INTRODUCTION
Owing to their high functionalization, sugars are difficult
compounds to be synthetically manipulated, and their employment in
organic synthesis is often associated with long multistep schemes
required for setting the right profile of functional groups in each
residue. Additionally, oligosaccharide synthesis is still the object of
intense investigation, and the functionalization of each saccharide
building block plays a fundamental role in tuning several parameters
such as the donor/acceptor reactivity and the regio- and stereocontrol
of glycosidations [1]. Despite the remarkable progress in the
development of straightforward procedures for both the rapid functional
differentiation of sugars [2] and the oligosaccharide assembly [1], the
request of ever more practical procedures is still an issue of high
interest. Indeed, most advanced procedures devised for these tasks still
suffer from practical drawbacks such as the recourse to unstable
reagents, lengthy reactions, use of expensive stoichiometric reagents,
etc.
In this regard, this contribution is focused on the numerous
synthetic opportunities offered by the [I.sub.2]/[Et.sub.3]SiH combined
system in synthetic carbohydrate chemistry. Both reagents are of
practical interest as they are easily handled and storable, and upon
mixing they quickly generate HI and [Et.sub.3]SiI in situ by a process
that can occur under anhydrous conditions [3]. As will be shown, a
different stoichiometric combination of the reagents can be addressed to
varied applications in carbohydrate chemistry ranging from
glycosidations to functional group elaboration of sugars.
Besides our contributions, the combined use of [I.sub.2] and
[Et.sub.3]SiH is not frequently reported in the literature; in some
applications these reagents are employed in combination with further
catalysts, for example, in the hydroiodination of alkenes and alkynes
[4] and in the anomeric iodination of sugars [5]. In the absence of
other co-reagents, [I.sub.2] and [Et.sub.3]SiH have been reported to
promote the reductive Ferrier rearrangement of glycals [6], the
Friedel-Crafts cyclization of aryl-substituted propargyl alcohols [7],
the reductive opening of benzylidene acetals [8], and the reduction of
an aromatic formyl substituent to a methyl group [9].
RESULTS AND DISCUSSION
Use of the [I.sub.2]/[Et.sub.3]SiH system as a glycosidation
promoter
Our first contribution on the [I.sub.2]/[Et.sub.3]SiH system
stemmed from an attempt to extend the scope of iodine as a glycosidation
promoter after Field and co-workers had reported some examples of armed
glycosyl donors (including a tetra-O-benzylated glycosyl
trichloroacetimidate) activated by stoichiometric iodine [10]. This
effort led to a protocol for the activation of the less reactive
disarmed glycosyl imidates that relies on a combined activation system
composed of stoichiometric iodine and catalytic triethylsilane (Scheme
1) [11]. Interestingly, the use of catalytic amounts of both reagents
was not effective for the reaction to occur, suggesting that the sole HI
cannot catalyze the reaction at a practical rate. The reaction outcome
was critically dependent on the nature of the solvent, the best results
being achieved in 4:1 dichloromethane/acetonitrile. The nature of drying
agents had a critical role too; the use of commercially available
acid-washed molecular sieves (AW 300 MS) did indeed prevent the shutdown
of the catalytic cycle that was instead observed in the presence of
normal 4[Angstrom] molecular sieves.
[ILLUSTRATION OMITTED]
Interestingly, in the coupling with the selected model acceptor 1,
we observed an especially good performance of the
N-phenyltrifluoroacetimidate donor 2. This class of donors had been
proposed just a few months before in a paper by Yu [12], and in the
following years numerous contributions have confirmed its usefulness
[13] as well as its possible complementarity with standard
trichloroacetimidates [1].
Another interesting result of that investigation lies in the higher
yields obtained from donors 2 and 3 (equipped with an unusual
methoxycarbonyl group as a participating functionality) than when using
donors 4 and 5, protected with conventional acetyl and benzoyl groups.
In our experience, this trend was often found reproducible under a
variety of activation conditions for imidate donors, especially when
mild conditions were contemplated [14].
Gratifyingly, some years later, Tanaka, Takahashi, and co-workers
developed a very useful application of the analogous stoichiometric
combination even in the activation of highly armed glycosyl donors. In
particular, they achieved a highly selective synthesis of
[beta]-glycosides of 2,6-di-deoxy sugars by activating the corresponding
trichloroacetimidate donors at very low temperatures in toluene; the
wide scope of the procedure was demonstrated with the application to a
large number of compounds, including very demanding targets [15].
