showed promising inhibitory activities against a gram-positive bacterial strain.

TABLE OF CONTENTS

LIST OF SCHEMES ........................................................................... v
LIST OF FIGURES ........................................................................... vii
LIST OF TABLES .............................................................................. ix

CHAPTER 1. ANTIBACTERIAL OXAZOLIDINONES

l. 1. BACKGROUND ....................................................................... l
1.2. MECHANISM OF ACTION ......................................................... 4
1.3. STRUCTURE ACTIVITY RELATIONSHIP ...................................... 6
1.4. REFERENCE ......................................................................... 10

CHAPTER 2. SYNTHESIS OF OXAZOLIDINONE MOIETIES

2. 1. INTRODUCTION .................................................................... 12
2.2. CHIRAL 3-CARBON SYNTHON ................................................. 13
2.3. CHIRAL AUXILIARIES ............................................................ 17
2.4. ENZYMATIC AND BIOCATALYSIS ............................................. 19
2.5. CHIRAL POOL ....................................................................... 20
2.6. CARBONYLATION OF B-AMINONALCOHOLS ............................. 23
2.7. SOLID-PHASE SYNTHESIS ....................................................... 25
2.8. REFERENCE ......................................................................... 28

CHAPTER 3. AN EFFICIENT ROUTE TOWARDS HOMOCHIRAL
S-D-ERYTHRO-Z-OXAZOLIDINONES, IMPORTANT CHIRAL
PRECURSORS OF DIVERSE PHARMACEUTICALS ESPECIALLY
ANTIBIOTICS

3. 1. INTRODUCTION .................................................................... 32
3.2. DESIGN AND SYNTHESIS. . . . . . .'. ................................................ 34
3.3. RESULTS AND DISCUSSION ..................................................... 40
3.4. CONCLUSION ....................................................................... 44
3.5. EXPERIMENTAL .................................................................... 45
3.6. REFERENCE .......................................................................... 50

CHAPTER 4 IN SITU DEPROTECTION AND REARRANGEMENT OF

N-p—METHOXYBENZYL-B—HYDROXYAMIDES TO FORM
S-SUBSTITUTED-Z-OXAZOLIDINONES
4.1. INTRODUCTION .................................................................... 51
4.2. RESULTS AND DISCUSSION .................................................... 52
4.3. CONCLUSION ....................................................................... 62
4.4. EXPERIMENTAL .................................................................... 63
4.5. REFERENCE .......................................................................... 70

iii

CHAPTER 5 PREPARATION OF THIOALKYL SUBSTITUTED SMALL
RING NITROGEN HETEROCYCLES FOR GLYCOSIDASE INHIBITORS

5.1. INTRODUCTION .................................................................... 72
5.2. DESIGN AND SYNTHESIS ....................................................... 75
5.3. RESULTS AND DISCUSSION .................................................... 78
5.4. BIOLOGICAL TESTS ............................................................... 82
5.5. CONCLUSION ....................................................................... 83
5.6. EXPERIMENTAL ................................................................... 84
5.7. REFERENCE ........................................................................ 103

iv

LIST OF SCHEMES

Scheme 2.1 ....................................................................................... 12
Scheme 2.2 ....................................................................................... 13
Scheme 2.3 ....................................................................................... 14
Scheme 2.4 ....................................................................................... 15
Scheme 2.5 ....................................................................................... 15
Scheme 2.6 ....................................................................................... 16
Scheme 2.7 ....................................................................................... 17
Scheme 2.8 ....................................................................................... 17
Scheme 2.9 ....................................................................................... 17
Scheme 2.10 ..................................................................................... 18
Scheme 2.11 ..................................................................................... 18
Scheme 2.12 ..................................................................................... 19
Scheme 2.13 ..................................................................................... 20
Scheme 2.14 ..................................................................................... 21
Scheme 2.15 ..................................................................................... 21
Scheme 2.16 ..................................................................................... 21
Scheme 2.17 ..................................................................................... 22
Scheme 2.18 ..................................................................................... 23
Scheme 2.19 ..................................................................................... 23
Scheme 2.20 ..................................................................................... 24
Scheme 2.21 ..................................................................................... 24
Scheme 2.22 ..................................................................................... 24
Scheme 2.23 ..................................................................................... 25
Scheme 2.24 ..................................................................................... 26
Scheme 2.25 ..................................................................................... 27
Scheme 3.1 ....................................................................................... 41
Scheme 3.2 ....................................................................................... 43
Scheme 3.3 ....................................................................................... 43

Scheme 4.1 ...................................................................................... 52

Scheme 4.2 ...................................................................................... 53
Scheme 4.3 ...................................................................................... 53
Scheme 4.4 ...................................................................................... 54
Scheme 4.5 ...................................................................................... 55
Scheme 4.6 ...................................................................................... 57
Scheme 4.7 ...................................................................................... 59
Scheme 4.8 ...................................................................................... 59
Scheme 4.9 ...................................................................................... 61
Scheme 4.10 .................................................................................... 63
Scheme 5.1 ...................................................................................... 86
Scheme 5.2 ...................................................................................... 89

vi

LIST OF FIGURES

Figure 1.1. Template examples of antibacterial oxazolidinones ........................... 3
Figure 1.2. Mechanism of action ................................................................ 5
Figure 1.3. Comparison between Linezolid and Vancomycin .............................. 6

Figure 1.4. Structure-activity relationships (SAR) of antibacterial oxazolidinones

(Linezolid) ........................................................................... 7
Figure 1.5. Oxazolidinone modification trends in recent years ............................. 8
Figure 3.1. 5-Substituent modification ...................................................... 33
Figure 3.2. Chiral 4-carbon building blocks ................................................ 35
Figure 3.3. Strategies for preparing oxazolidinones ....................................... 36
Figure 3.4. Strategies for preparing oxazolidinones (continued) ......................... 37
Figure 3.5 Synthesis of novel oxazolidinones from 2-deoxypentoses ................... 39
Figure 3.6 Retrosynthetic analysis ............................................................. 39
Figure 4.1 Proposed mechanism of debenzylation of N-benzyl protection ............. 56
Figure 4.2 Mechanism of Hoffmann rearrangement ........................................ 56
Figure 4.3 in situ Debenzylation and Hofi'mann rearrangement .......................... 61
Figure 5.1 Nojirimycin ......................................................................... 72
Figure 5.2 General enzymatic glycoside hydrolysis mechanism ......................... 74

Figure 5.3 Mimic of pyranosyl oxacarbenium ion trigonal planar configuration. . . ....75

Figure 5.4 A comparison of novel thioalkyl heterocycles with the existing glycosidase

inhibitors ............................................................................ 77
Figure 5.5 Dehydrohalogenation .............................................................. 80
Figure 5.6 Selective hydrolysis ................................................................ 81

vii

Figure 5.7 Thioalkyl heterocycle candidate ................................................. 82

Figure 5.8 Thioalkyl substituted heterocycles for glycosidase inhibitors ................ 85

viii

LIST OF TABLES

Table 1. Introduction of new drug classes by the pharmaceutical industry .............. 2

ix

CHAPTER 1 ANTIBACTERIAL OXAZOLIDINONES

1.1. BACKGROUND

The development of bacterial resistance to almost all currently available
antibacterial agents is a growing global health problem. In particular, the most cited
gram-positive bacteria, including methicillin resistant Staphylococcus aureus (MRSA),
penicillin resistant Streptococcus pneumoniae, and vancomycin resistant Enterococci
(VRE), are of major concern because they have acquired multidrug resistance and
formidable treatment problems with these pathogens, which could easily result in
mortality. The molecular mechanisms by which bacteria become resistant usually involve
drug efflux, drug inactivation, or alterations in the antibiotic target site.I A number of
solutions to the problem of bacterial resistance are possible. The most common approach
is to modify the existing classes of antibacterial agents to provide new analogs with
improved attributes. More importantly, another approach is to explore novel classes of
antibacterial agents with new mechanisms of action because such agents would have the
advantage of lack of cross-resistance with existing drugs. However, over the past forty
years, very few new classes of antibacterial agents have been developed. (Table 1.1) The
large number of antibiotics available on the market were developed through extensive
effort in structure activity relationships (SAR), which led to a very large array of
antibiotics within existing classes when improvements in activity, or safety were realized
relative to pre-existing members of the classes. For example, the “new” macrolides and
fluoroquinolones, are modified variants of agents known since the 19508 and 19605,

respectively.2

Table l.1. Introduction of new drug classes by the pharmaceutical industry

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Year of launch Antibiotic Targets
1936 Sulphonamides (SMZ) Folic acid antgmists
1940 Penicillins (B-lactams) Cell-wall synthesis
1949 Tetracyclines Protein synthesis
1949 Chloramphenicol Protein Sflthesis
1950 Aminoglycosides Protein synthesis
1952 Macrolides (Erythromycin) Protein synthesis
1958 Glycopeptides (Vancomycinl Cell-wall synthesis
1962 Streptogramins Protein synthesis
1962 Quinolones (fluoroquinolone) DNA replication
2000 Oxazolidinones 3 (Linezolid) Initiation of protein synthesis

 

A novel class of antibacterial agents, oxazolidinones, 3 are being developed to fill the gap
caused by the increasing instances of resistance to the last-line defense drugs such as B-
lactams, vancomycin and a host of other agents. The oxazolidinones were originally
developed as monoamine oxidase (MAO-A) inhibitors for treatment of depression.4’ 5 The
first oxazolidinones to Show any appreciable activity against bacteria were described by a
South African group in a 1968 patent.6 In the late 19708, the oxazolidinones were
developed by E. I. DuPont de Nemours Company for control of bacterial and fungal
foliage diseases of tomatoes and other plants.5 By the late 19805, after chemical

modification of the early agents, the same company released two main candidates of (S)-

5-acetamidomethyl-3-aryloxazolidin-2-ones: DUP-721 and DUP-IOS (Figure 1.1), which
showed a broad in vitro spectrum against most gram-positive bacteria, as well as several
anaerobes.2 Toxic effects (myelosuppression caused by inhibition of protein synthesis) in
animal models prevented their clinical development. However, they served as the key
structural templates on which the first successful oxazolidinone candidates were based.
Scientists at Pharmacia continued to work with the oxazolidinones, exploring a series of
chemical modifications. Their efforts led to the discovery of two agents, Linezolid and
Eperezolid in 1994, with improved antibacterial activity and markedly diminished
toxicity compared with DUP-721. In April 2000, Linezolid (Zyvox®) was approved by
the Food and Drug Administration (FDA) as the first of this new class of antibacterial
agent for treatment of infections associated with vancomycin-resistant Enterococcus
faecium, E. faecium related bacterernia, complicated and uncomplicated skin and soft-

tissue infections, and nosocomial (hospital-acquired) pneumonia.7’ 8

Figure 1.1: Template examples of antibacterial oxazolidinones

o F N o
\__k/NHAC
DUP-721. x = c NHAC
DUP-105, X = S Linezolid (Zyvox®)

1.2. MECHANISM OF ACTION (Figure 1.2)

The oxazolidinones have excellent in vitro activity against most of the major
gram-positive bacteria that are pathogenic in humans. The oxazolidinones such as
linezolid are orally active and bear excellent pharmacokinetic behavior (lOO %
bioavailability). Also, the drug is a totally synthetic compound, which lessens the
likelihood of naturally occurring resistance mechanism. And more importantly, the
oxazolidinones have a unique mechanism of action so that cross-resistance with existing
resistance mechanisms in bacteria is believed to be unlikely. Based on the studies of
oxazolidinone binding sites, and the solution of the crystal structure of the entire
ribosome9 including high-resolution structure of the SOS subunit10 and a structure for the
308 subunit,ll it became mainly accepted that oxazolidinones inhibit bacteria protein
synthesis by interacting with the ribosome at the initiation step. Initiation of protein
synthesis occurs when the 303 ribosomal subunit, mRNA, fMet-tRNA, and the SOS
ribosomal subunit combine. Collective data suggest that the ribosomal interaction of
oxazolidinones is between these two subunits. For example, linezolid binding to the 23S
rRNA component of the SOS subunit distorts the binding site for fMet-tRNA, preventing
the formation of fMet-tRNA-mRNA-7OS (7OS initiation complex), thus inhibiting
initiation of protein synthesis.12 It is worth mentioning that most marketed antibiotics that
inhibit protein synthesis by binding to the 23 rRNA at the peptidyl transferase cavityl3
such as Chloramphenicol, clindamycin, erythromycin, clarithromycin and roxithromycin,
inhibit later steps in the pathway involving elongation of the growing peptide chain rather

than the initiation step. To date a crystal structure of an oxazolidinone bound to a

ribosomal subunit has not been described. But the availability of the ribosome

structures'4 will, no doubt, facilitate rational drug design.

Figure 1.2: Mechanism of action'5

 

Initiation factors fMet-tRNA

fibggém.
, SOS
ribosome 70$
mRNA (3) initiation
complex

  

 

 

 

 

 

 

~ Termination Elon tion Elon tion
. 4————
Peptide product of translation $33,,
. T Fusidic acid
Linoosamides Chloramphenicol

Macrolides Aminoglycosides
Tetracyclines Pristinamycin

 

 

 

 

 

 

Schematic representation of oxazolidinone mechanism of action in bacterial cell (fMet-tRNA: N-

formylmethionyl transfer RNA)

1.3. STRUCTURE-ACTIVITY RELATIONSHIP (SAR) STUDY

Figure 1.3: Comparison between Linezolid and Vancomycin

‘3 o 0
JK °
F N 0 H H0 \ Cl
\—k/NTCH3 NH2
Linezolid 0
HO““
o
Vancomycin

 

One of the striking features of the antibacterial oxazolidinones is their structural
simplicity compared to vancomycin, the previous last-resort drug. (Figure 1.3) The
oxazolidinone drugs are totally synthetic. The structural simplicity of oxazolidinones
could lead to a good understanding of structure-activity relationships (SAR) and it is
possible to see how new oxazolidinone drugs have evolved based on the structures of

earlier drugs.

Figure 1.4: Structure-activity relationships of antibacterial oxazolidinones (Linezolid)

Electron-donating N Atom

Flexible in well tolerated & improved

substitution safety profile

pattern 3-N-Aryl required for activity
\ O/\\ O

/
U O A.
/" LR»...

F improves antibacterial

activity
5-(S)-acetamidomethyl
necessary for antibacerial
activity

During the development of oxazolidinones as antimicrobial drugs, some structure-activity
trends began to emerge. Activity was determined to be a fiinction of the size and nature
of the S-Substituent, the nature of the 3-substituent and the configuration at the 5-
position.'(”[7 The 5-substituent was mostly small, non-basic and the (S)-acetamido methyl
group was thought to be optimum.l8 This dogma was mainly limited to carbonyl
fimctionalities. The 3-substituent was invariably an aryl group with different substitution
pattern.'5 Among those variations, the 3—substituent was most extensively investigated
from different perspectives including antibacterial activity,”2| safety profile,22
pharmacokinetic issue,'9 and antibacterial spectra.”15b For example, the linezolid and
eperezolid candidates were selected after rigorous SAR study of the 3-substituent, the
modification of structural template DUP-721 with the finding that para electron-donating
nitrogen heterocycle improves safety profile by marked diminution of bone marrow

toxicity, meta— fluorination of phenyl ring not only increased antibacterial potency but

also enhanced pharmocokinetic performance.” (Figure 1.4) Recent years there are some
distinct departures in structure from the earlier candidates. But most of these departures
still occurred in the 3-substituent on the oxazolidinone ring which was believed to
tolerate more functional modification. There are fewer variations in the 5-substituent
(Figure 1.5). However, recent studies have reported the introduction of substituents such
as the S-thiourea and S-thiocarbamate, S-dithiocarbamate and N- or O-substituted
heterocyclic aryl moieties as potential replacements for the acetamidomethyl side chain at
the C-5 position?“26 Three noteworthy publications were launched in a row discussing
comprehensively the S-substituents on oxazolidinones beyond the conventional
acetamidomethyl function on the basis of structure-activity relationship (SAR).”29 The
study revealed that replacing sulfur function S-substituents may increase the antibacterial
activity. And the balance of lipophilicity between the 5-hydrophilic substituents and
hydrophobic substituents on the aromatic ring is very important for the antibacterial
activity. These attempts led to more active compounds with increased range of

therapeutic spectrum.

Figure 1.5: Oxazolidinone modification trends in recent years

 

 

retain against NS aureus modest
NHAc
o 3”“ 9”” Y “39 N
T o N N o=(N l \
O
o \ =(o o /
better pharmacokinetics
IB-rlng modification]
6 O
0 A0 ”°Vu‘~ 07"" I
F NW
WNHAC potent pom

Better MIC (24 ug/ml) against gram negative

\ H. influenza and M. cant-titans
NCQ‘ F, pic-(1’9”
S

_~ N aegequlc (<0.l125~O.5 ug/ml) against (+) Better MIC ((0.125 “9"”) against H)
NC-{bl‘ H inliuengfgtcl $922: 3:121:80) NC 4/\ IN MIC (0.25-2 mgiml) against gram (~)
' ' H. influenza and M. caninhaiis
PNU-172576

o
O / \ /=\ l—\ HOJN’H

LL ° \
O
NHAc ° 0
good but inactive In vrvo good more potent but poor oral modest
[c-rlng modification I

83:151. as; US! “/5535; Q; Q}. ”0,;
Wm... E W

N-O
enhanced actlvly

[ D-rlng modification]

\ /—\
N o -N N O
#an% owu’lo
“Lair, “Eur,
O S

a = m" “kw Of “ROXY ‘40” ICING '0' 8. ST. E. “I0 (OAR-0.5%”
"Julianna MIC (0.5-1 point!)

mm MCL 35,0}, 6" ~‘.
K’:U\Nk0 O potent;
A20 2563

x oom [Lime OY CA; I]; [N NJ)? (2:;

ls-aubatltuanta moduicauofl

 

1.4. REFERENCE

1.

2.

10.

ll.

12.

l3.

14.

15.

16.

C. Walsh, Nature, 2000, 406, 775-781

R. Moellering, Annals of Internal Medicine, 1999, 130, 155-157

S. J. Brickner et al, J. Med. Chem. 1996, 39, 673-679

See for MAO activity: “Monoamine Oxidase and Disease: Prospects for Therapy with
Reversible Inhibitors” Edited by R. F. Tipon, P. Dostert and M. Strolin Benedetti,
Academic Press, London, 1984

a) S. J. Brickner, Current Pharmaceutical Design, 1996, 2, 175-194; b) C. Z. Ding,
and R. B. Silverman, J. Med. Chem. 1993, 36, 3606-3610

W. Hoyle, J. R. Green and G. P. Roberts South African Patent ZA 6703694 (1968)

http://www.fdagoWbbs/topics/ANSWERS/ANS0l010.html;
http://www.ph_annacia.com; http://www.zvvox.com

M. J. Rybak, D. M. Cappelletty, T. Moldovan, J. R. Aeschlimann, and G. W. Kaatz, J.
Antimicrob. Chemotherapy 1998, 42, 721-728

J. H. Cate, M. M. Yusupov, G. Z. Yusupova, T. N. Ernest and H. F. Noller, Science
1999, 285, 2095-2104

N. Ban, P. Nissen, J. Hansen, P. B. Moore and T. A. Steitz, Science, 2000, 289, 905-
920

A. P. Cartet and W. M. Clemens Jr., et al., Science, 2001, 291, 498-501

U. Patel, Y. P. Yan, F. W. Hobbs Jr., J. Kaczmarczyk, A. M. Slee, D. L. Pompliano,
M. G. Kurilla and E. V. Bobkova, J. Bio. Chem. 2001, 276, 40, 37199-37205

F. Schlunzen, R. Zarivach, J. Harms, A. Bashan, A. Tocilj, R. Albrecht, A. Yonath
and F. Franceschi, Nature, 2001, 413, 814-821

J. R. Colca, W. G. McDonald, D. J. Waldon, L. M. Thomasco, R. C. Gadwood, E. T.
Lund, G. S. Cavey, W. R. Mathews, L. D. Adams, E. T. Cecil, J. D. Pearson, J. H.
Book, J. E. Mott, D. L. Shinabarger, L. Xiong, and A. S. Mankin, .1. Bio. Chem. 2003,
278, 21972-21979

a) G. E. Zurenko, et al., Curr. Opin. in Pharm, 2001, I, 470-476; b) D. Shinabarger,
Exp. Opin. Invest. Drugs, 1999, 8, 1195-1202

M. R. Barbachyn and C. W. Ford, Angew. Chem. Int. Ed. 2003, 42, 2010-2023

10

17. B. Gopalakrishnan, A. Khandelwal, S. A. Rajjak, N. Selvakumar, J. Das, S. Trehan, J.
Iqbal, and M. S. Kumar, Bioorg. & Med. Chem. 2003, 11, 2569-2574

18. W. A. Gregory, D. R. Brittelli, C. J. Wang, M. A. Wuonola, R. J. McRipley, D. C.
Eustice, V. S. Eberly, P. T. Bartholomew, A. M. Slee, and M. Forbes, J. Med. Chem.
1989, 32, 1673-1681

19. J. A. Tucker, D. A. Allwine, K. C. Grega, M. R. Barbachyn, J. L. Klock, J. L.
Adamski, S. J. Brickner, D. K. Hutchinson, C. W. Ford, G. E. Zurenko, R. A. Conradi,
P. S. Burton, and R. M. Jensen, J. Med. Chem. 1998, 41, 3727-3735

20. M. R. Barbachyn, S. J. Brickner, D. K. Hutchinson, US Patent, 5,688,792, 1997

21. S. J. Brickner, D. K. Hutchinson, M. R. Barbachyn, P. R. Manninen, D. A. Ulanowicz,
S. A. Garmon, K. C. Grega, S. K. Hendges, D. S. Toops, C. W. Ford, G. E. Zurenko,
J. Med. Chem. 1996, 39, 673

22. S. J. Brickner, US Patent, 5,164,510, 1992

23. C. S. Lee, D. A. Allwine, M. R. Barbachyn, K. C. Grega, L. A. dolak, C. W. ford, R.
M. Jensen, E. P. Seest, J. C. Hamel, R. D. Schaadt, D. Stapert, B. H. Yagi, G. E.
Zurenko, and M. J. Genin, Bioorg. & Med. Chem. 2001, 9, 3243-3253

24. O. A. Phillips, E. E. Udo, A. A. M. Ali, and N. Al-Hassawi, Bioorg. & Med. Chem.
2003, 11, 35-41

25. a) M. B. Gravestock, D. G. Acton, M. J. Betts, M. Dennis, G. Hatter, A. McGregor,
M. L. Swain, R. G. Wilson, L. Woods, and A. Wookey, Bioorg. &. Med. Chem. Lett.
2003, 13, 4179-4186; b) M. B. Gravestock, D. A. Roberts, M. J. Betts, S. D. Mills, G.
Hatter, PCT WO 2000021960

26. N. Selvakumar, D. Srinivas, M. K. Khera, M. S. Kumar, R. N. Marnidi, H. Sarnaik, C.
Charavaryamath, B. S. Rao, M. A. Raheem, J. Das, J. Iqbal, and R. Rajagopalan, J.
Med. Chem. 2002, 45, 3953-3962

27. R. Tokuyama, Y. Takahashi, Y. Tomita, T. Suzuki, T. Yoshida, N. Iwasaki, N. Kado,
E. Okezaki, and O. Nagata, Chem. Pharm. Bull. 2001, 49, 347-352

28. R. Tokuyama, Y. Takahashi, Y. Tomita, M. Tsubouchi, T. Yoshida, N. Iwasaki, N.
Kado, E. Okezaki, and O. Nagata, Chem. Pharm. Bull. 2001, 49, 353-360

29. R. Tokuyama, Y. Takahashi, Y. Tomita, M. Tsubouchi, N. Iwasaki, N. Kado, E.
Okezaki, and O. Nagata, Chem. Pharm. Bull. 2001, 49, 361-367

11

CHAPTER 2 SYNTHESIS OF OXAZOLIDINONE MOIETIES

2.1. INTRODUCTION

The oxazolidinone nucleus is a simple heterocyclic system that is not frequently
found in drug structure. The 4-substituted systems are better known because they are
readily obtained from amino acids by reduction and car'bonylation.35'36 (Scheme 2.1)
The 5-substituted oxazolidinones are more difficult to prepare but are of special

relevance to the area of antibacterials.

