Two new fluorous photolabile-protecting groups (FNBC and FNB) and a new base-labile protecting group (FOC) for the masking of amines are reported. The protecting groups survive a wide range of common reaction conditions used in oligosaccharide synthesis and render the attached molecules amenable to fluorous solid-phase extraction (FSPE). A glycosyl acceptor containing the FNB group is shown to be useful in the synthesis of carbohydrates tagged with free deactivated secondary amines.
Two new fluorous photolabile-protecting groups (FNBC and FNB) and a new base-labile protecting group (FOC) for the masking of amines are reported. The protecting groups survive a wide range of common reaction conditions used in oligosaccharide synthesis and render the attached molecules amenable to fluorous solid-phase extraction (FSPE). A glycosyl acceptor containing the FNB group is shown to be useful in the synthesis of carbohydrates tagged with free deactivated secondary amines.
The synthesis of oligosaccharides
is complex yet often unavoidable given that the isolation of these
ubiquitous structures from natural sources, in pure form, is extremely
expensive, challenging, and time-consuming.[1] The use of soluble light fluorous tags (those that contain fluorine
content equal or less than 40% by molecular weight) has shown promise
in simplifying the process of oligosaccharide synthesis; the fluorous-tagged
carbohydrate structures can easily be purified from nonfluorous-tagged
side products and excess reagents by fluorous solid-phase extraction
(FSPE) before continuation of each synthesis cycle.[2]Given these advantages, a range of alcohol protecting
groups with
minimal fluorous content have been designed and applied to carbohydrate
synthesis.[3] Apart from hydroxyl groups,
amines are the next most common functional group found in glycans
that requires masking. The N-sulfate and N-acetyl groups found in natural products are too polar
or basic to undergo most protection or other reaction steps in a typical
oligosaccharide synthetic sequence and therefore require amine blocking.
However, relatively little work has been reported using fluorous variants
for nitrogen protection in the realm of carbohydrates. Herein we report
the design of a set of three new fluorous amine-protecting groups
and demonstrate their tolerance to various standard conditions used
in oligosaccharide synthesis.The most common amine-containing
sugars, glucosamine or galactosamine,
are often protected by 2,2,2-trichloroethyl chloroformate (TrocCl),
allyloxycarbonyl chloride (AllocCl), phthalic anhydride (Phth), and
trichloroacetyl chloride (TCACl) to form carbamates or amides when
neighboring group participation is needed in the glycosylation step.[4] After the required oligosaccharide is obtained,
the protecting groups are cleaved and then the freeamine is acetylated
to obtain the final N-acetyl or N-sulfate moiety. Trichloroacetyl chloride, one of the most common
protecting groups used to this end, is converted to an N-acetyl group in one step using AIBN and tributyltin hydride.[5] Unfortunately, this conversion step is often
incomplete when applied to higher oligosaccharides (e.g., hexamers)
leading to inseparable mixtures and, in the presence of uronic acids,
often leads to elimination products. Also, removal of tin byproducts
can be extremely challenging. Phthalimido groups can pose challenges
in base-catalyzed reactions,[6] while an
Alloc group prohibits the use of N-iodosuccinimide
and related halonium ion-mediated activation strategies of thioglycosides.In light of these limitations, fluorous protecting group versions
that provide the added benefit of using the masking group for purification
are of particular interest. To date, the synthesis and use of a fluorousalkyl carbamate-type (Troc-type) nitrogen-protecting group has been
described for the protection and purification of amino sugars.[7a] However, this protecting group synthesis involves
the use of toxic chemcals such as mercury acetate and phosgene, thereby
limiting its practical use to small scales.Our first goal then
was to design a new carbamate-type amine-protecting
group that could be made from nontoxic reagents amenable to scale
up. A fluorous alkyl carbamate-type protecting group was chosen because
alkyl carbamates not only can be deprotected using mild conditions
like methyltrichlorosilane[8a] or trimethylsilyl
iodide[8b] but also are known to survive
most protection/deprotection conditions in carbohydrate synthesis
as well as to provide stereoselectvity during glycosylations through
neighboring group participation when placed next to the anomeric center.Imizadolinium carbamates are known for their use in the protection
of amines.[9] We hypothesized that attaching
a fluorous chain five bonds distant from the reactive center would
not alter the reactivity of this type of carbamate, and therefore,
a new fluorous version of this protecting group could be produced.
