Stereoselective U-4CRs with 1-amino-5-desoxy-5-thio-2,3,4-O-isobutanoyl-β-D-xylopyranose - An effective and selectively removable chiral auxiliary

Article abstract of DOI:10.1016/S0040-4020(02)00484-2

The Ugi four component reaction (U-4CR) offers a short and direct route for the synthesis of α-amino acid and peptide derivatives. With 1-amino-5-desoxy-5-thio-2,3,4-O-isobutanoyl-β-D-xylopyranose as a chiral amine component excellent chemical yields and stereoselectivities are obtained. After the U-4CR the chiral auxiliary can be cleaved off selectively under mild conditions. The configuration of one of the products was confirmed by X-ray analysis.

Full text of DOI:10.1016/S0040-4020(02)00484-2

TETRAHEDRON
Pergamon
Tetrahedron 58 ,2002) 6127±6133
Stereoselective U-4CRs with 1-amino-5-desoxy-5-thio-2,3,4-O-
isobutanoyl-b-d-xylopyranoseÐan effective and selectively
removable chiral auxiliary
GuÈnther F. Ross, Eberhardt Herdtweck and Ivar Ugip
È È
Organische Chemie 1, Technische Universitat Munchen, Lichtenbergstr. 4, 85747 Garching, Germany
Dedicated to M. Bodanszky
Received 13 December 2001; revised 11 March 2002; accepted 7 May2002
AbstractÐThe Ugi four component reaction ,U-4CR) offers a short and direct route for the synthesis of a-amino acid and peptide
derivatives. With 1-amino-5-desoxy-5-thio-2,3,4-O-isobutanoyl-b-d-xylopyranose as a chiral amine component excellent chemical yields
and stereoselectivities are obtained. After the U-4CR the chiral auxiliarycan be cleaved off selectivelyunder mild conditions. The
con®guration of one of the products was con®rmed byX-rayanalysis. q 2002 Published byElsevier Science Ltd.
1. Introduction
carbon±carbon bond are formed and a new chiral center is
created.
The four component reaction of isocyanides ,U-4CR) was
introduced in 1959, and it was shortlylater recognized that it
In 1966 Bodanszkyand Ondetti mentioned the potential
preparative advantages of U-4CRs for the preparation of
a-amino acid and peptide derivatives.8 Especiallyif
products contain a-amino acids, which cannot be obtained
is also possible to prepare derivatives of a-amino carboxylic
acids bystereoselective U-4CRs.
1±3
In the early1960s it
was realized that such stereoselective U-4CRs can best be
accomplished byusing suitable chiral primaryamine
components.4±6
The reaction mechanism of a stereoselective U-4CR was
determined in 1967, and it was found that four pairs of
reaction mechanisms compete, so that one product or the
other one can preferentiallybe formed. Even in the same
solvent and at the same temperature one or the other dia-
stereomeric product can be formed, just byusing different
concentrations of the educts. These results were obtained
from experimental and mathematicallyoriented combina-
torial investigations.3,7
To carryout a U-4CR, it is generallyadvantageous to
precondense the carbonyl compound 1 and the chiral
primaryamine 2 into the imine 3 ,Fig. 1). This is mixed
with the isocyanide 4 and the carboxylic acid 5. Therebythe
a-adduct 6 is formed. This rearranges irreversiblyinto the
a-amino carboxylic acid derivative 7.
In the course of the reaction two peptide bonds and one
Keywords: stereoselective Ugi four component reaction; chiral auxiliary;
chiral amine component; amino thiocarbohydrate; a-amino acid deriva-
tives.
Corresponding author. Tel.: 149-89-289-13229; fax: 149-89-289-
13249; e-mail: ivar.ugi@ch.tum.de
Figure 1. Synthesis of a-amino carboxylic acid derivatives by a stereo-
selective Ugi four component reaction with a chiral amine component 2.
The chiral auxiliary Rp should be cleaved off and regenerated after the
synthesis.
p
0040±4020/02/$ - see front matter q 2002 Published byElsevier Science Ltd.
PII: S0040-4020,02)00484-2

6128
G. F. Ross et al. / Tetrahedron 58 '2002) 6127±6133
Figure 2. Synthesis of 1-amino-5-desoxy-5-thio-2,3,4-O-isobutanoyl-b-d-xylopyranose hydrochloride 14.
from easilyavailable natural sources or biochemical pro-
cedures, the application of U-4CRs would be advantageous.
deformylation and cleavage of the chiral auxiliary could
be achieved under mild acidic conditions. In a subsequent
N-acylation step products identical to the U-4CR products 9
,Rc presumablynot H) were obtained. This methodology
avoids the harsh conditions necessaryto cleave N-acylated
and therebydeactivated N-glycosides by switching to a two
step process via the readilycleavable formamide. Recently
this research group has also reported the use of O-acyl-
pyranosylamine derivatives in U-4CRs12 on solid support.