 

 

Scheme 2.1:
O O
LiAIH4 or 8H3 R 00x2 )L
H2N\)L 4, Y\OH O NH
; OH
amino acid X = CI or similar leaving group

A growing effort focused on the synthesis of 5-substituted oxazolidinones started
around 1987, when Dupont Company formally reported the structure and anti-bacterial
activity profiles of two agents, DuP- 105 and DuP-721. Subsequently, a great variety of
methodologies including asymmetric catalysis (transition metal and enzymatic),
bio-transforrnations, and chiral pool approaches such as chiral 3-carbon fragment or
synthon were developed to meet the needs of structure diversity of the oxazolidinone
candidates for drug discovery. Later on, SAR-guided oxazolidinone liberaries based on
solid phase synthesis were generated to accelerate the quality leads synthesis. More
recently, with the success of the first marketed antibacterial oxazolidinone drug

Linezolid, the development of more efficient, cost-effective, and large-scale synthesis

12

approaches is becoming imperative.

2.2. CHIRAL 3-CARBON SYNTHON

Among the numerous methodologies, the chiral 3-carbon fragment or synthon
condensed with an activated nitrogen compound dominates the literature on
antibacterial oxazolidinone synthesis. In the preparation of DUP-721, an aryl
isocyanate was reacted with commercially available (R)-glycidyl butyrate in the
presence of tri-n-butylphosphine oxide to yield the N-arylated oxazolidinones.
(Scheme 2.2) The hydroxymethyl group was then transformed to an acetamido group
by mesylation, displacement by azide, reduction to amino group and acylation to form
the S-acetamidomethyl group."2 This methodology was effective but not general
because of the unavailability of a large variety of isocyanates starting material.

Scheme 2.2:

0
he
wow—b xylene reflux
0
lNaOMe
O
0 o
NWOH

A more effective approach was developed by Manninen and co-workers at Pharmacia
and used in the preparation of Linezolid in 1996.3’4 In this approach, an aniline was
transformed to a carbamate by reaction with benzyloxycarbonyl chloride. The
carbamate was then lithiated and reacted with (R)-glycidyl butyrate to directly
generate (R)-3-aryl-5-(hydroxymethyl) oxazolidinones. (Scheme 2.3) The

13

hydroxymethyl substituents were then elaborated in a linear manner to the

5-acetamidomethyl group with the same process.

 

 

 

Scheme 2.3:
- OR H -
A’ H i
CBz-Cl 1) n-BuLi, THF, -78 °C \N/ .
ArNHz —> Al’NHCOzBfl > ‘0 0
2’ 0 0 H 0. I ,o ' "’
h K M ‘Li'
(ArNH2 = GUN-QNHZ) o
F
_ u
i- O - I- 1

O o
y... JL )k
Ar-N DU 0 _> Ar-N __, Ar-” 0
o

 

 

OH

 

 

It was subsequently found in an effort to develop the synthesis on a process scale that
(S)-3-chloro—l,2-propanediol could be used as a substitute for (R)-glycidyl butyrate
with lithium tert-butoxide as base and DMF and DMAc as solvent.5 From the
commercial standpoint, the resulting hydroxymethyl intermediates were activated by
nosylation and a large excess of ammonia (100 equiv) was used for the subsequent
displacement reaction.4 The excess of aqueous ammonia suppressed competing dimer
formation and accelerated the displacement reaction. To reduce the number of steps
from the aryl carbamate to the desired 5(S)-acetamido oxazolidinones, convergent
early amination approaches using chiral oxiranylmethylacetamide were explored.6’9
The chiral oxiranylmethylacetamide was obtained by dynamic kinetic resolution of
cheap racemic epichlorohydrin using Jacobsen’s chiral transition metal complex with

azide as the nucleophile.6 However, the 3-carbon fragment oxiranylmethylacetamide in

14

the Manninen conditions resulted in only a moderate yield (70%) due to its competitive
formation of the oxazoline byproduct. (Scheme 2.4) Also, the
oxiranylmethylacetamide was relatively unstable and could not be isolated on large
scale.

Scheme 2.4:

\—_<—NH O

/ A o
O / r \Ar—NH 0-H )k
0 > Ar—N O
\ / LiOlBu MeCN/THF \_k/NHAC

\o O

A\_ )\ Lio'Bu, THF

NH \ "Om

In 1998, Jacobsen developed a highly effective hydrolytic kinetic resolution of racemic

 

epichlorohydrin7 using chiral Mn-salen complexes (Scheme 2.5) and directed towards
the large-scale synthesis.8 The cost of (S)-epichlorohydrin became less prohibitive. In
1999, Chirex began making tonne quantities of (S)-epichlorohydrin from cheap
racemic epichlorohydrin using Jacobsen’s methodology. A large portion of the chiral
epichlorohydrin starting material used in the manufacture of antibacterial

oxazolidinones relies on this methodology.

 

Scheme 2.5:
o 9“
X (R, R)- -.(Salen)CoOAc(0 2- -2mol%) X to 5 g W0“
0.5 eq H20
racemic (R \/)\/

Jacobsen Hydrolytic Kinetic Resolution

15

Scheme 2.6: Chiral 3-Carbon Synthon

L\fix°Y\/ LNLIY LAN/“fix 13\,~.

(1) O (2) 0 (a) 0 (4)
J. Med. Chem. 1992. 35. 1156 Tetrahedron Lett. 1990. 37. 7937 WC 9940094. Aug. 12. 1999 Tetrahedron Lett. 1990. 37. 7937
J. Med. Chem. 1996. 3. 673 WO 9937641 . Jul. 29. 1999
’ Ac 0 OH OH
0 H >~o xA/x Cl Won
Ac

OCH3 x = Cl or Br.NH2 (7) (8)

(5) (6)
0,9. pm R. a D. 2000. 7’ 533 Tetrahedron “3,1993. 39, 4453 Tetrahedron Lett. 2000. 41. 4301 Tetrahedron Lett. 2000. 41.4301
Heterocycies 2001. 55. 353

It was found that the chiral 3-carbon intermediate (5) could afford
oxiranylmethylacetamide in situ by deacetylation in the Manninen condition and (5)
was crystalline in enantiopure form making it an attractive synthetic target for
large-scale synthesis. (Scheme 2.6) Thus a shortened route of one-pot synthesis of
5(S)-acetamidomethyl-2-oxazolidinone was developed with N-aryl carbamate and
(5).9

In 2001, E. Steckhan et a1lo reported an even more direct and concise route for
preparing 5-substituted oxazolidinones by starting with the same chiral synthon
epichlorohydrin, which is now commercially available in both enantiomerically pure
forms. Either enantiomer can be transformed in one step into the respective chiral
5-chloromethyl-2-oxazolidinone in nearly quantitative yield by using potassium
cyanate as nucleophile for the opening of the epoxide. This synthesis has been
described previously for racemic 5-chloromethyl-2-oxazolidinone from racemic
epichlorohydrin.ll (Scheme 2.7) It is noteworthy that the reaction was carried out in

aqueous solution and the potassium cyanate is considerably cheap.

16

Scheme 2.7:

o o
0 2 eq KOCN i 0 2 eq KOCN )k
z 5 ————> —>
-., CI H t, HN 0 ROI H O. uant. RN 0
’II/ 20.01113" 2 q
(S) L—Jm/CI (a) CI

Some other interesting 3-carbon chiral synthon manipulations to afford 5-substituted
oxazolidinones have been reported such as the reaction of cyclic carbonate with
carbamate ester,12 Vilarrasa reductive cyclization of hydroxyl azide,” (Scheme 2.8)
iodolactonization of allylic olefin,l4 Sharpless epoxidation followed by tandem Payne
rearrangement and cyclization of carbamate,l5 (Scheme 2.9) etc. These methodologies
demonstrate the versatility of synthesis of antibacterial oxazolidinone moieties from

various perspectives.

 

Scheme 2.8:
o o
O 0&0
PMe
1. NaN3. NH4CI NH N3 Na 3 co Na 1
H 2. n-BuLi. C02. PMreg A/Bz "" ‘> 32 T2» A/82 -> “N O
Vilarrasa Red. Cyc. . F11 ; n H, .
67 % We'd : ' é MeaPO :2 a
TMS 94 °/o ee OH 0.. OH 1 R2
TMS
Scheme 2.9:

.9“ o
that... o 1 cf
x/\=_/\X ——a» —> Ar—NMOH £11.. Ar—N Jp—o Af’ O
002R zl—on ayne rearr
2.3. CHIRAL AUXILIARIES
The use of chiral auxiliaries to form optically enriched oxazolidinones from achiral

3-synthons represents another approach to the synthesis of oxazolidinones. In one

17

example, prochiral 1,3-dibromO-2-propanol was converted into carbamate with
optically pure l-phenylethylisocyanate or l-phenylethylamine. Intramolecular
asymmetric desymmetrization with base afforded the protected oxazolidinone moieties
with moderate yield and diastereoselectivity. The diastereomers were separated by
chromatography and the 5-bromomethyl oxazolidinones were obtained after
debenzylation with anisole-methanesulfonic acid.” (Scheme 2.10)

Scheme 2.10:

    

i I JUL ML ....... it

   

OH O:C:NJ\

 

 

N 0 Ar rMsOI-I 50 °c 0‘ {NH
CuCl. DMF. rt. or A base X \“.'\—j X \ c.
t —. . \
Asym. Desym. 69 % 109'“

X X /I\ X X 50%de

X=CI or Br HZN Ar Ar=Ph or 1-naphthyl
tnphosgene. Py.

DMAP. 0-2500 0
OJLN/LAr
.34

However, the presence of chiral auxiliary may not necessarily generate
enantiomerically enriched oxazolidinones. In another example, the palladium
catalyst-mediated reaction of cis-2-butenylene dicarbonate and chiral amine failed to
give enantio-discriminated oxazolidinones, thus chiral ligand was added to promote
the enantio-differentiation and afford the 5-vinyl oxazolidinone moieties in moderate

yield and fair to good diastereoselectivity.l7 (Scheme 2.11)

 

Scheme 2.11:
. L‘s) “ch. OCOMe
PhANHz *
(S) Pd 0 L. 0 Ph2P
Me0200/\=/\0002Me + or ( )' 0 Fe
66°/o yield . Em) P02?
PhiNHz 900/0 d6 /\S\N /.\Ph
(F1) 0 1.1: (S)-(R)-BPPFOAc

l8

2.4. ENZYMATIC AND BIOCATALYSIS

Enzymatic resolution and biotransforrnation have been manipulated to generate chiral
synthon intermediate that can be used to form 5-substituted oxazolidinones. Lipase B
from Candida antarctica catalyzed the kinetic resolution of racemic
ethyl-4—chloro-3-hydroxybutanoate through an ammonolysis reaction, selectively
forming the (R)-B-hydroxyamide intermediate which could be processed to yield chiral
5-chloromethyl-2-oxazolidinone via a Hoffmann rearrangement.l8 (Scheme 2.12) Just
like the normal kinetic resolution, the maximum yield of the product was 50 %.
Another method employed bioreduction with S. cerevisiae (Baker’s yeast) to produce
homochiral B-hydroxy esters. The B-hydroxy esters were transformed into their
hydrazides and furnished optically pure 5-substituted oxazolidinones via a Curtius
rearrangement.‘9 (Scheme 2.13) It is worth mentioning that both of these concise
routes have the potential of process-scale synthesis due to the highly effective catalytic

ability of enzymatic transformation.

Scheme 2.12:
OH 0 OH 0 OH 0
En me '5
GM + NH3 _z_y_’ 0M + CI\/:\/"\
oer NH2 oer
racemic
NH3
0
PIDA: PhI(0Ac>2 0% PIDA,THF/MeOH 2“ °

 

:

C' \\\““K/NH fHoffmann Rearr. Cl

NH2

19

Scheme 2.13:

 

 

 

 

 

 

 

H.
S. Cerevisiae § '5' l
0028 = ““0025: “”22"“?!I “x |\
m baker's yeast reduction Fad: -EtOH mfi- NHNH2
R2 R2 Hz
Curtius Rearr. NaNOle“
R1 H2 -H20, -N2
CH3 H ..
CHZCI H O P H’il)
CF3 H 0/ C
\(CH2)3" Q g "
~(CH2>.J = ,NH ‘— - “(N
Fl2 u R2 ..
2.5. CHIRAL POOL

The chiral pool represents another important access to optically pure oxazolidinones.
Carbohydrates and their derivatives are known for their ready availability and chiral
richness in the syntheses of antibacterial oxazolidinones. In 1995, Danielmeier and
Steckhan reported synthesis of optically pure 5-hydroxymethyl-2-oxazolidinones
starting from D-mannitol (6 steps with overall yield 19%), L-ascorbic acid (9 steps,
7%), and (R)- or (S)-malic acid (5 steps, 42%) respectively.2°~2‘ (Scheme 2.14) The
syntheses involved multiple steps and complicated conversions of functional groups
from chiral natural precursors. However, this was the first example of exploiting
stereochemistry from the chiral natural resources for the preparation of 5-substituted
oxazolidinones. A related synthesis was reported in which the Cz-symmetry character
of the D—mannitol was manipulated in the reactions.22 Two directional syntheses
starting from l,2,5,6-dianhydro-3,4-O-isopropylidine-D-mannitol with aniline and

then with carbonyldiirnidazole formed a Cz-symmetric bis-oxazolidinone. Removal of

20

acetal group followed by periodate oxidation of the diol formed two molecules of
identical S-formyloxazolidinone. Reduction of the aldehyde group yielded the
enantiomerically pure (S)-5-hydroxymethyl-2-oxazolidinone.22'24 (Scheme 2.15)

Scheme 2.14:

 

Ho 0 0H 97”
2 e TFAA -‘ BnOH H020 pH DPPA, Et3N CON :0“
R ———~ A —- a ——-» H
COzH O O O 0023" 0028n
(S)-malic acid 1 A
it it
HN p <— HN p
. OH \ .
R/ 0028n
Scheme 2.15:
HC'fi O O
Ho—+
o p o
HO—l __> _. -‘ _._ __’ Ar-NJLO QAN-Ar k
r—OH .- \—-k_(—/ _. Af—N 0
-0. x
D-mannitol

Another report exploiting Cz-symmetry in the natural precusor is from dimethyl
L-tartrate. The starting L—tartrate underwent mono-functionalization according to
Font’s methodology to afford a tosylate derivative.25 Aryl amination followed by
deprotection of acetal, and carbonylation reached the 3-N-aryl-5-(l ’-hydroxyethyl)

oxazolidinones.‘5 (Scheme 2.16)

Scheme 2.16:
Me02C 002m TSOH2C ArNHg. Et3N Ari-{NV "’OH
"I —-> —b .1 —vb o '0 —* —> _
HO OH o><o x A’ "Y0
o

(L)-tartrate

21

Though abundantly available from nature as the chiral precursor in the oxazolidinone
synthesis, regio- and stereocontrolled functionalization of carbohydrate derivatives in
a nimble and efficient way is still an extremely important issue to commercialize the
synthesis. On the other hand, the complexity and chiral richness of carbohydrates
allow for a high degree of flexibility in the functional group transformation.

The introduction of commercially important methods for preparing optically pure
3,4-dihydroxybutyric acid and various 3- and 4-carbon derivatives has opened up ways
to the preparation of optically pure oxazolidinones.2("31 The ready availability of
enantiopure 3,4-dihydroxybutyramides by treatment of 3-hydroxybutyrolactone by
ammonia has facilitated several routes. In one method the isopropylidene acetal of
3,4-dihydroxybutyramide was subjected to Curtius rearrangement to afford the
protected 3-amino-1,2-dihydroxypropane. This latter intermediate can be deprotected
and converted to 5-hydroxymethyl-2-oxazolidinone by treatment with
carbonylirnidazole.32 (Scheme 2.17) The 3-amino-1,2-dihydroxypropane can also be
converted to l-halo-2,3—dihydroxypropanes which can be used in alternative synthesis
of oxazolidinones.

Scheme 2.17:

O O
D-sugar _> _’ fio 1. NH3. H20 NH2 OCI'IOH' +0
HO 2. dimethoxypropane O OWNHZ
acetone. H“ )(o

 

 

1. H+
2. “Mb-CO

i
O N

HO

 

H

22

Enantiopure 3,4-dihydroxybutyrarnide can also be converted to its 4-trityloxy
derivative that readily yields 5-tn'tyloxymethyl-2-oxazolidinone when subjected to
Hoffmann rearrangement.33 (Scheme 2.18) The resultant can be used for the quick and
efficient synthesis of oxazolidinone template familes as antibacterial drug candidates.
One-pot synthesis by means of dynamic protection of enantiopure
3,4-dihydroxybutyramide with alkyl or arylboronic acids followed by Hoffmann
rearrangement yields the free 5-hydroxymethyloxazolidinones directly in high yield.34
(Scheme 2.19) This is an important new development since enantiopure
3,4-dihydroxybutyramides are available in only 2 steps from starch, lactose, maltose

and hemicelluloses. The boronic acid derivatives can be generated in situ and

 

 

transformed directly.
Scheme 2.18:
0H 0 TrCl/Pv 0“ ° Hoffinann Rearr. ’R 0

H0 “OM i/ ’IC’IO + —* 0%

NH2 NHz H0 N \J H “(RA/NH
Scheme 2.19:

NH2 ,a—o o ,8-0 *0 We 0‘2 ‘0

R H 0H
j...
R = n-Bu, Phenyl. 2.5-dimethoxy phenyl, 2,6-dimethoxy phenyl O

2.6. CARBONYLATION OF fl-AMINOALCOHOLS

One of the conventional approaches to make oxazolidinones is the carbonylation of

35-36

B-amino alcohols with toxic phosgene-based reagents or cyanates”. Recently,

23

several groups have reported on the use of carbon monoxide and carbon dioxide with
B—amino alcohols including palladium-catalyzed oxidative carbonylation with carbon
monoxide,”39 (Scheme 2.20) sulfur-assisted thiocarboxylation with carbon monoxide
followed by oxidative cyclization with oxygen,40 (Scheme 2.21) and dehydrative
condensation with carbon dioxide catalyzed by dibutyltin oxide“ (BuzsnO) or
1,3-Dicyclohexylcarbodiimide (DCC) (Scheme 2.22).42 Due to the prohibitive cost of
chiral B-amino alcohols that could be used in the synthesis of enantiomerically pure
5-substituted oxazolidinones, these methods were complementary to the preparation of

antibacterial oxazolidinones.