Synthesis of glycosyl iodides and synthetic elaborations thereof
In a subsequent investigation, the use of a different
stoichiometric combination of [I.sub.2] and [Et.sub.3]SiH was addressed
toward alternative useful transformations of carbohydrate chemistry.
Remarkably, per-O-acetylated sugars can be smoothly converted into the
corresponding glycosyl iodides upon exposure to a moderate excess of
both reagents (1.4 equiv each) in refluxing dichloromethane (Scheme 2)
[16]. The reaction takes very short times (5-10 min), and the outcome is
independent of the anomeric composition of the starting material (with
just an exception being found to date with the N-Troc glucosamine
precursor [17]). This approach can be considered a useful alternative to
the use of iodotrimethylsilane (TMSI, sensitive reagent) [18], or the
combined propandithiol/[I.sub.2] [19] and hexamethyldisilane/[I.sub.2]
[20] systems. Glycosyl iodides generated with [I.sub.2]/[Et.sub.3]SiH
are obtained after a simple extractive work-up just contaminated by
inert nonsaccharidic triethylsilylated side-products; the resulting
crude mixtures can thereby be directly submitted to further elaborations
typically applied to glycosyl bromides. For example, 1,2-orthoesters,
glycals, and ethylidenes can be readily accessed via glycosyl iodides
thus obtained [16]. In some cases, the extractive work-up following the
anomeric iodination can be even suppressed; for example, 1,2-orthoesters
can be accessed by simply adding the requisite reagents to the
iodination mixture (one-pot procedure), whereas the ethylidenation step
can be carried out by just replacing the solvent of the iodination step
(dichloromethane) with acetonitrile. All these methodologies are
generally effective in terms of yields, and, not unexpectedly, glycosyl
iodides often reacted in shorter times than the brominated counterparts
under otherwise identical reaction conditions.
[ILLUSTRATION OMITTED]
The ready access to glycosyl iodides and their high reactivity
elicited the development of other useful synthetic schemes in
carbohydrate chemistry. Practically important applications are the
streamlined approaches for the synthesis of alkyl-, aryl thioglycosides,
and selenoaryl glycosides (Scheme 3) [21]. Alkyl thioglycosides are
readily obtained from crude iodides through the generation of the
corresponding thiouronium intermediate and its direct treatment, in a
one-pot fashion, with a mild base as triethylamine (TEA) and the
suitable alkylating agent (Scheme 3). Alternatively, direct exposure of
crude glycosyl iodides (from the extractive work-up) to the appropriate
phenyl chalcogenide (easily generated in situ under reductive
conditions) led to the smooth generation of phenylthio- and
phenylselenoglycosides (Scheme 3).
[ILLUSTRATION OMITTED]
Interestingly, the thioalkylation step can also be used to attach
functionalized chains, and this chemistry was recently employed by
Comegna et al. for the streamlined synthesis of aminoethyl
thioglycosides useful for the assembly of glycosylated peptoids via a
solid-phase approach [22]. Another application was also described with
the preparation of suitable intermediates for oligosaccharide synthesis
[23].
The scheme of thioglycosidation has been more recently adapted to
the generation of phenylselenyl thioglycosides, a class of compounds
recently introduced by Davis and co-workers for the direct
thioglycosidation of proteins through the generation of a disulfide
linkage with cysteine residues [24]. The S-Se linkage can be generated
on addition of phenyl diselenide and TEA to the reaction vessel where
the glycosyl thiouronium intermediate 9 has been formed (Scheme 4) [25].
Experiments conducted by changing the equivalents of [(PhSe).sub.2] in
this step indicated that overall yields of phenylseleno sulfide 10 are
always moderate (never exceeding 45 %); on the other hand, an advantage
of the scheme is that this product can be purified with a single
chromatography and without isolation of any intermediate.