For the synthesis of chiral a-amino acid derivatives by
stereoselective U-4CRs it is most reasonable to use a chiral
amine component. Then the chiral template is covalently
bound in close vicinityto the newlysynthesized chirality
center. The amine residue Rp of the product 7 must be
removable under mild conditions such that the desired
®nal product 9 is not destroyed.
In 1994 Goebel et al.13 accomplished high stereoselec-
tivities ,up to 98% de) and high yields by using per-O-alkyl-
ated pyranosylamines. But these chiral templates could still
not be removed under suf®cientlymild conditions to obtain
the desired end products directly.
A chiral amine component of the U-4CR has ideal proper-
ties if all of the following conditions are ful®lled: the amine
component must form the desired diastereomeric products
exclusivelyor at least with veryhigh stereoselectivityand
excellent yields of very pure products. The chiral auxiliary
must be selectivelycleavable from the product under mild
conditions. The ideal compound should be readilyavailable
and a resynthesis of the chiral amine component from its
cleavage product ,8!2) would be desired.
Lehnhoff et al.14 produced a great collection of U-4CR
products of the 1-amino-2-deoxy-2-N-acetyl-3,4,6-tri-O-
acetyl-b-d-glucopyranose. However, also in these cases no
satisfactorycleavage of the chiral auxiliarycould be
achieved.
For best results the most suitable solvent, optimal concen-
trations of educts and a Lewis-acid catalyst must be used at
the optimal temperature.
Zychlinski et al.15 performed U-4CRs with an amine
component derived from xylopiperidinose, in which the
endocylic O-atom was replaced by an acetylated N-atom.
This auxiliarygroup could alreadybe removed in neutral
solutions. Since the synthesis of these amine components
required a minimum of 11 steps, and since its products were
not suf®cientlystable, this chemistrywas not optimal.
More than 40 years ago, the search for a suitable amine
component began. Since 1969 ef®cient wayof preparing
chiral a-ferrocenylalkylamines were developed and it was
found that some of them have in principle all desired proper-
ties, but overall these amines were not suf®cientlystereo-
selective ,up to 76% de) and had too low ®nal yields
,30±40%).5,6,9 Therefore, the development of the a-ferro-
cenylalkylamines was given up.10
The previous development of various chiral amine com-
ponents led to the assumption, that pyranosylamines in
U-4CRs generallygive excellent iyelds and stereoselec-
tivities, especiallyin the presence of Lewis-acid catalysts.
In 1988 Kunz et al.11 introduced per-O-pivaloyl pyranosyl-
amines as chiral amine components for the U-4CR. In the
presence of zinc chloride etherate the condensations
proceeded with high stereoselectivityin excellent yields.
If formic acid was used as the acid component ,RCˆH),
If suitable ring-heteroanalogous pyranosylamines are used,
the chiral auxiliaryshould be selectivelycleavable under
mild conditions. The homologous sulfur is an especially
interesting candidate for the replacement of the endocyclic

G. F. Ross et al. / Tetrahedron 58 '2002) 6127±6133
6129
Figure 3. Stereoselective U-4CR with 1-amino-5-desoxy-5-thio-2,3,4-O-isobutanoyl-b-d-xylopyranose 15 and cleavage of the chiral auxiliary.
oxygen. The resulting condensation products can be seen as
S/N analogous ketals, which can be attacked bysoft electro-
philes for cleavage. The activation of thioanalogous ketals
bymercurysalts and other soft Lewis acids is well known,
for a review and mechanistic studies see, e.g. Satchell et
al.21 Johnston and Pinto22 have used mercury,II)chloride
as a catalyst when preparing S/N heteroanalogous 1,2-and
1,3-connected mannopyranose-disaccharides.
11 can be prepared in six steps from d-xylose 10. This
product can be peracylated into 12 byan excess of isobuta-
noylchloride in pyridine17 ,Fig. 2). Because of the anomeric
effect mainlythe a-anomer is generated. 12 can be
converted into 13 byreplacing the anomeric aclyoxy
group of 12 with an azido function by trimethylsilylazide
in the presence of tintetrachloride.18
The result is an anomeric mixture, in which the b-anomer
slightlyprevails. This means a distinct difference to the oxo-
analogous compounds, where onlythe b-anomer is formed.
In these, during the course of the reaction an intermediate
anomeric carbenium ion is formed byabstraction of the
anomeric acyloxy group. It is stabilized by the neighboring
acyloxy group as a cyclic acylium ion which only allows
the b-attack of the azide.18 Obviouslythis effect is not as
dominant in the thioanalogous system, probably due to the
reduced electronegativityand enhanced donor properties of
Thus the title compound 1-amino-5-desoxy-5-thio-2,3,4-O-
isobutanoyl-b-d-xylopyranose 15 has a particularlysuitable
combination of desirable properties.
