Synthesis and conformational analysis of cyclic homooligomers from pyranoid ε-sugar amino acids

DOI: 10.1002/chem.201303841

Full Paper

&

Macrocyclic Peptides

Synthesis and Conformational Analysis of Cyclic Homooligomers

from Pyranoid e-Sugar Amino Acids

Andrꢀs Feher-Voelger,^[a]Jorge Borges-Gonzꢁlez,^[a]Romen Carrillo,^[a]Ezequiel Q. Morales,^[a]

Javier Gonzꢁlez-Platas,^[b]and Tomꢁs Martꢂn*^{[a, c]}

Dedicated to Professor Victor S. Martꢀn on the occasion of his 60th birthday

Abstract: New pyranoid e-sugar amino acids were designed

as building blocks, in which the carboxylic acid and the

amine groups were placed in positions C2 and C3 with re-

spect to the tetrahydropyran oxygen atom. By using stan-

dard solution-phase coupling procedures, cyclic homo-

oligomers containing pyranoid e-sugar amino acids were

synthesized. Conformation analysis was performed by using

NMR spectroscopic experiments, FTIR spectroscopic studies,

X-ray analysis, and a theoretical conformation search. These

studies reveal that the presence of a methoxy group in the

position C4 of the pyran ring produces an important struc-

tural change in the cyclodipeptides. When the methoxy

groups are present, the structure collapses through interresi-

due hydrogen bonds between the oxygen atoms of the

pyran ring and the amide protons. However, when the cyclo-

dipeptide lacks the methoxy groups, a U-shape structure is

adopted, in which there is a hydrophilic concave face with

four oxygen atoms and two amide protons directed toward

the center of the cavity. Additionally, we found important

evidence of the key role played by weak electrostatic inter-

actions, such as the five-membered hydrogen-bonded pseu-

docycles (C₅) between the amide protons and the ether

oxygen atoms, in the conformation equilibrium of the mac-

rocycles and in the cyclization step of the cyclic tetrapepti-

des.

Introduction

of natural and synthetic cyclic peptides have been reported to

display remarkable biological activity, such as cyclosporine A,^[3]

antamanide,^[2]microcystin LR,^[4]octapeptidic somatostatin ana-

logues,^[5]and so forth. All those peptides, and actually any po-

tential drug to be evaluated, must be not only bioactive, but

also bioavailable, and therefore these drugs must reach the

therapeutic target once introduced into a living organism,

which usually implies going through the cellular membrane. In

fact, all the above-mentioned cyclic peptides display a good

absorption through the gastrointestinal tract. However, oddly

In the last few years, macrocyclic peptides have attracted con-

siderable attention as potential scaffolds in medicinal chemis-

try for the development of new drugs.^[1]Cyclization of the

linear precursors provides, on one hand, a valuable tool to in-

troduce steric and angular constraints into the macrocycles,

which restrict their conformational degrees of freedom and po-

tentially can impart higher target binding and selectivity; on

the other hand, cyclization also improves the resistance to pro-

teolytic degradation and oral bioavailability.^[2]A large number

[6]

enough, the “rule of five” stated by Lipinski et al. states that

most of the cyclic peptides should not be able to cross the

membrane due to a high number of hydrogen-bond donors

and acceptors and low theoretical octanol/water (o/w) parti-

tion coefficients (ClogP_o/w). This outcome should result in

a very low passive permeability (transcellular pathway) of the

cyclic peptides through the cellular membrane, and therefore

a scarce bioavailability. This apparent contradiction between

the Lipinski rules and experimental outcomes is due to particu-

lar three-dimensional arrangements of the cyclic peptides. In

general, cyclic peptides are not rigid compounds; instead, they

have a compromise between preorganization and structural

flexibility, which can increase membrane permeability provid-

ing that there are conformations that maximize the internal hy-

drogen bonds (folded conformers).^[7]In this way, in nonpolar

environments, the polar groups interact with each other to

form internal backbone hydrogen bonds and the nonpolar

[a] Dr. A. Feher-Voelger, J. Borges-Gonzꢁlez, Dr. R. Carrillo, Dr. E. Q. Morales,

Dr. T. Martꢀn

Instituto de Productos Naturales y Agrobiologꢀa-CSIC

Avda. Astrofꢀsico Francisco Sꢁnchez, 3

38206-La Laguna, Tenerife (Spain)

Fax: (+34)922-260135

E-mail: tmartin@ipna.csic.es

[b] Dr. J. Gonzꢁlez-Platas

Servicio de Difracciꢂn de Rayos X

Departamento de Fꢀsica Fundamental II, Universidad de La Laguna

Avda. Astrofꢀsico Francisco Sꢁnchez, 2

38206-La Laguna, Tenerife (Spain)

[c] Dr. T. Martꢀn

Instituto Universitario de Bio-Orgꢁnica “Antonio Gonzꢁlez”

Avda. Astrofꢀsico Francisco Sꢁnchez, 2

38206-La Laguna, Tenerife (Spain)

Supporting information for this article is available on the WWW under

http://dx.doi.org/10.1002/chem.201303841.

Chem. Eur. J. 2014, 20, 4007 – 4022

4007

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

groups would be pointing outward. In this regard, recent stud-

ies show that solvent shielding of the amide bonds by N-meth-

ylation plays an important role in imparting permeability to

the cyclic peptides.^[5,8]However, it is quite difficult to predict

the impact of N-methylation on the conformation equilibrium

of cyclic peptides. Therefore, the search for new strategies in

the conformational control of cyclic peptides is of significant

interest. The development of suitable models to understand

and modulate the formation of an internal network of hydro-

gen bonds is vital to understand the passive permeability of

cyclic peptides better and to develop future smart applications

of such an exciting prospect.

tion, with the NH of the amide group helping to create an in-

ternal backbone of hydrogen bonds in the cyclic peptides.^[15]

Finally, we introduce a methoxy group at C4 as an additional

appendix, which can act as a control element of the conforma-

tion equilibrium (Figure 1). Taking all these design features into

account, we describe herein the synthesis of cyclic homoo-

ligomers from pyranoid e-SAAs analogues. Additionally, a de-

tailed conformation analysis was made both in solution and

the solid state to understand and measure the effect of all the

above-mentioned conformational control factors.

With this aim in mind, sugar amino acid units (SAAs) were

considered as an optimal choice for the synthesis of model

cyclic peptide homooligomers to study if weak noncovalent in-

teractions can modulate the internal backbone of hydrogen

bonds. SAAs are hybrid scaffolds that combine elements from

carbohydrates and amino acids in their structure. There are

several advantageous features displayed by SAAs concerning

our above-mentioned purpose, among which it is worth men-

tioning a controllable and partially predictable conformational

restriction and the ability of SAAs to modulate chemical func-

tionality and stereochemistry precisely. Since these compounds

were first employed as useful peptide building blocks, diverse

types of furanoid^[9]and pyranoid^[10]a-, b-, g-, and d-SAAs have

been described as conformationally constrained scaffolds for

the formation of cyclic peptides.^[11]

Results and Discussion

Synthesis

Usually the synthetic strategies to obtain cyclic peptides re-

quire the use of orthogonally protected amino acid units.

Taking this approach into account, we chose tert-butyl ester

and azide groups as precursors to the acid and amine groups,

respectively. This strategy was previously described by Koert

and co-workers.^[15d]A commercially available starting material,

tri-O-acetyl-d-glucal, was used as the common precursor for

the synthesis of both analogues of the e-SAA units, that is,

compounds 5 and 6 (Scheme 1).

Previous studies in our group have shown that some deriva-

tives of the cis-2-oxymethyl-3-oxytetrahydropyran unit, al-

though quite flexible, display inherent conformational prefer-

ences.^[12]Particularly interesting are some macrooligolides

made from such a scaffold, modified to display an e-hydroxy

acid unit.^[13]Thus, it is reasonable to base the design of our

model cyclic peptide homooligomers on a e-SAA unit by

simply changing the hydroxy group for an amine group. Curi-

ously, there are scarce precedents that use e-SAAs in cyclic ho-

mooligomers,^[14]and in all these cases the amine and carboxyl-

ic acid groups were linked through the adjacent positions to

the oxygen atom in such a way that the oxygen atom of the

ring is part of the macrocycle. In our case, the carboxylic acid

and amine groups are placed at C2 and C3, thus leaving the

tetrahydropyran oxygen atom in a suitable position to act as

a weak hydrogen-bond acceptor, which can have a modulating

effect over other noncovalent interactions (Figure 1). Moreover,

the carboxylic acid and sugar moieties are linked through an

ether group, the oxygen atom of which might participate in

a five-membered hydrogen-bonded pseudocyclic (C₅) interac-

Scheme 1. Retrosynthetic analysis for the synthesis of analogues of the

e-SAA units 5 and 6.

Tri-O-acetyl-d-glucal was converted into the corresponding

triol 7^[16]and diol 8^[17]by using previously described proce-

dures. The benzylidenation of triol 7 followed by methylation

gave compound 9.^[18]With the first step, we achieved one im-

portant goal: the chemical differentiation between the two

secondary hydroxy groups. With the benzylidenacetal 9 in

hand, the next step was a regioselective reduction by using

the BH₃·THF complex and a catalytic amount of Cu(OTf)₂to

afford the secondary benzyl ether,^[19]which was submitted to

O-alkylation of the primary alcohol with tert-butyl 2-bromoace-

tate to obtain the tert-butyl ester 10.^[20]Cleavage of the benzyl

protecting group under hydrogenation conditions provided

the secondary alcohol, which was used to obtain the masked

e-SAA unit 5 by using a Mitsunobu reaction with diphenyl-

Figure 1. Structures of the cyclic peptides designed based on e-SAA units.

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4008

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Scheme 2. Synthesis of the masked e-SAA units 5 and 6. Reagents and con-

ditions: a) 1) BH₃·THF, Cu(OTf)₂, THF, RT (98%); 2) NaH, BrCH₂CO₂tBu, THF, RT

(89%); b) 1) H₂, Pd(C), MeOH, RT; 2) DIAD, DPPA, PPh₃, THF (95% over

2 steps); c) NaH, BrCH₂CO₂tBu, THF, 08C (60%); d) DIAD, DPPA, PPh₃, THF

(83%); e) 1) BzCl, Et₃N, CH₂Cl₂; 2) MsCl, Et₃N, CH₂Cl₂(92% over 2 steps);

f) NaN₃, DMF, 1008C (85%); g) 1) K₂CO₃, MeOH; 2) NaH, BrCH₂CO₂tBu, THF, RT

(75% over 2 steps). Bz=benzoyl, DIAD=diisopropyl azodicarboxylate,

DPPA=diphenyl phosphorazidate, Ms=methanesulfonyl.

Scheme 3. Divergent–convergent synthesis of linear precursors 16 and 22 of

the cyclic dipeptides and linear precursors 19 and 25 of the cyclic tetrapep-

tides. Reagents and conditions: a) TFA, CH₂Cl₂, RT; b) H₂, Pd(C), MeOH, RT;

c) EDCI, HOBt, DIPEA, CH₂Cl₂, RT. DIPEA=N,N-diisopropylethylamine,

EDCI=1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide hydrochloride,

HOBt=1-hydroxybenzotriazole.

phosphoryl azide (DPPA),^[21]which introduced an azide group

with complete inversion of the configuration (Scheme 2).

To carry out the synthesis of the masked e-SAA unit 6, a simi-

lar strategy was attempted. Thus, selective O-alkylation of the

diol 8 with tert-butyl 2-bromoacetate was performed to afford

alcohol 11 in moderate yield. Unfortunately, this approach

failed because the Mitsunobu reaction provided an equimolar

mixture of the desired compound 6 along with elimination

product, which were difficult to separate.^[22]As a result of this

outcome, a new approach was developed for the synthesis of

6. We selectively protected the primary alcohol of diol 8 as

a benzoate group, and then the secondary alcohol was con-

verted to give the mesylate compound 12, which was trans-

formed into azide 13 by nucleophile displacement with NaN₃.

Finally, deprotection of the benzoate and further O-alkylation

with tert-butyl 2-bromoacetate afforded the e-SAA unit 6 in

a good yield (Scheme 2).

nomer 6 and applying the same sequence of reactions de-

scribed above, the linear precursors 22 and 25 were obtained

(Scheme 3).