 

 

 

Scheme 2.20:
31 R2 R R
R1 R2 Pdl2(cat.) H 1H 2
l‘ < H0 NH O NH
Ho NH2 00.02 T
IN 0 0
Scheme 2.21:
CO+S 020:3
O O
NH2 NHJLS' 02 NH’LS ii
+ 0=C=S ——* E —" E '—"’ E>=O +H2Sn
OH 0
OH OH 2
Scheme 2.22:
R2 Rs
R2 93 Gog/“BqunO/NMP 7—(
fi- R1HN o
R1HN OH 5Mpa,180°C \[r
0
Ph Ph

COg/DCC H
H2N 0“ 70°C.300kg/cm2 ”NYC

0

 

24

2.7. SOLID-PHASE SYNTHESIS

SAR studies have revealed that substitution features surrounding the oxazolidinone
nucleus, especially the 3 and 5 attachments are responsible for the biological acitivity
of the entire molecule. The character of oxazolidinones inspired the preparation of
combinatorial libraries of oxazolidinones to discover the next generation leads with
enhanced potency and improved activity ranges compared to Linezolid. In one
example, a set of commercially available chiral 1,2-diols was immobilized by reaction
with polymer-bound sulfonyl chloride which also activated the primary alcohol sites.
The subsequent reaction of the secondary alcohol with p-toluenesulfonyl isocyanate
gave carbamates that were then susceptible to base-promoted cycloelirnination and
afford the enantiopure oxazolidinones with concurrent detachment from the resin. The

43‘“ (Scheme

detached sulfonate could be readily regenerated using thionyl chloride.
2.23) This methodology could provide the preparation of optically pure libraries
because it started from chiral diol.

Scheme 2.23:

 

O
HO/YOH % 0-80 80-2 _O/\'c,OH__. TsNCO 0-802- O/YOO NHTs DBN/CHZCIZ Ts- N63)

R ElaN. 01-12012 CH 2012 CH2C|2
' R

a = Me, Et. tBu. CHzCHzF’h 0—3020:

0......

 

Another solid-phase synthesis of oxazolidinones involving a cyclization/cleavage
reaction is reported earlier. Resin-bound carbamates were alkylated with

glycidyltosylate to the corresponding epoxides. Nucleophilic opening of the epoxides

25

with pyrrolidine and subsequent cyclization with concurrent detachment of the resin
yielded functionalized oxazolidinones.45 (Scheme 2.24)

Scheme 2.24:

OHOA H
BOA LiN(TMS) UC'04 ”’3?
-=- 0A°rr "(jar—fl 2 Q

OCN H.331): CHzcr2 o Rglyoidyltosylate QAOT'L/LA :(fl {NOR

 

Gordeev from Versicor Inc. reported one of the most practical approaches for the
solid-phase syntheses of oxazolidinone libraries. A library of
3-aryl-5-aminomethyl-oxazolidinone templates with various functionalities in the aryl
group was first prepared via conventional solution-phase reaction described in Scheme
2.3. Then the oxazolidinones were immobilized with a resin via their amino groups and
modification of the 5-position was made by solid-phase synthesis followed by

detachment from the resin.4647

(Scheme 2.25) This methodology also allows the
preparation of enantiopure libraries because the oxazolidinone skeleton is prepared
from optically pure glycidylbutyrate. This approach has been used to prepare quite a
few oxazolidinone libraries and led to the discovery of some new antibacterial
oxazolininones with improved potency and broadened spectrum compared with

. . 9
Linezolrdflg‘1

26

Scheme 2.25:

R2 “3
RFQ'
R4 Rs

08201
NHz —’ Rt

35ch

H

"Pr

0
R2 R3 LVNOR
«Q—NHCOZBn "BU“
R4 Rs

930

NOX 1) N50!
R5N\/‘\/0H_. 2) NH3 H20

92 R2 R3 0

)k
£§~JL~H
Fla Rs
cerc
(4,00vo

 

HcozH. HATU. 0'“ 0'

 

"\gn CHZCI2

RNCO.DMFor R2 9 o o g)
RNCS.DMFor 3
ROCOCI.DIEA.DMF k0 neo
a N\/K/N
R4 R5

27

2.8. REFERENCE

1.

10.

ll.

12.

13.

a) W. A. Gregory, D. R. Brittelli, C. L. Wang, M. A. Wounola, R. J. Mcripley, D. C.
Eustice, V. S. Eberly, P. T. Bartholomew, A. M. Slee and M. Forbes, J. Med. Chem.
1989, 32, 1673-1681; b) C. L. Wang, W. A. Gregory, M. A. Wuonola, Tetrahedron,
1989, 45, 1323-1326

C. H. Park, D. R. Brittelli, C. L. Wang, F. D. Marsh, W. A. Gregory, M. A. Wuonola,
R. J. Mcripley, V. S. Eberly, A. M. Slee, M. Forbes, J. Med. Chem. 1992, 35,
1156-1165

a) S. J. Brickner, D. K. Hutchinson, M. R. Barbachyn, P. R. Manninen, D. A.
Ulanowicz, S. A. Gannon, K, C. Grega, S. K. Hendges, D. S. Toops, C. W. ford and
G. E. Zurenko, J. Med. Chem. 1996, 39, 673-679; b) S. J. Brickner, Curr: Pharma.
Design, 1996, 2, 175-194

M. R. Barbachyn and C. W. Ford, Angew. Chem. Int. Ed, 2003, 42, 2010-2023

B. A. Pearlman, W. R. Perrault, M. R. Barbachyn, P. R. Manninen, D. S. Toops, D.
J. Houser, T. J. Fleck, US 5837870, 1998 [Chem Abstr 1998, 130, 25061]

S. E. Schaus and E. N. Jacobsen, Tetrahedron Lett. 1996, 37, 7937-7940

M. E. Furrow, S. E. Schaus, and E. N. Jacobsen, J. Org. Chem. 1998, 63,
6776-6777

a) E. N. Jacobsen, M. Tokunaga, WOOOO9463, 2000; b) E. N. Jacobsen, M.
Tokunaga, J. F, Larrow, US 5929232, 2001

. a) W. R. Perrault, B. A. Pearlman, D. B. Godrej, A. Jeganathan, K. Yamagata, J. J.

Chen, C. V. Lu, P. M. Herrinton, R. C. Gadwood, L. Chan, M. A. Lyster, M. T.
Maloney, J. A. Moeslein, M. L. Greene, and M. R. Barbachyn, Org. Process R & D,
2003, 7, 533-546; b) W. R. Perrault, PCT. Int. Appl. WO 02085849, 2002

K. S. Amdt, D. Kolter, K. Danielmeier, and E. Steckhan, Eur. J. Org. Chem. 2001,
13, 2425-2433

a) E. Patemo, and E. Cingolani, Gazz. Chim. Ital. 1908, 38, 243; b) A. L. Thomsen,
Ber: Dtsch. Chem. Ges. 1878, I], 2136

S. Jegham, A. Nedelec, P. Burnier, Y. Guminski, F. Puech, J. J. Koenig and P.
George, Tetrahedron Lett. 1998, 39, 4453-4454

a) K. S. Feldmarr, J. C. Saunders, and M. L. Wrobleski, J. Org. Chem. 2002, 67,

28

14.

15.

l6.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28

29

30

7096-7109; b) X. Ariza, O. Pineda, F. Urpi, J. Vilarrasa, Tetrahedron Lett. 2001, 42,
4995-4999

G Cardillo, M. Orena, S. sandri, and C. Tomasini, Tetrahedron, 1987, 43,
2505-2512

J. P. Genet, A. Denis, and A. Vilar, Tetrahedron Lett. 1990, 31, 515-518
S. Sugiyama, K. Morishita, and K. Ishii, Heterocycles, 2001, 55, 353-364
S. Tanimon', U. Inaba, Y. Kato, and M. Kirihata, Tetrahedron, 2003, 59, 3745-3751

E. Garcia-Urdiales, F. Rebolledo, and V. Gotor, Tetrahedron Asymm. 1999, 10,
721-726

M. Bertau, M. Burli, E. Hungerbuhler, and P. Wagner, Tetrahedron Asymm. 2001,
12, 2103-2107

K. Danielmeier, and E. Steckhan, Tetrahedron Asymm. 1995, 6, 1181-1190
R. Andruszkiewicz, and M. Wyszogrodzka, Synlett, 2002, 12, 2101-2103

B. B. Lohray, S. Baskaran, B. S. Rao, B. Y. Reddy and I. N. Rao, Tetrahedron Lett.
1999, 40, 4855-4856

M. Zelgert, M. Nieger, M. Lennartz, and E. Steckhan, Tetrahedron, 2002, 58,
2641-2646

For the synthesis of l,2,5,6-dianhydro-3,4-O-isopropylidine-D-mannitol from
D-mannitol, see: B. Lohray, M. Chatterjee, and Y. Jayamma, Synth. Commun. 1997,
27,1711-1724

R. M. Ortuno, D. Alonso, J. Cardellach, and J. Font, Tetrahedron, 1987, 43,
2191-2198

R. Hollingsworth, Biotech. Ann. Rev. 1996, 2, 281-291

G. Wang and R. I. Hollingsworth, Tetrahedron Asymm. 1999, 10, 1895-1901
R. I. Hollingsworth and G Wang, Chimica Oggi, 2000, 9, 40-42

R. I. Hollingsworth,J. Org. Chem. 1999, 64, 7633-7634

R. I. Hollingsworth and G Wang, Chem. Revs. 2000, 100, 4267-4282

29

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49

G. Huang and R. Hollingsworth, Tetrahedron Asymm. 1999, 9, 4113-4115
G. Wang and R. I. Hollingsworth, J. Org. Chem. 1999, 64, 1036-1038

G. Wang and R. I. Hollingsworth, Tetrahedron Asymm. 2000, 12, 4429-4432
R. I. Hollingsworth and G Wang, US patent 2001, 6,288,238

M. P. Sibi and P. A. Renhowe, Tetrahedron Lett. 1990, 31, 7407-7410

0. Tamura, M. Hashimoto, Y. Kobayashi, T. Katoh, K. Nakatani, M. Kamada, I.
Hayakawa, T. Akiba, S. Terashima, Tetrahedron Lett. 1992, 33, 3487-3490

T. Ishimaru, Nippon Kagaku Zasshi, 1960, 8], 1589-1592

B. Gabriele, G Salerno, D. Brindisi, M. Costa, and G. P. Chiusoli, Org. Lett. 2000,
2, 625-627

B. Gabriele, R. Mancuso, G Salerno, and M. Costa, J. Org. Chem. 2003, 68,
601-604

T. Mizuno, J. Takahashi, and A. Ogawa, Tetrahedron, 2002, 58, 7805-7808
K. Tominaga, and Y. Sasaki, Synlett, 2002, 2, 307-309

K. Hajirne, I. Yutaka, H. Kiyotaka, and T. Kazuo, Jpn. Kokai Tokkyo Koho, JP
2003064062, Mar. 05, 2003

P. Holte, L. Thijs and B. Zwanenburg, Tetrahedron Lett. 1998, 39, 7407-7410

P. Holte, B. Esseveldt, L. Thijs, B. Zwanenburg, Eur. J. Org. Chem. 2001,
2965-2969

H. P. Buchstaller, Tetrahedron, 1998, 54, 3465-3470

M. F. Gordeev, G. W. Luehr, D. V. Patel, and R. C. Gadwood, US. Patent 2002,
6,441,005

M. F. Gordeev, Curr. Opin. Drug Dis. & Dev. 2001, 4(4), 450-461

M. F. Gordeev, A. Renslo, G Luehr, S. Lam, N. E. Westlund, and D. V. Patel, US.
Patent, WO 2003062231

U. Singh, B. Raju, S. Lam, J. Zhou, R. C. Gadwood, C. W. Ford, G E. Zurenko, R.

30

D. Schaadt, S. E. Morin, W. J. Adams, J. M. Friis, M. Courtney, J. Palandra, C.
Hackbarth, S. Lopez, C. Wu, K. H. Mortell, J. Trias, Z. Yuan, D. V. Patel, M. F.
Gordeev, Bioorg. & Med. Chem. Lett. 2003, 13, 4209-4212

31

CHAPTER 3 AN EFFICIENT ROUTE TOWARDS HOMOCHIRAL 5-D-
ERYTHRO-Z-OXAZOLIDINONES, IMPORTANT CHIRAL PRECURSORS OF

DIVERSE PHARMACEUTICALS ESPECIALLY ANTIBIOTICS

3.1. INTRODUCTION

The synthesis of biologically active oxazolidinones has been of keen interest in
the last two decades because they have been shown to be multitherapeutic candidates
including antibiotics,l antifungal,2 and antidepressants.3 For example, Linezolid
(Zyvox®), has been approved in 2000 as a new antibiotic drug with a completely
different mechanism of action since 19508. Numerous methods and approaches have been
developed thus far including asymmetric catalysis, biotransformations, chiral pool
approaches, etc.
The 3-carbon chiral synthon has been extensively used in the synthesis of 5-substituted
oxazolidinones and results in one carbon unit in the 5-sidechain. Based on the former
structure-activity relationship (SAR) studies, limited to the carbonyl functionalities, most
of the functional group transformation on the 5-substituents was l-carbon unit such as the
acetamido-methyl group. (Figure 3.1: 1) Although recent studies have reported variation
of the 5-substituents such as S-thiourea, 5-thiocarbamate, S-dithiocarbamates and N- or
O-substituted heterocyclic rings as potential replacements for the acetamidomethyl side
chain and found quite a few antibacterial candidates with increased potency and
broadened therapeutic spectrum, but a concise and efficient synthetic route of novel
oxazolidinones with more than one carbon unit in the 5-substituents was still not

available. (Figure 3.1: 2) In comparison with the extensively studied antibacterial

32

oxazolidinone series 1, the new molecules 2 are a series of 5-substituted-2-
oxazolidinones in which one of the hydrogen atoms is replaced by a carbon unit, a
heteroatom and consequently an additional chiral center compared to the earlier
candidates. This novel substitution pattern has never been explored before. According to
the generally accepted mechanism of antibacterial action, oxazolidinones bind to the 23S
ribosomal subunit and prevent the formation of the functional initiation complex (fMet-
tRNA-mRNA-70S), thus inhibiting protein synthesis at the initiation step. It is expected
that the extended substitutions in the 5-side chain of oxazolidinones could have the
chance to contribute extra substrate-acceptor binding to the ribosomal binding site, and
change the subtle conformation required for increased antibacterial activities. Also, this
modification could associate with the 3-substituent and optimize to obtain the desired
lipophilicity. More importantly, this structural modification could trigger a new round of
SAR study of various R1, R2, R3, X, Y functional groups and establish oxazolidinone

libraries that would lead to new drug discovery based on functional group diversity.

Figure 3.1: 5-Substituent modification

 

 

 

----------- ->
H“"
H
1
® Aromatic or aliphatic (including heterocyclic) ® Aromatic or aliphatic (including heterocyclic)
9 One chiral center 0 Two chiral centers
® Various heteroatoms ® CD Various heteroatoms

Alkyl 9' 86" With a wide 8086101?“ 9' 1110015003“)! 69 Alkyl or aryl with a wide spectrum of functionality
'00“!de heterocycles 30d fused ""9 systems including heterocycles and fused ring systems

33

3.2. DESIGN AND SYNTHESIS

The efficient synthesis of the target oxazolidinones 2 is challenging because the small
molecules incorporate highly condensed firnctionalities and two stereogenic centers.
Carbohydrate based chiral synthons have the advantage in the preparation of the
oxazolidinone 2 series molecules because a number of chiralities and functionalities have
already integrated into the molecular target. Moreover, carbohydrate abundance could
lead to the substantial large-scale synthesis in industry.4

Chiral 4-carbon building blocks such as tartaric acid, butadiene diepoxide, 3,4-0-
isopropylidine—L-erythronate, D-erythronate, L-threose, D-erythrose, L-threonolactone,
D-erythronolactone, etc, could play an important role in the synthesis of oxazolidinones.
3. (Figure 3.2) While the former compounds tartaric acids 4 and their derivative
diepoxides 5 are available in abundance as C2 symmetric D, L, or meso forms, the latter
compounds 6-7 and their isopropylidene derivatives are the most versatile four-carbon
entities being prepared by degradative processes starting with L-rhamnose, D-ribose, D-

ribonolactone, potassium glucuronate, ascorbic acid, arabinose, etc.5’6

34

Figure 3.2: Chiral 4-carbon building blocks

 

 

 

 

CHO
nozc 002R I
* * CH2
F
HO OH x X * N"‘OH
O N—R1
—OH
O * Y * 7
. t: ( .
O
5 Z 3
X = O or S
0 Y = O, NH, or other function
Z = O. NH, or other function
*
*
HO 6

 

For the C2 symmetric four-carbon units, the synthesis involves monofunctionalization of
the precursor termini; Threo- series oxazolidinones could then be prepared after
carbonylation of the corresponding chiral amino alcohol. For example, commercially
available L-tartrate could afford 5-L-threo-2-oxazolidinones in this fashion. (Figure 3.3)
However, the erythro- series oxazolidinones require the introduction of chiral factors to
enantioselectively differentiate between the termini of the mesa precursors thus
complicating the process. An alternative route that circumvents the desymmetrization
process could start from the L-threonolactone and D-erythronolactone. This process
involves aminolysis of the starting lactones, correct protection with isopropylidenes,
reduction, and carbonylation. A related synthesis could begin from cheaper 5, 6-0-
isopropylidene L- and D-ascorbic acids.7 Oxidative degradation with hydrogen peroxide

in basic condition affords 3,4-O-isopropylidine-L-threonate and D-erythronate

35

respectively.(Figure 3.4) Similar process involving aminolysis, reduction and
carbonylation could lead to the desired oxazolidinones.8 The latter route could effectively
circumvent the need for regioselectivity in the acetal protection step by protection of
ascorbic acid at an early stage.9 However, this approach has several limitations and
drawbacks. The chirality at the (it-hydroxyl function adjacent to the carbonyl could be
scrambled during the reaction and lose the diastereoselectivity. Only two diastereomers,
D-erythro- and L-threo-2-oxazolidinones would be afforded due to the availability of the
starting ascorbic acids. The other two configurations would need to be prepared by other
approaches. Furthermore, the prohibitive cost of the starting L-threono- and D-
erythronolactone and the requirement of the toxic phosgene or phosgene related
derivatives in the carbonylation step hamper these methodologies to be applied in large-

scale synthesis.

Figure 3.3: Strategies for preparing oxazolidinones

 

O O
E O mono functionalization
i; Nu‘ > 0 0H
m , ’I ,,
(3‘ W NNU
W" OH
0
0°: O N
g N ‘ asymmetric desymmetrization N”

 

' "’0 . > OH
\JNU 0,,“

o"... ‘3on9 002m
Ph"< NHS (19(1) ph AV<OHM [H] ohm CHOHZ 1) H+
oi ' PM< H2) co “0

CC32M" 0 CONH2 o
D-tartrate D- threo-2-oxazoolidinon;D

36

Figure 3.4: Strategies for preparing oxazolidinones (continued)

OK 0
0 Hg 0 HO 4
0 -ooc o 2) NH3 H N\/—\/ ' o
_ 3) LAH 2 O ><
HO OH O

D-asoorbic acid D-erythro-Z-oxazolidinone
><° °
0 X
2) Mia MAO 0”,,
— 3) LAH >< '

HO OH

L-ascorbic acid L-threo-Z-oxazolidinone

On the other hand, the exploitation of the natural 5-carbon pentoses as synthons would
probably be the optimum choice as chiral raw materials. For instance, 2-deoxy-D-ribose
could afford D-erythro-2-oxazolidinones in only four steps with high yield including
oxidation, arrrinolysis, isopropylidene protection and Hoffrnann rearrangement. The
advantage of this approach includes, firstly, the reagents used are all conventional and
generally inexpensive. Secondly, during the reaction, the correct C3 and C4
stereochemistry would remain intact throughout the reaction sequence and therefore
ensure both chiralities being inherited by the final product, which guarantees high
enantiomeric purity. Thirdly, the resulting isocyanate intermediate formed during the
Hoffmann rearrangement could trap a vicinal hydroxyl group as a nucleophile in
cyclization to form an oxazolidinone system so that a separate carbonylation reaction
using phosgene, ethyl-chloroformate or some similar reagent would be avoided. Thus, an
atom economical “green” synthetic methodology could be developed from 2-

deoxypentoses and used in the large-scale synthesis. Fourthly, the availability of all the

37

four 2-deoxypentose sugar from a variety of cheaper precursors allow the transformation
to all the four configurations of the desired products with the same methodology. (Figure
3.5) The novel oxazolidinone products would be greatly amenable to functional
modification. The hydroxyl function could be transformed into a wide variety of
substituents. And N-arylation of the corresponding function by means of Buchwald and
Hartwi g protocol could give the novel bioactive oxazolidinone candidates.IO

From a broader perspective, this methodology is not only important to the target
oxazolidinones themselves, but even more to optically pure 4-carbon B-amino alcohols
such as 9, which can easily be obtained from the corresponding 2-oxazolidinones by
basic hydrolysis. For instance, the four-carbon chiral synthon l-amino-erythronotriol
converted from asymmetric desymmetrization of the corresponding tetraol 10 could be
complex.l "'4 However, Hoffrnann rearrangement from the corresponding 5-carbon amide
1 1 followed by base hydrolysis (acyl carbon is lost) could easily reach the target. (Figure
3.6) The 4-carbon B-amino alcohol subunits are important structural element in a number
of different enzyme inhibitors, B-blockers, and chiral synthon used in the natural product
synthesis. The amino and hydroxy functional groups are very amenable to

transformations to other functionalities.

