Fernando Fernández1, Alberto G Fernández1, Rosalino Balo1, Víctor M Sánchez-Pedregal2, Miriam Royo3,4, Raquel G Soengas5, Ramón J Estévez1,2, Juan C Estévez1,2. 1. Centro Singular de Investigación en Química Biolóxica e Materiais Moleculares (CIQUS), Universidade de Santiago de Compostela, c/Jenaro de la Fuente s/n, 15782 Santiago de Compostela, Spain. 2. Departamento de Química Orgánica, Universidade de Santiago de Compostela, Avda. das Ciencias s/n, 15782 Santiago de Compostela, Spain. 3. Centro de Investigación Biomédica en Red Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), c/ Jordi Girona 18-26, 08034 Barcelona, Spain. 4. Instituto de Química Avanzada de Cataluña (IQAC-CSIC), c/ Jordi Girona 18-26, 08034 Barcelona, Spain. 5. Departamento de Química Orgánica e Inorgánica, Universidad de Oviedo, c/ Julián Clavería s/n, 33006 Oviedo, Spain.
Abstract
A stereoselective synthesis of polyhydroxylated cyclopentane β-amino acids from hexoses is reported. The reaction sequence comprises, as key steps, ring-closing metathesis of a polysubstituted diene intermediate followed by the stereoselective aza-Michael functionalization of the resulting cyclopent-1-ene-1-carboxylic acid ester. Examples of synthesis of polysubstituted 2-aminocyclopentanecarboxylic acid derivatives starting from protected d-mannose and d-galactose are presented. A general protocol for the incorporation of these highly functionalized alicyclic β-amino acids into peptides is also reported.
A stereoselective synthesis of polyhydroxylated cyclopentane β-amino acids from hexoses is reported. The reaction sequence comprises, as key steps, ring-closing metathesis of a polysubstituted diene intermediate followed by the stereoselective aza-Michael functionalization of the resulting cyclopent-1-ene-1-carboxylic acid ester. Examples of synthesis of polysubstituted 2-aminocyclopentanecarboxylic acid derivatives starting from protected d-mannose and d-galactose are presented. A general protocol for the incorporation of these highly functionalized alicyclic β-amino acids into peptides is also reported.
The enantioselective
synthesis of β-amino acids has received
great attention in recent times,[1−5] mainly because peptidomimetics[6] based
on these amino acids may overcome the pharmacological limitations
of natural peptides.[7−10] They are more resistant than α-peptides to protease and peptidase
degradation,[11−13] and their conformational properties and stability
facilitate their interaction with receptors and enzymes, which usually
results in improved activity and no side effects.[14] More recently, α,β-peptides have evidenced
promising applications in material sciences, mainly as nanomaterials.[15]Among the many β-amino acids that
have been studied, cyclopentane-based
β-amino acids are particularly attractive building blocks because
their peptides exhibit specific folding properties. For instance,
their homo-oligomers show a high propensity to fold in well-defined
secondary structures in short peptide sequences, a structural property
that often gives them enhanced biostability and activity.[16,17] Thus, oligomers that contain at least four units of trans-2-aminocyclopentanecarboxylic acids (trans-ACPC)
adopt a stable 12-helix with topological dimensions similar to those
of the α-helix in α-peptides,[18−20] while their cis-homo-oligomers adopt β-sheet secondary structures.[21] Homo-oligomers with alternating heterochiral cis-ACPC sequences form a 10/12 helix, while those with
alternating heterochiral trans-ACPC tend to attain
a polar-strand secondary structure in solution.[22] In contrast with their homo-oligomers, we demonstrated
that short peptides based on alternating trans-ACPC
and trans-2-aminocyclohexane adopt a 14-helix fold
in aqueous SDS solution but not in organic solvents.[23] Moreover, cis-ACPC can satisfactorily
replace prolines as inducers of β-turns in α-peptides.[24,25] Controlled self-assembly of helical homo-oligomers of trans-ACPC in the presence of surfactant gives rise to 3D nanostructures
of different shapes.[26,27] Accordingly, cyclopentane β-amino
acids proved to be ideal candidates for the stabilization of conformations
in peptides.The development of methodologies for the stereo-
and regioselective
synthesis of polysubstituted cyclopentane rings continues to be a
challenge in synthetic chemistry.[28−31] A specific goal of this significant
area of research is to increase the limited number of known polyhydroxylated
cyclopentane β-amino acids[3,32,33] that would enable access to a larger variety of hydro- or liposoluble