2. Synthesis of 1-amino-5-desoxy-5-thio-2,3,4-O-isobuta-
noyl-b-d-xylopyranose
Ingles and Whistler16 found, that 5-desoxy-5-thio-d-xylose

6130
G. F. Ross et al. / Tetrahedron 58 '2002) 6127±6133
Figure 4. ORTEP drawing of the main part of the molecular structure of N-benzoyl-N-,5-desoxy-5-thio-b-d-xylopyranosyl)-R-leucine-tert-butylamide 19 in
the solid state. Thermal ellipsoids are drawn at the 50% probability level. Only the hydrogen atoms bound to the chiral centers are shown, all other hydrogen
atoms together with the atoms of the minor part of the disordered phenyl group are omitted for clarity.
the sulfur atom. This makes a neighboring group stabili-
zation of the intermediate cation less necessary; the azide
can attack from both sides.
removing its O-acyl groups by aminolysis with methyl-
amine. The absolute con®guration of the R-leucine deriva-
tive 19 was determined byX-rayanalysis ,Fig. 4). The right
enantiomer is proved byFlack's parameter , xˆ20.03,5)).23
The phenyl group appears disordered over two positions
with a ratio of 82:18.
The anomericallypure
b-amine hydrochloride 14 is
obtained from the a/b-azide mixture 13 byreduction with
1,3-propanedithiol ,Fig. 2).19 The b-azide is reduced much
quicker than the a-azide. So during the workup the b-amine
15 can be precipitated from an etheric solution as the hydro-
chloride salt 14, while the a-azide stays in solution. We
assume, that the relative stabilization of the a-azide bythe
anomeric effect is the reason for this kinetic discrimination.
The formation of the product with the R-con®guration is
stronglypreferred. This is in agreement with the assumed
steric course of this reaction via the seven membered imine
chelate 17.
The complexation with zinc chloride increases the rigidity
of this intermediate compound, so that in the chirogenic
a-addition step the prochiral imine function is ®xed rela-
tivelyto the chiral auxiliary. Therebya distinct steric shield-
ing of the Re-side of the imino function takes place and thus
a distinct evident stereoselection is achieved. Besides, the
zinc chloride activates the imine for the a-addition of the
isocyanide. A related reaction mechanism was already
3. Sereoselective U-4CRs with 1-amino-5-desoxy-5-thio-
2,3,4-O-isobutanoyl-b-d-xylopyranose and cleavage of
the chiral auxiliary
The thiopyranosylamine 15 is relased from 14 with triethyl-
amine. The 15 and isovaleraldehyde are subsequently
converted into the imine 16 ,Fig. 3). At 2408C in THF
the latter is mixed with zinc chloride diethyl etherate, tert-
butylisocyanide and benzoic acid.
11
discussed byKunz et al.
The O-deacylated chiral auxiliary group can be cleaved off
under mild acidic conditions bya dilute methanolic solution
of tri¯uoroacetic acid in the presence mercury,II)acetae.
During the aqueous workup of the product the mercury
compounds are precipitated byinducing hydrogen sul®de
or byadding sodium hydorgensul®de. The N-benzoyl-R-
leucine-tert-butylamide 20 and 5-desoxy-5-thio-d-xylose
After 72 h the product 18 is formed bythe U-4CR with
a stereoselectivityof 92% de ,diastereomeric ratio 24:1
determined byHPLC-MS) and a yield of 91.8%.
20
The readilycrystallizing product 19 is obtained from 18 by

G. F. Ross et al. / Tetrahedron 58 '2002) 6127±6133
6131
11 are therebyobtained. After 3 h at 20 8C the cleavage of 19
into 11 and 20 is complete, and no byproducts are then
found. From 11 the chiral amine 15 can be resynthesized.
stituents of the sugar ring can easilybe varied in order to
match reactants with different steric demands, thus increas-
ing the versatilityof this chiral auxiliary. The use of the
readilyavailable d-form of the sugar template induces the
formation of R-amino acid derivatives. This is particularly
useful since these `unnatural' amino acids are rather rare in
nature. If desired, also the l-xylose derivative is accessible,
although at a higher price. This means a major advance in
the development of the stereoselective U-4CR.
The O-deacylation facilitates the subsequent acidolytic
cleavage of the N-glycosidic bond in two ways: ®rstly, the
steric shielding of the reaction center is therebyreduced.
Secondly, the electron density of the anomeric carbon is
enhanced bythe removal of the electron withdrawing acyl
groups ,see also the armed/disarmed concept byFraser-Reid
et al.25).