At this point of the synthesis, hydrogenation of the azide

group followed by acidic hydrolysis of the tert-butyl ester

moiety of all the linear precursors 16, 19, 22, and 25 yielded

the corresponding amino acids 26–29, respectively. Cyclization

of 26 to the cyclic dipeptide 1 proceeded under high-dilution

conditions (0.001m) in CH₂Cl₂with EDCI, HOBt, and DIPEA over

a period of 72 hours. The reaction of amino acid 27 under the

conditions described above furnished the cyclic tetrapeptide 3

in 43% yield (Scheme 4).

Macrolactamization of amino acid 28 worked better when

HBTU was used as a coupling agent, thus affording the cyclic

dipeptide 2 in moderate yield. Unfortunately, these experimen-

tal conditions failed when cyclization of amino acid 29 was

performed. We also explored the use of different coupling

agents (i.e., N,N-dicyclohexylcarbodiimide (DCC), EDCI, diethyl-

phosphoryl cyanide (DEPC), and benzotriazol-1-yloxytris(dime-

thylamino)phosphonium hexafluorophosphate (BOP)) and dif-

ferent experimental conditions (i.e., temperature and high-dilu-

tion conditions in CH₂Cl₂, DMF, and a mixture of both); howev-

er, all these attempts failed (see below), and we were unable

to obtain the cyclic tetrapeptide 4 (Scheme 4).

The cyclic homooligomers were synthesized from their linear

precursors through a head-to-tail cyclization. The linear dimers

and tetramers were prepared by using conventional solution-

phase peptide synthetic methods by means of a divergent–

convergent strategy. Thus, with the protected monomer 5 in

hand, acidic hydrolysis of the tert-butyl ester or hydrogenation

of the azide group gave the primary coupling partners carbox-

ylic acid 14 and amine 15, respectively. Standard peptide-bond

formation conditions by using EDCI with stoichiometric

amounts of HOBt and DIPEA were employed, thus affording

the pseudo-dipeptide 16 in good yield (Scheme 3). The

pseudo-dipeptide 16 served as a common intermediate to pre-

pare the carboxylic acid 17 and amine 18 required for the

pseudotetrapeptide synthesis. A portion of 16 was converted

into the free carboxylic dipeptide 17, and the rest of 16 was

converted into the unmasked dipeptide amine 18. Finally, cou-

pling of 17 and 18 by using the above-described procedure af-

forded the pseudo-tetrapeptide 19 in good yield. By using mo-

Conformational analysis

NMR and IR spectroscopic studies

The NMR spectra (¹H and ¹³C) of all the cyclic peptides dis-

played symmetric structures with only one set of resonances

from their constituent e-SAA units in all the solvents in which

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4009

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

drogen-bonded amide groups, thus yielding further evidence

of the presence of hydrogen-bonded conformations.

Conformational analysis of 1

The cyclic dimer 1 is a 14-membered macrocyclic peptide. The

¹H NMR spectrum of 1 recorded in CDCl₃, C₆D₆, [D₆]DMSO, and

D₂O at 300 K showed a symmetrical set of high-resolution sig-

nals, which is consistent with the macrocycle either having

a rigid conformation with C₂symmetry or a structure involved

in a fast conformation equilibrium. Chemical shifts of the

amide proton in CDCl₃(d_NH=7.65 ppm) remained constant as

the sample concentration was diluted from 25 to 1.0 mm,

which clearly suggested that NÀH bonds are involved in intra-

molecular hydrogen bonds. Also, the amide proton displays

a large coupling constant with a proton at C3 (³J_NHÀC3H

=

10.4 Hz), which corresponds to an antiperiplanar arrangement

(H-N-C3ÀHꢀ1808). COSY, HSQC, and NOESY experiments were

performed at 300 K to assign all the signals. The NOESY spec-

trum shows several well-resolved cross-peaks, most of which

are either intraresidue or sequential. Coupling constants and

NOE interactions between protons of the pyranose rings con-

firmed a chair conformation with substituents at the C2 and

C4 equatorial positions and a substituent at the C3 axial posi-

tion. Diastereotopic assignment of the prochiral methylene

Scheme 4. Synthesis of cyclic dipeptides 1 and 2 and cyclic tetrapeptides 3

and 4. Reagents and conditions: a) 1) H₂, Pd(C), MeOH, RT; 2) TFA, CH₂Cl₂, RT;

b) EDCI, HOBt, DIPEA, CH₂Cl₂, RT (38 and 43% for 1 and 3, respectively);

c) HBTU, DIPEA, CH₂Cl₂, RT (31%). HBTU=O-(benzotriazol-1-yl)-N,N,N’,N’-tet-

ramethyluronium hexafluorophosphate.

3

protons at C7 was carried out on the basis of the J_2,7coupling

constants and NOE interactions. Thus, the coupling constant

values were ³J_C2HÀC7HproR=10.7 and ³J_C2HÀC7HproS=4.6 in CDCl₃,

they were studied. The NMR

spectra of the cyclic products 2

and 3 in C₆D₆were better dis-

Table 1. ¹HNMR chemical shifts (d in ppm) and coupling constants (J in Hz) of cyclic peptide 1 in CDCl₃and

cyclic peptides 2 and 3 in C₆D₆.

persed and resolved than in

CDCl₃. Therefore, two-dimension-

al NMR spectroscopic experi-

ments for the cyclic peptides 2

and 3 (i.e., COSY, HSQC, HMBC,

and NOESY) were performed in

C₆D₆, and most of the spectral

parameters could be obtained

and are reported in Table 1.

Protons^[a]

1 (CDCl₃)

2 (C₆D₆)

3 (C₆D₆)

NH

C2ÀH

7.65 (d) (10.4)

7.72 (bd) (7.0)

7.22 (d) (10.1)

However, cyclodipeptide

1 in

3.69 (ddd) (10.7, 4.6, 2.92 (ddd) (4.6, 4.6, 1.5) 3.20 (ddd) (8.5, 2.4,

1.0)

4.53 (dd) (10.4, 3.8) 3.99 (m)

3.47 (ddd) (11.7, 4.9, 1.19 (dddd) (13.5, 13.5, 2.89 (ddd) (11.9, 4.5,

3.9)

–

1.74 (dddd) (13.2,

13.2, 11.8, 5.3)

1.87 (m)

3.61 (ddd) (12.7,

12.6, 2.6)

4.11 (ddd) (12.1, 5.2, 3.56 (dd) (11.2, 5.0)

1.2)

3.34 (s)

3.77 (dd) (10.3, 4.6) 3.33 (dd) (10.3, 4.0)

3.39 (dd) (10.7, 10.3) 2.99 (dd) (10.3, 5.0)

3.87 (d) (16.4)

4.26 (d) (16.4)

CDCl₃showed better resolution

in the NMR spectra than in C₆D₆.

Also, variable-temperature stud-

ies were carried out in CDCl₃in-

stead of C₆D₆to increase the

range of low temperatures.

These experiments were used to

measure the temperature coeffi-

cients (Dd/DT) of the amide-

proton chemical shifts, which

were used to determine their im-

plications in intramolecular hy-

drogen bonds. FTIR spectroscop-

ic analysis was performed in

CH₂Cl₂to differentiate the hydro-

gen-bonded from the non-hy-

1.6)

C3ÀH

4.50 (dd) (10.0, 4.0)

C4ÀH_ax

3.5, 3.5)

2.06 (bd) (13.5)

1.54 (ddddd) (13.7, 13.5, 1.67 (dddd) (13.0,

12.8, 4.5, 4.5)

0.82 (bdd) (13.7, 2.0)

2.97 (ddd) (12.7, 11.3,

2.5)

4.3)

–

C4ÀH_eq

C5ÀH_ax

12.9, 12.6, 5.1)

1.39 (dd) (13.0, 4.2)

3.03 (ddd) (12.9,

11.7, 2.4)

C5ÀH_eq

C6ÀH_ax

C6ÀH_eq

3.70 (dd) (11.7, 4.1)

OMe

–

3.22 (s)

C7ÀH_proS

C7ÀH_proR

C9ÀH_proS

C9ÀH_proR

3.51 (dd) (11.0, 2.5)

3.61 (dd) (11.0, 8.5)

3.88 (d) (15.8)

4.05 (d) (15.8)

3.69 (d) (14.7)

3.88 (d) (14.7)

[a] ax=axial, eq=equatorial.

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4010

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

which could not be used to calculate the rotamer populations

around the torsional angle (O1-C2-C7-O8) by applying the Ser-

ianni equation system,^[23]probably because the most stable

conformer is not totally staggered. However, the coupling-con-

stant values imply a great population of one conformer similar

to the gt staggered conformer (Figure 2). The presence of

FTIR spectroscopic analysis was performed with cyclic dimer

1 in CH₂Cl₂at different concentrations to obtain more informa-

tion about the hydrogen-bonding pattern. The IR spectrum of

1 in CH₂Cl₂displays a sharp amide A absorption (NÀH stretch-

ing mode) at n˜ =3336 cm^À1, amide I absorption (mainly C=O

stretching mode) at n˜ =1674 cm^À1, and amide II absorption

(mainly NÀH bending and CÀN stretching mode) at n˜ =

1533 cm^À1, which are independent of the sample concentra-

tion in the range 100–5 mm. These results confirm the forma-

tion of intramolecular hydrogen bonds.

Cyclodipeptide 1 yielded a single crystal suitable for X-ray

analysis,^[25]and the crystal structure strongly resembles the so-

lution-phase conformer. The tetrahydropyran rings displayed

a chair conformation with substituents in the C2 and C4 equa-

torial positions and a substituent in the C3 axial position. The

dihedral angle between the amide and C3 protons (H-N-C3ÀH)

is close to an antiperiplanar arrangement with an average

value of +1708. Additionally, the torsional angle (O1-C2-C7-O8)

with an average value of +468 confirms a not completely stag-

gered conformation similar to the gt conformer (Figure 2).^[26]

However, the most interesting feature of the crystal structure

emerged from the close contact established between the

amide proton of one residue and the pyran oxygen atom of

the other residue, thus forming an eight-membered hydrogen-

bonded pseudocycle (C₈), which helps to keep the center ring

collapsed and thereby generating a folded structure (Fig-

ure 4a). Moreover, this folded structure is also upheld by two

Figure 2. Rotamers around the C7ÀC2 bond.

a strong NOE interaction between the amide proton and

C5ÀH_axconfirms that the amide proton is situated above the

3

sugar ring, which is in agreement with the observed J_NHÀ

_C3Hvalue mentioned above (Figure 3).

Also, strong NOE interactions between the amide proton

and C7ÀH_proR, which could be intra- or interresidue, and be-

tween C3ÀH and C7ÀH_proSand between C3ÀH and C7ÀH_proR

,

confirm an analogue gt staggered rotamer around the C7ÀC2

bond. Additionally, the stereospecific assignment of C9ÀH_proS

and C9ÀH_proRwas accomplished from the strong NOE interac-

tions of C9ÀH_proSwith C7ÀH_proRand C7ÀH_proS(Figure 3).^[24]

Figure 3. a) Schematic representation of observed NOE interactions of cyclodipeptide 1 in CDCl₃. The solid arrows represent strong NOE interactions. Sequen-

tial NOE interactions have been omitted for clarity. b) Two-dimensional NOESY sections that show key intraresidue NOE interactions for cyclodipeptide 1 in

CDCl₃at 300 K.

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4011

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

from 25 to 1.0 mm, which suggested that NÀH bonds are in-

volved in intramolecular hydrogen bonding. However, the ob-

served ³J_NHÀC3Hvalue was 7.0 Hz, which corresponds to a torsion

angle of H-N-C3ÀHꢀ Æ1428. Therefore, the amide protons are

pointing to the pyran oxygen atoms of the same residue, thus

probably establishing C5Àtype electrostatic interactions as de-

scribed above. However, the presence of the methoxy group

at C4 in the cyclic peptide 1 competes with the pyran oxygen

atom through an interaction of the same type, thereby forcing

the amide protons to adopt an antiperiplanar arrangement

with respect to the proton at C3 (H-N-C3ÀHꢀ1808). Coupling

constants and NOE interactions between protons of the pyra-

nose rings confirmed a chair conformation with substituents in

the C2 and C4 equatorial positions and a substituent in the C3

axial position. Diastereotopic assignment of the prochiral

methylene protons at C7 was carried out on the basis of the

Figure 4. X-ray structure of cyclodipeptides 1 and 2.^[27]a) Side and top views

of cyclopeptide 1. b) Side and top views of cyclopeptide 2. The dashed lines

represent hydrogen bonds.

