Stereoselective Synthesis of Carbon-Linked Analogues of α- And β-Galactoserine Glycoconjugates Using Asymmetric Enolate Methodology

Full text of DOI:10.1021/jo980026x

J . Org. Chem. 1998, 63, 4817-4820
4817
Sch em e 1. Retr osyn th etic An a lysis
Ster eoselective Syn th esis of Ca r bon -Lin k ed
An a logu es of r- a n d â-Ga la ctoser in e
Glycocon ju ga tes Usin g Asym m etr ic En ola te
Meth od ology1
Robert N. Ben,2a Arturo Orellana,2b and Prabhat Arya*
Steacie Institute for Molecular Sciences, National Research
Council of Canada, 100 Sussex Drive,
Ottawa, Ontario K1A 0R6, Canada
Received J anuary 8, 1998
In recent years, glycopeptides have become the focus
of considerable bioorganic/medicinal research due to their
involvement in various cellular biological and pathologi-
cal processes. Glycosylation of proteins not only affects
their physical properties (i.e., folding and conformation)
but also influences biological function.3 Since glycosides
anchored to proteins form highly branched structures, it
is not surprising that they are involved in cell surface
recognition processes such as interactions with bacteria,
viruses, and toxins as well as tumor metastasis.4 Gly-
cosylation has also been implicated in the modulation of
protein function (i.e., cellular uptake, proteolytic stability)
and immune responses.5
Oxygen-linked glycosylation (i.e., glycosylation of serine
and threonine amino acid moieties) is one of the primary
modes for the attachment of glycosides to proteins.
However, a fundamental problem in using O-glycocon-
jugate-based therapeutic approaches to the treatment of
disease is the inherent lack of in vivo stability of such
compounds since native O-glycopeptides are easily de-
graded in both acidic and basic media.3 A solution to this
problem lies in obtaining carbon-linked isosteric deriva-
tives of O-glycopeptides.6 Biologically stable carbon-
linked glycopeptides may have applications in the areas
of cancer research, immunotherapy, and treatment of
inflammatory responses as well as various infection and
pathogenic processes.
(S)-serine in a number of steps. Following Bednarski’s
approach, a catalytic asymmetric hydrogenation meth-
odology for the synthesis of C-glycosyl amino acid deriva-
tives was recently achieved by Toone et al.7 As an
alternative to the previously reported approaches, herein,
we describe a novel strategy involving the application of
well-established asymmetric enolate methodology for the
synthesis of carbon-linked isosteres of R- or â-glycocon-
jugates bearing the amino acid serine. The key step in
our approach involved the reaction of a chiral oxazolidi-
none enolate with an electrophilic azide source or elec-
trophilic amination using a dialkyl azodicarboxyl ester
derivative (Scheme 1: retrosynthetic analysis, 1-4).8 In
either instance, reduction of the resulting R-azido ester
or the fundamental functional group manipulation of the
hydrazide derivative would furnish the desired C-linked
galactoserine. To our knowledge, a chiral auxiliary-based
enolate strategy for the synthesis of carbon-linked gly-
cosylated amino acids has never been explored. One
significant advantage of this methodology is that it could
easily be extended to a variety of electrophiles which
allows for the synthesis of other analogues of C-glycosyl
derivatives. To demonstrate the utility of this approach,
we have chosen to develop the methodology using D-
galactose.
In 1992, Bednarski et al. reported the first synthesis
of carbon-linked â-galactoside serine.6a A key step in
their approach was the coupling of C-glycosyl aldehyde
with the chiral Wittig reagent which was obtained from
As illustrated in Scheme 2, a two-carbon homologation
of R-C-galactose derivative 5, using commercially avail-
able (carbethoxymethylene)triphenylphosphorane, pro-
duced a mixture of geometric olefin isomers, which upon
hydrogenation using Pd/C as a catalyst furnished 6.
After conversion of the acetate protecting groups to
benzyl ethers on galactose moiety, the free carboxylic acid
derivative was then coupled to (4S)-4-isopropyl-2-oxazo-
lidinone to furnish 7 in 55% isolated yield. Surprisingly,
reaction of the potassium enolate derived from 7 with
trisyl azide failed to produce the expected R-azido prod-
uct.9 In contrast, di-tert-butyl azodicarboxylate (DBAD)
underwent reaction with the enolate of 7 to give the
* Corresponding author. Tel: (613) 993 7014. Fax: (613) 952-0068.