38

Figure 3.5: Synthesis of novel oxazolidinones from 2-deoxypentoses

 

 

 

 

 

 

 

 

 

 

 

 

 

(IDHO CIIHO
CH2 CH2
It L—"OH or HO— It
t —0H t r—OH
O (S)
—OH —OH
2-deoxy-D-nbose 2-deoxy-D-xylose 0 N—R1
-—>
CHO CHO —>
I I V .
CH2 CH2 ( ‘
H0—‘ at or e —0H 7' 3
HO—‘ a: HO— 4:
HO-—-‘ HO---
2-deoxy-L-ribose 2-deoxy-L-xylose
Figure 3.6: Retrosynthetic analysis
—OH —NH2 CONH2
Asymmetric Hoffmann I
—OH Desymmetrization ——OH Rearrangement CH2
—OH (5 J i—"OH r :) —OH
—-OH
10 9 11

39

3.3. RESULTS AND DISCUSSION

The reactions are summarized in Scheme 3.1. Commercially available 2-deoxyribose was
first oxidized to 5-membered lactone 12 by bromination in darkness at 0 °C for 36 h using
standard procedure.'5 Aminolysis of lactone 12 with 28-30 % ammonium hydroxide at
room temperature opened the lactone ring quantitatively to give 2-deoxy-erythro-
pentyrarnide 11, which upon treatment with 2,2-dimethoxypropane in dry DMF and
catalytic amount of p-toluenesulfonic acid afforded the desired 4,5-isopropylidene acetal
15 and 3,5-O-isopropylidene 14 in about 3:1 ratio (separated by recrystallization),
together with trace amount (5-6 mol%) of bisacetal 13. The problem with this step is the
lower regioselectivity in the reaction. A pure starting material 11 is required because any
moisture or inorganic salt from the earlier steps could make the reaction inoperative. And
a slight excess of acid and the 2,2-dimethoxypropane could result in appreciable
byproduct 13, which is thermodynamically more stable. Selective hydrolysis of bisacetal
13 gave undesired aminal product. The acetal 15 was then subjected to Hoffmann
rearrangement condition to result in D-erythro-Z-oxazolidinones 16 in high yield. And
3,5-O-isopropylidene 14 could be hydrolyzed to amide 11 with 70 % acetic acid and went
back to the reaction route. Decarbonylation of the resulting oxazolidinones 16 with
potassium hydroxide can give corresponding chiral B-amino alcohol 9. This reaction
sequence involved efficient four-step synthesis of enantiomerically pure D-erythro-2-
oxazolidinones as antibacterial oxazolidinone precursors using very mild reaction
condition and inexpensive, environmentally tolerated reagents. The overall yield was 51

%. Using the same methodology, the other three configurations of the desired

40

oxazolidinones should be obtained from 2-deoxy-D-xylose and their L-pentose

counterparts.
Scheme 3.1:
(IJHO o 0
CH2
a o b NH2
OH ——-> ———>
OH 12 1 1
2-deoxy-D-ribose
e c
O
OH O

NH2 WL
o/k + 0 “\OH + O E NH2
0
2 'O 14 X 15

(ratio 14: 15 = 1 : 3) d H011mm"

 

Rearrangement
O
- O' O ..
o-i .. .
NH Néc

vH+

 

 

Scheme 3.1. Synthesis of D-erythro-Z-oxazolidinones 16. Reagents, conditions and
yields: (a) Br2 / H20, 0 °C, 36 h, 92.4 %; (b) 28-30 % ammonium hydroxide, methanol, 0
°C to r.t., overnight, quantitative; (c) 2,2-dimethoxypropane, DMF (anhydrous), p-
toluene-sulfonic acid (cat.), r.t., overnight, l4+15: 86%, ratio: 14:15 = 1:3, trace bisacetal

13; (d) NaOCl, THF / H20, 35 °C, 12 h, 85 %; (e) AcOH (70 %), r.t., 12 h., 95 %

41

A one-pot synthesis of 5-hydroxymethy1-2-oxazolidinone using arylboronic acid
protection and in situ Hoffrnann rearrangement process was developed and patented in
2001.16 In this invention, an unfavorably positioned free hydroxyl group was dynamically
masked from participating in the Hoffinann reaction and yielded the preferred
oxazolidinone. (Scheme 3.2) An attempt was made in order to see whether the same
methodology could be used to mask one of the two secondary hydroxyl groups and gave
desired oxazolidinones. The synthesis was summarized in Scheme 3.3. The primary
hydroxyl group in 2-deoxy-D-ribonolactone 12 was protected with trityl group followed
by aminolysis to open the lactone and the resulting secondary hydroxyl groups was
protected with phenylboronic acid. However, the following Hoffrnann conditions did not
produce any detectable amount of desired oxazolidinone 20, but went back to amide 18.
This result showed that the similarity of the two secondary hydroxyl groups of compound
18 caused the masking and umnasking to take place simultaneously, thus would not allow
a dynamic differentiation in the standard Hoffinann rearrangement conditions to form

desired oxazolidinone product.

42

Scheme 3.2: Process for the preparation of 5-hydroxymethyl 2-oxazolidinone via aryl-

boronic acid protectionl6

H+
NH
CW 2 CIO. o/\=/\N‘c 0H CNN“):
: —> =
\e—-0 o 8’0 7‘0 /e\ o ‘0
Ph Ph Ph OH
0 l
> '9.
= OH
Scheme 3.3:
o o 0
a b NH2
5 O, —> E O, —>
H O ”90/ OH HO "’1’, / OTI' HO OTT
12 17 OH
18
c; td
JZ
H5

Ph/

Scheme 3.3. Attempted synthesis of oxazolidinone 20 via phenylboronic acid protection.
Reagents, conditions and yields: (a) Trityl chloride, DMF / pyridine, r.t., overnight; (b)
28-30 % ammonium hydroxide, methanol, 0 °C to r.t., 12 h, 81 % for two steps; (c)
phenylboronic acid, toluene, azeotrope, quantitative; ((1) sodium hypochlorite, sodium

hydroxide, THF / H20

43

3.4. CONCLUSION

In conclusion, a concise and efficient synthese towards the novel enantiomerially pure
oxazolidinones as pharrnacologically active precursors from naturally occuning
carbohydrates has been developed. The route developed with Hoffinann rearrangement of
deoxypentoses as key step could be used to generate oxazolidinones with novel, more
complex 5-substitution. It is expected that the additional molecular interaction offered by
such modification would enhance the protein-substrate binding. These structure
modifications could trigger a new round of SAR study at the various sites around the core
oxazolidinone periphery and possible new drug discovery. Also, the short reaction
sequence, mild reaction condition, cheap and environmentally friendly reagents
employed great potential for large-scale synthesis of 5-oxazolidinones and chiral amino

alcohol analogs as chiral synthon.

44

3.5. EXPERIMENTAL

General Techniques

All reagents used were reagent grade. Reaction temperatures were measured
externally. Column chromatography was performed on Aldrich silica gel (60 A 200-400
mesh). Yields refer to chromatographically (isolated yields) and spectroscopically ('H
NMR) homogeneous materials.

NMR spectra were recorded on a Varian 300 MHz or 500 MHz VXR
spectrometer at ambient temperature. Chemical shifts were reported relative to the
residue solvent peak. Melting points were measured on a Fischer-Johns melting point
meter and uncorrected. IR spectra were recorded on a Nicolet 710 FT-IR spectrometer.
MS spectra were performed on an AX-SOSH mass spectrometer operating in positive ion

mode.

Synthesis

2-Deoxy-D-ribonolactone (12).

13.5 g (0.1 mol) of 2-deoxy-D-ribose was stirred with 19.2 g (0.12 mol) of
bromine in 1.8 L of water at O 0C in darkness for 36 hours. The product was concentrated
in vacuo with careful heating (< 40 0C). The residue was dissolved in 50 mL of methanol
and undissolved substance was filtered. The filtrate was concentrated again to give crude

product. Small amounts of impurities were removed with short flash column to result in a

45

colorless syrup product 12. Isolated yield: (12.2 g, 92.4 %). 1H NMR (500MHz, D20): 8
4.52-4.56 (1H, m), 4.49-4.51 (1H, m), 3.82 (1H, dd, J = 12.5, 3 Hz), 3.73 (1H, dd, J =
12.5, 5 Hz), 3.01 (1H, dd, J = 18.5, 7 Hz), 2.54 (1H, dd, J = 18.5, 3.5 Hz); 13C NMR (125

MHz, D20): 5 179.67, 89.09, 68.45, 61.28, 38.00.

2-Deoxy-erythro-pentyramide (11).

10 g (75.8 mmol) of 2-deoxy-D-ribonolactone 12 was dissolved in 100 mL of
methanol and cooled with ice—bath. 19 g (152 mmol) of ammonium hydroxide (28-30 %)
was added dropwise with vigorous stirring. The reaction mixture was stirred at room
temperature for 12 hours. The liquid was concentrated under vacuum to remove most of
the solvent and the residue was rinsed with a small portion of acetone. The acetone was
decanted and the residue was dissolved in 50 mL DMF and concentrated to dryness to
give a light yellow syrup product 11 in quantitative yield. Yield: (11.2 g). 1H NMR
(300MHz, D20): 6 3.95-4.08 (1H, m), 3.72-3.80 (1H, m), 3.56-3.72 (2H, m), 2.50-2.65
(1H, m), 2.26-2.48 (1H, m); I3C NMR (75 MHz, D20): 8 177.15, 74.41, 69.02, 62.52,

38.86. FT IR cm’1 3331, 1660, 1419, 1069, 1036, 738.

2-Deoxy-4, 5-O-isopropylidene-erythro-pentyramide (15).

11 g (73.8 mmol) of amide 11 and 1 g (cat.) of p-TsOH were dissolved in 200 mL
of anhydrous DMF. 10 g (96.0 mmol) of 2, 2-dimethoxypropane was added dropwise to
the solution while stirring at room temperature. The mixture was kept stirring overnight.
5 g of sodium bicarbonate was added to quench the reaction. The mixture was

concentrated to dryness and the residue was taken up twice with 40 mL of acetone and

46

filtered. The filtrate was concentrated to give light yellow syrup, which was subject to
column chromatography with ethyl acetate to remove small amount of bis-acetal by-
product. NMR indicated a mixture of 14:15 in about 1:3 ratio (86 % total yield). Careful
recrystallization over EtOAc / hexanes gave pure colorless needles 15. Isolated yield: (8.8
g, 63.3 %). lH NMR (300MHz, CD30D): 8 4.05-4.09 (1H, m), 3.85-3.97 (3H, m), 2.48-
2.56 (1H, m), 2.24-2.32 (1H, m), 1.36 (6H, d, J = 12.0 Hz); 13C NMR (75 MHz, CD30D):

8 176.77, 1 10.59, 79.70, 70.66, 67.64, 40.56, 26.90, 25.47. Mp: 78-79 °C

5-D-erythro-2-oxazolidinones (16).

0.25 g (1.32 mmol) of 2-deoxy-4, 5-O-isopropylidene-erythro-pentyramide 15
was dissolved in 5 mL of THF. 5 mL (10-13 %) of NaOCl was added dropwise into the
solution with vigorous stirring to get a fine mixture of two phases. The stirring was kept
at 30-35 °C for about 12 h until TLC indicated that the conversion was complete. The
THF phase was partitioned and the water phase was extracted with 5 mL of THF twice.
The THF was combined and evaporated to dryness. 10 mL of chloroform was applied to
remove the inorganic salt and the organic phase was concentrated to afford the crude
product 6. Yield: (0.21 g, 85 %). The crude product could be recrystallized in ether to
give colorless needles. IH NMR (500 MHz, CDC13): 8 6.56 (1H, s), 4.34-4.42 (1H, m),
4.06-4.12 (1H, m), 4.03 (1H, dd, J = 9.0, 6 Hz), 3.81 (1H, dd, J = 9.0, 5 Hz), 3.60 (1H, t, J
= 9.0 Hz), 3.43 (1H, dd, J = 9.0, 6 Hz); 1.31 (3H, s), 1.24 (3H, s); 13c NMR (125 MHz,
CDC13): 8 159.65, 109.75, 75.92, 75.44, 66.29, 42.76, 26.43, 24.68. Mp: 106-107 °C. MS
(EI+, CHC13) m/z 57, 69, 83, 101, 129, 172; MS (FAB+, CHC13) MH+, m/z 188.19;

Theoreticals: C3H13NO4, mass: 187.0845

47

Trityl lactone (17).

In a 250 mL flask, 5.0 g (38 mmol) of 2-deoxy-D-ribonolactone 12 was dissolved
in 30 mL of pyridine and 30 mL of DMF mixed solvents and stirred at room temperature.
10.5 g (38 mmol) of trityl chloride was added. The reaction was kept stirring overnight at
r.t. until TLC indicated that most of the starting lactone was converted. Rotary
evaporation removed most of the solvents and the residue was partitioned with 50 mL of
chloroform and 25 mL of 0.5 N HCl. The organic phase was collected and washed with
25 mL water, brine respectively and dried over sodium sulfate. Then the organic phase
was concentrated. The residue was triturated with hexanes and ether to remove most of
the impurities and filtration to yield crude white product 17. The crude product was used
directly without further purification. lH NMR (300 MHz, CD3C1): 8 7.22-7.46 (15H, m,
broad), 4.40-4.50 (1H, m), 3.55 (1H, dd, J = 10.8, 4 Hz), 3.22 (1H, dd, J = 10.8, 3 Hz),
3.03-3.13 (1H, m), 2.52 (1H, dd, J = 29.5, 3 Hz), 2.14 (1H, d, J = 3 Hz); l3C NMR
(75MHz, CD3C1): 8 175.87, 143.39, 128.76, 128.33, 127.61, 86.57, 70.23, 63.45, 38.76.

Mp: 126-127 °C.

Trityl amide (18).

The crude product 17 was used directly without further purification. 30 mL of
MeOH was added and stirred to form a clear solution. The solution was cooled to 0 °C
with ice bath. 9.5 g (76 mmol, 28-30 %) of ammonium hydroxide was added dropwise to
the solution. The reaction mixture was allowed to stir at room temperature for 12 h for
the complete conversion. The solvent was removed and the residue was washed with 30

mL of ether twice. Recrystallization over EtOAc / ether result in white crystal 18. Overall

48

yield, two steps: (12.0 g, 81.0 %). 'H NMR (500 MHz, c0301): 5 7.43 (6H, d, J = 7.0
Hz), 7.20-7.34 (9H, m), 5.95 (1H, s), 5.80 (1H, s), 3.97-4.01 (1H, m), 3.73 (1H, q), 3.28-
3.33 (2H, m), 2.31-2.42 (2H, m); 13C NMR (75MHz, c1330): 8 175.34, 143.83, 128.84,

128.19, 127.46, 87.35, 72.82, 69.92, 64.83, 37.79.

Phenyl Boronic ester (19).

2.0 g (5.1 mmol) of trityl amide 18, 0.62 g (5.1 mmol) of phenyl boronic acid and
100 mL of toluene were placed in a 250 mL of flask and stirred at room temperature for 6
h to form a clear solution. The water formed was vacuum distilled azeotropically with
toluene to give phenyl boronic ester 19 in quantitative yield (2.5 g). |H NMR (500 MHz,
CD3C1): 8 7.37-7.58 (11H, m), 7.19-7.35 (9H, m), 5.62 (1H, s), 5.43 (1H, s), 4.98-5.06
(1H, m), 4.73-4.79 (1H, m), 3.57 (1H, dd, J = 17.5, 6 Hz), 3.07 (1H, dd, J = 17.5, 2 Hz);
2.79 (1H, dd, J = 25.0, 13 Hz), 2.52 (1H, dd, J = 25.0, 7.5 Hz); l3C NMR (75MHz,

CD3C1): 8 175.34, 143.83, 128.84, 128.19, 127.46, 87.35, 72.82, 69.92, 64.83, 37.79.

49

3.6. REFERENCE

1. M. J. Rybak, D. M. Cappelletty, T. Moldovan, J. R. Aeschlimann, and G. W. Kaatz, J.
Antimicrob. Chemotherapy 1998, 42, 3, 721-724

E"

S. J. Brichner, Current Pharmaceutical Design, 1996, 2, 175-194

S”

C. Z. Ding, and R. B. Silverman, J. Med. Chem. 1993, 36, 3606-3610

5‘

R. I. Hollingsworth and G. Wang, Chem. Rev. 2000, 100, 4267-4282

S"

M. C. B. V. de Souza, M. N. da Silva, V. F. Ferreira, Synlett, 1998, 12, 1339-1340

6. N. Cohen, B. L. Banner, A. J. Laurenzano, and L. Carozza, Org. Syn. Call. 7, 297

>‘

M. E. Jung, T. J. Shaw, J. Am. Chem. Soc. 1980, 102, 6304-6311

9°

L. V. Nechev, M. Zhang, D. Tsarouhtsis, P. J. Tamura, A. S. Wilkinson, C. M. Harris,
and T. M. Harris, Chem. Res. T oxicol. 2001, 14, 379-388

9. C. C. Wei, S. D. Bernardo, J. P. Tengi, J. Borgese, and M. weigele, J. Org. Chem.
1985, 50, 3462-3467

10. D. J. Madar, H. Kopecka, D. Pireh, J. Pease, M. Pliushchev, R. J. Sciotti, P. E.
Wiedeman, and S. W. Djuric, Tetrahedron Lett. 2001, 42, 3681-3684

11. M. Flasche and H. Scharf, Tetrahedron. Asym. 1995, 6, 1543-1546
12. H. J. Gais, H, Hemmerle, S. Kossek, Synthesis, 1992, 169-173

13. M. Pottie, J. V. Pottie, M. Vandewalle, H. Roper, Tetrahedron Asym. 1991, 2, 329-
330

14. H. J. Bestrnann, U. C. Philipp, Angew. Chem. Int. Ed. Engl. 1991, 30, 86-88

15. B. S. Isbell, Methods in Carbohydrate Chemistry, Academic Press Inc.: New York
and London, 1963, 2, 13-14

16. R. I. Hollingsworth and G. Wang, US patent, 2001, 6,288,238

50

CHAPTER 4 IN SI T U DEPROTECTION AND REARRANGEMENT
OF N-p-METHOXYBENZYL-B-HYDROXYAMIDES TO FORM

S-SUBSTITUTED-Z-OXAZOLIDINONES

4.1. INTRODUCTION

During the synthesis of 5-D-erythro-2-oxazolidinones, the major obstacle was the
isopropylidene protection step due to the interfering protons of the amide. Very subtle
differences in reactivity of the hydroxyl and amide groups need to be exploited in case the
bisacetal-protected product prevails. Regioisomers may predominate if a small excess of
acid catalyst or isopropylidene reagent was presented. 2-deoxy-erythro-pentyramide 1,
together with inorganic salt carried from the previous steps, was a syrup soluble in water
and other polar solvents. However a small amount of moisture could make the acetal
protection inoperative. The crude product was sticky syrup, and separation by
recrystallization was not very efficient. All these difficulties could be solved by
introduction of N-benzyl or N-4-methoxybenzy1 protection groups. The benzyl protection
could effectively suppress formation of the unwanted regioisomers and greatly simplify
the after-treatment process and increase the reaction yield because of the fine crystalline
products obtained in each step. The question remained was: how to deprotect the
protection groups for further transformation? Numerous debenzylation reagents were
screened and an in situ deprotection and rearrangement method was achieved to give

5-substituted-2-oxazolidinones in fair yield.

51

4.2. RESULT AND DISCUSSION

Using a general protection strategy described in Scheme 4.1 ' resulted in the formation of
thermodynamically more stable bisacetal 2 as major product and minor lactone acetal 3.
The desired 4,5-O-isopropylidene product was not afforded in appreciable amout.
Selective hydrolysis of bisacetal 2 gave lactam 4 in quantitative yield. (Scheme 4.1)
Besides, trace impurities or moisture would make the starting amide 1 as a sticky syrup
with limited solubility in less polar and aprotic solvent, which would not be favored in the
large-scale synthesis.

Scheme 4.1:

/\/'\/|k 2,2-dimethoxypropane (2.5 eq) 0 o? + 0
Ho NH ’ 0 °
- 2 acetone / CuSO4 / H2304 I

H
1 2 (65-75 °/o) 3 (242%)

 

TFA.
MeOH / H20

Ho H
Ho 0;“
\M' o

”4

quantitative

To block the interfering amide hydrogen from participating in acetal formation, and to
simplify the workup due to the high water solubility of starting material and the desired
product, a benzyl-protecting group was introduced. While introduction of dibenzylarnine
by aminolysis approach was not successful, the benzyl or p-methoxybenzyl protected
compounds 6b and 6c were prepared with surprising ease and excellent yield. (Scheme
4.2) Both compounds were crystalline and after-treatment was simply filtration. In the
subsequent isopropylidene protection step, no bisacetal product was formed. Isomers 7

(solid) and 8 (liquid) were obtained in about 2:1 ratio and easily separated by

52

recrystallization. Compound 8 could be recycled to 6 with 70% acetic acid and returned to

the reaction route.