³J_2,7coupling constants and NOE interactions. Thus, the cou-

3

additional five-membered hydrogen-bonded pseudocycles (C₅)

between the amide protons and the ether oxygen atoms of

the 2-oxyacetamide groups. This type of hydrogen bonding

has been traditionally been considered to be a weak electro-

static interaction. However, there have been some reported ex-

amples in which these interactions play a key role in the con-

formation equilibrium and in the formation of secondary struc-

tures.^[15]These intramolecular hydrogen bonds explain perfect-

ly the results of the NMR and FTIR spectroscopic experiments.

The conformational change of cyclodipeptide 1 in a more

polar environment was checked by dissolving 1 in D₂O and

performing ¹H NMR, COSY, and NOESY experiments.^[24]The

¹H NMR spectrum of 1 in D₂O is similar to that observed in

CDCl₃. However, the most notable change was seen in the cou-

pling-constant values of ³J_C2HÀC7HproR=5.0 and J_C2HÀC7HproS

=

4.0 Hz were used to calculate the rotamer populations around

the torsional angle (O1-C2-C7-O8) by applying the Serianni

equation system.^[23]Thus, the calculated populations of the

three staggered conformers were 48, 35, and 17% for gg, gt,

and tg, respectively (Figure 2). In the two-dimensional NOESY

experiment, NOE cross-peaks between the amide proton in

C5ÀH_axand C7ÀH_proRwere observed. The former cross-peak

confirms that the amide proton is situated above the sugar

ring, and the latter peak indicates the contribution of the gt

conformer around the C2ÀC7 bond in the conformational pop-

ulation in solution. The stereospecific assignment of C9ÀH_proS

and C9ÀH_proRwas accomplished from strong NOE interactions

between C9ÀH_proSand C7ÀH_proRand between C9ÀH_proRand

C7ÀH_proS(Figure 5).

3

pling-constant values J_C2HÀC7HproR=8.1 and J_C2HÀC7HproS=4.5 Hz.

These parameters are quite relevant because they can be con-

sidered to be a probe for the degree of the molecular folding.

The rotamer populations around the torsional angle (O1-C2-

C7-O8) were calculated from the coupling-constant values by

applying the Serianni equation system,^[23]thus yielding the

population of the three staggered conformers of 12, 63, and

25% for gg, gt, and tg, respectively (Figure 2). By comparing

these results with those obtained in CDCl₃, it is clear that there

is an increase in the population of gg and tg conformers and

a decrease in the population of the gt conformer in D₂O. Bear-

ing in mind that the gg and tg conformers are representative

of the unfolded structure, it can be assumed that there is a sig-

nificant conformational change on going from an aqueous to

a lipophilic medium.^[28]However, the folded conformer gt re-

mains the most abundant in the aqueous medium.

The IR spectrum of cyclopeptide 2 in CH₂Cl₂displays two

sharp amide A absorptions (NÀH stretching mode) at n˜ =3425

(free) and 3331 cm^À1(bond), an amide I absorption at n˜ =

1671 cm^À1, and an amide II absorption at n˜ =1536 cm^À1, which

are independent of the sample concentration in the range

100–5 mm. These results confirm the formation of intramolecu-

lar hydrogen bonds. It is noteworthy that there was an absorp-

tion band at n˜ =3186 cm^À1, which is assigned to water in-

volved in the formation of hydrogen bonds.

The X-ray structure of cyclodipeptide 2 strongly resembles

one of the conformers found in the solution phase.^[25]The tet-

rahydropyrans rings display the chair conformation with sub-

stituents at the C2 and C4 equatorial positions and a substitu-

ent at the C3 axial position. The dihedral angle between the

amide and C3 protons (H-N-C3-H) is approximately equal to

À1528.^[26]Additionally, the crystal structure exclusively dis-

played the gg conformer around the C2ÀC7 bond (Figure 4b).

However, the most important feature is that cyclodipeptide 2

adopts a U-shaped structure, in which there is a hydrophilic

concave face with four oxygen atoms and two amide protons

directed toward the center of the cavity. In the center of the

cavity, a water molecule is complexed by two hydrogen bonds

established between a water oxygen atom and both amide

protons (Figure 4b). This U-shaped structure is further stabi-

lized by a trifurcated hydrogen bond between each amide

Conformational analysis of 2

The cyclic dimer 2 is a 14-membered macrocyclic peptide,

which lacks the methoxy groups at C4. However, this small

structural change implies large conformational changes. The

¹H NMR spectrum of 2 recorded in CDCl₃, C₆D₆, and D₂O at

300 K showed a symmetrical set of high-resolution signals. The

chemical shifts of the amide proton in C₆D₆(d_NH=7.72 ppm)

remained constant as the sample concentration was diluted

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4012

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Figure 5. a) Schematic representation of observed NOE interactions of cyclodipeptide 2 in C₆D₆. The solid arrows represent strong NOE interactions. Sequen-

tial NOE interactions have been omitted for clarity. b) Two-dimensional NOESY sections that show key intraresidue NOE interactions for cyclodipeptide 2 in

C₆D₆at 300 K.

proton and three hydrogen-bond acceptors: the pyran oxygen

atom (C₅) and O8 (C₆) both from the same residue and O8 of

the other residue (C₅). This complexation with water explains

the results obtained from the NMR and FTIR spectroscopic ex-

periments in the solution phase.

n˜ =3438 cm^À1and a chemical shift of the amide proton at d=

5.62 ppm, which corresponds to a non-hydrogen-bonded NÀH

bond. These results clearly indicate that model compounds 30

and 31 have a five-membered hydrogen-bonded pseudocycle

(C₅) between the amide protons and ether oxygen of the

2-methoxy acetamide group.^[30]Furthermore, the amide proton

of 30 displays a large coupling constant with the proton at C3

(³J_NHÀC3H=10.1 Hz), which corresponds to an antiperiplanar ar-

rangement (H-N-C3-Hꢀ Æ1728). This dihedral angle indicates

an arrangement of the amide proton almost identical to that

observed in the X-ray structure of cyclodipeptide 1 (Figure 6).

Conversely, amide 31 displays a coupling constant of the

Model acyclic compounds 30–32 were synthesized^[24]and

1

the FTIR and H NMR spectra were studied to confirm the im-

portant role of the methoxy group in the conformation equi-

librium observed. These compounds represent a structural

simplification of cyclic dipeptides 1 and 2. Additionally, N-iso-

propyl-2-methoxyacetamide was used as a model to study the

five-membered-ring hydrogen bond between the amide

proton and ether oxygen atom. The IR spectrum of this com-

pound displays one absorption maxima in the NÀH stretching

region at n˜ =3411 cm^À1in CH₂Cl₂(7 mm) and a chemical shift

of the amide proton at d_NH=6.31 ppm in CDCl₃at 7 mm.^[29]

Compound 30 displays one absorption maxima in the NÀH

stretching region similar to that observed in N-isopropyl-2-me-

thoxyacetamide (n˜ =3417 cm^À1in CH₂Cl₂at 7 mm) and a chemi-

cal shift of the amide proton (d_NH=6.74 ppm) in CDCl₃at the

same concentration (i.e., 7 mm). Similarly, compound 31 shows

one NÀH stretching maximum at n˜ =3412 cm^À1in CH₂Cl₂

(7 mm), and the NÀH signal was slightly shifted downfield with

respect to 30 (d_NH=6.98 ppm in CDCl₃at 7 mm). Nonetheless,

model compound 32 shows an NÀH stretching maximum at

3

proton at C3 of J_NHÀC3H=8.3 Hz, which corresponds to a dihe-

dral angle of H-N-C3-Hꢀ Æ1518. According to these values

and by taking into account the lack of the methoxy group at

C4, the amide proton is placed closer to the pyran oxygen

atom, thus favoring an additional C₅interaction^[15e,g](Figure 6).

Therefore, the results obtained from the acyclic models are in

perfect agreement with those obtained from the cyclic pep-

tides. These models also allow us to explain the role of the me-

thoxy group in the conformation of the cyclopeptides: When

there is no methoxy group, the amide protons form a bifurcat-

ed hydrogen bond (two C₅), thus forcing the structure to

remain unfolded, as in cyclodipeptide 2; on the contrary, the

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4013

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Conformational analysis of 3

The cyclic tetramer 3 is a 28-membered macrocyclic peptide.

1

In this case, the H NMR spectrum in C₆D₆also showed a sym-

metrical set of high-resolution signals, which could be due to

a rigid conformation with C₄symmetry, two interconvertible

structures with C₂symmetry, or a structure involved in a fast

conformation equilibrium on the NMR timescale. Coupling

constants indicated a chair conformation of the pyranose rings

with substituents at the C2 and C4 equatorial positions and

a substituent at the C3 axial position. The chemical shift of the

amide proton (d_NH=7.22 ppm in C₆D₆) suggested that the NÀH

bond could be involved in weak hydrogen bonds. Also, this

chemical shift remained constant as the sample concentration

was diluted in CDCl₃from 22.8 to 0.68 mm (d_NH=7.12 ppm),

thus suggesting there is no aggregation in CDCl₃and ruling

out the presence of intermolecular hydrogen bonds. Addition-

ally, the values of the temperature coefficient of the NÀH bond

in CDCl₃in a range from 249 to 300 K (Dd/DT=À3.0 ppb K^À1)

confirms the formation of intramolecular hydrogen bonds.^[31]

The amide proton exhibits a large coupling constant with the

proton at C3 (³J_NHÀC3H=10.1 Hz), which corresponds to an anti-

periplanar arrangement (H-N-C3-Hꢀ Æ1728). The coupling-

constant values of ³J_C2HÀC7HproR=8.5 and ³J_C2HÀC7HproS=2.5 Hz

were used to calculate the rotamer populations around the

torsional angle (O1-C2-C7-O8) by applying the Serianni equa-

tion system.^[23]Thus, the calculated populations of the three

staggered conformers were 22, 76, and 2% for gg, gt, and tg

(Figure 2). Due to the symmetry of the spectrum, it is not pos-

sible a priori to discriminate between the intra- and interresi-

due NOE interactions. In addition to the typical NOE interac-

tions from the pyranose chair conformation, there were other

strong NOE interactions between the amide proton and

C5ÀH_axand C7ÀH_proRand between C3ÀH and C7ÀH_proS

(Figure 7). The former result indicated that the amide proton is

positioned above the sugar ring, which is in agreement with

Figure 6. Structures, coupling constants (³J_NHÀC3H), dihedral angles (H-N-C3-

H), and chemical shifts of the amide proton in CDCl₃(7 mm) of the model

compounds 30–32 and N-isopropyl-2-methoxyacetamide.

methoxy group at C4 induces a change in the dihedral angle

(H-N-C3-H), thereby favoring the folding of cyclodipeptide 1.

On the other hand, the existence or otherwise of additional

interactions between the NH bond and pyran oxygen atom

could also help to explain why the macrocyclization reaction

failed to provide cyclic tetrapeptide 4, whereas this reaction

worked well to give macrotetrapeptide 3. The cyclization pre-

3

cursor 25 displayed coupling values of J_NHÀC3H=8.7, 9.1, and

9.3 Hz (dihedral angles: H-N-C3-HꢀÀ154–À1608), a chemical

shift of d_NH= >7.15 ppm in CDCl₃, and a major band at n˜ =

3415 cm^À1in the FTIR spectrum in CH₂Cl₂at 7 mm. These

values clearly indicate the formation of a bifurcated hydrogen

bond (2ꢄC₅) of the NH bond in each residue, in which two hy-

drogen-bond acceptors participate, namely, the pyran oxygen

atom of residue i and O8 of residue iÀ1. These interactions

could stabilize a twist–turn conformation, which puts the ex-

tremes of the molecule far away, therefore increasing the pro-

pensity for oligomerization instead of cyclization. Conversely,

3

the cyclization precursor 19 displayed coupling values of J_NHÀ

_C3H=9.8, 10.2, and 10.3 Hz (dihedral angles: H-N-C3-Hꢀ +167–

1798), a chemical shift of d_NH= <6.86 ppm in CDCl₃, and

a major band at n˜ =3422 cm^À1in CH₂Cl₂at 7 mm. Thus, the ab-

sence of bifurcated hydrogen bonds produces a more flexible

molecule, in which both ends could be found, thus leading to

the cyclized product 3.

3

the observed J_NHÀC3Hvalue mentioned above. The latter data

confirmed the high percentage of the gt conformer. Also, weak

NOE interactions between C3ÀH and C7ÀH_proRcorroborated

the presence of the gg conformer. Probably, the most flexible

part of the cyclic tetramer 4 is at position C9. Nonetheless,

medium NOE interactions were observed between C2ÀH and

C9ÀH_proR, between C7ÀH_proSand C9ÀH_proR/C9ÀH_proS, and be-

tween C7ÀH_proRand C9ÀH_proS, which indicates that the C9ÀH_proS

bond is on the bisector of the angle between the protons lo-

cated at C7 and also that C9ÀH_proRis antiperiplanar to C7ÀH_proR

(Figure 7). Unfortunately, all attempts to crystallize macrotetra-

peptide 3 were unfruitful.