E-mail: prabhat.arya@nrc.ca.
(1) (a) NRCC Publication No. 40865. (b) Presented at the 15th
American Peptide Symposium, Nashville, TN J une 14-19, 1997.
(2) (a) NRC Research Associate, 1996-present. (b) NRC Summer
Student, May-August 1997.
(3) Kihlberg, J .; Elofsson, M. Curr. Med. Chem. 1997, 4, 85-116
and references therein.
(4) (a) Hakomori, S. Adv. Cancer Res. 1989, 52, 257-331. (b) Varki,
A. Glycobiology 1993, 3, 97-130. (c) Sharon, N.; Lis, H. Sci. Am. 1993,
268 (1), 82-89. (d) Dwek, R. A. Chem. Rev. 1996, 96, 683-720. (e)
Sears, P.; Wong, C.-H. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 12086-
12093.
(5) (a) Arsequell, G.; Haurum, J . S.; Elliot, T.; Dwek, R. A.; Lellouch,
A. C. J . Chem. Soc., Perkin Trans. 1 1995, 1739-1745. (b) Kihlberg,
J .; Ahman, J .; Walse, B.; Drakenberg, B.; Nilsson, A.; So¨derberg-Ahlm,
C.; Bengtsson, B.; Olsson, H. J . Med. Chem. 1995, 38, 161-169.
(6) (a) Bertozzi, C. R.; Hoeprich, P. D., J r.; Bednarski, M. D. J . Org.
Chem. 1992, 57, 6092-6094. (b) Bertozzi, C. R.; Cook, D. G.; Kobertz,
W. R.; Gonzalez-Scarano, F.; Bednarski, M. D. J . Am. Chem. Soc. 1992,
114, 10639-10641. (c) Kessler, H.; Wittmann, V.; Ko¨ck, M.; Kotten-
hahn, M. Angew. Chem., Int. Ed. Engl. 1992, 31, 902-904. (d) Gurjar,
M. K.; Mainker, A. S.; Syamala, M. Tetrahedron: Asymmetry 1993, 4,
2343-2346. (e) Frey, O.; Hoffmann, M.; Kessler, H. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2026-2028. (f) Burkhart, F.; Hoffmann, M.;
Kessler, H. Angew. Chem., Int. Ed. Engl. 1997, 36, 1191-1192.
(7) Debenham, S. D.; Debenham, J . S.; Burk, M. J .; Toone, E. J . J .
Am. Chem. Soc. 1997, 119, 9897-9898.
(8) Evans, D. A.; Nelson, S. G. J . Am. Chem. Soc. 1997, 119, 6452-
6453 and references therein. For other applications of Evans’ chiral
enolate methodology for the synthesis of unnatural amino acids, see:
Gilbertson, S. P.; Chen, G.; McLoughlin, M. J . Am. Chem. Soc. 1994,
116, 4481-4482.
S0022-3263(98)00026-7 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/23/1998

4818 J . Org. Chem., Vol. 63, No. 14, 1998
Notes
Sch em e 2a
a
Reagents and conditions: (a) (i) Ph3PdCHCOOEt, benzene, reflux, (ii) H2, 10% Pd/C, EtOH; (b) (i) 1 M LiOH, MeOH/H2O, (ii) p-TSA,
MeOH, (iii) NaH, BnBr, Bu4NI, 15-crown-5, DMF, (iv) 1 M LiOH, H2O/THF, (v) 2,4,6-trichlorobenzoyl chloride, Et3N, CH2Cl2, for 30 min
then lithio-(S)-isopropyl-2-oxazolidinone, THF; (c) KHMDS (1.1 equiv), THF, -78 °C, 30 min, MeI (5 equiv) or DBAD (1.3 equiv).
Sch em e 3a
a
Reagents and conditions: (a) LDA, THF, -78 °C, 30 min, DBAD (1.3 equiv); (b) (i) 2.5 M LiOH, THF/H2O, 0 °C, (ii) CH2N2, THF, (iii)
TFA/CH2Cl2 (1:1), (iv) Raney-Ni, H2, 500 psi.