Scheme 4.2:

o 0
NR R . NR R NRtfiz
O NHR1R2 1 2 2.2-drmethoxypropane l 2
_, : + OH
-. ”‘30” H+ DMF 8“
HO ’e/OH Ho OH ' 0

0H
Ba-c

HO 0
"7L *0
Ba-c

 

73": 2 : 1
R1 = R2 = benzyl Ga: 0 °/o 78, 6:: N/A
R1 = H. R2 = benzyl 6b: 95 °/o 7b: 65 %; 8b: 33%
R1 = H, R2 = p-Methoxybenzyl 6c: 85 % 70: 65 %; BC: 32%
Scheme 4.3:

 

O

o
NAPh Pd/C. HCOzNH4 NH2

H or cyclohexene
Ho 0 PdIC.l'?2((10081m7) Ho 0
°7L 7‘—

7b 9

The debenzylation was then attempted at first by hydrogenolysis with catalytic 10 % Pd/C,
but the starting material 7b was recovered unchanged. (Scheme 4.3) Among reported
N-debenzylation of amide or carbamate, the most commonly employed methods besides
hydrogenolysis are acid hydrolysis in refluxing TFA;2’3 methanesulfonic acid or
p-toluenesulfonic acid in the presence of benzyl cation scavengers such as anisole or
toluene;‘”’ formic acid;7 Birch reduction in Na or Li / NH3 condition;8‘IO oxidative
N-debenzylation using CAN,I "'2 DDQ;'3"4 and radical oxidation using Br2l5’ '6 or NBS.l7

Among these methodologies, the TsOH-toluene or MsOH-anisole conditions are

53

considered effective and mild for the debenzylation of compounds in the presence of
oxidation sensitive groups such as alcohols, and base sensitive or surprisingly, even acid
sensitive functionalities such as t-BOC, N-trityl, N-Fmoc and carbamate/1'5
p-Methoxybenzyl, 2,4-methoxylbenzyl and 2,4,6-trimethoxylbenzyl groups with
increased acid sensitivity were often incorporated for N-debenzylation of amide rather
than benzyl group.

Thus, a model reaction with simpler starting material, B-butyrolactone, was used for the
debenzylation. The N-benzyl and N-p-methoxybenzyl protected B-hydroxylamides 12
were prepared by aminolysis of B-butyrolactone 10 with the corresponding aryl amine
hydrochloric acid salt 11 in good yield.18 Debenzylation in TsOH/toluene was achieved for

the p-methoxybenzyl derivatives, but not benzyl ones. (Scheme 4.4)

Scheme 4.4

O

OH O b OH 0
a
o NHR NH2
12 13

10 11

11a: R = benzyl
116: R = p-Methoxybenzyl

Scheme 4.4: Synthesis of 3-hydroxybutyramide 13. Reagents, conditions and yields: (a)
THF, Sodium 2-ethyl-hexanoate, 12 h, r.t. 1) R = benzyl, 90 %; 2) R = p-Methoxybenzyl,

81 %. (b) TsOH, toluene, reflux, 2 h, 1) R = benzyl, unaffected; 2) R = p-Methoxybenzyl.

The N-p-methoxybenzyl protected amide 16 was then prepared from
3-hydroxy-y-butyrolactone 14. However, the O-trityl group could not survive the

debenzylation with TsOH/toluene and was readily cleaved before the p-methoxybenzyl

54

group. (Scheme 4.5)

Scheme 4.5

o
o 8 0H 0 b OH 0 c .
. HUM —"> TrO —> NOdGSll'ed
NHPMB NHPMB ”mu“
H0 14

15 16
Scheme 4.5: Attempted synthesis. Reagents, conditions and yields: (a)
4-methoxybenzylamine, CH2C12, r.t., overnight, quantitative; (b) TrCl, pyridine/DMF, r.t.

15 h, 92 %; (c) TsOH, toluene, reflux, 2 h.

It was found that free radical oxidation by bromine could be used for the debenzylation of
N-benzylacetamides in a two-phase system”, or NBS debenzylation, which was formerly
used for oxidation of O-benzyl group, '7 shared a similar radical mechanism and could be
used for removal of p-methoxybenzyl protected amide.

According to the debenzylation mechanism, Br° is initiated by incandescent or ambient
light and reacts with benzylic hydrogen to form benzylic radical, then oxidative radical
addition of another Br° forms benzyl bromide. After that, base such as saturated sodium
carbonate, calcium hydroxide, or barium carbonate is generally used to displace the Br
group with hydroxyl group, and subsequent hydrolysis of the resulting unstable
hemiacetal forms benzaldehyde derivative and the free amide. The base also functions as
hydrogen bromide scavengers. (Figure 4.1)

On the other hand, the Hoffrnann rearrangement mechanism shows that the reaction of

primary amide with equal molar quantities of Br2 and alkali yields N-bromoacetamide that

55

reacts with another equivalent of alkali to give an unstable salt. The salt readily
decomposes to give isocyanate that is converted to urethane immediately in alcoholic

solution.20 (Figure 4.2)

Figure 4.1 Proposed mechanism of debenzylation of n-benzyl protection
(ah Br
0 'Br 0
H C H ('Iir H ( OH
Br /hv Br /hv ‘OH
Riff )4, 2 [lb 2 TLJquat 2°pr
H H (
’OH

//|L OH H ( OH
R N <—> -> ArCHO + RCONH2
H” R E

OCH3

Bf OCH3 ¢ Bl’2

Ar = p-Methoxybenzyl Ar002H

Figure 4.2 Mechanism of Hoffrnann rearrangement

0
°“ i
—-> _H
R N’

O

O
sz iN/ H OH
—> R -[Br —> RN=C=O
|
Br

RUN

 

soul

0
R /u\
\N on
H

56

Scheme 4.6 Proposed mechanism of in situ deprotection and Hoffmann

rearrangement of N-p-methoxybenzyl-B-hydroxyamides to form 5-substituted-2-

oxazolidinones
o H (H‘G' P o H ,Jsr _ O .
“O “ct it;
H —-> H —>

HOJ:k OCH3 "V Hofi OCH3 "V :HO’<:L:1( QCH3
OP _ OP _

 

 

 

 

S I
O
O
I
0
l
0
12,3:

 

o 0 {-OH 1
OH
_ N
+ 4— <—
l 1 HO OCH3 H0 (796%
.. OP ..

OP
POM b
NH
OH C/lo
OH 0 Br—Br POMS; ,H POM

 

 

 

ll?

PO ‘,'H -
1 N’

2|

3‘
3.: he)
(2%)“)

 

NH
PO

i.

Based on the understanding of debenzylation and Hoffrnann rearrangement mechanism,
we found it is possible to establish an in situ reaction towards antibacterial oxazolidinone
moieties by Hoffrnann rearrangement immediately following debenzylation because of the
compatibility of the reaction conditions. (Scheme 4.6) This methodology has the
advantage that the deprotection and product formation could be accomplished in one step.
Thus, in conjunction with the benzyl protection, which greatly simplifies the after

treatment, a novel route to obtain antibacterial oxazolidinone moieties without increasing

57

a single step could be established. However, there were several problems that need to be
tackled. Firstly, the oxidative cleavage of the p-methoxybenzyl group initiated from a
radical mechanism, while the subsequent Hoffrnann rearrangement followed an ionic
mechanism. Moreover, it is well known that by raising the pH above 5, Br2 is converted to
OBr'.21 Whether enough Br° could be generated in the basic medium was not yet known.
And how many equivalents of base would be enough need to be figured out. Secondly, it
was previously reported that the reaction of N-(p-methoxybenzyl)acetamide 18 with Br2
led to four compounds. '9 Three competitive processes were involved including free radical
oxidation (19+20), electrophilic ring bromination (20+21) and ipso substitution (22).22
(Scheme 4.7) The desired radical oxidation process could be increased by appropriate
solvent selection and irradiation with incandescent light source. The percentage of
hydrogen abstraction product by Br° versus electrophilic ring bromination product was
determined by the relative reaction rates. A higher concentration of En was necessary for
the former product to prevail. In this case, appropriate solvent systems should be found to
cater to the radical oxidation. Thirdly, in principle, 20 could be formed either by
electrophilic bromination of anisaldehyde 19 or by fi'ee radical oxidation of 21. Fourthly,
in excess bromine, the free-radical chain reaction could further oxidize the resulting
anisaldehyde to the benzoic acid and a-hydroxyamide to the imide. (Scheme 4.8) If the
formation of imide prevailed before cleavage, it would greatly lower the reaction yield. It
was indicated that the different electronic configurations of the alkyl (it-type) and acyl
(O-type) radicals determine the different reactivity (For R = aryl, ka << kc; and for R =
alkyl, ka >> kc).23’ 24 However, what mole percentage of bromine would be optimum still

requires further exploration. Moreover, the starting material contains B-hydroxyl function.

58

Whether it would be labile or not in the oxidizing and basic condition also needs to be
concerned. Finally, the Hoffrnann rearrangement product in the strong base may cause
decarbonylation to yield B-aminoalcohol instead of oxazolidinones. With all those
questions in concern, the reaction was tried in different conditions and it was found that it

worked in two-solvent system.

       

 

 

Scheme 4.7
CHZNHCOCHg CH0 CHO CHQNHCOCHa Br
0 C O O O
—_’ + + +
SOIvents ,/ /’
Br ” Br "
0MB 0M6 0MB OMB OMB
18 19 20 21 22
Scheme 4.8
II __k_a-_’ . R 372 / -Bf C')H (a
Ar-CHz-NHC-R B.r /—HBr Ar-CH-NHC-R H20 / —H.Br Ar-CH-NHC-R -> ArCHO + RCONH2
Br 4 kn - Brl kc
II Br2 / H20 ?H R Ar'CEO _> ArCOzH
Ar-C-NHC-R ‘_— Ar-(E-NHC-R ' ‘

 

 

 

For reasons of solubility in organic solvent and facile detection, the starting material was
chosen as compound 22 which could be easily prepared from 3-hydroxy-y-butyrolactone
by aminolysis. The trityl group was used to selectively protect the primary hydroxyl group
because of the Hoffrnann rearrangement reaction in the excess of base.

The attempted solvent systems included CC14, MeOH, DME, and two-solvent systems
THF/HzO, DME/HZO, ether/HzO, CHzClz/I-IzO, hexanes/HZO, EtOAc/HZO, CCl4/H20.

The attempted bases were NaOH, K2CO3, CaO, and NaHCO3. The ratio of bromine and

59

base was initially chosen as l: 2 according to the in situ reaction mechanism. (Figure 4.3)
It was found that less than 15 mol% of trityl group was cleaved in Br2/NaOH (l :3) while
the trityl group was quite labile in Br2/NaOH (l: 2). Also, the weaker bases such as K2CO3,
CaO, and NaHCO3 could allow more Br' to be generated, but the resulting HBr could not
be scavenged quick enough to avoid trityl group cleavage. A two-solvent system was
better applied than homogeneous solvent for the same reason. The concentration of HBr
was very low in the organic phase because of its much higher solubility in the aqueous
phase. (Figure 4.3) On the other hand, the Br“ could be generated transiently in the
organic phase. Thus, the debenzylation happened in two stages: First, radical oxidation in
the organic phase; then, cleavage of benzyl group in the basic aqueous solution. The
subsequent Hofiinann rearrangement was carried out in the aqueous phase as well, but the
desired product would stay in the organic phase to prevent further decarbonylation in the
basic condition. Furthermore, the reaction rate was directly related to the concentration of
Br2, Br' and NaOH, and a faster reaction was shown to increase the conversion and lessen
the by-product formation. However, increasing the reaction temperature would also
accelerate the reaction rate, but higher temperature would give more byproduct. The
optimized condition was SM/Brz/NaOH (1:15: 40) in equalmolar solvents THF/H20. The
reaction was carried out under a 160 W Tungsten light at 35-40 °C with vigorous stirring
for 3 h. The result was shown in Scheme 4.9. The desired product 17 was obtained in
around 25-35 % yield. There was 25-35 % of unreacted starting material 16 retrieved.
Cleavage of the trityl group still happened and resulted in about 15 % of TrOH. The
cleaved p-methoxybenzaldehyde showed that no electrophilic bromination happened. The

result showed that in NaOH solution, Br2 was largely in the form of 810', and Br was not

60

generated in sufficient amount to completely oxidize the benzylic hydrogen. It could be
possible that the solvent THF also consumed a small amount of Br2 by radical oxidation to

form y-butyrolactone,21 But, the Hoffrnann rearrangement for the preparation of

Figure 4.3 in situ Debenzylation and Hoffrnann rearrangement

incandescent light

 

.Br Br2 <3 Organic Phase

 
  

 

 

BrO' C3 Aqueous Phase

N-Debenzylation:
SM (1eq) + Br2 (1eq) + OH‘ (1eq) —» TM + Br' (2eq) + H20
Hoffrnann rearrangement:

SM (1eq) + Br2 (1eq) + OH' (2eq) + R’OH (1eq) -> TM + Br' (Zeq) + H20

Scheme 4.9
0
OH 0 Br2/NaOH(1:2.75) A
no“ _—’ O NH + SM
THF/H o
NHPMB h 2 T,o\)_/
16 V 17

35 °/o 25-35 °/o

61

oxazolidinones was carried out best in the THF/H20,26’27 which explained that the best
yield was obtained in this solvent system. On the other hand, an experiment using
N-benzyl protection instead of N-p-methoxybenzyl group at the same condition did not
afford any appreciable N-debenzylation products, though there were reports of effective

O-benzyl group deprotection.'7’25

4.3. CONCLUSION

An in situ deprotection and rearrangement of N-p-methoxybenzyl-B-hydroxyamides to
form 5-substituted-2-oxazolidinones was established. The yield was moderate and was
influenced by multi-factors including concentration of Br2, Br' and OH', solvents, and
temperature. A practical application could involve the solid-phase synthesis and fast
establishment of the 5-substituted-2—oxazolidinone family 26 from B-hydroxycarboxylic
acid 23. (Scheme 4.10) In the in situ experiment, the major impurities were attributed to
the p-methoxybenzyl deprotection. When this moiety was immobilized in a solid state, the
3-hydroxyamide moieties could be converted by Hoffrnann rearrangement without much
difficulty. The purification process is expected to be simple filtration. The exploration of

this solid-phase synthesis approach is underway.

62

Scheme 4.10

QC < > O
H
OH O NH2 0 O )k
24

Br 2 I NaOH

xi __. .MN _» . N.
R . 0H EDC/CHZCIZ H D )_/
O t

23 25 R 26
EDC: 1-Ethyl~3-(3-dimethylaminopropyl)carbodiimide

4.4. EXPERIMENTAL

General Techniques

All reagents used were reagent grade. Reaction temperatures were measured
externally. Column chromatography was performed on Aldrich silica gel (60 A ZOO-400
mesh). Yields refer to chromatographically (isolated yields) and spectroscopically ('H
NMR) homogeneous materials.

NMR spectra were recorded on a Varian 300 MHz or 500 MHz VXR spectrometer at
ambient temperature. Chemical shifts were reported relative to the residue solvent peak.
Melting points were measured on a Fischer-Johns melting point meter and uncorrected. IR
spectra were recorded on a Nicolet 710 FT-IR spectrometer. MS spectra were performed

on an AX-SOSH mass spectrometer operating in positive ion mode.

Bisacetal (2) and lactone (3)

In a 250 mL of flask with 15 g (0.1 mol) of 2-deoxy-D-erythro-pentyramide l was

63

added 8 g (0.05 mol) of CuSO4, 100 mL of anhydrous acetone and 5 drops of sulfuric acid.
The suspension was stirred and 25 g (0.25 mol) of 2, 2-dimethoxypropane was added
dropwise at room temperature. The amide 1 was gradually dissolved in acetone (several
hours) and the stirring was continued overnight. Then 5 g of sodium bicarbonate was
added and the stirring was kept for 30 min. The solids were filtered and the filtrate was
concentrated. The residue was partitioned with 150 mL of CHC13 and 75 mL H20. The
organic layer was extracted and brine, and dried over Na2804. CHC13 was evaporated and
the residue was recrystallized over acetone / hexanes to yield light yellow needles 2
(isolated yield: 16.8 g, 73 %) and light brown solid 3 (isolated yield: 2 g, 12 %). 'H NMR
for 2 (300 MHz, CDC13) 1H NMR (300 MHz, CDC13) 8 8.06 (1H, s), 4.15-4.19 (1H, m),
3.96-4.01 (1H, m), 3.80-3.86 (2H, m), 2.44 (1H, dd, J = 11.5, 2 Hz), 2.29 (1H, dd, J = 11.5,
7.5 Hz), 1.45 (6H, d, J = 2 Hz), 1.35 (6H, d, J = 22.5 Hz); 13C NMR (75 MHz, CDC13) 8
168.79, 109.83, 84.89, 77.45, 69.40, 66.89, 33.80, 30.55, 26.91, 26.66, 25.04. FTIR cm'1
2968, 2935, 2896, 1670, 1400, 1377, 1206, 1195, 1077. Mp: 154-155 0C. 1H NMR for 3
(300MHz, CDC13) 5 4.69-4.77 (1H, m), 4.45-4.51 (1H, m), 4.41 (1H, dd, J = 13.8, 2 Hz),
4.22 (1H, dd, J = 13.8, 3 Hz), 2.87 (.lH, dd, J = 16.2, 3 Hz), 2.54 (1H, dd, J = 16.2, 4 Hz),
1.40 (6H, d, J = 33.0 Hz). I3C NMR (75 MHz, CDC13) 8 169.90, 109.67, 71.73, 71.45,
68.67, 34.95, 26.18, 24.29. FTIR cm’1 2991, 1741, 1261, 1222, 1097, 736, 489. Mp: 119

oC.
Lactam (4)

The bisacetal 2 was stirred in methanol containing trifluoroacetic acid at room temperature

for 5 h. The solvent was removed and the pure product 4 was obtained as white solid in

64

quantitative yield. 'H NMR (300 MHz, D20) 5 4.01-4.10 (1H, m), 3.56-3.68 (1H, m),
3.50-3.56 (1H, m), 3.40-3.48 (1H, m), 2.15-2.37 (2H, m); 13c NMR (75 MHz, 1320) 5

171.50, 85.31, 72.96, 67.88, 61.96, 31.19, 29.00, 25.53.

N-Benzyl-Z-deoxy-erythro-pentyramide and N-p-methoxybenzyl-Z-deoxy-erythro-
pentyramide (6b and 6c)

In a 50 mL round-bottom flask, 1.0 g (8 mmol) of 2-deoxy-D-ribonolactone 5 was
dissolved in 10 mL of methanol. At room temperature, 1.7 g (16 mmol) of benzylamine
was added dropwise. White precipitate was formed gradually. The reaction was stirred
overnight to completion. The reaction mixture was filtered and the filter cake was
triturated with acetone and filtered to dryness. A white solid 6b was obtained in pure form.
Isolated yield: (1.72 g, 95 %). lH NMR (300 MHz, d6-DMSO): 5 8.20 (1H, s), 7.16-7.30
(5H, m), 4.68 (2H, dd, J = 32.5, 5 Hz), 4.27-4.50 (1H, m), 3.75-3.85 (1H, m), 3.36-3.56
(1H, m), 3.15-3.35 (1H, m), 2.43 (1H, d, J = 15.0 Hz), 2.19 (1H, dd, J =15.0, 6 Hz); l3C
NMR (75MHz, d6-DMSO): 6 172.31, 140.29, 128.88, 127.81, 127.30, 75.42, 69.64, 63.84,
42.64, 39.32. Mp: 174-175 °C. Amide 6c followed the same procedure. Isolated yield:
(1.54 g, 85 %) 1H NMR (300 MHz, d6-DMSO): 6 8.21 (1H, t, broad), 7.18 (2H, d, J = 8.7
Hz), 6.82 (2H, d, J = 8.7 Hz), 4.71 (1H, d, J = 5 Hz), 4.60 (1H, d, J = 5 Hz), 4.38 (1H, t, J =
4 Hz), 4.18 (2H, d, J = 5 Hz), 3.66 (3H, s), 3.41-3.49 (1H, m), 3.18-3.26 (1H, m), 2.41 (1H,
dd, J = 15.5, 2H2), 2.15 (1H, dd, J = 15.5, 10 Hz). 13C NMR (75MHz, d6-DMSO): 6

172.17, 158.76, 132.19, 129.15, 114.26, 75.40, 69.63, 63.82, 55.70, 42.10, 39.07.

Dioxolane (7c)

65

1.0 g (3.7 mmol) of 6c, 4 mL of dry DMF and 0.04 g (cat.) of p-toluenesulfonic acid
was placed in a 25 mL vial and stirred at room temperature. 0.8 g (7.7 mmol) of 2,
2-dimethoxypropane was added dropwise to the solution. The stirring was continued at the
same temperature for 1 h. 1.0 g of sodium bicarbonate was added and stirred for 15 min.
Then the reaction mixture was concentrated. The residue was dissolved and partitioned
with 15 mL of CHC13 and 10 mL of H20. Water phase was extracted twice with 10 mL of
CHC13. The CHC13 was combined, washed with brine, dried and concentrated to give a
mixture of 7c and 8c in 2:1 ratio. Recrystallization from EtOAc / hexanes afforded 7c.
Isolated yield: (0.75 g, 65 %) as colorless needles. lH NMR (300 MHz, CDC13): 8 7.18
(2H, d, J = 9.0 Hz), 6.82 (2H, d, J = 9.0 Hz), 6.20 (1H, s, broad), 4.36 (1H, d, J = 5.5 Hz),
4.01-4.13 (1H, m), 3.84-4.00 (2H, m), 3.78 (3H, s), 2.55 (1H, dd, J = 17.0, 2 Hz), 2.34 (1H,
dd, J = 17.0, 11 Hz), 1.34 (6H, d, J = 18.5 Hz); l3C NMR (75MHz, CD3C1): 8 172.13,
159.20, 130.06, 129.35, 114.36, 109.65, 76.86, 70.23, 67.04, 55.54, 43.20, 38.69, 26.89,

25.37.