Conformational changes induced in cyclodipeptide 2 by

changing from CDCl₃to D₂O were determined by comparison

3

of the coupling-constant values J_C2HÀC7HproRand J_C2HÀC7HproSob-

tained from the ¹H NMR spectra, which yielded three staggered

conformers around the torsional angle (O1-C2-C7-O8) by using

the Serianni equation system.^[23]The populations obtained in

CDCl₃(³J_C2HÀC7HproR=6.4 and ³J_C2HÀC7HproS=4.1 Hz) were 33, 48

and 19% for the gg, gt, and tg conformers, whereas the equa-

tions provided 70, 16, and 14% for the gg, gt, and tg conform-

ers in D₂O (³J_C2HÀC7HproR=3.0 and ³J_C2HÀC7HproS=3.9 Hz).^[24]The

conformation of cyclodipeptide 2 was more sensitive to the

change of the medium than it’s the analogous cyclodipeptide

1; that is, the folded and unfolded conformers of cyclodipep-

tide 2 are in equal proportion in CDCl₃, whereas 84% (gg+tg)

of the conformers are unfolded and only 16% (gt) are folded

in D₂O.

FTIR spectroscopy was performed with cyclic tetramer 3 in

dilute CH₂Cl₂(7 mm). Under these conditions, hydrogen bond-

ing is directly detectable from the NÀH stretching region. The

IR spectrum of 3 displays three absorption maxima in the NÀH

stretching region at n˜ =3414, 3388, and 3350 cm^À1. The new

model acyclic compounds 33 and 34 were synthesized,^[24]and

1

the H NMR and FTIR spectra of the reference compounds 30

and 32–34 were examined at the same concentration

(Figure 8) to determine the type of hydrogen bond formed.

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4014

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Figure 7. a) Schematic representation of observed NOE interactions of macrotetrapeptide 3 in C₆D₆. The solid arrows represent strong NOE interactions. Se-

quential NOE interactions have been omitted for clarity. b) Two-dimensional NOESY sections showing key intraresidue NOE interactions for macrotetrapeptide

3 in C₆D₆at 300 K.

These compounds represent a structural simplification of cyclic

tetrapeptide 3. As it was mentioned before, acetamide 32

shows only one NÀH stretching maximum at n˜ =3438 cm^À1

and a chemical shift of the amide proton of d_NH=5.62 ppm,

which corresponds to a non-hydrogen-bonded NH group.

Also, amide 30 displays one NÀH stretching maximum at n˜ =

33 only has the chance to form a conventional nine-mem-

bered-ring hydrogen bond between the amide proton of the

acetamide group and the carbonyl group of the piperidine

amide. However, in the NÀH stretching region only one maxi-

mum appears at n˜ =3437 cm^À1and a chemical shift of the

amide proton of d_NH=5.76 ppm, which corresponds to a non-

hydrogen bonded NH group. Furthermore, the model com-

pound 34 has two amide protons that can form conventional

9- and 11-membered hydrogen-bonded pseudocycles (C₉or

C₁₁) and weak electrostatic interactions (C₅). The FTIR spectrum

in the NÀH stretching region shows three maxima: one major

band at n˜ =3415 cm^À1, which corresponds to the weak electro-

3417 cm^À1and a chemical shift of the amide proton at d_NH

=

6.74 ppm, which may be attributed to a five-membered hydro-

gen-bonded pseudocycle (C₅) between the amide proton and

the ether oxygen atom of the 2-metoxyacetamide. Compound

static interactions, and two bands at n˜ =3351 and 3325 cm^À1

,

which arise from the 9- and 11-membered hydrogen-bonded

pseudocycles (Figure 8). Therefore, these results confirm the

existence of weak electrostatic interactions, such as C₅, be-

tween the NH bond and the ether oxygen atoms in cyclic tet-

rapeptide 3, and they also indicate the presence of intramolec-

ular conventional hydrogen bonds.

Molecular modeling

A theoretical conformation analysis was performed to gain in-

sight into the conformation equilibrium observed in these

cyclic peptides. A conformation search was done by using

a mixed torsional/low-mode conformational search,^[32]which

has proved to be very useful in conformation searches of

Figure 8. Structures, partial IR spectra in CH₂Cl₂(7 mm), and chemical shifts

in CDCl₃(7 mm) of macrotetrapeptide 3 and the reference compounds 30

and 32–34.

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4015

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

medium-size rings. It was used as implemented in the Maestro

version 9.3^[33]/Macromodel version 9.9^[34]by using either CHCl₃

or water as specified in the text. The best force field (FF) to

suit our molecules was OPLS 2001^[35]among all those imple-

mented on MacroModel with only two medium quality torsion

parameters (Osp²-Csp²-Csp³-Osp³) against all the other 156/

114/60 high-quality torsion/bending/stretching parameters.

Charges were provided from the FF with an extended cutoff.

Conformers were minimized by using the truncated Newton

conjugate gradient (TNCG) method with a gradient conver-

gence within a threshold of 0.005. A maximum numbers of

steps of 10000 was the limit within an energetic range of

21.0 kJmol^À1(5.02 kcalmol^À1) to ensure that all the conforma-

tional space was mapped. Significant conformation differences

between cyclodipeptides 1 and 2 were observed when the sol-

vent was changed from CHCl₃to water, but these compounds

are better discussed if a definition of folded geometry in such

systems is proposed. First, a reference system was taken. It is

reasonable to consider the crystal structure of cyclic peptide

1 as the folded reference (Figure 4a). Second, a good parame-

ter to follow the folding of the cyclic peptides is required. The

crystal structure of cyclic peptide 1 displays a folded structure,

and the collapsed structure is mainly due to the formation of

two eight-membered hydrogen-bonded pseudocycles (C₈) be-

tween the amide proton of one residue and the pyran oxygen

atom of the other residue. Consequently, the distance between

the amide proton of one residue and the pyran oxygen atom

of the other residue defines quite well the folding degree, and

therefore it was considered to be the geometrical parameter

of folding, which is on average 2.28 ꢅ measured on the refer-

ence system. This value was considered to be the geometrical

reference of folding. Finally, a borderline was set at 2.7 ꢅ,

which includes a top margin of almost 20% with respect to

the folding reference value, to define what a folded conformer

is. Thus, the cyclodipeptide is considered to be folded if the

distance between those selected positions is smaller than

2.7 ꢅ, and it is considered to be unfolded if the distance is

larger. According to this parameter, a prevalence of folded con-

formers in the cyclic peptide 1 either in CHCl₃or water with re-

spect to the cyclic peptide 2 is clearly seen. Indeed, among the

conformers found within a range of 21 kJmol^À1from the

global minimum (Figure 9) in CHCl₃, 30 out of 39 conformers

of cyclic peptide 1 (77%) are folded, whereas they represent

only 10 out of 22 conformers of cyclic peptide 2 (45%). How-

ever, among the conformers found within a range 21 kJmol^À1

from the global minimum in water (Figure 9), 32 out of 88 con-

formers of cyclic peptide 1 (36%) are folded, whereas they rep-

resent only 8 out of 40 conformers of cyclic peptide 2 (20%).

Moreover, cyclic peptide 1 displays the first unfolded structure

in CHCl₃at 9.80 kJmol^À1with respect to its global minimum,

scopic experiments, in which it was established that the pro-

portions between the populations of the gg, gt, and tg con-

formers can be used as a parameter to determine the degree

of folding.

In the case of macrotetrapeptide 3, the conformation search

displays a collapsed structure as the most stable conformer

(Figure 9d). Also, the conformers found within a range of

10 kJmol^À1from the global minimum show similar three-di-

mensional structures (Figure 9), all of them with the pyranose

rings in a chair conformation, in which the substituents at C2

and C4 are in the equatorial position and the substituent at C3

is in the axial position. The structure is collapsed mainly by the

formation of an 11-membered hydrogen-bonded pseudocycle

(C₁₁) between the amide proton of residue i and the carbonyl

group of residue iÀ2, which is present in all the conformers in

the range of 10 kJmol^À1. This hydrogen bond explains the ab-

sorption maximum in the NÀH stretching region at n˜ =

3350 cm^À1(Figure 8). The second most stable conformer dis-

plays an eight-membered hydrogen bonded pseudocycle (C₈)

between the amide proton of residue i and the pyran oxygen

atom of residue iÀ1, which explains the absorption maximum

in the NÀH stretching region at n˜ =3388 cm^À1(see Figure S1 in

the Supporting Information). Additionally, other types of intra-

molecular hydrogen bonds, such as interresidue five-mem-

bered hydrogen-bonded pseudocycle (C₅) interactions be-

tween the oxygen atom of the linker and the NÀH bond of the

amide groups, are present. These theoretical results confirm

those obtained by IR spectroscopic analysis and can also ex-

1

plain how the H NMR spectrum shows a symmetrical set of

high-resolution signals, which corresponds to a structure in-

volved in a fast conformation equilibrium on the NMR time-

scale.

Conclusion

In summary, new pyranoid e-sugar amino acids (e-SAAs) were

designed as building blocks for the synthesis of cyclic peptide

homooligomers. Those e-SAAs have proven to be privileged

scaffolds for the synthesis of conformationally modulated

cyclic peptides. Indeed, the sugar unit allows control of the

disposition and configuration of several moieties that are sus-

ceptible to acting as conformational modulators. Remarkably,

the carboxylic acid and the amine groups are placed at C2 and

C3 with respect to the pyran oxygen atom in these e-SAAs.

This disposition allows such an oxygen atom to participate ac-

tively in the control of the intramolecular hydrogen-bonding

pattern, and therefore to affect the conformation equilibria.

Also, the carboxylic acid and sugar moieties are linked through

an ether group. The ether oxygen atom and pyran oxygen

atom participate in five-membered hydrogen-bonded pseudo-

cycle (C₅) interactions, and the NÀH bonds of the amide

groups help to create an internal backbone of hydrogen

bonds in the cyclic peptides. The influence of an additional

control element on the sugar moiety in the conformation equi-

librium was also studied, namely, a methoxy group at C4,

which plays an important role in the populations of the folded

and unfolded conformers. Thus, when the methoxy group is

whereas cyclic peptide

2

displays this structure at

1.48 kJmol^À1. Nonetheless, although the folded structures still

are the most stable in water, the energy gap between the first

folded and the first unfolded conformers decreases noticeably.

Thus, this value is 3.91 kJmol^À1for cyclic peptide 1 and is just

0.88 kJmol^À1for cyclic peptide 2. These theoretical calculations

clearly confirm the results obtained from the NMR spectro-

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4016

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

formation of an additional intra-

residue five-membered hydro-

gen-bonded pseudocycle (C₅)

with the pyran oxygen atom is

favored,

thus

generating

a system of bifurcated hydrogen

bonding (2ꢄC₅), and this inter-

nal network of noncovalent in-

teractions helps to stabilize the

unfolded structures. However,

when the methoxy group is

present, the torsion angle (H-N-

C3-H) changes and the pyran

oxygen atom is instead involved

in an interresidue eight-mem-

bered hydrogen-bonded pseu-

docycle (C₈), and therefore

a folded structure is formed. This

methoxy group not only influen-

ces the conformation equilibri-

um, it also affects the reactivity

of the molecules. Indeed, the

macrocyclization reaction takes

place in the synthesis of the cy-

clotetrapeptides when such ap-

pendix on the sugar moiety is

present; however, the absence

of this unit prevents the reac-

tion. These results highlight how

subtle structural changes can

completely alter the pattern of

noncovalent interactions and

induce conformational changes

that affect the properties of cy-

clopeptides. Therefore, the intro-

duction of this type of appendix

as a conformation-control ele-

ment can be considered to be

a useful tool in the design of

novel cyclic peptides because

the formation of folded confor-

mations, which usually have

higher passive permeability, is

promoted, therefore improving

bioavailability.