Sch em e 4a
a
Reagents and conditions: (a) (i) Ph3PdCHCOOEt, benzene, reflux; (ii) H2, 10% Pd/C, EtOH, (iii) 1 M LiOH, MeOH/H2O, (iv) 2,4,6-
trichlorobenzoyl chloride, Et3N, CH2Cl2, for 30 min then lithio-(S)-isopropyl-2-oxazolidinone, THF; (b) LDA (1.1 equiv), THF, -78 °C, 30
min, DBAD (1.3 equiv).
hydrazide product 9, but the selectivity of the reaction
using DBAD was not very high (72% de). On the basis
of literature precedents of similar reactions, the observed
low diastereoselectivity was quite surprising.8 In view
of the experimental results, it seemed probable that the
benzyl protecting group at the 2-position of the galactose
moiety hindered the approach of the electrophile (see
enolate derived from 7, Scheme 2). In an effort to enhance
the diastereoselectivy of the electrophilic amination
reaction, the benzyl group at the 2-position was replaced
by the less sterically demanding methyl group to give the
permethylated derivative 10, which was synthesized from
6 (Scheme 3). As anticipated, reaction of the enolate of
10 (generated under standard conditions, LDA, THF, -78
°C) with DBAD furnished 11 with a high diastereoselec-
tivity (92% de).10 Compound 11 was subjected to reaction
with 2.5 M LiOH, esterification with diazomethane, and
treatment with TFA to affect the removal of the Boc
group. In situ reduction of the hydrazine derivative with
Raney-Ni under a hydrogen atmosphere then furnished
the desired product 12.
In a separate attempt to investigate the relative
orientation of the enolate and attendant benzyloxy group
at the 2-position of the galactose moiety, the perbenzy-
lated derivative of â-galactose 14 was synthesized from
1311 in a number of steps: (i) two-carbon homologation
by coupling with the (carbethoxymethylene)triphenyl-
phosphorane as described earlier for R-galactose deriva-
tive 5, (ii) hydrogenation using Pd/C, (iii) hydrolysis of
the carboxyl ester group to obtain free carboxyl acid, and
(iv) coupling of the (4S)-4-isopropyl-2-oxazolidinone to
furnish 14 in 45% isolated yield (Scheme 4). It seemed
likely that with the â-configuration at the anomeric
(10) Diastereomeric excess (% de) was measured using HPLC and
HPLC-MS techniques. Structures of all the compounds were fully
assigned using 1H and 13C NMR and MS experiments. In compounds
8, 11, and 15, the stereochemistry of new stereogenic centers was
assigned as it would be expected from Evans’ asymmetric oxazolidinone
enolate methodology.
(9) The enolate of 7 with methyl iodide gave 8 in excellent yield
with a high diastereoselectivity (>95% de).

Notes
J . Org. Chem., Vol. 63, No. 14, 1998 4819
hexanes-ethyl acetate (3:1) as an elutant to give 55 mg (72%
yield) of 9 as a mixture of diastereomers. 1H NMR (500 MHz,
CDCl3) δ: 0.80 (d, 6H, J ) 13 Hz), 1.50-1.31 (m, 18H), 1.60 (br
s, 1H), 1.80 (br s, 1H), 2.20 (br s, 2H), 3.40 (br s, 1H), 3.80-3.61
(m, 3H), 3.90 (s, 1H), 4.10-3.95 (m, 5H), 4.44 (d, 1H, J ) 14
Hz), 4.60-4.50 (m, 3H), 4.85-4.65 (m, 4H), 5.71 (br s, 1H), 6.74
(br s, 1H), 7.42-7.10 (m, 20H). 13C NMR (50 MHz, CDCl3) δ:
19.2, 21.2, 27.8, 27.9, 28.0, 40.2, 64.0, 73.1, 78.2, 79.1, 80.5, 82.2,
131.8, 132.5, 132.8, 131.0, 144.0, 143.2, 148.3, 150.2, 154.2, 154.8,
154.9, 170.2. MS (ES+): m/z (relative intensity) 952.0 (50), 463.5
(28), 346.0 (37), 332 (71), 233 (100).