N-Benzy1-3-hydroxybutyramide and N-p-Methoxybenzy1-3-hydroxybutyramide (12a
and 12b)

To the mixture of 2.0 g (23.2 mmol) of B-butyrolactone (from Aldrich Co.), 6.0 g
(34.8 mmol) of benzylamine, and 9.6 g (58.0 mmol) of sodium 2-ethylhexanoate (NaEH)
was added 120 mL of THF, then the reaction mixture was stirred at room temperature for
12 h. Afier the lactone was consumed, 50 mL of saturated sodium chloride solution and
100 mL of ethyl acetate were added. The two layers were separated, and the aqueous layer

was back-extracted four times with 100 mL ethyl acetate. The organic phase was

66

combined and dried over Na2804. After concentration, the product 12a was obtained by
recrystalization over ether. Yield (2.0 g, 90 %). lH NMR (300 MHz, CD3OD): 8 7.27 (5H,
s), 4.96 (1H, s), 4.38 (2H, s), 415-423 (1H, m), 2.29-2.43 (2H, m), 1.20 (3H, d, J = 7.5
Hz); The same procedure was applied for the preparation of 12b. Yield (2.1 g, 81 %). 1H
NMR (300 MHz, CD3OD): 8 7.20 (2H, d, J = 12.0 Hz), 6.84 (2H, d, J = 12.0 Hz), 4.83 (1H,
s), 4.29 (2H, s), 4.14-4.20 (1H, m), 3.76 (3H, s), 2.26-2.42 (2H, m), 1.18 (3H, d, J = 7.5

Hz).

N -p-Methoxybenzyl-3,4-dihydroxybutyramide (15)

In a 500 mL flask, 5.0 g (4.9 mmol) of 3(S)-hydroxybutyrolactone 14 was dissolved in
150 mL of CHzClz. 7.5 g ofp-methoxybenzylamine dissolved in 50 mL of CHzClz was
added dropwise into the solution with stirring at room temperature. The precipitate was
formed gradually. The reaction was allowed to continue overnight to ensure completion.
Then dichloromethane was filtered to dryness and the white filter cake was collected to
give the pure product 15 in quantitative yield (12.3 g). 1H NMR (300 MHz, CD3OD): 6
7.21 (2H, d, J = 8.5 Hz), 6.84 (2H, d, J = 8.5 Hz), 4.30 (2H, s), 3.98-4.10 (1H, m), 3.75 (3H,
s), 3.48 (2H, d, 3 Hz), 2.28-2.46 (2H, m); l3C NMR (75MHz, CD3C1): 6 172.57, 159.10,

130.67, 128.77, 113.74, 69.36, 65.77, 54.58, 42.47, 39.99. Mp: 119-120 °C.

Trityl protected N-p-Methoxybenzy1-3,4-dihydroxybutyramide (16)
In a 250 mL flask, 12.3 g (0.05 mol) of amide 15 was dissolved in 50 mL of pyridine
and 50 mL DMF mixed solvents. The solution was stirred while 15.3 g (0.055 mol) of

trityl chloride was added portionwise to it. The reaction was continued for 15 h while TLC

67

indicated a complete conversion. Rotary evaporation removed most of the solvents and the
residue was dissolved in 100 mL of CHC13. The chloroform solution was washed with 30
mL of 0.5 N HCl, 30 mL of sodium bicarbonate solution, brine, and dried over Na2SO4.
Then the chloroform was concentrated and the residue was recrystallized over ether to
afford compound 16 in good yield (22.0 g, 92 %). 1H NMR (500 MHz, CDCl3): 8 7.43 (6H,
d, J = 6.8 Hz), 7.22-7.32 (9H, m), 7.16 (2H, d, J = 8.0 Hz), 6.84 (2H, d, J = 8.0 Hz), 6.37
(1H, t, J = 5.0 Hz), 4.32 (2H, d, J = 5.0 Hz), 4.18-4.22 (1H, m), 3.78 (3H, s), 3.59 (1H, d, J
= 3.5 Hz), 3.16-3.23 (2H, m), 2.42 (1H, dd, J = 15.0, 3.5 Hz), 2.35 (1H, dd, J = 15.0, 8.5
Hz); 13C NMR (125MHz, CD3C1): 8 171.68, 159.27, 143.98, 130.43, 129.32, 128.87,

128.13, 127.39, 114.35, 87.05, 68.31, 67.12, 55.53, 43.17, 40.01.

5-TrityloxymethyI-2-oxazolidinone (17)

NaOH (1.6 g, 40 mmol) was dissolved in 8 mL of H20 and put in an Erlenmeyer flask, in
which was added dropwise 2.4 g (15 mmol) of Br2 when the solution was chilled in ice
bath. 0.5 g (1 mmol) of trityl protected amide 16 was dissolved in 10 mL of THF and
poured all at once into the Br2/Na0H solution while it was stirred vigorously. 160 W
incandescent light was applied and the temperature was quickly raised to 35-40 °C with
heating. Afier about 1 h, the yellow color was retreated and the heating was kept at 40 0C
for 2 h more. After standing still for a couple of minutes, the organic layer was separated,
and the water layer was extracted twice with 10 mL of THF. The combined organic layer
was concentrated and taken up by dichloromethane (20 mL), and the solution was dried
over Na2S04. It was concentrated again and the residue was purified by column

chromatography. Yield: (0.12 g, 35 %). lH NMR (300 MHz, CDC13): 8 7.38-7.52 (5H, m),

68

7.39-7.18 (10H, m), 5.96 (1H, s), 4.71-4.79 (1H, m), 3.56-3.62 (1H, m), 3.39-3.45 (1H, m),
3.28 (1H, dd, J = 9.6, 4.2 Hz), 3.23 (1H, dd, J = 9.6, 4.5 Hz); 13C NMR (75 MHz, CDC13):
5 159.74, 143.45, 128.60, 127.94, 127.21, 86.90, 75.40, 64.25, 42.63. The NMR agreed

with the reported Data.26

69

4.5. REFERENCE
1. P. Rollin and J. R. Pougny, Tetrahedron, 1986, 42, 3479-3490

2. S. S. Shimshock, R. E. Waltermire, P. Deshong, J. Am. Chem. Soc. 1991, 113,
8791-8796

3. G M. Brooke, S. Mohammed, and M. C. whiting, Chem. Commun. 1997, 1511-1512
4. C. Chem, Y. Huang, and W. Kan, Tetrahedron Lett. 2003, 44, 1039-1041

5. S. Sugiyama, K. Morishita, and K. Ishii, Heterocycles, 2001, 55, 353-364

6. J. L. Wood, B. M. Stoltz, S. N. Goodman, J. Am. Chem. Soc. 1996, 118, 10656-10657
7. J. E. Semple, P. C. Lysenko, and M. M. Joulie, J. Am. Chem. Soc. 1980, 102, 7505

8. T. (1 Back, K. Brunner, P. W. Codding, and A. W. Roszak, Heterocycles, 1989, 28,
219-238

9. G Cardillo, M. Orena, S. Sandri, and C. Tomasini, Tetrahedron, 1987, 43, 2505-2512

10. F. X. Webster, J. G. Millar, and R. M. Silverstein, Tetrahedron Lett. 1986, 27,
4941-4944

11. L. E. Overman and T. Osawa, J. Am. Chem. Soc. 1985, 107, 1698-1701

12. F. L. Buenadicha, M. T. Bartolame, M. J. Aguirre, C. Avendano, M. Sollhuber,
Tetrahedron Asym. 1998, 9, 483-501

13. Y. 0ikawa, T. Tanaka, K. Horita, T. Yoshioka, and 0. Yonemitsu, Tetrahedron Lett.
1984, 25, 5393-5396

14. F. L. Buenadicha, M. T. Bartolame, M. J. Aguirre, C. Avendano, M. Sollhuber,
Tetrahedron Asym. 1998, 9, 483-501

15. A. Bravo, F. Fontana, B. Dordi, and F. Minisci, J. Org. Chem. 2000, 65, 3880-3882
16. B. Classon, P. J. Garegg, and B. Samuelsson, Acta. Chem. Scand. B, 1984, 38, 419-422

17. H. Hashimoto, M. Kawa, Y. Saito, T. Date, 8. Horito, and J. Yoshimura, Tetrahedron
Lett. 1987, 28, 3505-3508

18. W. Liu, D. D. Xu, 0. Repic, and T. J. Blacklock, Tetrahedron Lett. 2001, 42,
2439-2441

70

19. A. Bravo, F. Fontana, B. Dordi, and F. Minisci, J. Org. Chem. 2000, 65, 3880-3882
20. E. S. Wallis and J. F. Lane, Org. Reactions, 1946, 3, 267-306

21. N. C. Deno and N. H. Potter, J. Am. Chem. Soc. 1967, 89, 3550-3554

22. C. L. Perrin and G A. Skinner, J. Am. Chem. Soc. 1971, 93, 3389-3394

23. H. R. Bjorsvik, F. Fontana, L. Liguori, and F. Minisci, Chem. Commun. 2001, 523-524
24. F. Minisci, Top. Curr: Chem. 1976, 62, 1-48

25. R. W. Binkley and D. G. Hehemann, J. Org. Chem. 1990, 55, 378-380

26. G Wang and R. I. Hollingsworth, Tetrahedron Asym. 2000, 11, 4429-4432

27 . R. I. Hollingsworth and G Wang, US patent 2001, 6,288,238

71

CHAPTER 5 PREPARATION OF THIOALKYL SUBSTITUTED
SMALL RING NITROGEN HETEROCYCLES FOR

GLYCOSIDASE INHIBITORS

5.1. INTRODUCTION

Interest in glycosidase inhibitors was stimulated by the finding that plants and
microorganisms produce a variety of monosaccharide-like alkaloids such as
nojirimycin (discovered from the cultured broth of the Streptomyces species) that are
involved in a number of physiologically important processes including intestinal
digestion, catabolism and post-translational process of glycoproteins.l During the past
decade the interest has grown due to the chemotherapeutic potential of the glycosidase
inhibitors in the prevention and treatment of a variety of diseases, including lowering
blood glucose levels by affecting carbohydrate metabolismf’3 inhibition of tumor
metastasis by interruption of the glycosylated expression productf"5 inhibition of viral
replication by interfere the co- and post-translational process of glycoproteinsf"7 and
treatment of genetic disorder by quality control of the endoplasmic reticulum of the
N-linked glycoprotein.8 Mreover, some inhibitors could be useful tools in preparing

affinity ligands for the purification of specific glycosidases or glycosyltransferases.9

 

Figure 5.1.
CHZOH
NH
OH OH
0“ Nojirimycin

 

 

 

72

Most of the naturally occurring and synthetic glycosidase inhibitors are carbohydrate
derivatives or their structural analogs. According to their way of action, they can be
classified into reversible and irreversible inhibitors by the dissociation abilities of the
inhibitor from the enzyme-inhibitor complex. According to the mechanism, there are
inverting glycosidase and retaining glycosidase.IO Retaining glycosidases convert the
substrate to product with retention of anomeric configuration. This mechanism has two
separate steps and therefore, involving two oxacarbenium-ion-like transition states.
The inverting glycosidase involves only one transition state and the product is obtained
with inversion of configuration at anomeric center. The generally accepted enzymatic
glycoside hydrolysis mechanism shows that most (1- and B-glycosidases have an active
site proton donating group (usually a carboxylic acid) in the proximity of glycosidic
oxygen atom and another carboxylate group that can stabilize the partial positive
charge developing on the anomeric carbon. And then a covalent glycosyl-enzyme
intermediate with cation/oxocarbenium ion may form which is subsequently
hydrolyzed by cleavage of the glycosyl bond.”’12 (Figure 5.2) The relationship
between structure and inhibitory activity in glycosidase inhibitors based on X-ray
crystallography studies on complexes with substrate analogs and inhibitors show that
good glycosidase inhibitors should have a flattened half-chair conformation with a
positive charge character around the trigonal anomeric carbon center and ring
heteroatom, and proper hydroxyl group configurations. The structures, shapes and
charge of these inhibitors closely resemble the glycosyl cation transition state involved

in glycosidase mechanism. (Figure 5.3)

73

Figure 5.2. General enzymatic glycoside hydrolysis mechanism

(A) Inverting glycosidase

 

 

 

to o O o O O
I 5* l
O (H O 5‘ o

O-Fi \ 06:8

H H "

 

 

(B) Retaining glycosidase

 

 

 

 

 

 

 

 

ll
ll
1

 

 

 

74

Figure 5.3. Mimic of pyranosyl oxacarbenium ion trigonal planar configuration

 

 

 

 

r m - 2F - mm — 4;
A . a
a -O O . -O O
+ ,O-R , V >\ +'.Y-Fl
M ‘—V—' / ' h—v—I
(H0) 1 _ b (HO) :. b
0 O O O a: glycopyranosyl binding site
Y b: aglycon binding site
JVVVV‘ W J X.Y=O.N,orS
1 2

Figure 5.3: Analogy between (1) the transition state in the glycosidase mechanism and (2) the

possible binding mode of a glycosidase inhrbitor with the active catalytic site of a glycosidase

5.2. DESIGN AND SYNTHESIS

A large fraction of glycosidase inhibitors have general structural similarity with natural
glycosides and are designed based on the glycoside transition state of the reaction
catalyzed by glycosidase. In the past decade, several new types of synthetic inhibitors
have emerged that are very potent.'3 Development of novel structures as glycosidase
inhibitors is of considerable interest that could lead to new insight in the glycoside
cleavage process and therapeutically active leads for drug discovery.

Earlier studies have indicated various degrees of sp2 character at the anomeric carbon
with the formation of an oxacarbeniurn-ion—like transition state. A good transition-state
analogue must necessarily be a strong inhibitor. The design of the desired

transition-state analogue focused on forming the generally accepted flattened

75

half-chair conformation, the spz-hybridized anomeric carbon, and mimicking partial
charge build-up on the endo- and exocyclic heteroatoms. In 1990, the Ganem group
synthesized the first amidine (4) analogue of D-glucose, which was shown to be a
strong competitive inhibitor of B-glucosidase.l4 Later on, amidrazones and
amidoximes (Figure 5.3) have also been prepared to show broad-spectrurn inhibition
of glycosidases.'5"6 Those glycosidase inhibition families were considered as
transition-state analogues that mimic both partial positive charges on the endo- and
exocyclic oxygens, the spz-hybridized anomeric carbon and the flattened chair
conformation. Various substituents at the exocyclic nitrogen were also synthesized as
mimic of aglycon moieties to enhance the selectivity and inhibitory properties of the
compounds. 0n the other hand, several classes of thio-containing carbohydrates,
including thiodisaccharides (5),”l9 thioglycosides (8),20 thioureas (6),21 and
thioalditols (7)22 were synthesized as the transition-state analogs of glycosidase
inhibitors. Since the report of the synthesis of 5-thio-D-xylopyranose, the first example
of thiosugars ever described, there have been a large number of thiosaccharides and
their analogs of biological interest being synthesized and characterized. However, the
thioalkyl heterocycles or their pseudo-disaccharides with endocyclic nitrogen atom
and sulfur in the pseudo-interglycosidic linkage as glycosidase inhibition candidates

have never been explored so far. (Figure 5.4)

76

Figure 5.4. A comparison of novel thioalkyl heterocycles with the existing
glycosidase inhibitors

H
/+ N NRR‘ /5 0
J HO 4AM Hm
(H0)n/ ( )" (OH)n

 

 

   

4 5
Amidines: R = R': H or alkyl Disaccharide analogs
Amidrazones: R = H or alkyl
H': NH2
Amidoximes: R = H or alkyl
R' = OH H
N+
HO O > / \ SR
HO n
Thioalkyl heterocycles: R = alkyl or sugar (H0)n V 3
Target Molecules 6
/0 SR Thioureas
(H0)n/\/
8 S
Thioglycosides /VA
R = alkyl or sugar (HO)n
7
Thioalditols

The target molecule thioalkyl heterocycles (3) could be prepared from carbohydrate
precursor 2-deoxy-D-ribose as the transition-state analogs of such candidates. The
reason to introduce sulfur linkage includes: Firstly, the sulfur moieties were believed to
participate more readily than the corresponding oxygen moieties in heterolytic
processes so that the chemistry of these thiosugar analogs is more subtle and delicate
than that of the corresponding sugar analogs.”23 Secondly, the thiolactam
functionality was a weak nucleophile, which could be alkylated with a variety of
proper alkylating agents (mimicking the aglycon part), including sugar moieties such
as glucose, galactose, fructose, ribose, etc., to quickly establish a template family for
biological screening. By introduction of various structural features of the alkyl or sugar
groups that mimic the aglycon moieties, it is possible to modify the transition state to

more efficient and specific inhibition. Thirdly, the amidines and their derivatives have

77

controversial endo- or exocyclic double bond, upon which has cast doubt as the
transition-state analogs.”25 Unlike amidines, the alkylated thiolactams have
well-defined endocyclic double bond location. This is important because the location
of the double bond has a large effect on the conformation of the compound and thereby
on the interpretation of the inhibitory properties. Also, unlike amidine and their
derivatives, alkylated thiolactams are more stable in the neutral and relatively basic
condition, which allow the compounds to adopt a half-chair-like conformation without

the protonation step.

5.3. RESULTS AND DISCUSSION.

The reactions are summarized in Scheme 5.1. (see pp 85) Starting from the same
commercially available 2-deoxyribose as described in the antibacterial oxazolidinones
synthesis. (Chapter 3) The same oxidation followed by aminolysis gave
2-deoxy-D-erythro-pentyramide (9) in high yield. The crude product (9) could be used
after being lyophilized to dryness. Treatment with 2,2-dimethoxypropane, 1 eq of
cupric sulfate and catalytic amount of sulfuric acid (1 %) in acetone gave
thermodynamic bisacetal (10) as major product and trace lactonoacetal by-product (11).
The other possible acetal regioisomers were not detected. The lactam function of
bisacetal (10) was then converted to thiolactam (12) with Lawessons’ reagent. And the

resulting thiolactam (12), a mimic of glycopyranoside, was alkylated with a variety of

78

alkylating agents (mimic of aglycon) to give a variety of compounds (28-45), which
after hydrolysis, served as glycosidase inhibitor candidates. The representative alkyl
groups were aliphatic acyclic and cyclic alkanes, aromatic benzyls, and sugars.

Upon treatment with NaH in DMF at room temperature, the alkylation reaction was
successful for most of the iodoalkanes and substituted benzyl halides, except
p-nitrobenzyl and cyclohexyl because of the electronic or steric effect, and ethylamide
because of the protic amide hydrogen.

Alkylation with sugars was successful for B-methyl-Z,3-0-isopropylidene-5-tosylated
riboside in NaH/DMF at 80 oC. The corresponding iodo-riboside was unaffected in the
same condition. The tosylated riboside could be synthesized according to the reported

2 27
procedure. 6'

However, alkylation with acetobromo—a-D-glucose under the optimized
condition (NaH/DMF) gave a series of inseparable acyl hydrolysis products due to the
labile acyl protection group in basic condition. A synthesis of
pivaloylbromo-a-D-glucose (18) was achieved from D-glucose in three steps:
(Scheme 5.2, see page 88) Glycobenzylation followed by pivaloyl protection of all the
free hydroxyl groups, then a one-pot radical oxidation of the benzylic proton with NBS
and rearrangement to give desired glucosyl bromide (18).28 The replacement of the
acyl groups with pivaloyl groups could solve the protection group problem but gave
predominantly a glycal product (15), presumably following an E2 mechanism.29
(Figure 5.5) The desired product was not detected. Blank reaction with

pivaloylbromo-a-D-glucose (18) in the same condition afforded the same E2 glycal

product (15), which indicated that the elimination of the B-bromo group at the

79

anomeric carbon of glucosyl bromide happened much faster than the Sn2

displacement.”3 '

Figure 5.5. Dehydrohalogenation

OPiv f H. OPiv
“.0

O
EZ -
p
M —- "Blvd /
PivO
OPivBh 15 °in
18 Dehydrohalogenation Glycal

Alkylation with 2,3:4,5-di-0-isopropylidene-B-D-bromofructopyranose (23) (Scheme
5.2, see page 88) at room temperature gave full recovery of the starting materials.
Raising the temperature to reflux (in DMF) resulted in the decomposition of both
starting materials and no desired product was detected. This could be explained as an
overwhehning steric bulk of both isopropylidenes in the vicinity of the methylene
bromide.

The selective hydrolysis with TFA was first carried out in methanol. It was found that
the ring aminal function was labile in acidic methanol condition and complete
decomposition of the desired product was observed within 8-12 h at room temperature.
The decomposition product was detected as mainly 2-deoxy-D-ribonolactone. (Figure
5.6) Dichloromethane was replaced and the hydrolysis was done smoothly with
protonation of the aminal to stabilized the ring structure followed by selective
hydrolysis of the acetal function with TFA to give desired product in quantitative yield.

(Scheme 5.1, see page 86)

80

Figure 5.6. Selective hydrolysis

H20

 

 

 

 

 

   

0 + RSH

'o,’/OH

H0

81

5.4. BIOLOGICAL TEST

Among the eighteen compounds prepared, sixteen of them were tested against three
bacterial strains Enterococcus faecalis 29212, Staphylococcus aureus 29213,
Staphylococcus aureus 43300, respectively. The concentrations used 250 11ng down
to 0.49 ug/ml. The absorbance to monitor bacterial grth was 600 nm. All of the
compounds were taken up in 5 % DMSO, except (32), which was taken up in 10 %
DMSO. Two DMSO controls were done. (5 % and 10 %) The known antibiotic
Chloramphenicol was used as a positive control for inhibition.