Figure 9. a) Clustering of conformers within 10 kJmol^À1from the global minimum (for ease of visualization only).

b) Representation of all the conformers found within 21 kJmol^À1from the global minimum versus the geometrical

parameter of folding d (the interresidue distance [ꢅ] between the pyran oxygen atom and amide proton). The

borderline was set at 2.7 ꢅ, thus all the conformers below that value are considered to be folded and vice versa.

c) A stereoview of the minimum-energy structure of cyclotetrapeptide 3. The dashed line represents a hydrogen

bond (C₁₁).

present in cyclodipeptides, the folded conformation is the

most abundant in nonpolar and polar environments, with only

small differences in the conformation population between

Experimental Section

both media. However, when there is no methoxy group, the

folded and unfolded conformers are in equal proportion in

a nonpolar environment, but the unfolded conformers are

present almost exclusively in an aqueous medium. These re-

sults are mainly due to a change in the dihedral angle be-

tween the amide proton and the proton at C3 (H-N-C3-H) in-

duced by the methoxy group. In all the cases, the oxygen

atom of the linker forms an interresidue five-membered hydro-

gen-bonded pseudocycle (C₅) interaction with the NÀH bond

of the amide group. When the methoxy group is absent, the

Materials and general methods

The ¹H NMR spectra were recorded at 500 or 400 MHz; ¹³C NMR

spectra were recorded at 100 or 125 MHz; and the chemical shifts

are reported in ppm and referenced to the solvent peak. Melting

points were taken on a capillary melting-point apparatus and are

uncorrected. Optical rotations were obtained on a polarimeter at

l=589 nm at 258C with a path length of 10 cm and a volume of

1.0 mL. Concentration (c) is reported in grams per 100 mL of the

solvent specified. Infrared (FTIR) spectra are reported in wavenum-

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4017

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

bers (cm^À1). Low- and high-resolution mass spectra were recorded

on TOF analyzer mass spectrometers by using electrospray ioniza-

tion (ESI+) or electron impact (EI) at 70 eV, as specified in each

case. Column chromatography was performed on silica gel (60 ꢅ

and 0.2–0.5 mm). Compounds were visualized by use of UV light,

2.5% phosphomolybdic acid in ethanol, or vanillin with acetic and

sulfuric acid in ethanol with heating. All the solvents were purified

by using standard techniques.^[36]Reactions requiring anhydrous

conditions were performed under nitrogen. Anhydrous magnesium

sulfate was used for drying solutions.

H₂pressure. The reaction mixture was filtered through Celite,

washed with methanol (2ꢄ25 mL), and the filtrate was collected.

Removal of the solvents by rotary evaporation afforded the expect-

ed secondary alcohol, which was used without purification. DIAD

(4.1 mL, 20.6 mmol), DPPA (4.4 mL, 20.6 mmol), and triphenyl phos-

phine (5.4 g, 20.6 mmol) were added successively to a solution of

secondary alcohol in THF (70 mL) at 08C. The reaction mixture was

stirred for 10 min and then warmed to reflux for 1 h. The reaction

mixture was cooled to room temperature and quenched by the ad-

dition of water. The reaction mixture was extracted with EtOAc (3ꢄ

50 mL), the organic layer was dried over MgSO₄, the solvent was

removed under reduced pressure, and the crude product was puri-

fied by column chromatography to give 5 (3.92 g, 95% yield) as an

oil. [a]²_D⁵= +2.9 (c=1.4 in CHCl₃); ¹H NMR (400 MHz; CDCl₃): d=

1.46 (s; 9H), 1.77 (m; 1H), 1.95 (dddd, J=12.6, 12.6, 12.6, 5.1 Hz;

1H), 3.38 (m; 1H), 3.43 (s; 3H), 3.48 (ddd, J=11.6, 4.6, 3.4 Hz; 1H),

3.54 (ddd, J=6.2, 6.2, 1.2 Hz; 1H), 3.64 (ddd, J=14.0, 9.4, 6.1 Hz;

2H), 3.97 (d, J=16.5 Hz; 1H), 4.00 (m; 1H), 4.02 (d, J=16.5 Hz;

1H), 4.03 ppm (m; 1H); ¹³C NMR (100 MHz; CDCl₃): d=27.0 (t), 28.1

(q), 55.8 (q), 58.7 (d), 66.1 (t), 69.1 (t), 71.2 (t), 76.3 (d), 79.2 (d), 81.7

(s), 169.4 ppm (s); IR (CH₂Cl₂): n˜ =2983, 2106, 1746, 1265,

1103 cm^À1; HRMS (ESI): m/z: calcd for C₁₃H₂₃N₃O₅Na [M+Na]⁺:

324.1535; found: 324.1529.

tert-Butyl 2-{[(2R,3S,4R)-3-(benzyloxy)-4-methoxytetrahydro-

2H-pyran-2-yl]methoxy}acetate (10)

A 1m solution of borane tetrahydrofuran complex in THF (87.9 mL,

87.9 mmol) was added to benzylidene acetal 9^[18](4.4 g,

17.58 mmol) dissolved in CH₂Cl₂(176 mL) at room temperature

under nitrogen. The reaction mixture was stirred for 10 min, and

freshly dried copper(II) trifluoromethanesulfonate (318 mg,

0.88 mmol) was added portionwise to the solution. After stirring

for a period of time (2 h), the mixture was cooled to 08C, and the

reaction was quenched by sequential additions of triethylamine

(2.45 mL, 17.58 mmol) and methanol (caution: hydrogen gas was

evolved). The resulting mixture was concentrated at reduced pres-

sure followed by coevaporation with methanol. The residue was

purified by flash column chromatography on silica gel to give the

expected primary alcohol (4.35 g, 98% yield). ¹H NMR (400 MHz;

CDCl₃): d=1.46 (ddd, J=12.9, 11.0, 5.1 Hz; 1H), 2.00 (ddd, J=13.1,

4.2, 9.0 Hz; 1H), 3.14 (ddd, J=8.9, 5.0, 2.7 Hz; 1H), 3.24 (dd, J=

17.7, 8.7 Hz; 1H), 3.3 (m; 2H), 3.37 (s; 3H), 3.58 (dd, J=11.7, 5.1 Hz;

1H), 3.74 (dd, J=11.7, 2.9 Hz; 1H), 3.88 (ddd, J=11.7, 5.0, 1.7 Hz;

1H), 4.55 (d, J=11.1 Hz; 1H), 4.81 (d, J=11.1 Hz; 1H), 7.23 ppm

(m; 5H); ¹³C NMR (100 MHz; CDCl₃): d=30.9 (t), 56.9 (q), 62.7 (t),

65.5 (t), 74.9 (t), 79.8 (d), 82.9 (d), 127.7 (d), 128.0 (d), 128.4 (d),

138.6 ppm (s).

tert-Butyl 2-{[(2R,3S)-3-hydroxytetrahydro-2H-pyran-2-yl]me-

thoxy}acetate (11)

A solution of the diol 8 (1.00 g, 7.57 mmol) in THF (5 mL) was

added to a suspension of NaH (60% in mineral oil, 303 mg,

7.57 mmol) in dry THF (10 mL) at 08C, and the reaction mixture

was stirred for 5 min.

A solution of tert-butylbromoacetate

(1.12 mL, 7.57 mmol) in THF (20 mL) at 08C was added portionwise,

and the reaction mixture was stirred for a further 2 h (monitored

by TLC analysis). The reaction mixture was quenched by the slow

addition of water. The reaction mixture was extracted with EtOAc

(3ꢄ20 mL), the organic layer was dried over MgSO₄, the solvent

was removed under reduced pressure, and the crude product was

purified by column chromatography to give 11 (1.12 g, 60% yield)

NaH (60% in mineral oil, 0.76 g, 19.02 mmol) was added to a solu-

tion of the primary alcohol (4.00 g, 15.85 mmol) in THF (20 mL) at

08C stirring for 5 min and was added portionwise to a solution of

tert-butylbromoacetate (3.51 mL, 23.78 mmol) in THF (60 mL) at

08C. The reaction mixture was allowed to warm to room tempera-

ture and stirred for 2 h (monitored by TLC analysis). The reaction

mixture was cooled to 08C and quenched by a slow addition of

water. The reaction mixture was extracted with EtOAc (3ꢄ50 mL),

the organic layer was dried over MgSO₄, the solvent was removed

under reduced pressure, and the crude product was purified by

column chromatography to give 10 (5.17 g, 89% yield) as an oil.

1

as an oil. [a]²_D⁵= +47.8 (c=1.2 in CHCl₃); H NMR (400 MHz; CDCl₃):

d=1.46 (s; 9H), 1.46 (m; 1H), 1.61–1.78 (m; 2H), 2.13 (m; 1H), 3.15

(ddd, J=9.0, 3.3, 3.3 Hz; 1H), 3.35 (ddd, J=11.6, 11.6, 3.0 Hz; 1H),

3.67 (dd, J=9.8, 3.2 Hz; 1H), 3.80 (dd, J=9.9, 3.4 Hz; 1H), 3.80 (m;

1H), 3.88 (d, J=17.1 Hz; 1H), 3.92 (m; 1H), 3.99 (d, J=3.5 Hz; 1H),

4.09 ppm (d, J=17.1 Hz; 1H); ¹³C NMR (100 MHz; CDCl₃): d=26.0

(t), 28.5 (q), 31.8 (t), 67.5 (d), 68.4 (t), 68.6 (t), 71.9 (t), 81.7 (d), 82.9

(s), 171.2 ppm (s); IR (film, NaCl plates): n˜ =3483, 2978, 2944, 2859,

1731, 1371, 1266, 1152 cm^À1; HRMS (ESI): m/z: calcd for C₁₂H₂₂O₅Na

[M+Na]⁺: 269.1365; found: 269.1368.

1

[a]²_D⁵= +3.1 (c=1.1, CHCl₃); H NMR (400 MHz; CDCl₃): d=1.46 (s;

9H), 1.59 (m; 1H), 2.07 (m; 1H), 3.37 (m; 4H), 3.45 (s; 3H), 3.73

(dd, J=10.4, 4.5 Hz; 1H), 3.80 (dd, J=10.4, 1.6 Hz; 1H), 3.99 (s;

2H), 4.00 (m; 1H), 4.68 (d, J=11.1 Hz; 1H), 4.90 (d, J=11.1 Hz; 1H),

7.27 (m; 1H), 7.34 ppm (m; 4H); ¹³C NMR (100 MHz; CDCl₃): d=

28.1 (q), 30.7 (t), 56.8 (q), 65.6 (t), 69.3 (t), 71.0 (t), 74.8 (t), 78.2 (d),

79.3 (d), 81.4 (s), 83.1 (d), 127.5 (d), 128.0 (d), 128.3 (d), 138.8 (s),

169.5 ppm (s); IR (film, NaCl plates): n˜ =3061, 1746, 1722, 1605,

{(2R,3S)-3-[(Methylsulfonyl)oxy]tetrahydro-2H-pyran-2-yl}-

methyl benzoate (12)

Triethylamine (10.5 mL, 75.7 mmol) was added to

8 (5 g,

37.8 mmol) dissolved in CH₂Cl₂(190 mL, 0.2m) at 08C. Benzoyl

chloride (4.4 mL, 37.8 mmol) was added to the mixture. After stir-

ring for 30 min, triethylamine (10.5 mL, 75.7 mmol) and methane-

sulfonyl chloride (2.93 mL, 37.8 mmol) were sequentially added

and the mixture was stirred overnight at room temperature. The

reaction mixture was quenched by the addition of water (100 mL).

The reaction mixture was extracted with CH₂Cl₂(3ꢄ50 mL), the or-

ganic layer was dried over MgSO₄, the solvent was removed under

reduced pressure, and the crude product was purified by column

chromatography to give 12 (10.8 g, 92% yield) as an oil. [a]²_D⁵= +

1088 cm^À1

;

HRMS (ESI): m/z: calcd for C₂₀H₃₀O₆Na [M+Na]⁺:

389.1940; found: 389.1940.

tert-Butyl 2-{[(2S,3R,4R)-3-azido-4-methoxytetrahydro-2H-

pyran-2-yl]methoxy}acetate (5)

Compound 10 (5 g, 13.7 mmol) was added into a suspension of

5 wt% Pd/C (catalytic) in methanol (90 mL), stirring for 1–3 h

(monitored by TLC analysis) at room temperature under 1 atm of

1

7.9 (c=1.7 in CHCl₃); H NMR (400 MHz; CDCl₃): d=1.79 (m; 3H),

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4018

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

2.47 (m; 1H), 3.01 (s; 3H), 3.42 (m; 1H), 3.60 (ddd, J=9.5, 4.6,

2.2 Hz; 1H), 3.98 (m; 1H), 4.44 (dd, J=12.2, 4.6 Hz; 1H), 4.61 (dd,

J=12.2, 2.1 Hz; 1H), 4.69 (ddd, J=9.8, 9.8, 4.8 Hz; 1H), 7.44 (dd,

J=7.5, 7.5 Hz; 2H), 7.56 (ddd, J=7.3, 7.3, 1.3 Hz; 1H), 8.06 ppm

(m; 2H); ¹³C NMR (100 MHz; CDCl₃): d=25.5 (t), 31.3 (t), 39.2 (q),

63.9 (t), 68.2 (t), 75.4 (d), 77.7 (d), 128.9 (d), 130.2 (d), 130.3 (s),

133.5 (d), 166.7 ppm (s); IR (film, NaCl plates): n˜ =3063, 2951, 2862,

1723, 1176, 957 cm^À1; HRMS (ESI): m/z: calcd for C₁₄H₁₈O₆SNa

[M+Na]⁺: 337.0722; found: 337.0729.