(4S)-4-Isop r op yl-3-[1-oxo-4-(2,3,4,6-t et r a -O-m et h yl-r-D-
ga la ctop yr a n osyl)bu tyl]-2-oxa zolid in on e (10). For a gen-
eral experimental procedure, refer to 7. 1H NMR (200 MHz,
CDCl3) δ: 0.90 (t, 6H, J ) 8.6 Hz), 1.90-1.51 (m, 5H), 2.42 (m,
2H), 2.94 (m, 3H), 3.73-3.19 (m, 14H), 3.82 (m, 1H), 4.05 (m,
1H), 4.28 (m, 2H), 4.40 (m, 1H). 13C NMR (50 MHz, CDCl3) δ:
32.9, 36.2, 39.0, 43.7, 46.6, 53.5, 76.9, 78.3, 81.6, 88.4, 89.2, 89.9,
93.8, 94.7, 95.3, 95.9, 96.1, 97.2, 172.3, 191.3. MS (ES+): m/z
(relative intensity) 418.2 (100): HRMS calcd for C20H35NO8 (M+)
417.2362, found 417.2365.
(4S)-4-Isop r op yl-3-[2-[N-(N-b u t yloxyca r b a m id o)b u t yl-
oxyca r ba m id o]-1-oxo-4-(2,3,4,6-tetr a -O-m eth yl-r-D-ga la cto-
p yr a n osyl)bu tyl]-2-oxa zolid in on e (11). For a general ex-
perimental procedure, refer to 9. 1H NMR (400 MHz, CDCl3) δ:
0.80 (d, 3H, J ) 10.0 Hz), 0.84 (d, 3H, J ) 10 Hz), 1.40 (s, 18H),
1.82 (m, 2H), 1.86 (br s, 1H), 2.30 (br s, 2H), 3.60-3.20 (m, 18H),
3.82 (br s, 1H), 3.90 (br s, 1H), 3.95 (q, 1H, J ) 8.0, 3.1 Hz),
4.10 (m, 2H), 5.38 (br s, 1H). 13C NMR (50 MHz, CDCl3) δ: 14.5,
17.4, 17.7, 20.6, 23.2, 25.6, 27.9, 32.4, 58.1, 58.7, 60.1, 63.7, 68.3,
70.7, 72.4, 75.8, 78.7, 80.2, 81.6, 153.1, 154.2, 155.7, 160.1.
HRMS: calcd for C30H54N3O12 (M+) 648.3709, found 648.3701.
Meth yl 2-Am in o-4-(2,3,4,6-tetr a -O-m eth yl-r-D-ga la ctop y-
r a n osyl)bu ta n a te (12). Compound 11 (32 mg, 0.06 mmol) was
placed in a round-bottom flask, 8 mL of a TFA-CH2Cl2 (1:1)
mixture was added, and the reaction was stirred at room
temperature. After 1 h Raney-Ni (300 mg, damp weight) was
added, and the reaction flask was pressurized (550 psi) with
hydrogen gas. After 3 h the solution was filtered through Celite,
and the solvent was removed to produce a pale-green oil which
was purified by column chromatography to give 12 (10 mg, 54%
yield) as a colorless oil. 1H NMR (400 MHz, CDCl3) δ: 1.80 (m,
3H), 2.30 (t, 2H, J ) 5.0 Hz), 2.9 (s, 2H), 3.50-3.31 (m, J ) 16
Hz), 3.70 (s, 3H), 4.12 (m, 1H), 4.52 (t, 1H, J ) 6.2 Hz). 13C
NMR (50 MHz, CDCl3) δ: 17.7, 18.2, 21.5, 32.8, 33.9, 51.7, 53.2,
59.3, 68.7, 70.2, 71.3, 75.8, 76.4, 128.1, 170.2. MS (ES+): m/z:
335 [M+].
center, the benzyl group at the 2-position of the saccha-
ride moiety of 14 would be unable to interfere during the
reaction of the enolate with di-tert-butyl azodicarboxylate
in contrast to R-galactose derivative 7 (see Scheme 2 for
enolates derived from R- and â-galactose derivatives, 7
and 14). As expected, reaction of the enolate derived
from 14 with DBAD gave the expected product 15, with
excellent diastereoselectivity (98% de).