cyclopentane-based β-peptides. This latter goal can be achieved
by protection or deprotection of the hydroxyl substituents in polyhydroxylated
cyclopentane rings. In addition, it is feasible that these substituents
on the cyclopentane rings could result in novel folding properties
in β-peptides, which is a matter of evident interest in materials
chemistry. Furthermore, polyhydroxylated cyclopentane β-amino
acids have potential as clinical drugs[33,34] and biological
tools.[14,35] Also, other molecules containing the polyhydroxylated
cyclopentane ring, like some 4-amino-5-(hydroxymethyl)-1,2,3-cyclopentanetriols,
have been described as potent glycosidase inhibitors.[36−38]The first reported polyhydroxylated cyclopentane β-amino
acid was the trans-2-aminocyclopentanecarboxylic
acid derivative 3a, which was obtained in our laboratory
by a novel approach involving the key stereocontrolled cyclization
of d-glucose nitrosugar derivative 1 to bicyclolactone 2 (Scheme ).[39,40] Amino acid 3a was converted
into its derivative 3b, which is suitably functionalized
for incorporation into peptides.[40] Applications
of this approach to l-idose nitrosugar derivative 4 provided the first polyhydroxylated cis-2-aminocyclopentanecarboxylic
acid 6 (Scheme ).[41] Nevertheless, this strategy
turned out unsuitable for preparing peptides based on these β-amino
acids due to the low global yields achieved for 3a (12%
yield, seven steps), 3b (8% yield, 10 steps), and 6 (15% yield, seven steps). Furthermore, the scope of this
synthetic strategy is relatively limited because it can provide direct
access to only eight polyhydroxylated cyclopentane β-amino acids,
i.e., only those arising from the eight hexoses that meet the stereochemical
requirements for the key intramolecular alkylation leading to bicyclic
lactones like 2 or 5 (i.e., d-glucose, d-idose, d-allose, d-talose, l-glucose, l-idose, l-allose and l-talose).[42]
Scheme 1
First Syntheses of Polyhydroxylated Cyclopentane
β-Amino Acids
Here, we report a
more general and efficient method for the stereocontrolled
synthesis of polyhydroxylated cyclopentane β-amino acids from
hexoses. This approach is, in principle, of general application to
all hexoses and, in consequence, should give access to a larger variety
of relative configurations of these β-amino acids. Starting
from a conveniently protected hexose, the strategy involves the ring-closing
metathesis (RCM)[43] reaction of a richly
functionalized diene intermediate A leading to cyclopentenol B (Scheme ), which is then transformed into cyclopentene carboxylic acid derivative C, followed by an aza-Michael amination[44] of the α,β-unsaturated carboxylic moiety to
give the target highly functionalized β-amino acid D.
Scheme 2
Outline of the Synthesis of Functionalized Cyclopentane β-Amino
Acids from Hexoses
P: protecting group.
Outline of the Synthesis of Functionalized Cyclopentane β-Amino
Acids from Hexoses
P: protecting group.In order to demonstrate the generality of the method,
we synthesized
protected polyhydroxylated cyclopentane β-amino acids starting
from two hexoses (d-mannose and d-galactose) that
cannot give access to them using the previous strategy via nitrosugars.
Specifically, starting from d-galactose, we synthesized the
derived cyclopentane β-amino acid with two alternative protecting
group schemes suitable for the incorporation into peptides. In one
case, we observed an unwanted elimination reaction when trying to
couple these β-amino acids into peptides as already described
in a previous work.[40] Finally, we devised
an alternative and more general procedure for the successful incorporation
of this type of amino acids into peptides.[45]
Results and Discussion
Synthesis of Polyhydroxylated Cyclopentane
β-Amino Acid
Derivative 12
In order to demonstrate the feasibility
of this strategy with hexoses other than those suitable for the already
described intramolecular nitronate cyclization strategy, we synthesized
polyhydroxylated cyclopentane β-amino acid derivative 12 from d-mannose (Scheme ). Selective protection of the primary hydroxyl
group of d-mannose derivative 7a(46) with TBDPS and oxidation of its C5 free hydroxyl
group with Dess–Martin reactive gave ketone 8. When 8 was submitted to Wittig reaction conditions,
a double olefination occurs, one at the ketone group and the other
one at the anomeric position, which spontaneously deacetylated in
the basic medium of the reaction to give the expected diolefin 9a. Its free hydroxyl group was methylated, and then its silylether
was deprotected to give diolefin 9c, which is suitably
protected for the RCM reaction.