169.4 ppm (s); IR (CH₂Cl₂): n˜ =3026, 3011, 2982, 2933, 2859, 2105,

1742, 1370, 1148 cm^À1; HRMS (ESI): m/z: calcd for C₁₂H₂₁N₃O₄Na

[M+Na]⁺: 294.1430; found: 294.1434.

Dipeptide(OMe) (16)

Compound 5 was separated into two portions. One portion was

used to convert the tert-butyl ester moiety into the carboxylic acid

group under the following conditions:

General procedure for the deprotection of the tert-butyl

[(2S,3R)-3-Azidotetrahydro-2H-pyran-2-yl]methyl benzoate

ester

(13)

Trifluoroacetic acid (2 mL) was added to a solution of 5 (1.9 g,

6.3 mmol) in CH₂Cl₂(10 mL) with stirring for 2 h (monitored by TLC

analysis) at room temperature. The solvent and trifluoroacetic acid

were removed under reduced pressure, and the expected carboxyl-

ic acid 14 was used without purification.

Sodium azide (10.3 g, 159.1 mmol) was added to 12 (10 g,

31.8 mmol) dissolved in DMF (80 mL, 0.4m) at room temperature.

The reaction mixture was stirred for 10 min and then warmed to

110 8C overnight. The reaction mixture was cooled to room tem-

perature and quenched by the addition of water. The reaction mix-

ture was extracted with CH₂Cl₂(3ꢄ50 mL), the organic layer was

dried over MgSO₄, the solvent was removed under reduced pres-

sure, and the crude product was purified by column chromatogra-

phy to give 13 (7.1 g, 85% yield) as an oil. [a]²_D⁵=À16.9 (c=1.4 in

The second portion was used to convert the azide group into the

amine group under the following conditions:

General procedure for the reduction of the azide

1

CHCl₃); H NMR (400 MHz; CDCl₃): d=1.49 (m; 1H), 1.87 (dddd, J=

13.6, 13.6, 4.2, 3.3 Hz; 1H), 2.00 (ddddd, J=13.0, 13.0, 13.0, 4.3,

4.3 Hz; 1H), 2.20 (m; 1H), 3.51 (ddd, J=11.9, 2.4 Hz; 1H), 3.74 (m;

1H), 3.80 (ddd, J=6.7, 5.7, 1.8 Hz; 1H), 4.06 (m; 1H), 4.36 (dd, J=

11.5, 5.7 Hz; 1H), 4.43 (dd, J=11.5, 6.8 Hz; 1H), 7.44 (dd, J=7.9,

7.4 Hz; 2H), 7.57 (dddd, J=7.4, 7.4, 1.9, 1.4 Hz; 1H), 8.06 ppm (m;

2H); ¹³C NMR (100 MHz; CDCl₃): d=20.7 (t), 27.3 (t), 56.7 (d), 64.7

(t), 68.1 (t), 76.1 (d), 128.4 (d), 129.7 (d), 129.9 (s), 133.1 (d),

166.3 ppm (s); IR (film, NaCl plates): n˜ =3064, 2955, 2859, 2104,

1721, 1273, 1092, 952 cm^À1; MS (EI): m/z (relative intensity): 139

[MÀBzOH]⁺(7), 111 (12), 105 (100); elemental analysis (%) calcd for

C₁₃H₁₅N₃O₃: C 59.76, H 5.79, N 16.08; found: C 60.09, H 5.91, N

15.69.

Compound 5 (1.9 g, 6.3 mmol) was added to a suspension of

5 wt% Pd/C (catalytic) in methanol (50 mL) with stirring for 1–3 h

(monitored by TLC analysis) at room temperature under 1 atm of

H₂pressure. The reaction mixture was filtered through celite and

washed with methanol (3ꢄ25 mL). The filtrate was collected and

the solvents were removed by rotary evaporation, thus affording

the expected amine 15, which was used without purification.

General coupling procedure

The corresponding carboxylic acid 14 and amine 15 were dissolved

in CH₂Cl₂(60 mL). DIPEA (2.2 mL, 12.6 mmol), HOBt (1.7 g,

12.6 mmol), and EDCI (2.4 g, 12.6 mmol) were added to the reac-

tion mixture, which was stirred overnight. The reaction mixture

was quenched with water and extracted with CH₂Cl₂(3ꢄ20 mL).

The organic layer was dried over MgSO₄, filtered, concentrated,

and the crude product was purified by column chromatography to

give 16 (2.2 g, 70% yield) as an oil. [a]²_D⁵= +3.1 (c=1.3 in CHCl₃);

¹H NMR (400 MHz; CDCl₃): d=1.45 (s; 9H), 1.62 (dddd, J=12.3,

12.3, 12.3, 5.1 Hz; 1H), 1.80 (m; 2H), 1.96 (dddd, J=12.6, 12.6, 12.6,

4.9 Hz; 1H), 3.38 (s; 3H), 3.41 (m; 1H), 3.44 (s; 3H), 3.51 (m; 3H),

3.63 (m; 6H), 3.87 (m; 1H), 4.01 (m; 6H), 4.57 (dd, J=10.3, 3.4 Hz;

1H), 6.89 ppm (d, J=10.3 Hz; 1H); ¹³C NMR (100 MHz; CDCl₃): d=

26.9 (t), 28.1 (t), 28.1 (q), 45.8 (d), 55.8 (q), 56.2 (q), 58.8 (d), 66.0 (t),

66.3 (t), 69.1 (t), 71.0 (t), 72.0 (t), 72.2 (t), 76.2 (d), 77.3 (d), 77.7 (d),

79.4 (d), 81.6 (s), 169.5 (s), 169.8 ppm (s); IR (film, NaCl plates): n˜ =

3416, 3338, 2947, 2858, 2104, 1747, 1677, 1098 cm^À1; HRMS (ESI):

m/z: calcd for C₂₂H₃₈N₄O₉Na [M+Na]⁺: 525.2536; found: 525.2556.

tert-Butyl 2-{[(2S,3R)-3-azidotetrahydro-2H-pyran-2-yl]me-

thoxy}acetate (6)

Sodium carbonate (0.32 g, 3.1 mmol) was added to 13 (8.0 g,

30.6 mmol) dissolved in methanol (76 mL, 0.4m). After stirring for

2 h, the reaction mixture was filtered through celite and washed

with methanol (3ꢄ10 mL). The solvent was removed under re-

duced pressure and the crude product was purified by column

chromatography to give the expected primary alcohol.

NaH (60% in mineral oil, 1.37 g, 34.4 mmol) was added to a solution

of primary alcohol (4.5 g, 28.6 mmol) in THF (50 mL) at 08C with

stirring for 5 min. A solution of tert-butylbromoacetate (8.5 mL,

57.3 mmol) in THF (100 mL) was then added portionwise at 08C.

The reaction mixture was allowed to warm to room temperature

and stirred for 2 h. The reaction mixture was cooled to 08C and

quenched by the slow addition of water. The reaction mixture was

extracted with EtOAc (3ꢄ50 mL), the organic layer was dried over

MgSO₄, the solvent was removed under reduced pressure, and the

crude product was purified by column chromatography to give 6

Cyclodipeptide(OMe) (1): The general procedure for reduction of

azide was applied to 16 (110 mg, 0.22 mmol) to afford the expect-

ed amine, which was used without purification.

Then, the general procedure to deprotect the tert-butyl ester

moiety was applied to the amine obtained above to obtain the ex-

pected amino acid 26, which was used without purification.

1

(6.2 g, 75% yield) as an oil. [a]²_D⁵=À55.3 (c=1.6 in CHCl₃); H NMR

(400 MHz; CDCl₃): d=1.44 (m; 1H), 1.46 (s; 9H), 1.80 (dddd, J=

13.6, 13.6, 3.7, 3.7 Hz; 1H), 1.94 (ddddd, J=13.6, 13.4, 13.1, 4.4,

4.4 Hz; 1H), 2.14 (d, J=16.2 Hz; 1H), 3.48 (ddd, J=11.9, 11.9,

2.4 Hz; 1H), 3.58 (d, J=1.9 Hz; 1H), 3.60 (s; 1H), 3.67 (m; 1H), 3.70

(m; 1H), 3.96 (d, J=16.3 Hz; 1H), 4.01 (m; 1H), 4.02 ppm (d, J=

16.3 Hz; 1H); ¹³C NMR (100 MHz; CDCl₃): d=20.7 (s), 27.3 (s), 28.1

(t), 56.7 (s), 68.2 (s), 69.2 (s), 71. 6 (s), 77.0 (t), 77.3 (s), 81.7 (s)

General cyclization procedure: The corresponding amino acid 26

was dissolved in CH₂Cl₂(110 mL). DIPEA (0.06 mL, 0.34 mmol),

HOBt (46 mg, 0.34 mmol), and EDCI (66 mg, 0.34 mmol) were

added to the mixture, which was stirred for 5 days. The reaction

mixture was quenched with water and extracted with CH₂Cl₂(3ꢄ

20 mL). The organic layer was dried over MgSO₄, filtered, concen-

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4019

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

trated, and the crude product was purified by column chromatog-

raphy to give 1 (34 mg, 38% yield) as an oil. [a]²_D⁵= +5.9 (c=1.2 in

CDCl₃); ¹H NMR (500 MHz; CDCl₃): d=1.74 (dddd, J=13.2, 13.2,

11.8, 5.3 Hz; 2H), 1.87 (m; 2H), 3.34 (s; 6H), 3.39 (dd, J=10.7,

10.3 Hz; 2H), 3.47 (ddd, J=11.7, 4.9, 3.9 Hz; 2H), 3.61 (ddd, J=

12.7, 12.6, 2.6 Hz; 2H), 3.69 (ddd, J=10.7, 4.6, 1.0 Hz; 2H), 3.77 (dd,

J=10.3, 4.6 Hz; 2H), 3.87 (d, J=16.4 Hz; 2H), 4.11 (ddd, J=12.1,

5.2, 1.2 Hz; 2H), 4.26 (d, J=16.4 Hz; 2H), 4.53 (dd, J=10.4, 3.8 Hz;

2H), 7.65 ppm (d, J=10.4 Hz; 2H); ¹³C NMR (100 MHz; CDCl₃): d=

27.72 (t), 45.16 (d), 55.78 (q), 66.85 (t), 69.57 (t), 70.40 (t), 75.76 (d),

76.51 (d), 170.35 ppm (s); IR (CH₂Cl₂): n˜ =3336, 2957, 2933, 2866,

1674, 1533, 1266, 1127, 1087, 848 cm^À1; HRMS (ESI): m/z: calcd for

C₁₈H₃₀N₂O₈Na [M+Na]⁺: 425.1900; found: 425.1898.

boxylic acid group under was applied to the following conditions:

The general procedure to deprotect the tert-butyl ester moiety was

applied to 6 (2.0 g, 7.4 mmol) to obtain the expected carboxylic

acid 20, which was used without purification.

The second portion was used to convert the azide group into the

amine group under the following conditions: The general proce-

dure for reduction of azide was applied to 6 (2.0 g, 7.4 mmol) to

afford the expected amine 21, which was used without purifica-

tion.