Application of this synthetic sequence always led to the generation
of the corresponding symmetrical disulfide 11 as the main byproduct
(Scheme 4), and a suitable adjustment of the reaction conditions allowed
this latter product to be obtained in a synthetically useful yield. In
particular, treatment of the thiouronium intermediate 9 with excess TEA
(4 equiv) and a catalytic amount of phenyl diselenide (3 %) for an hour
at 50[degrees]C yielded the galactosyl disulfide 11 in a good overall
yield (69 % from the per-O-acetylated precursor) [26]. The product was
obtained with a large predominance of the [beta]-disulfide, owing to the
preferential generation of a 1,2-trans product in the anomeric
thio-functionalization step [21]. The catalytic role played by phenyl
diselenide suggests that the S-Se intermediate 10 is amenable to
nucleophilic attack by the glycosyl thiolate as this latter is gradually
generated in situ from the thiouronium precursor. The phenylselenolate
leaving group is then quickly re-oxidized under air to the corresponding
diselenide useful in turn for the regeneration of the S-Se reactive
intermediate 9. Interestingly, this mechanism presents some similarities
with that of some selenoenzymes promoting the synthesis of disulfides
[27].
The S-Se intermediates like 10 turned out to be useful
thioglycosylation agents in the presence of N-bromosuccinimide (NBS), to
yield 3-thioglycosylated derivatives of DAI (5,6-di-acetoxy indole) [25]
(Scheme 4). De-O-acetylation and oxidative polymerization of these
products led to a novel class of glycoconjugates, the water-soluble
thioglycosylated eumlanin polymers that offer a useful model for
investigating the properties of natural eumelanins. Indeed, these latter
are important biopolymers because thry are responsible for the black
color of human skin, hair, and eyes, but their investigation is very
difficult because of the insolubility in any solvent.
[ILLUSTRATION OMITTED]
Another synthetically useful application of crude glycosyl iodides
points to their possible role as glycosyl donors. Indeed, several
contributions by other groups have emphasized that glycosyl iodides can
be activated by stoichiometric promoters, including applications on
demanding oligosaccharide targets [28]. Alternatively, glycosyl iodides
can be activated by halide [29] and phosphine oxide [30] catalysis, or
undergo a direct substitution process with suitable nucleophiles [31].
The crude iodides generated from the [I.sub.2]/[Et.sub.3]SiH system
proved reactive enough for the selective glycosidation of the phenol
position of 17-[beta]-estradiol under basic two-phase conditions (Scheme
5) [17]. Yields were critically dependent on the nature of the
saccharidic precursors, owing to a competitive process elimination
providing 1,2-acetoxy glycals. In particular, best results were obtained
from galacosyl and N-Troc aminoglucosyl iodides as the donors.
[ILLUSTRATION OMITTED]
The rapid synthesis of per-O-acetylated glycosyl iodides recently
offered another interesting opportunity for the fast access to useful
saccharidic building blocks. In particular, the activation of glycosyl
iodides with sub-stoichiometric amounts of Bi[Br.sub.3] in the presence
of allyl alcohol resulted in the generation of the corresponding allyl
glycoside along with a product allylated at the anomeric position and
free at O-2 (Scheme 6) [32]. The latter product often predominated, and
therefore the procedure was useful for the quick generation of
saccharidic building blocks bearing a synthetically versatile allyl
group at the anomeric position and a free alcoholic functionality at
C-2, the other positions maintaining the initial protection.
Interestingly, all of the allyl glycosides thus obtained (either fully
protected or free at O-2) were generally characterized as anomeric
mixtures. Yields of the 2-O-deprotetcted allyl glycosides were strongly
dependent on the nature of the precursor; synthetically useful results
were recorded from gluco-, galacto-, lacto-configured precursors,
whereas disappointing results were obtained from mannose and 6-deoxy
sugar precursors (these latter essentially yielded fully protected allyl
glycosides). Some mechanistic experiment indicated that the reaction
might proceed via the slow generation of a 1,2-orthoester intermediate
that then evolves to the allyl glycosides. In addition, there is
evidence that the allyl moiety initially incorporated into the
orthoester intermediate is not necessarily transferred to the anomeric
position of the same molecule.