In general, the compounds did not inhibit the bacterial growth except (42) (Figure 5.7),
which appears to inhibit E. faecalis 29212. The greatest inhibition was at a

concentration of 62.5 ug/ml. The percent inhibition at this concentration is 77 %.

Figure 5.7. Thioalkyl heterocycle candidate

 

l 42

 

 

 

82

5.5. CONCLUSION

In summery, a facile and effective synthesis of thioalkyl substituted small ring nitrogen
heterocycles has been developed as potential glycosidase inhibitors starting from
carbohydrate precursor 2-deoxy-D-ribose. Eighteen compounds have been made thus
far and a preliminary biological test has been done. A large number of transition-state
analogues are being evaluated for inhibition of glycosidases and many potent
inhibitors with nanomolar affinity have been found for the last forty years.
Nevertheless, the transition state catalyzed by glycosidase must be bound much more
strongly by a glycosidase than any known inhibitor. Even the best glycosidase
inhibitors are “imperfect” transition-state analogues. However, while a small number
of glycosidase inhibitors owe their activities to coincidental bindings to the enzyme,
the transition-state mimicking is still the major guidance for discovery of the

glycosidase inhibitors.

83

5.6. EXPERIMENTAL

General Techniques

All reagents used were reagent grade. Reaction temperatures were measured
externally. Column Chromatography was performed on Aldrich silica gel (60 A
200-400 mesh). Yields refer to chromatographically (isolated yields) and
spectroscopically ('H NMR) homogeneous materials.

NMR spectra were recorded on a Varian 300 MHz or 500 MHz VXR spectrometer
at ambient temperature. Chemical shifts were reported relative to the residue solvent
peak. Melting points were measured on a Fischer-Johns melting point meter. IR spectra

were recorded on a Nicolet 710 FT-IR spectrometer operating in positive ion mode.

84

Figure 5.8. Thioalkyl substituted heterocycles for glycosidase inhibitor

3+ H»- St
H0 0 \ S\Me HO O \ S\E1 HO 0 SW
H0 ' H0 ' HO I
28 29 30
BUM

H = Methyl H = Ethyl H =
NW: 150 mg NW: 120 mg NW: 170 mg
Yield: 80 % Yield: 72 9’6 Yield: 92 %
H H H
+
H0 N\ S\/\/\/ HO N J 5 Ho N * s
o 0 MT, 0
HO ' HO HO
l I
31 32 33
R =Hexyl H = Hexadecyl H = Cyclohexyl
NW: 170 mg NW: 260 mg NW: trace
Yield: 84 % Yield: 90% Yield: < 10 96
CI
5 + 5 co H H + H +
H0 Cy \/ 2 HO N 8 Ho N 5
HO O O
1 HO HO .
H E 234 b . 35 as
= ‘ 'chlale H = Benzyl n = p-Chlorobenzyl
NW3 140 "‘9 NW: 160 mg NW: 190 mg
“91¢ 69 % Yield: 98 as Yield: 84%
N02
N02 H3CO
H H
+ '0'
HO o> /N\ 8 HO 05 )1: S “0 Dy" 8
”OM ”OM “OM
l l
37 38 39
R = p-Nitrobenzyl R = m-Nitmbenzyl R = o-Methoxybenzyl
NW: trace NW: 180 mg NW: 170 mg
Yield: < 15 % Yield: 77 86 Yield: 78%
OCH3
OCHa NC
H
i ,N "’ N + +
HO ' H0 ' HO
4o 41 ' 42
R = m-Methoxybenzyl R = eMethoxybenzyl R = o-Cyanobenzyl
NW: 223 mg NW: 170mg NW: 200 mg
Yield: 99% Yield: 76% Yield: 90%

 

ON
ON
a + H0 fl+ S\/O NW: 170 mg
HO 3 0:1/ ‘ {1+ Yield: 61%
O HO HO S
HO . 0 o
I 44 H0 OMa
‘3 l

 

 

 

R = C anobenzyl OH OH
R = m-Cyanobenzyi NW: 2'9; mg 48
NW5 21° "'9 Yield: 84%
Yleld: 95%

85

Scheme 5.1.

OH 0H 10 11 trace

Scheme 5.1 Synthesis of thioalkyl substituted heterocycles as glycosidase inhibitor
candidates. Reagents, conditions and yields: (a) Br2 / H20, 0 °C, 36 h, 92.4 %; (b)
28-30 % ammonium hydroxide, methanol, O 0C to r.t., overnight, quantitative; (c)
2,2-dimethoxypropane (2.5 ~ 3 eq), CuSOa, H2S04 (cat.), acetone, r.t., overnight, 73 %;
(d) Lawessons’ reagent, toluene/pyridine, 65-70 °C, 1.5 h, 80 %; (e) NaH, DMF, RX, 0

°C ~ r.t. (80 °C for compound 45), 2-12 h; (f) TFA, CH2C12, r.t., 5 min.

86

Synthesis

Bisacetal 10

In a 250 ml of flask with 15 g (0.1 mol) of 2-deoxy-erythro-pentyramide (9) was
added 8 g (0.05 mol) of CuSOa, 100 m1 of anhydrous acetone and 0.5 ml of sulfuric
acid. The suspension was stirred and 25 g (0.25 mol) of 2, 2-dimethoxypropane was
added dropwise into it at room temperature. The amide (9) was gradually dissolved in
acetone (several hours) and the stirring was continued overnight. Then 5 g of sodium
bicarbonate was added and the stirring was continued for 30 min. The solids were
filtered and the filtrate was concentrated. The residue was partitioned with 150 ml of
CHCl3 and 75 ml H20. The organic layer was extracted with brine, and dried over
Na2S04. CHC13 was evaporated and the residue was recrystallized over acetone /
hexanes to yield light yellow needles (10) (isolated yield: 16.8 g, 73 %) 'H NMR (300
MHz, CDC13) 8 4.17 (1H, m), 3.98 (1H, m), 3.83 (2H, m), 2.44 (1H, dd, J = 11.5, 2 Hz),
2.29 (1H, dd, J = 11.5, 7.5 Hz), 1.45 (6H, d, J = 2 Hz), 1.35 (6H, d, J = 22.5 Hz); 13C
NMR (75 MHz, CDCl3) 8 168.79, 109.83, 84.89, 77.45, 69.40, 66.89, 33.80, 30.55,
26.91, 26.66, 25.04. FT IR cm'l 2968, 2935, 2896, 1670, 1400, 1377, 1206, 1195, 1077.

Mp: 154-155 °C.

Thionolactam 12
11.4 g (0.05 mol) of bisacetal 10 was dissolved and stirred in 80 ml of anhydrous

toluene and 5 ml (1 eq) of pyridine mixed solvent. 10 g (0.25 mol) of Lawessons’

87

regent was added all at once into the solution. The mixture was quickly heated to 65 °C
while stirring and all the suspension was dissolved. The homogeneous mixture was
kept heating at 65-70 °C for 1.5 h until TLC indicated reaction complete. Solvent was
distilled azeotropically and the residue was purified by column chromatography
(EtOAc: Hexanes = 1:2) to afford 12 as light yellow needles. (Isolated yield: 10 g, 80
%) 1H NMR (300 MHz, CDC13) 8 9.64 (1H, br), 4.06 (1H, m), 3.98 (1H, m), 3.81 (1H,
m), 3.79 (1H, m), 3.09 (1H, dd, J = 18.3, 3 Hz), 2.67 (1H, dd, J = 18.3, 10.8 Hz), 1.48
(6H, d, J = 2 Hz), 1.36 (6H, d, J = 21.3 Hz); 13C NMR (75 MHz, CDC13) 8 197.22,

109.93, 87.10, 77.30, 68.78, 66.63, 40.49, 29.26, 26.65, 26.01, 24.93. Mp: 121-122 °C

88

Scheme 5.2.

 

OH
0 0H OPiv OPiv
Ho _a_, 0 030—.p 0 OBn __.p O
OH HO OH P'VO OPiv PM) PM) 8,
a-D-Glucose 16 17 18
OH
—-> O 03" —> O OBn -—'-> O
HO OH .
OH HO OH PWO OPiv P'VO Pivo B,
a-D-Galactose 19 20 21
—OH
:0 (I) 9
HO-- e 8‘ f 8‘
l-— —> —>
OH 0 CHZOH o CH,Br
—-OH 22 23
D-Fructose
HO HO HO Br
WOH 9 Wow, h Worm i Wow;
—> ——-> —>
OH OH OH OH ><° x
D-Hibose 24 25 26

T50
l_i- 19°79»
O O
)Q 27

Scheme 5.2 Synthesis of Sugar halides (RX). Reagents, conditions and yields: (a)
PhCH20H, Acetyl chloride, r.t., overnight, 75 %; (b) t-Bu(C=O)Cl (5 eq), pyridine, r.t.
3 d, 69 %; (c) NBS, CCla, BaC03, hv, reflux, 8 h; (d) HMPA; 75-80 °C, 24 h, 74 % in
two steps; (e) acetone, sulfuric acid (2 5 %), r.t., 1.5 h; (f) PPh3 (2 eq), CBra (3 eq),
pyridine, reflux, 36 h, 84 %; (g) Methanol, H2804, 0 °C, 15 h, 97 %, (a : 8: l : 3); (h)
2,2-Dimethoxypropane, acetone, r.t., overnight, quantitative; (i) PPh3, CBr4, pyridine,
60-65 °C, 5 h, 70 %; (j) TsCl, pyridine, CH2C12, r.t. 15 h, quantitative.

89

 

Benzyl-glucopyranoside (16)

To a suspension of glucose (10.0 g, 0.056 mol) in 80 ml of benzyl alcohol was added 4
ml of acetyl chloride. The mixture was stirred vigorously and a homogeneous solution
was formed. The solution was stirred overnight until TLC indicated the complete
conversion. The reaction mixture was diluted with 400 m1 of hexane/ether (l: 2) mixed
solvent and placed in fridge for 3 h. The top layer was decanted and the lower layer
was washed with another 200 ml of hexane to give a sticky liquid. Column
chromatograph with dichloromethane/methanol (6:1) afforded 11.3 g of white crystals.
(16) (a, [3 mixture, Yield: 75 %). 1H NMR (300 MHz, CDC13) 8 7.18-7.31 (5H, m),
5.54-5.62 (1H, m), 5.30 (1H, m), 5.07 (1H, s), 4.86 (1H, s), 4.58 (2H, m), 4.24 (2H, s),
3.75 (2H, s), 3.52 (3H, m), 3.26 (2H, 3); '3C NMR (75 MHz, CDC13) 8 137.54, 128.68,

128.57, 128.29, 102.20, 98.11, 74.00, 73.89, 72.10, 69.71, 61.05, 50.28.

Benzyl-2,3:4,5-pivaloy1-g1ucopyranoside (17)

11.0 g (42.0 mmol) of benzyl-glucopyranoside 16 was dissolved in 100 ml of pyridine
and stirred at room temperature. To the solution was added 25.5 g (210 mmol) of
pivaloyl chloride and the reaction mixture was kept stirring at room temperature for 3
days until TLC indicated that most of the starting material was converted. Solvent was
removed together with excess pivaloyl chloride by vacuum distillation. The residue
was partitioned by 200 ml of chloroform and 100 ml of sodium bicarbonate solution.
The organic layer was then washed twice with 50 ml of water, brine and evaporated.

The residue was crystallized in EtOAc / hexanes and used without further purification.

9O

(Yield: 17.5 g, 69 %). lH NMR (300 MHz, CDC13) 8 7.28 (5H, m), 5.58 (1H, t), 5.09

(2H, m), 4.83 (1H, m), 4.70 (1H, d), 4.57 (1H, t), 4.48 (1H, d), 4.10 (4H, m).

Pivaloylbromo-a-D-glucose (18)

The starting pivaloyl protected glucopyranoside 17 (14.5 g, 24 mmol) was dissolved in
300 ml of CC14. In the solution was added 8.5 g (48 mmol) of NBS and 10.0 g (50
mmol) of barium carbonate. The suspension was heated to reflux and irradiated with a
375-W Tungsten lamp while stirring vigorously for 8 h. Then 3.2 g (24 mmol) of
HMPA was added and the reaction mixture was kept at the same condition for 24 h
more. The suspension was filtered. The filtrate was concentrated in vacuo and the
residue was chromatographed to give 10.2 g (Yield: 74 %) of white crystals 18. IH
NMR (300 MHz, CDC13) 8 6.60 (1H, d, J = 4.2 Hz), 5.61 (1H, t, J = 9.6 Hz), 5.20 (1H,
t, J = 9.6 Hz), 4.79 (1H, dd, J = 10.2, 4.2 Hz), 4.29 (1H, d, J = 10.2 Hz), 4.15 (2H, s),
1.07-1.28 (36H, m); ”C NMR (75 MHz, CDC13) 8 177.91, 177.29, 176.75, 176.56,

86.85, 72.50, 70.83, 69.52, 66.48, 60.84, 38.72, 38.38.63, 27.12, 27.05, 27.01, 26.97.

2,3:4,5-di-O-isopropylidene-B-D-fructopyranose (22)
The preparation followed the procedure described by Brady.32 In this approach,

acetone and over 5 % of sulfuric acid was added to obtain the thermodynamically

favored bisacetal fructopyranose product 22.

2,3:4,5-di-0-isopropylidene-B-D-bromofructopyranose (23)
Bisacetal fructopyranose 22 (8.0 g, 32 mmol), triphenylphosphine (16.8 g, 64 mmol),

91

and carbon tetrabromide (31.2 g, 96 mmol) were placed in a 250 ml flask equipped
with a magnetic stirrer and ice bath. Anhydrous pyridine (120 ml) was poured slowly
into the flask with stirring while the mixture was cooled over ice bath. A reflux
condenser was then added and the homogeneous resultant was then heated over oil
bath to mild reflux. The reaction mixture turned into a dark brown solution and the
heating was continued for 36 h. TLC indicated the disappearance of starting material.
Methanol (300 ml) was added to quench the reaction. The reaction mixture was stirred
at room temperature for 30 min more. Then the solvent was removed by vacuum
distillation. The residue was taken up by 100 ml of dichloromethane and the solid
substance was filtered. The filtrate was concentrated and column chromatographed
with Ethyl acetate/hexanes (l: 5) to afford a golden liquid 23. (Yield: 8.3 g, 84 %). 1H
NMR (300 MHz, CDC13) 8 4.52 (1H, dd, J = 8.4, 3 Hz), 4.26 (1H, d, J = 3 Hz), 4.14
(1H, d, J = 8.4 Hz), 3.84 (1H, dd, J = 13.2, 1.5 Hz), 3.69 (1H, d, J = 13.2 Hz), 3.54 (1H,
d, J = 11.4 Hz), 3.40 (1H, d, J = 11.4 Hz), 1.45 (3H, s), 1.37 (6H, d, J = 6.6 Hz), 1.25
(3H, 3); '3C NMR (75 MHz, CDC13) 8 109.23, 109.06, 101.17, 71.63, 70.83, 70.36,

62.11, 35.82, 26.84, 26.05, 25.73, 24.21.

Methyl D-ribofuranoside (24)

1 ml of sulfuric acid was added to a solution of 5.0 g (33 mmol) of D-ribose in 100 ml
of methanol. The solution was stirred in the cold room (5 °C) for 15 h and then
quenched by stirring with Amberlyst resin. After filtration, the solution was

concentrated in vacuo to yield a thick golden syrup 24.(Yield: 5.1 g, 95 %, 01:8 = 1 : 3)

92

Methyl-2, 3-O-isopropylidene-D-ribofuranoside (25)

In a solution of 5.4 g (33 mmol) of methyl ribofuranoside 24 in 100 ml of anhydrous
acetone was added 8.0 g (76 mmol) of 2, 2-dimethoxypropane and 0.5 ml of sulfuric
acid. The solution was stirred overnight and 5 g of sodium bicarbonate was added to
quench the reaction. The stirring was kept for 30 min more and the suspension was
filtered. The filtrate was concentrated in vacuo to form a light yellow liquid near
quantitative yield. The a- and B- isomers could be further separated by column
chromatography to give B-isomer in 4.5 g isolated yield. (67 %), and a-isomer 1.6 g
(24 %). lH NMR for a—isomer (300 MHz, CDC13) 8 4.80 (1H, d, J = 2.4 Hz), 4.55 (2H,
d, J = 2.4 Hz), 4.05 (1H, t, J = 3.0 Hz), 3.68 (1H, dd, J = 12.0, 3 Hz), 3.58 (1H, dd, J =
12.0, 4 Hz), 3.56 (3H, s), 2.85 (1H, s, br), 1.45 (3H, s), 1.25 (3H, 8); '3C NMR for
a-isomer (75 MHz, CDC13) 8 115.16, 102.59, 81.39, 80.58, 80.14, 62.35, 55.52, 25.63,
25.46; 1H NMR for B-isomer (300 MHz, CDCl3) 8 4.84 (1H, s), 4.68 (1H, d, J = 6 Hz),
4.46 (1H, d, J = 6 Hz), 4.26 (1H, t, J = 3.6 Hz), 3.53 (1H, dd, J = 8.7, 3.6 Hz), 3.49 (1H,
dd, J = 8.7, 3.9 Hz), 3.29 (3H, s), 3.23 (1H, q, J = 4.8 Hz), 1.35 (3H, s), 1.18 (3H, s);
13C NMR for B-isomer (75 MHz, CDC13) 8 110.83, 109.60, 87.95, 85.46, 81.27, 63.63,

55.08, 26.10, 24.46.

Methyl-2, 3-0—isopropylidene-S-tosyl-B-D-ribofuranoside (27)
2.0 g (10 mmol) of Methyl-2,3-0-isopropylidenc—B-D-ribofuranoside and 2.2 g (12
mmol) of tosyl chloride were dissolved in 20 ml of CH2C12 at room temperature while

stirring. 6 m1 of pyridine was added and the reaction mixture was kept stirring for 15 h.

93

The solvent CH2Cl2 was evaporated in vacuo and the residue was poured in ice-water
mixture. Suction filtration followed by rinsing the filter cake twice with cold water

gave white solid 27 in quantitative yield. (3.6 g)

2,3:4,6-O-pivaloylglucose glycal (15)

Sodium hydride (0.09 g, 2.2 mmol) was suspended in 4 ml of anhydrous DMF. The
suspension was cooled to 0 °C and the pivaloyl-bromoglucose 18 (0.5 g, 0.87 mmol)
dissolved in 1 ml DMF was added dropwise to the suspension while stirring vigorously.
Temperature was raised slowly to r.t. in 1 h. The suspension was filtered and the filtrate
was concentrated in vacuo. The residue was purified by flash column directly and 0.3 g
of colorless needles 15 were obtained. (Y ield: 69 %). lH NMR (300 MHz, CDC13) 8
6.50 (1H, s), 5.50 (1H, d, J = 2.5 Hz), 5.26 (1H, t, J = 2.5 Hz), 4.35 (1H, dd, J = 6.0, 3
Hz), 4.27 (1H, m), 4.21 (1H, dd, J = 6.0, 1 Hz), 1.08-1.28 (36H, m); 13C NMR (75
MHz, CDC13) 8 177.82, 177.04, 176.89, 176.45, 138.96, 127.84, 74.14, 67.35, 66.33,

60.80, 38.73, 38.70, 38.67, 27.00, 26.98, 26.97, 26.85. Mp: 86 °C.

Selective hydrolyzed lactam bisacetal (46)

The bisacetal 10 was stirred in methanol containing trifluoroacetic acid at room
temperature for 5 h. The solvent was removed and the pure product was obtained as
white solid 46 in quantitative yield. 'H NMR (300 MHz, D20) 8 4.05 (1H, m), 3.62
(1H, m), 3.54 (1H, m), 3.44 (1H, m). 2.23 (2H, m); l3C NMR (75 MHz, D20) 8 171.50,

85.31, 72.96, 67.88, 61.96, 31.19, 29.00, 25.53.

94

Selective hydrolyzed thiolactam bisacetal (47)

The thiolactam 12 was stirred in methanol containing trifluoroacetic acid at room
temerature for 5 h. The solvent was removed and the product was crystallized from
methanol as colorless crystals 47. The yield was near quantitative. lH NMR (500 MHz,
CD30D) 8 4.07 (1H, m), 3.74 (1H, dd, J = 10.0, 3 Hz), 3.67 (2H, m), 3.01 (1H, dd, J =
18.0, 3 Hz), 2.79 (1H, dd, J = 18.0, 11 Hz), 1.56 (6H, d, J = 2.0 Hz); l3C NMR (125

MHz, CD3OD) 8 199.06, 87.95, 74.67, 68.83, 63.63, 40.89, 29.32, 25.67.

General Procedure for the Preparation of Thioalkyl Substituted Heterocycles
(28-44)

A suspension of 80 mg (60 %, 2 mmol) NaH in 8 ml of anhydrous DMF was cooled to
0 °C with sodium chloride ice-bath. 0.25 g (1 mmol) of thiolactam 12 dissolved in 5 m1
of DMF was added dropwise into the suspension while stirring. The stirring was kept
for 1 h at room temperature and cooled back to 0 °C. Alkyl halide (1.5-2.0 mmol)
dissolved in 3 ml DMF was added dropwise to the reaction mixture and the stirring
was kept at 0 °C to room temperature for 2-12 h as indicated by TLC. The desired
bisacetal product could be purified by column chromatography. (Ethyl acetate:
Hexanes = 1: 20). Treatment with TFA in CH2C12 gave first the protonated and then
selective hydrolyzed products in quantitative yield. Due to the labile character of the
hydrolyzed products (they were readily decomposed to 2-deoxy-D-ribonolactone in

methanol/TFA within 8-12 h), the products were stored in their bisacetal form.