The general coupling procedure described above was used with

carboxylic acid 20 and amine 21 to give 22 (2.2 g, 68% yield) as an

oil. [a]²_D⁵=À39.1 (c=1.4 in CHCl₃); ¹H NMR (400 MHz; CDCl₃): d=

1.40 (s; 9H), 1.64–1.92 (m; 7H), 2.18 (d, J=13.6 Hz; 1H), 3.45–3.65

(m; 8H), 3.75 (m; 1H), 3.92–4.15 (m; 7H), 7.33 ppm (d, J=8.8 Hz;

1H); ¹³C NMR (75 MHz; CDCl₃): d=20.7 (t), 20.9 (t), 27.2 (t), 28.1 (q),

28.6 (t), 44.7 (d), 56.5 (d), 67.9 (t), 68.6 (t), 69.1 (t), 70.8 (t), 72.2 (t),

72.4 (t), 76.8 (d), 78.3 (d), 81.4 (s), 169.2 (s), 169.4 ppm (s); IR (film,

NaCl plates): n˜ =3416, 3338, 3059, 2948, 2859, 2104, 1748, 1677,

1529, 1098 cm^À1; HRMS (ESI): m/z: calcd for C₂₀H₃₄N₄O₇Na [M+Na]⁺

: 465.2325; found: 465.2329.

Tetrapeptide(OMe) (19): Compound 16 was separated in two por-

tions. One portion was used to convert the tert-butyl ester moiety

into a carboxylic acid group under the following conditions: The

general procedure to deprotect the tert-butyl ester moiety was ap-

plied to 16 (1 g, 2.0 mmol), and the expected carboxylic acid 17

was used without purification.

The second portion was used to convert the azide group into the

amine group under the following conditions: The general proce-

dure for reduction of azide was applied to 16 (1 g, 2.0 mmol) to

afford the expected amine 18, which was used without purifica-

tion.

Cyclodipeptide (2): The general procedure for reduction of the

azide moiety was applied to 22 (140 mg, 0.32 mmol) to afford the

expected amine, which was used without purification.

Then, the general procedure to deprotect the tert-butyl ester

moiety was applied to the amine obtained above to give the ex-

pected amino acid 28, which was used without purification.

The general coupling procedure described above was used with

carboxylic acid 17 and amine 18 to give 19 (1.23 g, 68% yield) as

1

an oil. [a]²_D⁵= +4.9 (c=1.1 in CHCl₃); H NMR (400 MHz; CDCl₃): d=

The corresponding amino acid 28 was dissolved in CH₂Cl₂(64 mL).

DIPEA (0.164 mL, 0.96 mmol) and HBTU (146 mg, 0.38 mmol) were

added to the reaction mixture, which was stirred for 5 days. The re-

action mixture was quenched with water and extracted with

CH₂Cl₂(3ꢄ20 mL). The organic layer was dried over MgSO₄, filtered,

concentrated, and the crude product was purified by column chro-

matography to give 2 (34 mg, 31% yield) as an oil. [a]²_D⁵= +67.6

1.45 (s; 9H), 1.65 (m; 2H), 1.80 (m; 4H), 1.95 (m; 2H), 3.35 (s; 6H),

3.36 (s; 3H), 3.40 (m; 3H), 3.43 (s; 3H), 3.45 (m; 2H), 3.51 (m; 5H),

3.58 (m; 6H), 3.64 (m; 3H), 3.70 (m; 2H), 3.87 (m; 1H), 3.95 (d, J=

4.1 Hz; 2H), 4.00 (d, J=3.4 Hz; 2H), 4.05 (m; 7H), 4.54 (m; 3H),

6.83 (d, J=10.2 Hz; 1H), 6.85 (d, J=9.8 Hz; 1H), 6.86 ppm (d, J=

10.3 Hz; 1H); ¹³C NMR (100 MHz; CDCl₃): d=26.9 (t), 27.9 (t), 28.0

(t), 28.1 (q), 45.6 (q), 45.7 (q), 45.8 (q), 55.8 (d), 56.0 (d), 56.1 (d),

56.1 (d), 58.8 (d), 66.0 (t), 66.1 (t), 66.2 (t), 66.3 (t), 69.1 (t), 70.9 (t),

71.0 (t), 72.1 (t), 72.2 (t), 72.3 (t), 76.3 (d), 76.9 (d), 77.0 (d), 77.2 (d),

77.6 (d), 77.8 (d), 79.4 (d), 81.6 (s), 169.4 (s), 169.8 (s), 169.9 (s),

170.0 ppm (s); IR (film, NaCl plates): n˜ =3416, 2985, 2929, 2107,

1

(c=1.5 in CHCl₃); H NMR (500 MHz; C₆D₆): d=0.82 (bdd, J=13.7,

2.0 Hz; 2H), 1.19 (dddd, J=13.5, 13.5, 3.5, 3.5 Hz; 2H), 1.54 (ddddd,

J=13.7, 13.5, 12.8, 4.5, 4.5 Hz; 2H), 2.06 (bd, J=13.5 Hz; 2H), 2.92

(ddd, J=4.6, 4.6, 1.5 Hz; 2H), 2.97 (ddd, J=12.7, 11.3, 2.5 Hz; 2H),

2.99 (dd, J=10.3, 5.0 Hz; 2H), 3.33 (dd, J=10.3, 4.0 Hz; 2H), 3.56

(dd, J=11.2, 5.0 Hz; 2H), 3.69 (d, J=14.7 Hz; 2H), 3.88 (d, J=

14.7 Hz; 2H), 3.99 (m; 2H), 7.72 ppm (bd, J=7.0 Hz; 2H); ¹³C NMR

(100 MHz; C₆D₆): d=20.6 (t), 28.5 (t), 46.6 (d), 68.9 (t), 71.3 (t), 72.1

(t), 76.5 (d), 168.1 ppm (s); IR (CH₂Cl₂): n˜ =3425, 3331, 3186, 3066,

2932, 2859, 1671, 1536, 1265, 1134, 1097, 1070, 848 cm^À1; HRMS

(ESI): m/z: calcd for C₁₆H₂₆N₂O₆Na [M+Na]⁺: 365.1689; found:

365.1692.

1682, 1048 cm^À1

; HRMS (ESI): m/z: calcd for C₄₀H₆₈N₆O₁₇Na

[M+Na]⁺: 927.4539; found: 927.4516.

Cyclotetrapeptide(OMe) (3): The general procedure for reduction

of azide was applied to 19 (140 mg, 0.156 mmol) to afford the ex-

pected amine, which was used without purification.

Then the general procedure to deprotect the tert-butyl ester

moiety was applied to the amine obtained above to provide the

expected amino acid 27, which was used without purification.

Tetrapeptide (25): Compound 22 was separated into two portions.

One portion was used to convert the tert-butyl ester moiety into

a carboxylic acid group under the following conditions: The gener-

al procedure to deprotect the tert-butyl ester moiety was applied

to 22 (1 g, 2.3 mmol) to obtain the expected carboxylic acid 23,

which was used without purification.

The general cyclization procedure described above was applied to

the corresponding amino acid 27 to give 3 (54 mg, 43% yield) as

1

an oil. [a]²_D⁵= +6.1 (c=1.3 in CHCl₃); H NMR (500 MHz; C₆D₆): d=

1.39 (dd, J=13.0, 4.2 Hz; 4H), 1.67 (dddd, J=13.0, 12.9, 12.6,

5.1 Hz; 4H), 2.89 (ddd, J=11.9, 4.5, 4.3 Hz; 4H), 3.03 (ddd, J=12.9,

11.7, 2.4 Hz; 4H), 3.20 (ddd, J=8.5, 2.4, 1.6 Hz; 4H), 3.22 (s; 12H),

3.51 (dd, J=11.0, 2.5 Hz; 4H), 3.61 (dd, J=11.0, 8.5 Hz; 4H), 3.70

(dd, J=11.7, 4.1 Hz; 4H), 3.88 (d, J=15.8 Hz; 4H), 4.05 (d, J=

15.8 Hz; 4H), 4.50 (dd, J=10.0, 4.0 Hz; 4H), 7.22 ppm (d, J=

10.1 Hz; 4H); ¹³C NMR (100 MHz; C₆D₆): d=27.9 (t), 45.3 (d), 55.4

(q), 65.9 (t), 70.1 (t), 72.9 (t), 76.6 (d), 77.2 (d), 169.7 ppm (s); IR

(CH₂Cl₂, 7 mm): n˜ =3414, 3388, 3350, 2930, 2862, 1678, 1527, 1089,

848 cm^À1; HRMS (ESI): m/z: calcd for C₃₆H₆₀N₄O₁₆Na [M+Na]⁺:

827.3902; found: 827.3873

The second portion was used to convert the azide group into an

amine group under the following conditions: The general proce-

dure for reduction of the azide moiety was applied to 22 (1 g,

2.3 mmol) to afford the expected amine 24, which was used with-

out purification.

The general coupling procedure described above was used with

the corresponding carboxylic acid 23 and amine 24 to give 25

1

(1.1 g, 62% yield) as an oil. [a]²_D⁵=À12.3 (c=1.3 in CHCl₃); H NMR

(400 MHz; CDCl₃): d=1.45 (s; 9H), 1.48 (m; 4H), 1.75 (m; 6H), 1.90

(m; 5H), 2.18 (d, J=13.7 Hz; 1H), 3.44–3.77 (m; 17H), 3.90–4.20 (m;

15H), 7.14 (d, J=9.3 Hz; 1H), 7.18 (d, J=9.1 Hz; 1H), 7.29 ppm (d,

Dipeptide (22): Compound 6 was separated in two portions. One

portion was used to convert the tert-butyl ester moiety into a car-

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4020

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

J=8.7 Hz; 1H); ¹³C NMR (100 MHz; CDCl₃): d=20.7 (t), 20.8 (t), 20.9

(t), 27.2 (t), 28.1 (q), 28.5 (t), 28.6 (t), 44.1 (d), 44.2 (d), 44.4 (d), 56.5

(d), 67.9 (t), 68.5 (t), 68.6 (t), 69.0 (t), 70.7 (t), 70.9 (t), 72.4 (t), 72.4

(t), 72.5 (t), 73.0 (t), 76.8 (d), 78.0 (d), 78.4 (d), 81.5 (s), 169.2 (s),

169.2 (s), 169.4 ppm (s); IR (film, NaCl plates): n˜ =3413, 3338, 2944,

2859, 2104, 1746, 1677, 1525, 1099 cm^À1; HRMS (ESI): m/z: calcd for

C₃₆H₆₀N₆O₁₃Na [M+Na]⁺: 807.4116; found: 807.4137.

H. Kessler, E. Novellino, A. Lavecchia, J. H. van Boom, M. Overhand, J.

Am. Chem. Soc. 2003 , 125 , 1082 2 –1 0829; f) T. K. Chakraborty, D. Koley,

S. Prabhakar, R. Ravi, A. C. Kunwar, Tetrahedron 2005 , 61 , 950 6 –9 512;

g) B. Jagannadh, M. S. Reddy, C. L. Rao, A. Prabhakar, B. Jagadeesh, S.

Chandrasekhar, Chem. Commun. 2006 , 484 7 –4 849; h) T. K. Chakraborty,

S. Roy, D. Koley, S. K. Dutta, A. C. Kunwar, J. Org. Chem. 2006 , 71 , 624 0 –

624 3; i) F. Fujimura, T. Hirata, T. Morita, S. Kimura, Y. Horikawa, J. Sugiya-

ma, Biomacromolecules 2006 , 7 , 239 4 –2 40 0; j) T. Hirata, F. Fujimura, Y.

Horikawa, J. Sugiyama, T. Morita, S. Kimura, Biopolymers 2007 , 88 , 15 0 –

156; k) F. Fujimura, Y. Horikawa, T. Morita, J. Sugiyama, S. Kimura, Bio-

macromolecules 2007 , 8 , 61 1 –6 16; l) S. Chandrasekhar, B. Saritha, P.

Naresh, M. Udayakiran, C. R. Reddy, B. Jagadeesh, Helv. Chim. Acta 2008 ,

91 , 126 7 –1 276.

Acknowledgements

This research was supported by the Spanish MINECO and cofi-

nanced by the European Regional Development Fund (ERDF)

(CTQ2011–22653). A.F.-V. and J.B.G. thank the Spanish MEC for

an FPU fellowship.

[12] a) R. Carrillo, V. S. Martꢂn, T. Martꢂn, T e trahedron Lett. 2004 , 45 , 521 5 –

5219; b) R. Carrillo, V. S. Martꢂn, M. Lꢇpez, T. Martꢂn, Tetrahedron 2005 ,

61 , 817 7 –8 191; c) R. Carrillo, M. Lꢇpez-Rodrꢂguez, V. S. Martꢂn, T. Martꢂn,

Angew. Chem. 2009 , 121 , 794 3 –7 94 8; Angew. Chem. Int. Ed. 2009 , 48 ,

780 3 –7 808; d) R. Carrillo, A. Feher-Voelger, T. Martꢂn, Angew. Chem.