To this end, a fluorous protecting group 2, with imidazole
as a leaving group, was synthesized in one step starting from commercially
available perfluorooctylpropanol 1 (Scheme 1). Unfortunately, reagent 2 was found to efficiently
protect primary amino alcohols, but not glucosamine, even at elevated
temperatures.
Scheme 1
Synthesis of FOC-imidazolinium Triflate 3
We envisioned that due to the
relatively lower nucleophilicity
and steric hindrance of the secondary amine group in glucosamine compared
to the primary amino alcohol, a better leaving group than imidazole
was needed in compound 2. Fortunately, the addition of
a reaction step to our initial design to produce such an improved
leaving group was offset by the fact that, unlike previously reported
versions of fluorous protecting groups for amines,[7] the preparation of this new reagent did not require any
chromatographic separation steps. The resulting reagent 3 (FOC, 1-(((4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-heptadecafluoroundecyl)oxy)carbonyl)-3-methyl-1H-imidazol-3-ium trifluoromethanesulfonate) was found to
be stable at −20 °C for months under an argon atmosphere.
However, the FOC reagent 3 could be easily added to glucosamine,
and the resulting urethane was found to survive several common protection/deprotection
conditions used in oligosaccharide and other small molecule synthesis,
including basic and acidic conditions, as shown in Scheme 2. The group withstood a Lewis acid mediated glycosylation
reaction, standard ambient temperature Zemplén deacylation
conditions, and protic acid mediated benzylidine acetal formation
to eventually provide compound 6.[10] The free3-OH was then protected by a benzylchloromethyl
ether. Compounds 5–7 were all readily
purified by FSPE using the fluorophilic solvent (methanol) to successfully
elute the fluorous-tagged compounds from the C8F17-modified silica gel.
Scheme 2
Stability of the FOC Protecting Group under
Various Conditions Common
to Oligosaccharide Synthesis
Because the new protecting group showed good stability
under both
acidic and basic conditions, work next turned to finding suitable
conditions to remove the group. Several mild conditions known to remove
alkyl carbamates were initially attempted (using trimethylsilyl iodide
and methyl trichlorosilane respectively) to remove the FOC,[8] but none were successful in removing the carbamate.
After witnessing the stability of FOC under Zemplén conditions,
we realized a stronger basic medium was needed for a nucleophile-mediated
deprotection. Eventually, the freeamine was generated using 6 M aqueous
sodium hydroxide under reflux, which was subsequently acetylated to
provide compound 8 (Scheme 2).Our next objective was to design a fluorous amine-protecting group
amenable to the synthesis of amine-terminated oligosaccharides, commonly
used for attachment of these structures to arrays, and our first choice
was a photolabile group, owing to its convenient deprotection conditions.
Although o-nitrobenzyl carbamate groups are frequently
used for photolysis,[11] a fluorous version
of this type of protecting group amenable to oligosaccharide synthesis
is still unknown. The fluorous component of the acceptor will enable
us to synthesize various amine-terminated glycans in our automated
solution-phase synthesis platform, where FSPE is used for purification
of intermediates. Since upon photolysis the generated free primary
amine could recombine with the o-nitrosoaldehyde
side product,[12] we chose to refrain from
making an o-nitrobenzyl carbamate group. Instead
we decided to synthesize a fluorous o-nitrobenzylamine
with an additional Cbz (benzyl chloroformate) group on the amine to
prevent the lone pair of nitrogen from imine formation. With our design,
once the desired chain length is obtained, oligosaccharides with an
amine functional group at the reducing end protected with a Cbz group
could be made using UV light; the Cbz group would then be removed
during standard Pd-catalyzed hydrogenolysis of the benzyl ethers.[13] A fluorous protecting group FNB (fluorous o-nitrobenzaldehyde) was then synthesized in one step from
5-hydroxy-2-nitrobenzaldehyde that can be used to mask amines through
reductive amination. An acceptor 12 (Scheme 3) was made to probe the scope of photolysis of FNB
in the presence of the FOC group. Compound 13 containing
both protecting groups was synthesized from previously synthesized 4 and 12 and it was purified using gradation
FSPE.[3a] As expected, given that related
groups are known to be cleaved by light sources of wavelengths ranging
from 300 to 365 nm,[14] the Cbz-protected
amine could be generated using UV radiation at 330 nm to provide amine 14 (Scheme 4) in high yields while
keeping the FOC protecting group intact. Even without the FOC group,
the FNB-linker can render a molecule amenable to purification by FSPE
(Scheme S7, Supporting Information).