The divalent mercuryforms a complex with the endocyclic
sulfur of the thiosugar moiety. Thereby the sulfur is
activated as a leaving group and an open-chain form of
the sugar is stabilized. This promotes the acid catalyzed
solvolysis of the N-glycosidic bond.
5. Experimental
5.1. General
1H NMR spectra were recorded on a Bruker AC 200
,200 MHz), 13C NMR spectra were recorded on a Bruker
AC 250 ,63 MHz). The chemical shifts d are given in ppm
relative to tetramethylsilane as an internal standard.
If the use of mercurycompounds should be avoided for
safetyreasons, other reagents in general use for the cleavage
of thioketals could be tried. In literature, the use of different
substances such as copper or silver compounds, Chloramine
T, tert-butylhypochlorite and others has been reported.24
5.1.1. Compound 12. Isobutanoylchloride ,20.2 g, 190
mmol) is graduallydropped into an ice-cooled solution
of 5-desoxy-5-thio-d-xylopyranose 11 ,6.3 g, 38 mmol) in
20 mL of drypyridine. The resulting suspension is diluted
with dichloromethane till the mixture can be stirred well.
Stirring is continued for 20 h at room temperature. Then
50 mL water are added and stirred for further 60 min.
Subsequently100 mL dichloromethane are added. The
organic phase is separated and twice washed with 100 mL
1N hydrochloric acid and subsequently with 100 mL of
saturated sodium hydrogencarbonate solution. After drying
with magnesium sulfate the solvent is evaporated. The
bright brown product is obtained in a yield of 16.1 g
,95%) as an anomeric mixture ,a/bˆ3.3:1).
The application of alternative cleavage reagents and the
removal of the chiral auxiliaryin one step with or without
previous O-deacylation remains an interesting task for
future investigations.
In the experiments described here, the isobutanoyl sub-
stituents turned out to be the optimal compromise in regard
of synthesis, reactivity and stereoselectivity of the thiosugar
template. When pivaloyl substituents were used, the tetra-
O-pivaloyl thioxylose ,analogous to 12) was not obtained in
satisfying yields. With benzoyl substituents, the amine
,analogous to 14 and 15) could be synthesized without
problems, but with the components tested no U-4CR was
observed.20a In both cases sterical hindrance seems to be the
reason for problems during synthesis or application of these
compounds.
1H NMR ,a-anomer, CDCl3): dˆ6.11 ,d, 1H, Jˆ2.8 Hz);
3
5.52 ,t, 1H, 3Jˆ10.1 Hz); 5.24 ,dd, 1H); 5.14 ,m, 1H); 2.99
3
2
3
,dd, 1H, Jˆ11.0 Hz, Jˆ12.2 Hz); 2.78 ,dd, Jˆ3.7 Hz);
2.6 ,m, 4H); 1.15 ,m, 24H). 13C NMR ,a-anomer, CDCl3):
dˆ175.14; 174.96; 174.8; 172.76; 73.7; 72.44; 70.84;
69.83; 34.02; 33.87; 33.86; 33.84; 26.5; 19.33; 19.15;
19.13; 19.10; 19.06; 19.04; 18.99; 18.94.
The O-acyl substituents of the chiral template can easily be
varied in order to form different types of products using
components with different steric demands.
5.1.2. Compound 13. 1,2,3,4-Tetra-O-isobutanoyl-5-
desoxy-5-thio-d-xylose 12 ,4.3 g, 9.6 mmol) is dissolved
in 50 mL of drydichloromethane and at 0 8C trimethylsilyl-
azide ,1.3 g, 11.04 mmol) is added. Subsequentlytin-
tetrachloride ,2.1 g, 8.16 mmol) is added. The mixture is
stirred at 208C for 2 h. After adding 50 mL of icywater,
the organic phase is ®ltered via Celite. It is twice washed
with a saturated sodium hydrogencarbonate solution, dried
bymagnesium sulfate and evaporated. The yield of the
bright brown product is 3.46 g ,90%). Its ratio of anomers
is a/bˆ1:2.3.
The tri-O-isobutanoyl thioxylopyranosylamine 15 was also
used in the syntheses of a phenylglycine derivative and a
dipeptide derivative.20a,b
If desired, also the enantiomeric products can be prepared
bythis method, since also l-xylose is commercially
available, although the higher price of this compound has
to be considered.