2011 , 123 , 1080 4 –1 0808; Angew. Chem. Int. Ed. 2011 , 50 , 1061 6 –1 0620;

e) R. Carrillo, E. Q. Morales, V. S. Martꢂn, T. Martꢂn, Chem. Eur. J. 2013 , 19 ,

704 2 –7 048; f) R. Carrillo, E. Q. Morales, V. S. Martꢂn, T. Martꢂn, J. Org.

Chem. 2013 , 78 , 778 5 –7 795.

Keywords: conformation analysis

·

cyclic peptides

·

noncovalent interactions · solid-state structures · sugar amino

acids

[13] R. Carrillo, M. Lꢇpez-Rodrꢂguez, V. S. Martꢂn, T. Martꢂn, CrystEngComm

2010 , 12 , 367 6 –3 683.

[14] a) R. M. van Well, L. Marinelli, K. Erkelens, G. A. van der Marel, A. Lavec-

chia, H. S. Overkleeft, J. H. van Boom, H. Kessler, M. Overhand, Eur. J.

Org. Chem. 2003 , 230 3 –2 31 3; b) M. Mꢀnand, J.-C. Blais, L. Hamon, J.-M.

Valꢀry, J. Xie, J. Org. Chem. 2005 , 70 , 442 3 –4 430.

[1] a) E. M. Driggers, S. P. Hale, J. Lee, N. K. Terrett, Nat. Rev. Drug Discovery

2008 , 7 , 60 8 –6 24; b) E. Marsault, M. L. Peterson, J. Med. Chem. 2011 , 54 ,

196 1 –2 004; c) J. Mallinson, I. Collins, Future Med. Chem. 2012 , 4 , 140 9 –

1438; d) A. R. Bogdan, S. V. Jerome, K. N. Houk, K. James, J. Am. Chem.

Soc. 2012 , 134 , 212 7 –2 138.

[2] D. F. Veber, S. R. Johnson, H.-Y. Cheng, B. R. Smith, K. W. Ward, K. D.

Kopple, J. Med. Chem. 2002 , 45 , 261 5 –2 623.

[3] J. F. Tjia, I. R. Webber, D. J. Back, Br. J. Clin. Pharmacol. 1991 , 31 , 34 4 –

346.

[4] A. M. Dahlem, A. S. Hassan, S. P. Swanson, W. W. Carmichael, V. R. Beas-

ley, Pharmacol. Toxicol. 1989 , 64 , 17 7 –1 81.

[5] E. Biron, J. Chatterjee, O. Ovadia, D. Langenegger, J. Brueggen, D.

Hoyer, H. A. Schmid, R. Jelinek, C. Gilon, A. Hoffman, H. Kessler, Angew.

Chem. 2008 , 120 , 263 3 –2 637; Angew. Chem. Int. Ed. 2008 , 47 , 259 5 –

2599.

[15] a) I. A. Motorina, C. Huel, E. Quiniou, J. Mispelter, E. Adjadj, D. S. Grier-

son, J. Am. Chem. Soc. 2001 , 123 , 8 –1 7; b) H.-D. Arndt, A. Knoll, U.

Koert, Angew. Chem. 2001 , 113 , 213 7 –2 140; Angew. Chem. Int. Ed. 2001 ,

40 , 207 6 –2 078; c) J. T. Ernst, J. Becerril, H. S. Park, H. Yin, A. D. Hamilton,

Angew. Chem. 2003 , 115 , 55 3 –5 57; Angew. Chem. Int. Ed. 2003 , 42 , 53 5 –

539; d) H.-D. Arndt, B. Ziemer, U. Koert, Org. Lett. 2004 , 6 , 326 9 –3 272;

e) J. P. Saludes, J. B. Ames, J. Gervay-Hague, J. Am. Chem. Soc. 2009 , 131 ,

549 5 –5 505; f) W. Hu, Q. Yan, D. Zhao, Chem. Eur. J. 2011 , 17 , 708 7 –

7094; g) G. V. M. Sharma, K. S. Reddy, S. J. Basha, K. R. Reddy, A. V. S.

Sarma, Org. Biomol. Chem. 2011 , 9 , 810 2 –8 111; h) G. V. M. Sharma, P. S.

Reddy, D. Chatterjee, A. C. Kunwar, Tetrahedron 2012 , 68 , 439 0 –4 398.

[16] M. Kanai, Y. Hamashima, M. Shibasaki, Tetrahedron Lett. 2000 , 41 , 240 5 –

2409.

[6] C. A. Lipinski, F. Lombardo, B. W. Dominy, P. J. Feeney, Adv. Drug Delivery

Rev. 2001 , 46 , 3 –2 6.

[17] G. Snatzke, Z. Raza, I. Habus, V. Sunjic, Carbohydr. Res. 1988 , 182 , 17 9 –

196.

[18] R. M. Giuliano Jr., A. D. Jordan, A. D. Gauthier, K. Hoogsteen, J. Org.

Chem 1993 , 58 , 497 9 –4 988.

[19] C.-R. Shie, Z.-H. Tzeng, S. S. Kulkarni, B.-J. Uang, C.-Y. Hsu, S.-C. Hung,

Angew. Chem. 2005 , 117 , 169 3 –1 696; Angew. Chem. Int. Ed. 2005 , 44 ,

166 5 –1 668.

[20] In early experiments, the methyl ester unit was employed as the car-

boxyl protecting group, but the alkoxy group of the starting material

reacted with the final product, thus leading to transesterification; the

tert-butyl ester moiety proved to be hindered enough to effectively

prevent this deleterious side reaction.

[7] a) S. Gangwar, S. D. S. Jois, T. J. Siahaan, D. G. Vander Velde, V. J. Stella,

R. T. Borchardt, Pharm. Res. 1996 , 13 , 165 7 –1 662; b) T. Rezai, B. Yu, G. L.

Millhauser, M. P. Jacobson, R. S. Lokey, J. Am. Chem. Soc. 2006 , 128 ,

251 0 –2 511; c) T. Rezai, J. E. Bock, M. V. Zhou, C. Kalyanaraman, R. S.

Lokey, M. P. Jacobson, J. Am. Chem. Soc. 2006 , 128 , 1407 3 –1 4080;

d) S. B. Rafi, B. R. Hearn, P. Vedantham, M. P. Jacobson, A. R. Renslo, J.

Med. Chem. 2012 , 55 , 316 3 –3 169; e) J. E. Bock, J. Gavenonis, J. A. Kritzer,

ACS Chem. Biol. 2013 , 8 , 48 8 –4 99.

[8] a) J. Chatterjee, C. Gilon, A. Hoffman, H. Kessler, Acc. Chem. Res. 2008 ,

41 , 133 1 –1 342; b) L. Doedens, F. Opperer, M. Cai, J. G. Beck, M. Dedek,

E. Palmer, V. J. Hruby, H. Kessler, J. Am. Chem. Soc. 2010 , 132 , 811 5 –

8128; c) T. R. White, C. M. Renzelman, A. C. Rand, T. Rezai, C. M. McEwen,

V. M. Gelev, R. A. Turner, R. G. Linington, S. S. F. Leung, A. S. Kalgutkar,

J. N. Bauman, Y. Zhang, S. Liras, D. A. Price, A. M. Mathiowetz, M. P. Ja-

cobson, R. S. Lokey, Nat. Chem. Biol. 2011 , 7 , 81 0 –8 17; d) J. G. Beck, J.

Chatterjee, B. Laufer, M. U. Kiran, A. O. Frank, S. Neubauer, O. Ovadia, S.

Greenberg, C. Gilon, A. Hoffman, H. Kessler, J. Am. Chem. Soc. 2012 , 134 ,

1212 5 –1 2133; e) J. Chatterjee, F. Rechenmacher, H. Kessler, Angew.

Chem. 2013 , 125 , 26 8 –2 83; Angew. Chem. Int. Ed. 2013 , 52 , 25 4 –2 69.

[9] T. K. Chakraborty, S. Jayaprakash, P. V. Diwan, R. Nagaraj, S. R. B. Jampani,

A. C. Kunwar, J. Am. Chem. Soc. 1998 , 120 , 1296 2 –1 2963.

[21] B. Lal, B. N. Pramanik, M. S. Manhas, A. K. Bose, Tetrahedron Lett. 1977 ,

18 , 197 7 –1 980.

[22] We also tried a nucleophilic substitution of different sulfonate groups

of alcohol 11 (i.e., mesylate and nonaflate) with sodium azide; however,

an elimination product was also obtained, albeit in a smaller proportion

(the ratio of azide 6/elimination product was 3.9:1 for the mesylate de-

rivative).

[23] C. Thibaudeau, R. Stenutz, B. Hertz, T. Klepach, S. Zhao, Q. Wu, I. Carmi-

chael, A. S. Serianni, J. Am. Chem. Soc. 2004 , 126 , 15668—15685; equa-

tions: 1) 2.8P_gg+2.2P_gt+11.1P_tg=³J_C2H,C7HproS; 2) 0.9P_gg+10.8P_gt+4.7P_tg=

³J_C2H,C7HproR; 3) P_gg+P_gt+P_tg=1.

[10] E. G. von Roedern, H. Kessler, Angew. Chem. 1994 , 106 , 68 4 –6 86;

Angew. Chem. Int. Ed. 1994 , 33 , 68 7 –6 89.

[11] a) E. Locardi, M. Stçckle, S. Gruner, H. Kessler, J. Am. Chem. Soc. 2001 ,

123 , 818 9 –8 196; b) J. E. Campbell, E. E. Englund, S. D. Burke, Org. Lett.

2002 , 4 , 227 3 –2 275; c) M. Stçckle, G. Voll, R. Gꢆnther, E. Lohof, E. Locar-

di, S. Gruner, H. Kessler, Org. Lett. 2002 , 4 , 250 1 –2 50 4; d) T. K. Chakra-

borty, P. Srinivasu, E. Bikshapathy, R. Nagaraj, M. Vairamani, S. K. Kumar,

A. C. Kunwar, J. Org. Chem. 2003 , 68 , 625 7 –6 263; e) R. M. van Well, L.

Marinelli, C. Altona, K. Erkelens, G. Siegal, M. van Raaij, A. L. Llamas-Saiz,

[24] See the Supporting Information.

[25] Crystals of cyclodipeptides 1 and 2 were grown under identical condi-

tions, that is, from a solution of the cyclopeptide in dichloromethane

by vapor-phase equilibration with n-hexane.

[26] The structure of cyclodipeptides 1 and 2 in the solid state is not com-

pletely symmetrical; therefore, it is more appropriate to give an aver-

age of the dihedral angles.

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4021

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

[27] CCDC-949309 and -949308 (1 and 2, respectively) contains the supple-

mentary crystallographic data for this paper. These data can be ob-

tained free of charge from The Cambridge Crystallographic Data Centre

via www.ccdc.cam.ac.uk/data_request/cif.

[28] Chloroform is usually considered as a model for a lipophilic medium,

such as the cellular membrane.

[29] Similar values were previously reported, see: R. Vanderesse, V. Grand, D.

Limal, A. Vicherat, M. Marraud, C. Didierjean, A. Aubry, J. Am. Chem. Soc.

1998 , 120 , 944 4 –9 451.

[32] I. Kolossvꢁry, W. C. Guida, J. Am. Chem. Soc. 1996 , 118 , 501 1 –5 019.

[33] Maestro, Version 9.3 Schroꢈdinger, LLC, New York, NY, 2012.

[34] MacroModel, Version 9.9 Schroꢈdinger, LLC, New York, NY, 2012.

[35] G. A. Kaminski, R. A. Friesner, J. Tirado-Rives, W. L. Jorgensen, J. Phys.

Chem. B 2001 , 105 , 647 4 –6 487.

[36] W. L. F. Armarego, D. D. Perrin, Purification of Laboratory Chemicals, 4th

ed., Butterworth-Heinemann, Oxford, 1996.

[30] The 2-methoxy group in the acetamide represents the oxygen atom at

C7 of the adjacent residue.

Received: October 1, 2013

Revised: December 16, 2013

Published online on February 25, 2014

[31] S. H. Gellman, G. P. Dado, G.-B. Liang, B. R. Adam, J. Am. Chem. Soc.

1991 , 113 , 116 4 –1 173.

Chem. Eur. J. 2014, 20, 4007 – 4022

www.chemeurj.org

4022

ꢃ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Article Doi

Synthesis and conformational analysis of cyclic homooligomers from pyranoid ε-sugar amino acids

DOI: 10.1002/chem.201303841

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201303841

Full text of DOI:10.1002/chem.201303841

Products guided by the article

R&D Labs maybe for 1597406-78-4

Relevant to this article

Hot Product