[ILLUSTRATION OMITTED]
Other applications in protecting groups chemistry
Besides the application of glycosyl iodide intermediates in the
streamlined access to a very wide range of useful building blocks and
glycoconjugates, the [I.sub.2]/[Et.sub.3]SiH system proved very useful
in other deprotection procedures. Some years ago, the combination of
[I.sub.2] (ca. 0.4 equiv) and [Et.sub.3]SiH (0.05 equiv) in MeOH was
employed to remove the benzylidene protecting group from methyl
2,3-di-O-acetyl-4,6-O-benzylidene [alpha]-glucopyranoside, minimizing
the undesired processes of de-O-acetylation and acetyl migration
observed with other hydrolytic systems [33].
More recently, an especially relevant application was disclosed
with the regioselective deprotection of highly O-benzylated substrates
[34]. Exposure of a range of 2,3,4,6-tetra-O-benzylated glycosides to a
slight excess of both reagents in dichloromethane at low temperature
resulted in the preferential benzyl removal at position 4 (Scheme 7).
Regioselectivity was especially high when starting from galacto
precursors 12 and 13, whereas the corresponding gluco- and
manno-precursors (14 and 15, respectively) yielded minor amounts of the
inseparable 3-OH regioisomer. Very interestingly, the versatile allyl
protecting group survived unaffected under these conditions.
Very remarkably, application of the procedure to per-O-benzylated
disaccharides provided especially high yields of a single mono
de-O-benzylated product (Scheme 8), despite the larger number of
potential competitive events. Inspection of the results reveals a
well-defined regioselectivity trend for all the examined substrates with
the deprotection occurring on the reducing saccharidic terminus at a
secondary carbinol site flanking the glycosidation attachment site.
Interestingly, in the case of the 3-O-linked mannose disaccharide
(Scheme 8, last example), the axially oriented benzyl group at O-2 is
preferentially removed.
[ILLUSTRATION OMITTED]
[ILLUSTRATION OMITTED]
Collectively, all data highlight the unusual regioselectivity of
this procedure, which favors the removal of especially encumbered benzyl
groups. In addition, some mechanistic investigation indicated HI (and
not [Et.sub.3]SiI) as the actual promoter of the deprotection. Thus, the
reaction likely occurs via an initial fast (and reversible) HI-promoted
protonation of the benzyloxy group followed by the nucleophilic attack
of the nearby iodide anion to the benzyloxonium site. The resulting
transfer of the benzyl group (with formation of benzyl iodide as the
byproduct) might be kinetically favored with a crowded neighborhood
owing to the release of steric strain.
Another interesting result of this procedure was observed on
anomerically reactive substrates such as 6-deoxy sugar glycosides or
tetra-O-benzylated [beta]-hexopyranosides. In these cases, the anomeric
iodination was a faster process than de-O-benzylation at any site, and
then a higher amount of both [I.sub.2] and [Et.sub.3]SiH was needed for
achieving a concomitant deprotection (Scheme 9). The in situ generated
partially protected glycosyl iodides were found too reactive to be
isolated; nonetheless, they could be easily converted into useful and
workable building blocks under suitable quenching conditions (Scheme 9).
As shown in examples of Scheme 9, the initial allyl aglycon can be
either re-attached or even replaced by an acetyl group in dependence on
the quenching procedure. Also relevant is the preferential deprotection
at O-3 for highly O-benzylated 6-deoxy sugar derivatives rather than at
O-4 (see Scheme 7). In addition to the wide scope of the procedure for a
well-differentiated set of saccharidic precursors, the protocol lends
itself to incorporation into sequential one-pot schemes. An interesting
application of this idea was represented by the synthesis of a Lewis X
mimic starting from fully protected precursors, as illustrated in Scheme
10.
[ILLUSTRATION OMITTED]
[ILLUSTRATION OMITTED]
CONCLUSION
In this communication, we have shown the multiple synthetic
opportunities offered by the [I.sub.2]/[Et.sub.3]SiH combined system in
carbohydrate chemistry. The reactivity of this reagent is mainly
associated with the easy and fast generation in situ of HI. Easy
manipulation and storage of these reagents as well as the operational
simplicity of many of the developed protocols are expected to elicit in
the future new useful procedures for the streamlined preparation of
valuable "building blocks". More in general, the reagent is
expected to offer new opportunities in organic synthesis, as also
suggested by very recent contributions in the literature.
http://dx.doi.org/ 10.1351/PAC-CON-11-08-04
ACKNOWLEDGMENTS
We thank all coauthors in the references who contributed to
expanding the scope of the title reagent. We acknowledge the support
from the Italian Ministry of University (MIUR), the University of Naples
"Federico II" ("Programma Faro"), and facilities of
the "Centro Interdipartimentale di Metodologie Chimico-Fisiche
dell'Universita di Napoli".