Scheme 3
Synthesis of FNB and a FNB-Protected Acceptor That Can Be Unmasked
with Light
Scheme 4
Photodeprotection
of FNB in the Presence of FOC
Since more complex sugars such as heparin contain differentially
modified amines, we next set out to test a fluorous protecting group
for amino sugars orthogonal to FOC that could be removed by photolysis.
Given the precedent of o-nitrobenzyl carbamates,
which are known to generate freeamines under UV light in a traceless
manner,[11] we set out to test the feasibility
of making a fluorous variant of this group, which we named FNBC (fluorous o-nitrobenzyl carbamate). In addition, a carbamate group
was chosen in order to aid in forming stereoselective glycosylation
products through anchimeric assistance.Although removal of
the o-nitrobenzyl carbamates
is known to be a lower yielding reaction due to imine formation with
the o-nitrosobenzaldehyde side product produced,[11d] we decided to use a primary amine in the reaction
mixture. We hypothesized that owing to the relatively low nucleophilicity
and steric hindrance of an amine group in aminosugars, imine formation
would be slower than with primary amines. This, in turn, could make
the deprotection step higher yielding.[11d]Ideally, a scheme to produce such a protecting group would
also
require minimal or no chromatographic separation steps. As before,
we chose to have a methylimidazolinium leaving group on the protecting
group. Starting with a nucleophilic substitution reaction, followed
by a sodium borohydride-mediated reduction, a nucleophilic addition–elimination
reaction with carbonyl diimidazole, and formation of a methylimidazolinium
salt, we successfully synthesized FNBC 17 (Scheme 5).
Scheme 5
Synthesis of FNBC-imidazolinium Triflate
We next tested the stability
of this new protecting group under
common reaction conditions used for oligosaccharide synthesis (Scheme 6). Starting from the peracetylated glucosamine hydrochloride,
we carried out carbamate formation, glycosylation, deacetylation,
benzylidene acetal formation, and ether formation reactions; the protecting
group was found to be stable to all these conditions. Unfortunately,
this deprotection step proved to be as low yielding as prior reports
of such light-mediated cleavage reactions on nonfluorous compounds
with the formation of insoluble materials; using a primary amine as
a competing nucleophile in the reaction did not improve the outcome.[11d] Resubjecting the desired product 8 to the deprotection conditions resulted in complete reisolation
of the material, thereby demonstrating that the final product is stable
under the reaction conditions. The loss of material appears to be
the result of the formation of insoluble byproducts. To probe whether
the complex carbohydrate substrate was contributing to the low yields,
an additional simpler FNBC-protected amino alcohol was synthesized
(Scheme S8, Supporting Information). In
this case also, the deprotection turned out to be low yielding with
formation of insoluble byproducts. Realistically, this parent FNBC
group, like its nonfluorous analogues, should be limited to the protection
of secondary deactivated amines rather than primary amines. Given
literature precedents,[15] a second-generation
FNBC-type protecting group better suited for primary amines could
be envisioned with a methyl substitution at the benzylic position.
This added methyl group should render the molecule more effective
in the deprotection reaction owing to the low electrophilicity of
the generated o-nitrosoacetophenone product.
Scheme 6
Stability of FNBC under Various Conditions Common to Oligosaccharide
Synthesis
In summary, three
new fluorous protecting groups for amine functional
groups have been designed and synthesized. These new masking groups
should find use not only in oligosaccharide synthesis but also in
the design of small molecules and in peptide synthesis where amine
functional groups are also prevalent. Although the use of FNBC as
a protecting group for aminosugars is limited owing to low yields
during deprotections, an efficient photolabile amine-protecting group
FNB has been developed to aid in the construction of linkers for carbohydrates
or other molecules. FNB was found to be orthogonal to FOC, and the
deprotection step was found to be high yielding unlike that of FNBC.
Work is ongoing to apply FNB in the synthesis of a library of N-glycans.
Authors: Zhen Wang; Zoeisha S Chinoy; Shailesh G Ambre; Wenjie Peng; Ryan McBride; Robert P de Vries; John Glushka; James C Paulson; Geert-Jan Boons Journal: Science Date: 2013-07-26 Impact factor: 47.728
Authors: Arnaud Gautier; Duy P Nguyen; Hrvoje Lusic; Wenlin An; Alexander Deiters; Jason W Chin Journal: J Am Chem Soc Date: 2010-03-31 Impact factor: 15.419