Scheme 3
Synthesis of Polyhydroxylated Cyclopentane
β-Amino acid Derivative 12
Synthesis of Polyhydroxylated Cyclopentane
β-Amino acid Derivative 12
Conditions: (i) TBDPSCl, imidazole,
CH2Cl2, rt., 15 min. 98%. (ii) Dess–Martin,
CH2Cl2, rt., 2 h, 88%. (iii) n-BuLi, Ph3PCH3Br, THF, −78 °C to
rt., 2 h, 83%. (iv) NaH, MeI, THF, 0 °C to rt., 4 h, 95%. (v)
TBAF, THF, rt., 2 h, 85%. (vi) Grubbs 1st, CH2Cl2, rt., 24 h, 90%. (vii) a: TEMPO, BAIB, NBu4I, CH2Cl2/H2O, rt., 2 h. b: NaClO2, NaH2PO4·2H2O, 2-methyl-2-butene, tBuOH/H2O, rt., 1 h. (viii) NaHCO3, MeI,
DMF, rt., 12 h, 85% (from 10). (ix) NH2Bn,
DMF, rt., 48 h, 91%.Cyclopentenol 10 was formed in 90% yield from 9c under standard RCM
reaction conditions using the first-generation
Grubbs catalyst. Then, oxidation of the primary hydroxyl group of 10 gave cyclopentenecarboxylic acid 11a. Reaction
of 11a with NaHCO3 and MeI furnished its methyl
ester derivative 11b in 85% yield for the last three
steps. Finally, treatment of 11b with benzylamine resulted
in the expected stereoselective aza-Michael addition on the conjugated
double bond, which provided compound 12 in 91% yield.
The total yield for the transformation of 7a to 12 was 40% (nine steps). This yield is much higher than that
of the similar β-amino acid derivative 3b synthesized
from d-glucose by the nitrosugar strategy (8% yield, nine
steps).[40]
Synthesis of Polyhydroxylated
Cyclopentane β-Amino Acid
Derivative 19a
The satisfactory results of our
strategy for the transformation of d-mannose into β-amino
acid 12 prompted us to apply it to other hexoses, like
the transformation of d-galactose into β-amino acid 19a (Scheme ). The key reaction to build the cyclopentane ring of 19a was the RCM reaction of diolefin 16b, which was prepared
from the known d-galactose derivative 13a.[47] Olefination of the hemiacetal of 13a followed by the oxidation of the hydroxyl group of 14a gave ketone 15a, which was subjected to a second olefination
step to give diolefin 16a (Scheme ). Removal of the silylether group at the
C1 of 16a by treatment with TBAF gave the desired key
diolefin 16b. According to our synthetic plan, standard
RCM reaction conditions, using the second-generation Grubbs catalyst,
gave the expected cyclopentenol 17a in 89% yield. Oxidation
of this compound with TEMPO gave cyclopentenecarboxylic acid 18a through the spontaneous oxidation of the intermediate
aldehyde. Reaction of acid 18a with NaHCO3 and MeI furnished its methyl ester derivative 18b in
97% yield for the three last steps. The stereoselective aza-Michael
addition to the double bond of 18a was performed with p-methoxybenzylamine (PMBNH2), instead of benzylamine
(Scheme ), to enable
the selective deprotection of the amino group of 19a in
the presence of the OBn substituents. The total yield of the transformation
of 13a into 19a was 34% for the eight steps.
Scheme 4
Synthesis of Polyhydroxylated Cyclopentane β-Amino Acid Derivative 19a
Synthesis of
Polyhydroxylated Cyclopentane β-Amino Acid
Derivative 19b
Next, we devised a different
protection pattern for the same hexose that led to the β-amino
acid derivative 19b (Scheme ), which has its cis hydroxy
substituents protected with an isopropylidene substituent. This alternative
protecting scheme would open the possibility of selective deprotection
of chosen hydroxyl groups. Furthermore, this substitution pattern
allows us to compare the efficacy of this synthetic strategy with
the one previously reported by us, which had led to the enantiomer
(except for the protection of the N atom, Bn or Cbz) of 19b through a modification of the intramolecular nitronate cyclization
strategy.[48]
Scheme 5
Synthesis of Polyhydroxylated
Cyclopentane β-Amino Acid Derivative 19b
Synthesis of Polyhydroxylated
Cyclopentane β-Amino Acid Derivative 19b
Conditions: (i) NaH, MeI,
THF, 0 °C to rt., 4 h, 93%. (ii) NH4HCO2, Pd/C, MeOH, rt., 12 h, 86%. (iii) n-BuLi, Ph3PCH3Br, THF, −78 °C to rt., 12 h, 80%.