In conclusion, we have successfully demonstrated that
chiral auxiliary-based enolate methodology can be uti-
lized to synthesize carbon-linked analogues of both R- and
â-galactoserine glycoconjugates which are of tremendous
biological and medicinal importance. The effect of the
saccharide moiety as a remote chiral group on the
selectivity of the electrophilic amination is unprec-
edented. Although the mechanism by which the benzyl
protecting group at the 2-position of the R-galactose
moiety interferes with the enolate is not clear at this
stage, it has the ability to modulate the selectivity of the
electrophilic amination reaction. Further work is re-
quired to explore and exploit the full potential of this
surprising observation. We are currently extending the
enolate methodology and investigating other applications
of these C-linked glycoconjugates.
Exp er im en ta l Section
Gen er a l Rem a r k s. Proton NMR spectra were recorded
using a Bruker AM-200 or an AMX 400 MHz instrument using
CDCl3 as solvent. Chemical shifts are reported in ppm downfield
from TMS as an internal standard. Multiplicities are reported
as s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet. Mass
spectra were recorded using a J EOL J MS-AX 505H mass
spectrometer using mNBA as a matrix.
All reactions were carried out under a nitrogen atmosphere
using oven-dried glassware, and unless otherwise stated, typical
extractive workup procedures were applied in all cases. Chro-
matographic separations were performed using Merck grade 60
silica gel (230-400 mesh, 60A) in the solvent systems specified,
and all yields refer to isolated yields after purification. All
solvents were dried prior to use.
(4S)-4-Isop r op yl-3-[1-oxo-4-(2,3,4,6-t et r a -O-b en zyl-r-D-
ga la ctop yr a n osyl)bu tyl]-2-oxa zolid in on e (7). In an oven-
dried round-bottom flask under nitrogen the perbenzylated
carboxylic acid (188 mg, 0.31 mmol) was placed in 20 mL of THF
and cooled to -78 °C. Then diisopropylethylamine (0.06 mL,
1.1 equiv) and 2,4,6-trichlorobenzoyl chloride (0.05 mL, 1.1 equiv)
were added. After stirring for 1 h the lithio-(S)-isopropyloxazo-
lidinone was added via cannula and the solution was warmed
to room temperature over a 2-h period. Typical aqueous workup
and extraction with diethyl ether followed by treatment with
magnesium sulfate and removal of the solvent under reduced
pressure produced a yellow oil. After purification by column
chromatography using hexanes-ethyl acetate (2:1) as an elutant,
160 mg (72% yield) of 7 was obtained as a pale-yellow oil. 1H
NMR (200 MHz, CDCl3) δ: 0.85 (q, 6H, J ) 8, 13 Hz), 1.70 (m,
5H), 2.35 (m, 1H), 2.95 (s, 2H), 3.90-3.50 (m, 3H), 3.95 (s, 2H),
4.30-4.10 (m, 2H), 4.80-4.40 (m, 10H), 7.25 (m, 20H). 13C NMR
(50 MHz, CDCl3) δ: 18.0, 21.0, 23.2, 25.1, 26.0, 39.5, 63.2, 64.0,
70.0, 76.3, 79.1, 80.5, 81.3, 131.3, 132.1, 133.1, 143.0, 143.2,
157.0, 177.6. MS (ES+): m/z (relative intensity) 722.2 (40).
HRMS: calcd for C44H51NO8 (M+) 722.3695, found 722.3767.
(4S)-4-Isop r op yl-3-[2-[N-(N-b u t yloxyca r b a m id o]b u t yl-
oxyca r ba m id o-1-oxo-4-(2,3,4,6-tetr a -O-ben zyl-r-D-ga la cto-
p yr a n osyl)bu tyl]-2-oxa zolid in on e (9). Compound 7 (55 mg,
0.08 mmol) was placed in a flame-dried round-bottom flask with
25 mL of dry THF under a nitrogen atmosphere. This solution
was then cooled to -78 °C, and LDA (1.1 equiv) was added. After
45 min a THF solution of DBAD (22 mg, 1.2 equiv) was added,
and the solution was stirred for 5 min after which time glacial
acetic acid (0.01 mL, 2.6 equiv) was added and the solution was
allowed to warm to room temperature overnight. Extractive
workup furnished a crude yellow oil which was purified using
2-(2,3,4,6-Tetr a -O-ben zyl-â-D-ga la ctop yr a n osyl)a ceta ld e-
h yd e (13). For a general experimental procedure describing the
preparation of 13 see ref 11. 1H NMR (400 MHz, CDCl3) δ: 2.33
(m, 1H), 2.61 (m, 1H), 3.30 (td, 1H, J ) 8.0, 9.1 Hz), 3.54 (m,
1H), 3.71 (t, 1H, J ) 8.0 Hz), 3.98 (d, 1H, J ) 2.6 Hz), 4.45 (q,
2H, J ) 10.1, 11.2 Hz), 4.68 (m, 2H), 4.74 (d, 2H, J ) 11.2 Hz),
4.93 (d, 2H, J ) 9.0 Hz), 5.06 (m, 2H), 5.91 (m, 1H), 7.30 (m
20H). 13C NMR (50 MHz, CDCl3) δ: 36.9, 69.2, 72.8, 73.4, 73.6,
76.1, 76.9, 77.2, 78.1, 79.3, 85.3, 117.3, 128.3, 128.4, 128.7, 128.9,
130.1, 130.3, 132.4, 132.8, 136.2, 138.1, 138.4, 139.0, 139.2.