REFERENCES
[1.] X. Zhu, R.R. Schmidt. Angew. Chem., Int. Ed. 48, 1900 (2009).
[2.] For significant examples, see: (a) C.-C. Wang, J.-C. Lee,
S.-Y. Luo, S.S. Kulkarni, Y.-W. Huang, C.-C. Lee, K.-L. Chang, S.-C.
Hung. Nature 446, 896 (2007); (b) A. Francais, D. Urban, J.-M. Beau.
Angew. Chem., Int. Ed. 46, 8662 (2007); (c) C.-C. Wang, S.S. Kulkarni,
J.-C. Lee, S.-Y. Luo, S.-C. Hung. Nat. Protocols 3, 97 (2008); (d) Y.
Bourdreux, A. Lemetais, D. Urban, J.-M. Beau. Chem. Commun. 47, 2146
(2011).
[3.] D.R. Deans, C. Eaborn. J. Chem. Soc. 3169 (1954).
[4.] P.J. Campos, B. Garcia, M.A. Rodriguez. Tetrahedron Lett. 43,
6111 (2002).
[5.] H.H.A.M. Hassan, A.H.F. El-Husseiny. Pol. J. Chem. 75, 803
(2001).
[6.] J.S. Yadav, S.B.V. Reddy, K. Premalatha, T. Swamy. Tetrahedron
Lett. 46, 2687 (2005).
[7.] B.V. Subba Reddy, B. Brahma Reddy, K.V. Raghavendra Rao, J.S.
Yadav. Tetrahedron Lett. 51, 5697 (2010).
[8.] R. Panchadhayee, A.K. Misra. Synlett 1193 (2010).
[9.] S.R. Jayaraman, M. Sridharan, R. Nagappan.
"3-Methyl-4,5,6,7-tetrahydro-1-benzothiophene-2 carboxylic
acid", Molbank M648 (2010).
[10.] K.P. Ravindranathan Kharta, T.S. Karkkainen, S.J. Marsh, R.A.
Field. Synlett 260 (2001).
[11.] M. Adinolfi, G. Barone, A. Iadonisi, M. Schiattarella.
Synlett 269 (2002).
[12.] B. Yu, H. Tao. Tetrahedron Lett. 42, 2405 (2001).
[13.] B. Yu, J. Sun. Chem. Commun. 4668 (2010).
[14.] (a) M. Adinolfi, G. Barone, A. Iadonisi, L. Mangoni, M.
Schiattarella. Tetrahedron Lett. 42, 5967 (2001) ; (b) M. Adinolfi, G.
Barone, A. Iadonisi, M. Schiattarella. Tetrahedron Lett. 43, 5573 (2002)
; (c) M. Adinolfi, G. Barone, A. Iadonisi, M. Schiattarella. Org. Lett.
5, 987 (2003); (d) M. Adinolfi, A. Iadonisi, A. Ravida , M.
Schiattarella. J. Org. Chem. 70, 5316 (2005).
[15.] (a) H. Tanaka, A. Yoshizawa, T. Takahashi. Angew. Chem., Int.
Ed. 46, 2505 (2007); (b) H. Tanaka, A. Yoshizawa, S. Chijiwa, J. Ueda,
M. Takagi, K. Shin-Ya, T. Takahashi. Chem. Asian J. 4, 1114 (2009); (c)
H. Tanaka, S. Yamaguchi, A. Yoshizawa, M. Takagi, K. Shin-ya, T.
Takahashi. Chem. Asian J. 5, 1407 (2010).
[16.] M. Adinolfi, A. Iadonisi, A. Ravida, M. Schiattarella.
Tetrahedron Lett. 44, 7863 (2003).
[17.] M. Adinolfi, A. Iadonisi, A. Pezzella, A. Ravida. Synlett
1848 (2005).
[18.] J. Gervay, T. N. Nguyen, M. J. Hadd. Carbohydr. Res. 300, 119
(1997).
[19.] S.M. Chervin, P. Abada, M. Koreeda. Org. Lett. 2, 369 (2000).