(iv) Dess–Martin, CH2Cl2, rt., 24 h,
81%. (v) n-BuLi, Ph3PCH3Br,
THF, −78 °C to rt., 2 h, 93%. (vi) TBAF, THF, rt., 1 h,
82%. (vii) Grubbs 1st, CH2Cl2, rt., 24 h, 92%.
(viii) a: TEMPO, BAIB, NBu4I, CH2Cl2/H2O, rt., 2 h. b: NaClO2, NaH2PO4·2H2O, 2-methyl-2-butene. (ix) NaHCO3, MeI, DMF, rt.,
12 h, 82% (from 18c). (x) NH2Bn, DMF, rt.,
24 h, 80%.Accordingly, reaction of d-galactose derivative 13b(49) with methyl iodide gave its O-methylated derivative 13c, which was then
converted into the anomeric mixture 13d and then into
the key diene 16d, via compounds 14b, 15b, and 16c (Scheme ), following the protocol leading to its
analog 16b (Scheme ). Next, diene 16d was subjected to standard
RCM reaction conditions to yield the desired cyclopentenol 17b in 92% yield. In contrast to cyclization of diene 16b, this reaction was effective using the first-generation Grubbs catalyst,
probably because the steric hindrance is now lower. Compound 17b was next converted into cyclopentene carboxylic acid 18c and then into its ester 18d. The stereoselective
aza-Michael addition of BnNH2 led to the cyclopentane β-amino
acid derivative 19b. This synthesis is noticeably more
efficient (24% yield from 13b to 19b, 10
steps) than the previously described synthesis of the enantiomer (except
for the protection of the N atom, Bn or Cbz) of 19b from
its nitrosugar precursor 1 (8% yield, nine steps).[48]It is worth comparing the yields of the
two critical steps (ring-closing
metathesis and aza-Michael addition) in the above-described synthetic
sequences (Schemes to 5). Although all these yields are reasonably
high (80–92%), an attempt to justify the differences can be
done. Regarding the RCM reaction, the more reactive second-generation
Grubbs catalyst was needed for the transformation 16b → 17a (Scheme ), i.e., with the galactose derivative with its hydroxyls
protected with benzyl groups. The reason cannot be the configuration
of the starting hexose as the mannose 9c (Scheme ) and galactose 16d (Scheme ) derivatives,
which have less bulky protecting groups, reacted equally well with
the less reactive first-generation Grubbs catalyst. It is unclear
if the ultimate reason is the steric hindrance of the relatively bulky
benzyl groups of 16b or if it is a consequence of the
restrained conformational flexibility of intermediates 9c and 16d due to the protection of their cis-diols as cyclic acetonides; perhaps this might place the double
bonds in a position more favorable for the reaction with the less
reactive first-generation Grubbs catalyst.Regarding the aza-Michael
step, the yields are similar for the
transformations of the galactose derivatives 18b → 19a (80%; Scheme ) and 18d → 19b (80%; Scheme ), while the yield
of the mannose derivatives 11b → 12 reaches 91% (Scheme ). The amine approximates the double bond from the side opposite
to the C3 −OR substituent in all cases. That face of the double
bond is more hindered in the galactose derivatives (Schemes and 5) than in the mannose derivative (Scheme ), and this could explain the difference
in yield.
Synthesis of Tripeptide 21
Next, to demonstrate
the usefulness of the orthogonally protected polyhydroxylated cyclopentane
β-amino acids synthesized, we studied the feasibility of their
incorporation into short peptide chains by peptide coupling reactions
(Schemes and 7). With this purpose, removal of the PMB-protecting
group of 19a with CAN gave the free amine intermediate 19c, which was directly reacted with (Boc)2O to
furnish the orthogonally protected β-amino acid ester 19d in 75% yield in the two steps (Scheme ). Hydrolysis of the methoxycarbonyl group
of compound 19d under mild basic conditions was followed
by treatment of the resulting carboxylic acid 19e with
HATU as activating reagent and then with glycine hydrochloride. Dipeptide 20a was isolated in 60% yield (two steps). The N-Boc group was easily cleaved with TFA, and the resulting amine 20b was reacted with Boc-Gly-OH upon activation with HATU.
This furnished tripeptide 21 in 25% yield from 19a (six steps).
Scheme 6
Incorporation of Polysubstituted Cyclopentane
β-Amino Acid 19a into Peptide 21
Conditions: (i) CAN, CH3CN/H2O, 0 °C to rt., 6 h, (ii) (Boc)2O,
NaHCO3, rt., 18 h, 75% (from 19a), (iii)
Ba(OH)2·8 H2O, THF/H2O, rt.,
1 h. (iv) HCl·HGly-OMe, HATU, DIEA, CH2Cl2, rt., 14 h 60% (from 19d). (v) TFA, THF, rt., 1 h.