4. Conclusions
1H NMR ,b-anomer, CDCl3): dˆ5.39 ,t, 1H, Jˆ9.8 Hz);
3
3
2
A great varietyof products can now be formed stereoselec-
tivelybythe U-4CR of 1-amino-5-desoxy-5-thio-2,3,4- O-
isobutanoyl-b-d-xylopyranose 15. The chiral template is a
ring-thioanalogous carbohydrate. It combines excellent
stereoselectivities and chemical yields in the U-4CR with
a mild and speci®c cleavage mechanism. The O-acyl sub-
5.05 ,m, 2H); 4.46 ,d, 1H, Jˆ9.2 Hz); 2.97 ,dd, 1H, Jˆ
3
13.1 Hz); 2.7 ,dd, 1H, Jˆ3.7 Hz); 2.45 ,m, 3H); 1.1 ,m,
18H). 13C NMR ,b-anomer, CDCl3): dˆ175.49; 175.28;
175.13; 73.64; 73.33; 72.08; 63.24; 33.75; 33.73; 28.46;
18.74; 18.71; 18.64; 18.6; 18.56. IR ,KBr): 2100 cm21 ,s,
azide).

6132
G. F. Ross et al. / Tetrahedron 58 '2002) 6127±6133
5.1.3. Compound 14. 2,3,4-Tri-O-isobutanoyl-5-desoxy-5-
thio-d-xylopyranosylazide 13 ,5.82 g, 14.5 mmol) is dis-
solved in 50 mL dichloromethane. Subsequently50 mL
methanol, 1,3-propanedithiol ,3.3 g, 29 mmol) and triethyl-
amine ,3.0 g, 29 mmol) are added. This is stirred at 208C till
the formation of nitrogen has ended ,2±3 h). Then the vola-
tile compounds are removed in vacuo. The residue is
dissolved in 50 mL of diethyl ether and the insoluble 1,2-
dithiolane is removed. The ®ltrate is cooled and 2 mL
concentrated hydrochloric acid are added dropwise. The
b-amine hydrochloride 14 precipitates, while the a-ano-
meric azide remains in solution. The product is ®ltered off
and dried in a high vacuum. 2.09 g ,35%) of a white crystal-
line solid are obtained.
tion. It is dried over magnesium sulfate. The solvent is
removed in vacuo and a yield of 1.19 g ,91.8%) of the
white solid product 18 is obtained. According to the
HPLC-MS two diastereomers are present, the diastereo-
meric ratio is 24:1 ,92% de). The NMR shows two rotamers
in a ratio of 2.7:1. The data presented here refer to the major
rotamer.
1H NMR ,CDCl3): dˆ7.95 ,s, 1H); 7.6±7.45 ,m, 5H); 5.31
3
3
,d, 1H, Jˆ10.2 Hz); 5.18 ,t, 1H, Jˆ9.3 Hz); 5.08 ,t, 1H);
4.97 ,dt, 1H, 3Jˆ4.2, 10.3 Hz); 4.01 ,dd, 1H, 3Jˆ4.9,
2
9.7 Hz); 2.83 ,dd, 1H, Jˆ13.3 Hz); 2.53±2.32 ,m, 6H);
1.72 ,m, 1H); 1.34 ,s, 9H); 1.19±0.99 ,m, 24H). 13C NMR
,CDCl3): dˆ175.71; 175.61; 175.6; 173.93; 169.49; 135.28;
131.12; 128.73; 126.65; 74.01; 72.94; 72.02; 64.00; 63.10;
50.87; 40.51; 33.96; 33.9; 33.79; 28.68; 28.31; 25.67; 23.55;
22.22; 18.93±18.53. MS ,EI): 649 M1H; 671 M1Na; 687
M1K.
3
1H NMR ,DMSO-d61D2O): dˆ5.26 ,t, 1H, Jˆ9.2 Hz);
5.17 ,dd, 1H, 3Jˆ9.5 Hz); 4.96 ,ddd, 1H, 3Jˆ10.7,
4.8 Hz); 4.75 ,d, 1H, 3Jˆ9.5 Hz); 3.09 ,dd, 1H, 2Jˆ
13.4 Hz); 2.95 ,dd, 1H); 2.7±2.4 ,m, 3H); 1.05 ,m, 18H).
13C NMR ,DMSO-d61D2O): dˆ175.36; 175.28; 174.98;
72.41; 71.95; 71.46; 50.54; 33.17; 33.02; 27.24; 18.67;
18.51; 18.34; 18.31; 18.12; 17.73.
5.1.7. Compound 19. The condensation product 18 ,3.0 g,
4.6 mmol) is dissolved in 100 mL THF and 60 mL of a 40%
aqueous solution of methylamine are added. This is stirred
at 208C till the educt has reacted completelyaccording to
the TLC-control ,ethyl acetate/hexane 1:2) ,20 h). Then the
solvent is removed and the residue is washed with ether
several times. 1.94 g ,96%) of 19 are received as a white
powder that can be recrystallized from methanol.