(4S)-4-Isop r op yl-3-[1-oxo-4-(2,3,4,6-t et r a -O-b en zyl-â-D-
ga la ctop yr a n osyl)bu tyl]-2-oxa zolid in on e (14). For a gen-
eral experimental procedure, see 7 and 10. 1H NMR (400 MHz,
CDCl3) δ: 1.80 (m, 3H), 2.30 (t, 2H, J ) 5.0 Hz), 2.9 (s, 2H),
3.50-3.31 (m, J ) 16 Hz), 3.70 (s, 3H), 4.12 (m, 1H), 4.52 (t,
1H, J ) 6.2 Hz). 13C NMR (50 MHz, CDCl3) δ: 17.7, 18.2, 21.5,
32.8, 33.9, 51.7, 53.2, 59.3, 68.7, 70.2, 71.3, 75.8, 76.4, 128.1,
170.2. MS (ES+): m/z: 335 [M+].
(4S)-4-Isop r op yl-3-[2-[N(N-b u t yloxyca r b a m id o)b u t yl-
oxyca r ba m id o]-1-oxo-4-(2,3,4,6-tetr a -O-ben zyl-â-D-ga la cto-
p yr a n osyl)b u t yl]-2-oxa zolid in on e (15). For a general ex-
(11) Synthesis of 13 was achieved from 2,3,4,6-tetra-O-benzyl-R-D-
galactopyranosyl chloride by the reaction with allylmagnesium bromide
(0 °C), followed by ozonolysis and the reductive opening of the ozonide.
For a similar series of reactions, see: Uchiyama, T.; Woltering, T. J .;
Wong, W.; Lin, C.-C.; Kajimoto, T.; Takebayashi, M.; Weitz-Schmidt,
G.; Asakura, T.; Noda, M.; Wong, C.-H. Bioorg. Med. Chem. 1996, 4
(7), 1149-1165.

4820 J . Org. Chem., Vol. 63, No. 14, 1998
Notes
perimental procedure, see 9. 1H NMR (400 MHz, CDCl3) δ: 0.86
(br s, 5H), 1.50-1.05 (m, 19H), 1.84-1.60 (m, 3H), 2.10 (br s,
2H), 2.32 (br s, 1H), 3.22 (br s, 1H), 3.52 (m, 6H), 4.20-3.98 (m,
4H), 4.72-4.42 (m, 6H), 4.95 (m, 2H), 7.43 (m, 20H). 13C NMR
(50 MHz, CDCl3) δ: 15.0, 18.3, 28.4, 30.1, 59.3, 63.8, 72.5, 73.8,
74.4, 74.9, 75.7, 79.8, 80.8, 81.6, 82.0, 85.0, 127.9, 128.2, 128.5,
128.7, 138.6, 138.9, 139.4, 153.5, 155.0, 155.4, 156.1, 175.0. FAB-
MS: m/z (relative intensity) 974.4 (70), 874.4 (30), 774.3 (20),
617.3 (15). HRMS: calcd for C54H69O12N3Na (M+ + Na) 974.4778,
found 974.4788.
Ack n ow led gm en t. Mr. Don Leek and Ms. Malgosia
Daroszewska are thanked for providing technical as-
sistance for running NMR and MS.
Su p p or tin g In for m a tion Ava ila ble: Spectral data of
compounds (18 pages). This material is contained in libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
J O980026X