(vi) Boc-Gly-OH, HATU, DIEA CH2Cl2, rt., 10
h, 55% (from 20a).
Scheme 7
Protocol for the
Incorporation of Polysubstituted Cyclopentane β-Amino
Acid 19b into Peptide 24
Incorporation
of 19b into peptides is more problematic, as hydrolysis
of its methyl ester in basic conditions is usually accompanied by
the beta elimination of the −OR substituent contiguous to the
carboxymethyl alpha position as we previously reported for two analogs
of the enantiomer of 19b.[40] The solution we devised here involves protecting the carboxylic
acid group as trimethylsilylethyl ester (18e; Scheme ) instead of the
methyl ester 18d shown in Scheme . This choice of protecting group is made
on intermediate 18c prior to the aza-Michael addition.
So, starting from carboxylic acid 18c, esterification
with trimethylsilylethanol provided the expected cyclopentenecarboxylic
acid ester 18e,[50] which furnished
β-amino acid derivative 19f when subjected to the
aza-Michael addition using benzylamine as the nucleophile. Hydrolysis
of this ester 19f under mild basic conditions with TBAF
resulted in carboxylic acid 19g, which was efficiently
transformed into tripeptide 23a by direct coupling with
dipeptide 22a.[51] Removal of
the N-benzyl-protecting group of 23a, by catalytic hydrogenation, provided its free amino derivative 23b, which gave pentapeptide 24 when reacted
with dipeptide 22b under the stated coupling conditions.
The overall yield from 18c was 14% (six steps).In conclusion, we present here a promising approach to the stereocontrolled
synthesis of highly complex cyclopentane β-amino acids. This
method is more general and efficient than the previously reported
alternative from nitrosugars as it could be extended, in principle,
to the pool of hexoses. To demonstrate the generality of the method,
we applied it to two different hexoses (d-mannose and d-galactose) and with two alternative protecting patterns in
the case of d-galactose. This allowed us to synthesize, in
the gram scale, three new β-amino acids (12, 19a, and 19b), which are orthogonally protected
for their incorporation into peptides. This method opens up opportunities
for a new access to 4-amino-5-(hydroxymethyl)-1,2,3-cyclopentanetriols
(potent glycosidase inhibitors) by reduction of the methoxycarbonyl
group to hydroxymethyl. Furthermore, we have demonstrated how to incorporate
these β-amino acids into peptide chains using classical procedures.
In the case of those amino acids that present problems by classical
methods, we have also developed an alternative procedure for their
incorporation into peptides. The availability of more richly functionalized
cyclopentane β-amino acids, like the ones shown here, would
expand the opportunities of designing a larger variety of hydro- or
liposoluble β-peptides. We continue working in the synthesis
of monomers and peptides containing hydroxylated groups as well as
studying their potential applications in biological chemistry, new
materials, and catalysis. As preliminary studies, our immediate plans
are directed toward the synthesis of amphiphilic β-peptides
of this nature as potential ice recrystallization inhibitors and gelling
agents, which are two issues of great present interest.
Experimental
Section
General Information
All nonaqueous reactions were carried
out under a positive atmosphere of argon in flame-dried glassware
unless otherwise stated. Air- and moisture-sensitive liquid reagents
were added by dry syringe or cannula. Anhydrous tetrahydrofuran (THF)
was freshly distilled from sodium/benzophenone under argon, and all
other solvents and reagents were used as obtained from commercial
sources without further purification unless stated. Flash chromatography
was performed using 60 Merck 230–400 mesh (flash, 0.04–0.063)
silica. Thin-layer chromatography (tlc) was carried out on aluminium-backed
sheets coated with 60 GF254 silica. Plates were developed using a
spray of 0.2% w/v cerium(IV) sulfate and 5% ammonium molybdate in
2 M sulfuric acid or in 5% w/v ninhydrin in methanol. 1H and 13C NMR spectra were recorded on Bruker DPX 250
(250 MHz for 1H and 62.5 MHz for 13C) and Varian
Mercury 300 (300 MHz for 1H and 75 MHz for 13C) spectrometers at room temperature unless otherwise stated. All
chemical shifts are quoted on the δ scale using residual solvent
as internal standard; s, d, t, q, m, and br designate singlet, doublet,
triplet, quadruplet, multiplet, and broad, respectively. Coupling
constants (J) are measured in Hz. Mass spectra were
recorded on a Micromass VG-Autospec spectrometer [by chemical ionization
(NH3, CI) or electrospray techniques, as stated]. Infrared
spectra were recorded on a FT-IR Mattson Cygnus-100 spectrometer.