5.1.4. Compound 15. The hydrochloride 14 is suspended in
ether and an excess of triethylamine is added. This is stirred
for 15 min, ®ltered and the solvent is removed in vacuo.
1-Amino-5-desoxy-5-thio-2,3,4-O-isobutanoyl-b-d-xylo-
pyranose 15 is obtained as a white solid in a yield of 98%.
1H NMR ,MeOH-d4): dˆ7.52±7.40 ,m, 5H); 4.66 ,d, 1H,
3
3Jˆ9.7 Hz); 3.87±3.79 ,m, 2H); 3.72 ,t, 1H, Jˆ6.2 Hz);
1H NMR ,CDCl3): dˆ5.15±5.03 ,m, 3H); 4.04 ,d, 1H,
3
3
2
3Jˆ9.3 Hz); 2.77 ,dd, 1H, Jˆ4.9 Hz, Jˆ13.7 Hz); 2.68
3.51 ,dt, 1H, Jˆ4.4, 11.1 Hz); 2.69 ,m, 2H); 2.62 ,dd, 1H,
2Jˆ13.3 Hz); 2.31 ,dd, 1H); 1.9 ,m, 1H); 1.33 ,s, 9H); 1.01±
0.98 ,2£d, 6H, 3Jˆ6.6 Hz). 13C NMR ,MeOH-d4): dˆ174.39,
172.00; 137.1; 131.07; 129.52; 127.16; 79.37; 74.66; 73.75;
66.00; 57.68; 52.36; 40.79; 32.48; 28.67; 27.70; 23.68; 22.40.
MS ,EI): 439 M1H; 461 M1Na; 477 M1K.
3
,dd, 1H, Jˆ10.2 Hz); 2.58±2.43 ,m, 3H); 1.17±1.09 ,m,
18H). 13C NMR ,CDCl3): dˆ176.54; 175.84; 175.57; 75.33;
72.67; 72.63; 57.76; 34.02; 33.95; 33.93; 29.29; 19.01;
18.87; 18.83; 18.8.
5.1.5. Compound 16. The amine hydrochloride 14 ,2.06 g,
5 mmol) is suspended in 50 mL diethyl ether and isovaler-
aldehyde ,0.52 g, 6 mmol) as well as triethylamine ,0.61 g,
6 mmol) are added. 5 g Magnesium sulfate are added. This
is stirred for 20 h, ®ltered and dried in vacuo. The imine 16
is obtained in a yield of 1.97 g ,89%) as a white solid.
5.2. X-Ray analysis of 19
Single crystals of 19 suitable for a X-raydiffraction study
were grown byslow evaporation of a methanolic solution of
the compound. Preliminaryexamination and data collection
were carried out on a kappa-CCD device ,NONIUS Mach3)
at the window of a rotating anode ,NONIUS FR591: 50 kV,
80 mA, 4.0 kW) and graphite monochromated Mo Ka radia-
tion ,lˆ71.073 pm). Data collection was performed at
2100,^1)8C ,173,^1) K).
1H NMR ,CDCl3): dˆ7.83 ,t, 1H, 3Jˆ4.9 Hz); 5.43 ,t, 1H,
3
3Jˆ9.5 Hz); 5.14 ,m, 2H); 4.33 ,d, 1H, Jˆ9.2 Hz); 2.86
,dd, 1H, 3Jˆ4.3 Hz, 2Jˆ13.4 Hz); 2.73 ,dd, 1H, 3Jˆ
3
10.1 Hz); 2.24 ,m, 3H); 2.14 ,t, 1H, Jˆ5.8 Hz); 1.9 ,m,
3
1H); 1.08 ,m, 18H); 0.92 ,2£d, 6H, Jˆ6.4 Hz). 13C NMR
,CDCl3): dˆ175.38; 175.27; 174.96; 170.77; 74.96; 73.05;
72.89; 72.46; 58.15; 34.26; 34.24; 29.19; 26.32; 22.98;
22.67; 19.35±19.14.
Crystal data. C22H34N2O5S, Mˆ438.58, colorless crystal of
0.43£0.38£0.28 mm, orthorhombic, space group P212121
,No. 19), aˆ1039.63,1), bˆ1106.24,1), cˆ2007.99,2) pm,
Vˆ2309.35,4)£106 pm3 ,from 2441 re¯ections with 2.03,
Q,25.358), Zˆ4, Dcˆ1.261 g cm23, F,000)ˆ944, mˆ
5.1.6. Compound 18. The imine 16 ,0.89 g, 2 mmol) is
dissolved in 10 mL of dryTHF. Then zinc chloride diethyl
etherate ,2.2 mmol, 1 mL of a 2.2 M solution in dichloro-
methane) is added. The mixture is cooled to 2408C and tert-
butylisocyanide ,0.18 g, 2.2 mmol) is added. After 15 min
benzoic acid ,0.27 g, 2.2 mmol) is added. When TLC-
control ,ethyl acetate/hexane 1:2) indicates the complete
reaction of the imine ,72 h) the THF is removed and the
residue is dissolved in 50 mL diethyl ether. The organic
layer is washed with 50 mL of 0.5N hydrochloric acid and
two times with saturated sodium hydrogencarbonate solu-
0.175 mm21
.