[20.] (a) R.M. van Well, K.P.R. Kartha, R.A. Field. J. Carbohydr.
Chem. 24, 463 (2005); (b) B. Mukhopadhyay, K.P.R. Kartha, D.A. Russell,
R.A. Field. J. Org. Chem. 69, 7758 (2004).
[21.] S. Valerio, A. Iadonisi, M. Adinolfi, A. Ravida. J. Org.
Chem. 72, 6097 (2007).
[22.] D. Comegna, F. De Riccardis. Org. Lett. 11, 3898 (2009).
[23.] D. Comegna, E. Bedini, M. Parrilli. Tetrahedron 64, 3381
(2008).
[24.] D.P. Gamblin, P. Garnier, S. van Kasteren, N.J. Oldham, A.J.
Fairbanks, B.G. Davis. Angew. Chem., Int. Ed. 43, 828 (2004).
[25.] A. Pezzella, A. Iadonisi, S. Valerio, L. Panzella, A.
Napolitano, M. Adinolfi, M. d'Ischia. J. Am. Chem. Soc. 131, 15270
(2009).
[26.] M. Adinolfi, D. Capasso, S. Di Gaetano, A. Iadonisi, L.
Leone, A. Pastore. Org. Biomol. Chem. 9, 6278 (2011).
[27.] X. Huang, X. Liu, Q. Luo, J. Liu, J. Shen. Chem. Soc. Rev.
40, 1171 (2011).
[28.] (a) S.N. Lam, J. Gervay-Hague. J. Org. Chem. 70, 8772 (2005);
(b) J.R. Harding, C.D. King, J.A. Perrie, D. Sinnott, A.V. Stachulski.
Org. Biomol. Chem. 3, 1501 (2005).
[29.] (a) M.J. Hadd, J. Gervay. Carbohydr. Res. 320, 61 (1999); (b)
S.N. Lam, J. Gervay-Hague. Org. Lett. 4, 2039 (2002); (c) S.N. Lam, J.
Gervay-Hague. Carbohydr. Res. 337, 1953 (2002); (d) N. Miquel, S.
Vignando, G. Russo, L. Lay. Synlett 341 (2004); (e) W. Du, J.
Gervay-Hague. Org. Lett. 7, 2063 (2005); (f) W. Du, S.S. Kulkarni, J.
Gervay-Hague. Chem. Commun. 2336 (2007); (g) J.A. Morales-Serna, Y.
Diaz, I. Matheu, S. Castillon. Eur. J. Org. Chem. 3849 (2009); (h) M.
Schombs, F.E. Park, W. Du, S.S. Kulkarni, J. Gervay-Hague. J. Org. Chem.
75, 4891 (2010); (i) J.A. Morales-Serna, O. Boutureira, A. Serra, I.
Matheu, Y. Diaz, S. Castillon. Eur. J. Org. Chem. 2657 (2010).
[30.] (a) Y. Kobashi, T. Mukaiyama. Chem. Lett. 33, 10 (2004); (b)
Y. Kobashi, T. Mukaiyama. Chem. Lett. 33, 874 (2004).
[31.] (a) J. Gervay, M.J. Hadd. J. Org. Chem. 62, 6961 (1997); (b)
S. N. Lam, J. Gervay-Hague. Org. Lett. 5, 4219 (2003).
[32.] A. Pastore, M. Adinolfi, A. Iadonisi. Eur. J. Org. Chem. 6206
(2008).
[33.] M. Adinolfi, L. De Napoli, G. Di Fabio, A. Iadonisi, D.
Montesarchio, G. Piccialli. Tetrahedron 58, 6697 (2002).
[34.] A. Pastore, S. Valerio, M. Adinolfi, A. Iadonisi. Chem.--Eur.
J. 17, 5881 (2011).
Matteo Adinolfi, Alfonso Iadonisi ([double dagger]), Antonello
Pastore, and Silvia Valerio
Department of Organic Chemistry and Biochemistry, University of
Naples "Federico II", Via Cinthia 4, I-80126, Naples, Italy
* Pure Appl. Chem. 84, 1-106 (2012). A collection of invited papers
based on presentations at the 16th European Carbohydrate Symposium
(Eurocarb-16), Sorrento, Italy, 3-7 July 2011.
([double dagger]) Corresponding author