Only the characteristic peaks are quoted (in units of cm–1); st, m, and br designate strong, medium, and broad, respectively.
All the spectra were measured in KBr unless stated. Optical rotations
were measured on a Jasco DIP-370 polarimeter with a path length of
0.5 dm and in a Na (589 nm) lamp. Concentrations are given in g/100
mL. Elemental analyses were carried out on a Carlo Erba EA 1108 analyzer.
Synthesis of Polyhydroxylated Cyclopentane β-Amino Acid
Derivative 12
To a solution of compound 10 (0.55 g, 2.72 mmol) in
CH2Cl2/H2O (19 mL, 3:1), NBu4I (0.05 g, 0.14 mmol), TEMPO (0.09 g, 0.54 mmol), and DAIB
(2.19 g, 6.81 mmol) were added. The mixture was stirred at room temperature
for 2 h, quenched with saturated aq. Na2S2O3 (35 mL), and extracted with EtOAc (30 mL). The organic layer
was dried (anhydrous Na2SO4) and concentrated
in vacuo. The crude product was dissolved in BuOH (13.6 mL) and 2-methyl-2-butene (2,02 mL, 19.05 mmol),
and a solution containing NaClO2 (0.40 g, 3.54 mmol, 80%)
and NaH2PO4·H2O (0.55 g, 3.54
mmol) in water (13.6 mL) was added. The mixture was stirred at room
temperature for 1 h, quenched with 10% aq. HCl (20 mL), and extracted
with EtOAc (20 mL). The organic layer was dried (anhydrous Na2SO4) and concentrated in vacuo. To a solution of
the resulting carboxylic acid in dry DMF (13.6 mL), NaHCO3 (0.43 g, 5.17 mmol) and MeI (0.42 mL, 6.81 mmol) were added. The
mixture was stirred at room temperature for 12 h. The reaction mixture
was diluted with NH4Cl (20 mL) and extracted with EtOAc
(20 mL). The organic layer was dried (anhydrous Na2SO4) and concentrated in vacuo. Flash column chromatography of
the crude (EtOAc/hexane 1:4) furnished compound 11b (0.53
g, 2.31 mmol, 85% yield from 10) as a colorless oil.
[α]D18 = −16.1 (c 1.5, CHCl3). 1H NMR (250 MHz, CDCl3, ppm): δ 1.36 (s, 3H), 1.38 (s, 3H), 3.51 (s, 3H), 3.80 (s,
3H), 4.54 (d, 1H, J5,4 = 1.6 Hz), 4.61
(d, 1H, J3,4 = 5.2 Hz), 5.43 (dd, 1H, J4,3 = 5.2 Hz, J4, = 1.6 Hz), 6.80 (s, 1H). 13C{1H} NMR (62.5
MHz, CDCl3, ppm): δ 25.1, 26.7, 51.2, 57.4, 81.9,
82.8, 86.9, 111.2, 136.4, 144.0, 163.0. IR (NaCl, cm–1): ν 1726 (st, C=O). MS (CI, m/z, %): 229 (7, [M + H]+), 197 (8), 186 (100). Anal. calc.
for C11H16O5: C, 57.89; H, 7.07.
Found: C, 57.73; H, 7.01.
(3aS,4R,6R,7R,7aS)-4-(((tert-Butyldimethylsilyl)oxy)methyl)-7-methoxy-2,2-dimethyltetrahydro-4H-[1,3]dioxolo[4,5-c]pyran-6-ol and (3aS,4R,6S,7R,7aS)-4-(((tert-butyldimethylsilyl)oxy)methyl)-7-methoxy-2,2-dimethyltetrahydro-4H-[1,3]dioxolo[4,5-c]pyran-6-ol (13d)
Pd/C (0.64 g, 10%) and NH4HCO2 (4.58 g, 72.54 mmol) were added sequentially over a deoxygenated
solution of 13c (3.18 g, 7.25 mmol) in MeOH (51 mL),
and the resulting suspension was refluxed for 12 h. The reaction was
then filtered through Celite and washed with MeOH, and the solution
was concentrated to dryness under a vacuum. The residue was dissolved
in EtOAc (50 mL) and washed with water (50 mL); the organic layer
was dried (anhydrous Na2SO4) and filtered, and
the solvent was removed under a vacuum. The obtained residue was submitted
to flash column chromatography (EtOAc/hexane 1:2) to give compounds 13d (2.17 g, 6.24 mmol, 86%) as a yellow oil. Proportion 2:1
(d.e. 33%). 1H NMR (250 MHz, CDCl3, ppm): δ
0.08 (s, 12H), 0.89 (s, 9H), 0.90 (s, 9H), 1.35 (s, 6H), 1.51 (s,
3H), 1.54 (s, 3H), 3.19–3.24 (m, 1H), 3.36–3.43(m, 2H),
3.54 (s, 3H), 3.58 (s, 3H), 3.75–3.87 (m, 5H), 4.11–4.28
(m, 5H), 4.33–4.39 (m, 1H), 4.65–4.69 (m, 1H), 5.25–5.29
(m, 1H). 13C{1H} NMR (62.5 MHz, CDCl3, ppm): δ – 6.0, −5.9, −5.8, 17.8, 17.9,
25.4, 25.5, 25.6, 25.7, 27.5, 27.6, 58.0, 59.3, 61.5, 61.8, 67.2,
72.3, 72.5, 72.6, 74.6, 78.2, 79.1, 82.9, 89.9, 95.7, 108.4, 109.1.