Measurement and solution. A total number of 41,816 re¯ec-
tions were integrated. Data were corrected for Lorentz and
polarization effects. Corrections for absorption and decay
effects were not applied. After merging a sum of 4237
independent re¯ections remained and were used for all
calculations. The structure was solved bya combination
of direct methods and difference Fourier syntheses. All
non-hydrogen atoms of the asymmetric unit were re®ned

G. F. Ross et al. / Tetrahedron 58 '2002) 6127±6133
6133
2. Ugi, I. Angew. Chem. 1962, 74, 9; Angew. Chem., Int. Ed.
Engl. 1962, 1, 8.
with anisotropic thermal displacement parameters. A dis-
order ,82:18) of the phenyl group could be resolved and
re®ned. All hydrogen atoms were found in the difference
Fourier map and re®ned freelywith individual isotropic
thermal displacement parameters, except those of the
minor part of the disordered phenyl group, which were
placed in calculated positions ,riding model).
3. Ugi, I. Isonitrile Chemistry; Academic: New York, 1971.
4. Ugi, I.; Steinbrueckner, C. Chem. Ber. 1961, 94, 2802.
5. Ugi, I.; Marquarding, D.; Urban, R. Chemistry and Bio-
chemistry of Amino Acids, Peptides and Proteins; Weinstein,
B., Ed.; Marcel Dekker: New York, 1982; Vol. 6, p 245.
6. Ugi, I.; Offermann, K. Angew. Chem. 1963, 75, 917; Angew.
Chem., Int. Ed. Engl. 1963, 2, 624. Ugi, I.; Offermann, K.;
Herlinger, H. Angew. Chem. 1964, 76, 613; Angew. Chem.,
Int. Ed. Engl. 1964, 3, 656. Ugi, I.; Offermann, K.; Herlinger,
H.; Marquarding, D. Justus Liebigs Ann. Chem. 1967, 709, 1.
7. Ugi, I.; Kaufhold, G. Justus Liebigs Ann. Chem. 1967, 709, 11.
8. Bodanszky, M.; Ondetti, M. A. In Peptide Synthesis, Olah,
G. A., Ed.; Wiley: New York, 1966.
Re®nement. Full-matrix least squares re®nements were
P
carried out byminimizing
wꢀFo2 2 Fc2†2; converging at
R1 ,F, obs. data: 3692 I.2.0s,I))ˆ0.0267, wR2 ,F2, all
data)ˆ0.0640, goodness of ®t 1.024, 444 parameters,
maximum shift/err,0.001, residual electron density
Ê 23
Drmaxˆ0.14, Drminˆ20.14 e A . SIR-92, SHELXL-97,
and software for Nonius KappaCCD were used. Crystallo-
graphic data ,excluding structure factors) for the structure
reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary
publication no. CCDC-179-984. Copies of the data can be
obtained free of charge on application to CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK ,fax: 144-1223-336-033;
e-mail: deposit@ccdc.cam.ac.uk).
9. Wagner, G.; Herrmann, R. In Ferrocenes, Togni, A., Hayashi,
T., Eds.; VCH: Weinheim, 1995.
10. Demharter, A.; Ugi, I. J. Prakt. Chem. 1993, 335, 244.
11. Kunz, H.; Pfrengle, W. J. Am. Chem. Soc. 1988, 110, 651.
Kunz, H.; Pfrengle, W. Tetrahedron 1988, 44, 5487. Kunz,
H.; Pfrengle, W.; Sager, W. Tetrahedron Lett. 1989, 30, 4109.
Kunz, H.; Pfrengle, W.; RuÈck, K.; Sager, W. Synthesis 1991,
1039. Kunz, H.; Rueck, K. Angew. Chem. 1993, 105, 355.
12. Oertel, K.; Zech, G.; Kunz, H. Angew. Chem. 2000, 112, 1489;
Angew. Chem., Int. Ed. Engl. 2000, 39, 1431.
5.2.1. Compound 20. The O-deacylated product 19 ,0.5 g,
1.14 mmol) is dissolved in a mixture of 20 mL methanol and
10 mL tri¯uoroacetic acid. Subsequentlymercury,II)acetate
,0.36 g, 1.14 mmol) is added and stirred until the educt
has completelyreacted according to the TLC-control
,ethyl acetate) ,about 3 h). Then 150 mL of a saturated
sodium hydrogencarbonate solution and 50 mL of diethyl-
ether are prepared and the reaction mixture is gradually
added. After neutralizing this with sodium hydrogencarbo-
nate 2 g of sodium hydrogensul®de monohydrate are added.