95

Thioalkyl substituted heterocycles (45)

The treatment of NaH and thiolactam 12 followed the general procedure above.
methyl-2,3-0-isopropylidene-5-tosyl-B-D-ribofuranoside 27 (0.43 g, 1.2 mmol) was
dissolved in 3 ml DMF was added to the reaction mixture. The reaction was stirred at
80 0C for 36 h. Then the suspension was filtered and the filtrate was concentrated in
vacuo. The residue was place onto column chromatography directly (Ethyl acetate:
hexanes = 1:5) to give colorless needles 45. (Yield: 0.17 g, 61 %) 1H NMR (500 MHz,
CDC13) 8 4.90 (1H, s), 4.61 (1H, d, J = 6.0 Hz), 4.54 (1H, d, J = 6.0 Hz), 4.22 (1H, dd, J
= 9.0, 7 Hz), 4.03 (1H, dd, J = 8.0, 6 Hz), 3.85-3.90 (1H, m), 3.81 (1H, dd, J = 9.0, 5
Hz), 3.56-3.62 (1H, m), 3.30 (3H, s), 3.09 (1H, dd, J = 13.0, 9 Hz), 3.02 (1H, dd, J =
13.0, 6 Hz), 2.26 (1H, dd, J = 17.0, 4 Hz), 2.20 (1H, dd, J = 17.0, 10 Hz), 1.40 (3H, s),
1.35 (6H, s), 1.32 (3H, s), 1.28 (3H, s), 1.25 (3H, s); l3C NMR (75 MHz, CDC13) 8
158.36, 112.16, 109.59, 109.49, 89.72, 85.59, 85.30, 83.14, 77.70, 67.81, 67.01, 54.77,

33.28, 31.14, 29.96, 26.65, 26.30, 25.66, 24.94, 24.89. Mp: 76.5-77.5 0C.

Thioalkyl substituted heterocycles (28-44)

R = Methyl (28) (Yield: 150 mg, 80 %)

lH NMR (300 MHz, CDC13) 8 4.01-4.06 (1H, m), 3.87-3.91 (1H, m), 3.81-3.85 (1H,
m), 3.56-3.64 (1H, m), 2.14-2.31 (2H, m), 2.22 (3H, s), 1.35 (6H, d, J = 1 Hz), 1.30
(6H, d, J = 12.9 Hz); l3C NMR (75 MHz, CDC13) 8 159.75, 109.43, 89.67, 77.77, 67.82,

67.04, 33.25, 30.12, 26.66, 25.84, 24.96, 11.61.

96

R = Ethyl (29) (Yield: 120 mg, 72 %)

‘H NMR (300 MHz, CDC13) 5 4.06 (1H, dd, J = 9.0, 8 Hz), 3.82-3.94 (2H, m),
3.62-3.66 (1H, m), 2.83-2.91 (2H, m), 2.21-2.27 (2H, m), 1.38 (6H, d, J = 2.1 Hz), 1.36
(3H, s), 1.31 (3H, s), 1.21 (3H, t, J =-- 3.0 Hz); ”C NMR (75 MHz, CDC13) 5 159.04,

109.48, 89.58, 77.88, 67.94, 67.13, 33.42, 30.10, 26.74, 25.81, 25.06, 22.81, 14.29.

R = n-Butyl (30) (Yield: 170 mg, 92 %)

lH NMR (500 MHz, CDC13) 8 4.05 (1H, dd, J = 8.5, 6H2), 3.89 (1H, m), 3.83 (1H, dd,
J = 8.5, 5 Hz), 3.62 (1H, m), 2.92 (1H, q), 2.85 (1H, q), 2.26 (1H, dd, J = 17.0, 3.5 Hz),
2.19 (1H, dd, J = 17.0, 10.5 Hz), 1.54 (2H, q), 1.30-1.37 (14H, br, m), 0.88 (3H, t, J =
7.5 Hz); l3C NMR (125 MHz, CDC13) 8 159.22, 109.45, 89.55, 78.04, 68.11, 67.14,

33.60, 31.28, 30.12, 27.97, 26.72, 25.82, 25.10, 21.91, 13.53.

R = n-Hexyl (31) (Yield: 170 mg, 84 %)

'H NMR (500 MHz, CDC13) 5 4.02 (1H, dd, J = 8.5, 8 Hz), 3.84-3.88 (1H, m), 3.80
(1H, dd, J = 8.5, 5 Hz), 3.60 (1H, m), 2.88 (1H, q), 2.80 (1H, q), 2.24 (1H, dd, J = 17.0,
3.5 Hz), 2.16 (1H, dd, J = 17.0, 10 Hz), 1.54 (2H, q), 1.22-1.38 (18H, br, m), 0.84 (3H,
t, J = 7.0 Hz); '31: NMR (125 MHz, CDC13) 5 159.17, 109.34, 89.45, 77.91, 68.01,

67.09, 33.47, 31.25, 30.03, 29.05, 28.42, 28.23, 26.62, 25.72, 25.01, 22.34, 13.82.

R = n-Hexadodecyl (32) (Yield: 260 mg, 90 %)

1H NMR (500 MHz, CDC];) 8 4.05 (1H, dd, J = 8.5, 6 Hz), 3.89 (1H, m), 3.84 (1H, dd,

97

J = 8.5, 5.5 Hz), 3.60-3.66 (1H, m), 2.91 (1H, q), 2.85 (1H, q), 2.27 (1H, dd, J = 17.0, 4
Hz), 2.19 (1H, dd, J = 17.0, 10 Hz), 1.56 (2H, q), 1.24-1.73 (38H, br, m), 0.86 (3H, t, J
= 6.5 Hz); l3C NMR (125 MHz, CDC13) 5 159.24, 109.42, 89.52, 77.98, 68.07, 67.16,
33.54, 31.86, 30.10, 29.61, 29.59, 29.55, 29.43, 29.27, 29.15, 29.14, 28.86, 28.32,

26.68, 25.79, 25.06, 22.60, 13.98

R = Cyclohexyl (33) (Yield: 16 mg, 8 %)

lH NMR (500 MHz, CDC13) 8 4.05 (1H, dd, J = 8.5, 6 Hz), 3.85-3.91 (1H, m), 3.84
(1H, dd, J = 8.5, 5 Hz), 3.60-3.66 (2H, m), 2.22 (1H, dd, J = 13.5, 4 Hz), 2.17 (1H, dd, J
= 13.5, 10 Hz), 1.86-1.98 (4H, br, m), 1.62-1.74 (4H, br, m), 1.54 (2H, m), 1.37 (3H, s),
1.36 (3H, s), 1.34 (3H, s), 1.31 (3H, s); l3C NMR (125 MHz, CDC13) 8 158.86, 109.53,
89.53, 77.94, 67.99, 67.13, 48.69, 41.10, 33.51, 32.76, 30.10, 26.73, 25.93, 25.85,

25.78, 25.08.

R = Ethyl bromoacetate (34) (Yield: 140 mg, 69 %)

1H NMR (300 MHz, CDC13) 8 4.09 (1H, d, J = 9.0 Hz), 4.05 (1H, d, J = 9.0 Hz), 3.99
(1H, dd, J = 7.5, 6 Hz), 3.82-3.86 (1H, m), 3.77 (1H, dd, J = 7.5, 4 Hz), 3.57-3.59 (2H,
m), 3.55-3.57 (1H, m), 2.23 (1H, dd, J = 18.0, 15 Hz), 2.18 (1H, dd, J = 18.0, 6 Hz),
1.31 (3H, s), 1.26 (9H, d, J = 6.0 Hz), 1.18 (3H, t, J = 6.0 Hz), l3C NMR (75 MHz,
CDC13) 8 169.19, 157.70, 109.38, 89.55, 77.61, 67.72, 66.92, 61.14, 32.77, 31.07,

29.77, 26.56, 25.54, 24.86, 14.02.

98

R = Benzyl (35) (Yield: 195 mg, 95 %)

‘H NMR (500 MHz, CDC13) 5 722734 (5H, m), 4.18 (2H, dd, J = 30.0, 13.5 Hz),
4084 12 (1H, m), 3.93-3.96 (1H, m), 3.87-3.91 (1H, m), 3.66-3.70 (1H, m), 2.33 (1H,
dd, J = 16.5, 4 Hz), 2.26 (1H, dd, J = 16.5, 10 Hz), 1.43 (6H, J = 18.0 Hz), 1.39 (6H, d,
J = 37.5 Hz); ”C NMR (125 MHz, CDC13)8 159.11, 138.87, 129.29, 128.56, 127.10,

109.80, 90.02, 78.15, 68.27, 67.40, 33.51, 32.74, 30.40, 26.99, 26.15, 25.33.

R = p-Chlorobenzyl (36) (Yield: 190 mg, 84 %)

1H NMR (500 MHz, CDC13) 8 7.23 (4H, q), 4.10 (2H, dd, J = 26.5, 16 Hz), 4.04-4.08
(1H, m), 3.90-3.94 (1H, m), 3.84-3.88 (1H, m), 3.63-3.71 (1H, m), 2.30 (1H, dd, J =
17.0, 3.5 Hz), 2.23 (1H, dd, J = 17.0, 10 Hz), 1.40 (9H, d, J = 11.5 Hz), 1.36 (3H, s);
l3C NMR (125 MHz, CDC13) 8 158.55, 137.26, 132.53, 130.29, 128.28, 109.44, 89.66,

77.82, 67.97, 67.08, 33.21, 31.67, 30.05, 26.65, 25.79, 25.01.

R = m-Nitrobenzyl (38) (Yield: 180 mg, 77 %)

lH NMR (500 MHz, CDC13) 8 8.20 (1H, d, J = 2 Hz), 8.01 (1H, d, J = 8.0 Hz), 7.59
(1H, d, J = 8.0 Hz), 7.38 (1H, t, J = 8.0 Hz), 4.18 (1H, d, J = 14.5 Hz), 4.12 (1H, d, J =
14.5 Hz), 4.03 (1H, dd, J = 8.5, 6.5 Hz), 3.86-3.90 (1H, m), 3.81 (1H, dd, J = 8.5, 5 Hz),
3.60-3.66 (1H, m), 2.26 (1H, dd, J = 17.0, 4 Hz), 2.20 (1H, dd, J = 17.0, 10 Hz),
1.29-1.39 (9H, m), 1.28 (3H, s); ”C NMR (125 MHz, CDC13) 8 158.05, 147.96,
141.33, 135.02, 128.90, 124.16, 121.70, 109.44, 89.74, 77.71, 67.80, 66.99, 33.03,

31.34, 29.97, 26.59, 25.71, 24.90.

99

R = o-Methoxybenzyl (39) (Yield: 170 mg, 76 %)

IH NMR (500 MHz, CDC13) 8 7.38 (1H, d, J = 7.5 Hz), 7.21 (1H, t, J = 8 Hz), 6.87 (2H,
q), 4.24 (1H, d, J = 13.0 Hz), 4.19 (1H, d, J =13.0 Hz), 4.09 (1H, dd, J = 8.0, 2 Hz),

3.91-3.94 (1H, m), 3.88 (1H, dd, J = 8.0, 5 Hz), 3.85 (3H, s), 3.66-3.70 (1H, m), 2.31

(1H, dd, J = 17.0, 4 Hz), 2.24 (1H, dd, J = 17.0, 10 Hz), 1.48 (3H, s), 1.44 (3H, s), 1.41

(3H, s), 1.34 (3H, s); 13C NMR (125 MHz, CDC13) 8 159.06, 157.30, 130.62, 128.13,

126.71, 120.08, 110.24, 109.34, 89.57, 77.75, 67.86, 66.95, 55.22, 33.09, 30.10, 26.78,

26.59, 25.84, 24.94.

R = m-Methoxybenzyl (40) (Yield: 223 mg, 100 %)

1H NMR (500 MHz, CDC13) 8 7.16 (1H, t, J = 8.0 Hz), 6.89 (1H, d, J = 8.0 Hz), 6.88
(1H, s), 6.75 (1H, dd, J = 8.0, 1 Hz), 4.16 (1H, d, J =13.0 Hz), 4.11 (1H, d, J =13.0 Hz),
4.07 (1H, dd, J = 8.5, 6.5 Hz), 3.89-3.93 (1H, m), 3.86 (1H, dd, J = 8.5, 5 Hz), 3.76 (3H,
s), 3.65-3.69 (1H, m), 2.31 (1H, dd, J = 17.0, 4 Hz), 2.24 (1H, dd, J = 17.0, 10 Hz),
1.43 (3H, s), 1.40 (3H, s), 1.39 (3H, s), 1.32 (3H, s); l3C NMR (125 MHz, CDC13) 8
159.62, 158.90, 140.11, 129.18, 121.29, 114.55, 112.51, 109.45, 89.66, 77.86, 68.00,

67.09, 55.02, 33.22, 32.41, 30.10, 26.65, 25.84, 25.02.

R = p-Methoxybenzyl (41) (Yield: 170 mg, 76 %)
'H NMR (500 MHz, CDC13) 8 7.21 (2H, d, J = 8.5 Hz), 6.78 (2H, d, J = 8.5 Hz), 4.10
(1H, d, J = 13.5 Hz), 4.05 (1H, d, J = 13.5 Hz), 4.03 (1H, dd, J = 8.5, 2.5 Hz), 3.87-3.91

(1H, m), 3.83 (1H, dd, J = 8.5, 5 Hz), 3.73 (3H, s), 3.61-3.66 (1H, m), 2.26 (1H, dd, J =

100

 

16.5, 3.5 Hz), 2.20 (1H, dd, J = 16.5, 10 Hz), 1.40 (3H, s), 1.38 (3H, s), 1.36 (3H, s),
1.30 (3H, s); ”C NMR (125 MHz, CDC13) 5 158.96, 158.51, 130.50, 130.05, 113.70,

109.46, 89.67, 77.82, 67.92, 67.07, 55.09, 33.18, 31.90, 30.11, 26.66, 25.85, 24.98.

R = o-Cyanobenzyl (42) (Yield: 200 mg, 90 %)

lH NMR (500 MHz, CDC13) 8 7.51-7.59 (2H, m), 7.47 (1H, t, J = 7.5 Hz), 7.28 (1H, t,
J = 7.5 Hz), 4.35 (1H, d, J = 14.0 Hz), 4.25 (1H, d, J =14.0 Hz), 4.05 (1H, dd, J = 8.0, 6
Hz), 3.87-3.91 (1H, m), 3.83 (1H, dd, J = 8.0, 5 Hz), 3.60-3.64 (1H, m), 2.28 (1H, dd, J
= 11.0, 3.5 Hz), 2.21 (1H, dd, J = 11.0, 10 Hz), 1.36 (9H, d, J = 9.5 Hz), 1.30 (3H, s);
”C NMR (125 MHz, CDC13) 8 157.86, 142.75, 132.43, 132.36, 130.24, 127.16, 117.50,

112.81, 109.46, 89.76, 77.69, 67.82, 67.00, 32.94, 30.45, 29.83, 26.62, 25.66, 24.94.

R = m-Cyanobenzyl (43) (Yield: 210 mg, 95 %)

lH NMR (500 MHz, CDC13) 8 7.63 (1H, s), 7.52 (1H, d, J = 6.5 Hz), 7.45 (1H, d, J =
6.5 Hz), 7.34 (1H, t, J = 6.5 Hz), 4.15 (1H, d, J = 13.5 Hz), 4.07 (1H, d, J= 13.5 Hz),
4.03-4.08 (1H, m), 3.87391 (1H, m), 3.80-3.84 (1H, m), 3.61-3.64 (1H, m), 2.26 (1H,
dd, J = 15.0, 13.5 Hz), 2.10 (1H, dd, J = 15.0, 10 Hz), 1.36 (9H, d, J = 4.0 Hz), 1.29
(3H, 8); '3C NMR (125 MHz, CDCl;:) 8 158.09, 140.56, 133.32, 132.67, 130.23,
128.82, 118.50, 112.09, 109.40, 89.65, 77.63, 67.78, 66.96, 33.01, 31.41, 29.93, 26.57,

25.68, 24.89.

R = p-Cyanobenzyl (44) (Yield: 190 mg, 84 %)

101

'H NMR (500 MHz, CDC13) 5 7.51 (2H, d, J = 8.0 Hz), 7.39 (2H, d, J = 8.0 Hz), 4.16
(1H, d, J = 14.0 Hz), 4.06 (1H, d, J = 14.0 Hz), 4.04 (1H, dd, J = 8.5, 6 Hz), 3.87-3.91

(1H, m), 3.82 (1H, dd, J = 8.5, 4.5 Hz), 3.58-3.62 (1H, m), 2.25 (1H, dd, J = 17.0, 4 Hz),
2.19 (1H, dd, J = 17.0, 10 Hz), 1.35 (6H, d, J = 2 Hz), 1.33 (3H, s), 1.28 (3H, s); ”C
NMR (125 MHz, CDC13) 5 157.99, 144.55, 131.95,131.80, 129.77, 129.61, 118.74,

110.37, 109.46, 89.68, 77.63, 67.71, 66.85, 32.95, 31.78, 29.93, 26.59, 25.68, 24.84.

102

5.7. REFERENCE

1.

10.

ll.

12.

13.

14.

15.

16.

M. L. Sinnott, Chem. Rev. 1990, 90, 1171-1202
G Legler, Adv. Carbohydr. Chem. Biochem. 1990, 48, 319-385

(a) P. S. Liu, US Patent 4,634,765, 1987; (b) P. S. Liu, J. Org. Chem. 1987, 52,
4717—4721

J. W. Dennis, Cancer Res. 1986, 46, 5131-5136

V. V. Sazak, J. M. Ordovas, A. D. Elbein, R. W. Beminger, Biochem. J. 1985, 232,
759-766

B. D. Walker, M. Kowalski, W. C. Goh, K. Kozarsky, M. Krieger, C. Rosen, L.
Rohrschneider, W. A. Haseltine, J. Sodroski, Proc. Natl. Acad. Sci. U. S. A. 1987,
84, 8120-8124

L. R. Rohrschneider, US Patent 1997, 5,643,888

N. Asano, Glycobiology, 2003, 13, 93R-104R

B. Winchester, (1 W. Fleet, Glycobiology, 1992, 2, 199-210

C. Rye and S. Withers, Curr: Opin. Chem. Biol. 2000, 4, 573-580

D. L. Zechel and S. G Withers, Acc. Chem. Res. 2000, 33, 11-18

Y. Ichikawa, Y. Igarashi, M. Ichikawa, Y. Suhara., J. Am. Chem. Soc. 1998, 120,
3007-3018

V. H. Lillelund, H. H. Jensen, X. Liang, and M. Bols, Chem. Rev. 2002, 102,
515-553

M. K. Tong, G Papandreou, B. Ganem, J. Am. Chem. Soc. 1990, 112, 6137-6139
B. Ganem, G Papandreou, J. Am. Chem. Soc. 1991, 113, 8984-8985

S. Vonhoff, T. D. Heightrnan, and A. Vasella, Helv. Chim. Acta. 1998, 81,
1710-1725

17. Z. J. Witczak, P. Kaplon, P. M. Dey, Carbohydr. Res. 2003, 338, 11-18

103

 

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32

K. D. Randell, T. P. Frandsen, B. Stoffer, M. A. Johnson, B. Svensson, B. M. Pinto,
Carbohydr: Res. 1999, 321, 143-156

J. S. Andrews, T. Weirnar, T. P. Frandsen, B. Svensson, and B, M, Pinto, J. Am.
Chem. Soc. 1995, 117, 10799-10804

H. S. Byun, R. A. Bittman, Tetrahedron Lett. 1995, 36, 5143-5145
V. D. Le and C. H. Wong, J. Org. Chem. 2000, 65, 2399-2409

Y. L. Merrer, M. Fuzier, I. Dosbaa, M. J. Foglietti, J. C. Depezay, Tetrahedron,
1997, 53, 16731-16746

K. Randell, T. P. Frandsen, B. Svensson, B. M. Pinto, J. Am. Chem. Soc. 1995, 117,
10799-10804

P. Ermert, A. Vasella, M. Weber, K. Rupitz, S. G Withers, Carbohydr: Res. 1993,
250, 113-128

R. Hoos, A. B. Naughton, W. Thiel, A. Vasella, W. Weber, K. Rupitz, S. G Withers,
Helv. Chim. Acta. 1993, 76, 2666-2686

P. Sairam, R. Puranki, B. S. Rao, P. V. Swamy, S. Chandra, Carbohydr: Res. 2003,
338, 303-306

A. K. Ghosh and W. Liu, J. Org. Chem. 1996, 61, 6175-6182

H. Hashimoto, M. Kawa, Y. Saito, T. Date, S. Horito, and J. Yoshimura,
Tetrahedron Lett. 1987, 28, 3505-3508

C. L. Cavallaro and J. Schwartz, J. Org. Chem. 1995, 60, 7055-7057
F. Chretien, Syn. Comm. 1989, 19, 1015-1024

K. M. Khan, S. Perveen, S. T. A. Shah, M. S. Shekhani, and W. Voelter, New. J.
Chem. 2001, 25, 896-898

R. F. Brady, Carbohydr: Res, 1970, 15, 35-40

104

     

  

lllllllllllfllllllllllllllllljlllll

32