MS (CI, m/z, %): 349 (36, [M + H]+); 332
(63); 275 (100). IR (NaCl, cm–1): ν 3421 (br,
OH); 1105 (st, Si-O-C). Anal. Calc. for C16H32O6Si: C, 55.14; H, 9.26. Found: C, 55.27; H, 9.33.
CAN (4.19 g, 7.64 mmol) was added to a solution
of compound 19a (1.11 g, 1.91 mmol) in CH3CN/H2O (95.5 mL, 4:1) at 0 °C. The mixture was allowed
to warm up to room temperature and stirred for 6 h. The mixture was
quenched with saturated aq. Na2S2O3 (a few drops) and concentrated in vacuo. The crude product was dissolved
in dioxane (38.2 mL) and treated with (Boc)2O (2.08 g,
9.55 mmol) and saturated aq. NaHCO3 until basic pH was
reached. The mixture was stirred at room temperature for 18 h, diluted
with 10% aq. HCl (50 mL), and extracted with EtOAc (100 mL). The organic
layer was dried (anhydrous Na2SO4) and concentrated
to dryness under reduced pressure. The crude product was purified
by flash column chromatography (EtOAc/hexane 1:4) to give compound 19d (0.80 g, 1.43 mmol, 75% yield) as a yellowish oil. [α]D23 = +37.0 (c 1.6, CHCl3). 1H NMR (250 MHz, CDCl3, ppm): δ 1.41
(s, 9H), 3.16 (dd, 1H, J = 6.0, 4.1 Hz), 3.28 (dd,
1H, J = 7.8, 5.6 Hz), 3.72 (s, 3H), 3.80 (dd, 1H, J = 7.8, 7.1 Hz), 3.98 (dd, 1H, J = 5.6,
4.1 Hz), 4.18 (dd, 1H, J = 7.1, 6.0 Hz), 4.30–4.60
(m, 6H), 5.44 (br, 1H), 7.28–7.38 (m, 15H). 13C{1H} NMR (62.5 MHz, CDCl3, ppm): δ 29.0, 52.4,
54.2, 54.7, 72.0, 72.4, 72.7, 74.7, 80.0, 83.3, 84.8, 127.1, 127.2,
127.3, 127.4, 128.1, 128.2, 128.3, 137.0, 137.2, 138.7, 155.3, 174.2.
MS (CI, m/z, %): 562 (12, [M + H]+); 505 (49); 91 (100). IR (NaCl, cm–1): ν 3348 (br, NH), 1751 (st, C=O). Anal. calc. for C33H39NO7: C, 70.57; H, 7.00; N, 2.49.
Found: C, 70.37; H, 6.92; N, 2.62.
Dipeptide 20a
Ba(OH)2·8H2O (1.34 g, 4.26 mmol)
was added to a solution of compound 19d (0.80 g, 1.42
mmol) in a 1:2 THF/H2O mixture
(15 mL). The reaction was stirred at rt. for 1 h and then neutralized
with 50WX4-50 DOWEX resin, which was then filtered off and washed
with MeOH. The solvent was removed under vacuum on a rotary evaporator.
A solution of the resulting solid residue, HATU (0.57 g, 1.70 mmol),
and DIEA (0.72 mL, 4.26 mmol) in dry CH2Cl2 (10
mL) was stirred at rt. for 15 min. HCl-Gly-OMe (0.20 g, 1.56 mmol)
was then added, and the stirring was continued for 14 h. CH2Cl2 (15 mL) was added, the mixture was washed with 10%
aq. HCl (15 mL), and the organic layer was dried (anhydrous Na2SO4) and concentrated to dryness under a vacuum.