As an alternative hydrogen sul®de can also be induced. The
organic phase is separated, the aqueous phase is extracted by
50 mL of diethylether. The combined organic phases are
®ltered byCelite, washed bysaturated sodium hydrogen-
carbonate solution and dried bymagnesiumsulfate. A color-
less solid mixture of 20 and 11 is received. This mixture is
treated with chloroform, the organic phase is ®ltered over
Celite and evaporated in vacuo. 0.2 g ,60%) of 20 as a
transparent solid are obtained.
13. Goebel, M.; Ugi, I. Synthesis 1991, 1095.
14. Lehnhoff, S. PhD Thesis, Technical Universityof Munich,
È
1994. Lehnhoff, S.; Goebel, M.; Karl, R. M.; Klosel, R.;
Ugi, I. Angew. Chem. 1995, 107, 1208; Angew. Chem., Int.
Ed. Engl. 1995, 34, 1104.
15. Zychlinski, A. v. PhD Thesis, Technical University of
Munich, 1998.
16. Ingles, D. L.; Whistler, R. L. J. Org. Chem. 1962, 27, 3896.
Alternatives: Adley, T. J.; Owen, L. N. Proc. Chem. Soc.
1961, 418. Schwarz, J. C. P.; Yule, K. C. Proc. Chem. Soc.
1961, 417.
17. Behrend, R.; Roth, P. Liebigs Ann. Chem. 1904, 331, 359.
Fletcher, H. G. Methods in Carbohydrate Chemistry; Whistler,
R. C., Wolfrom, M. L., Eds.; Academic: New York, 1963;
Vol. 2, p 234.
18. Birkofer, L.; Ritter, A. Angew. Chem. 1965, 77, 414; Angew.
Chem., Int. Ed. Engl. 1965, 4, 417. Paulsen, H.; Gyorgydeak,
Z.; Friedmann, M. Chem. Ber. 1974, 107, 1568; Chem. Ber.
1974, 107, 1590. Strumpel, M. K.; Buschmann, J.; Szilagyi,
L.; Gyorgydeak, Z. Carbohydr. Res. 1999, 318, 91.
19. Bayley, H.; Standring, D. N.; Knowles, J. R. Tetrahedron Lett.
1978, 3633.
1H NMR ,CDCl3): dˆ7.79 ,d, 2H, 3Jˆ7.3 Hz); 7.51 ,t, 1H,
3
3Jˆ7.2 Hz); 7.43 ,t, 2H); 6.77 ,bd, 1H, Jˆ8.2 Hz); 6.00
,bs, 1H); 4.54 ,dt, 1H, 3Jˆ8.2, 5.8 Hz); 1.76±1.60 ,m,
3H); 1.35 ,s, 9H); 0.97 ,d, 6H). 13C NMR ,CDCl3): dˆ
171.75; 167.06; 133.86; 131.21; 128.15; 127.12; 52.67;
51.02; 41.22; 28.48; 24.76; 22.84; 22.10. MS ,EI): 291
M1H; 313 M1Na; 329 M1K.
20. ,a) Ross, G. F. PhD Thesis, Technical Universityof Munich
2001. ,b) Ross, G. F.; Ugi, I. Can. J. Chem. 2001, 79 ,12),
1934.
21. Satchell, D. P. N.; Satchell, R. S. Chem. Soc. Rev. 1990, 19,
55. Luh, T. Y. Acc. Chem. Res. 1990, 24, 257.
22. Johnston, B. D.; Pinto, B. M. J. Org. Chem. 1998, 63, 5797.
23. Flack, H. D. Acta Crystallogr. 1983, A39, 876.
24. El-Wassimy, M. T. M.; Jorgensen, K. A.; Lawesson, S. O.
J. Chem. Soc. Perkin, Trans. 1 1983, 2201. Lucchetti, J.;
Krief, A. Synth. Commun. 1983, 13, 1153. Emerson, D. W.;
Wynberg, H. Tetrahedron Lett. 1971, 3445.
Acknowledgements
The authors would like to thank Mrs G. Weidner and
Dr B. Werner for their assistance in the preparation of this
manuscript.
25. Fraser-Reid, B.; Wu, E.; Udodong, U. E.; Ottoson, H. J. Org.
Chem. 1990, 55, 6068.
References
1. Ugi, I.; Meyr, R.; Fetzer, U.; Steinbruckner, C. Angew. Chem.
1959, 71, 386.