Column chromatography of the solid residue (EtOAc/hexane 1:1) led
to the isolation of dipeptide 20a (0.33 g, 0.53 mmol,
60% overall yield from compound 19d) as a colorless oil.
[α]D21 = +68.2 (c 1.5,
CHCl3). 1H NMR (250 MHz, CDCl3, ppm):
δ 1.37 (s, 9H), 3.18 (dd, 1H, J = 5.8, 4.0
Hz), 3.27 (dd, 1H, J3 = 7.8, 5.6 Hz),
3.61 (s, 3H), 3.77–3.96 (m, 2H), 4.03 (s, 2H), 4.13 (dd, 1H, J = 7.0, 5.8 Hz), 4.28–4.48 (m, 6H), 5.67 (br, 1H),
6.91 (br, 1H), 7.28–7.41 (m, 15H). 13C{1H} NMR (62.5 MHz, CDCl3, ppm): δ 29.3, 39.7, 52.6,
54.2, 55.0, 72.2, 72.5, 73.6, 74.7, 80.7, 83.4, 85.4, 127.4, 128.3,
128.5, 128.6, 137.4, 137.7, 138.7, 157.3, 169.3, 172.5. MS (CI, m/z, %): 619 (56, [M + H]+);
588 (64); 91 (100). Anal. calc. for C35H42N2O8: C, 67.94; H, 6.84; N, 4.53. Found: C, 68.12;
H, 7.01; N, 4.29.
Tripeptide 21
TFA (2
mL) in THF (5 mL)
was added to a solution of compound 20a (0.33 g, 0.53
mmol), and the mixture was stirred at rt. for 1 h. The solvent was
then coevaporated with toluene (3 × 2 mL) under a vacuum in a
rotary evaporator. HATU (0.21 g, 0.64 mmol) and DIEA (0.27 mL, 1.59
mmol) were added to a solution of Boc-Gly-OH (0.10 g, 0.58 mmol) in
dry CH2Cl2 (5 mL), and the mixture was stirred
at rt. for 15 m. A solution of the crude amine from the previous transformation
in CH2Cl2 (10 mL) was added, and the resulting
mixture was stirred at rt. for 10 h. The reaction mixture was washed
with 10% aq HCl (20 mL), and the organic layer was dried (anhydrous
Na2SO4), filtered, and concentrated to dryness
under a vacuum. Column chromatography of the solid residue (EtOAc)
provided pure tripeptide 21 (0.20 g, 0.30 mmol, 55% overall
yield from compound 20a) as a colorless oil. [α]D18 = +21.7 (c 1.1, CHCl3). 1H NMR (250 MHz, CDCl3, ppm): δ 1.37
(s, 9H), 3.12–3.14 (m, 1H), 3.20 (dd, 1H, J = 7.4, 5.1 Hz), 3.66 (s, 3H), 3.79–3.91 (m, 2H), 4.03–4.09
(m, 4H), 4.31–4.43 (m, 6H), 4.55 (br, 1H), 5.55 (br, 1H), 6.93
(br, 1H), 6.96 (br, 1H), 7.27–7.39 (m, 15H). 13C{1H} NMR (62.5 MHz, CDCl3, ppm): δ 29.0, 40.0,
42.5, 52.8, 55.3, 56.3, 72.0, 72.2, 72.6, 75.0, 81.1, 84.7, 85.3,
127.9, 128.5, 128.7, 128.9, 138.3, 138.6, 139.0, 156.9, 166.0, 169.8,
172.2. MS (CI, m/z, %): 676 (18);
569 (64); 91 (100). Anal. Calc. for C37H45N3O9: C, 65.76; H, 6.71; N, 6.22. Found: C, 65.59;
H, 6.49; N, 5.98.
Authors: Fredy Sussman; Víctor M Sánchez-Pedregal; Juan C Estévez; Rosalino Balo; Jesús Jiménez-Barbero; Ana Ardá; Ana Gimeno; Miriam Royo; M Carmen Villaverde; Ramón J Estévez Journal: Chemistry Date: 2018-06-29 Impact factor: 5.236
Authors: Xingxian Gu; Vijayalaxmi Gupta; Yan Yang; Jin-Yi Zhu; Erick J Carlson; Carolyn Kingsley; Joseph S Tash; Ernst Schönbrunn; Jon Hawkinson; Gunda I Georg Journal: ChemMedChem Date: 2017-11-24 Impact factor: 3.466