A synthetic receptor for the Cbz-L-Ala-L-Ala-OH dipeptide sequence

Article abstract of DOI:10.1039/a903449h

A novel bowl-shaped macrobicyclic receptor has been prepared and is a particularly strong and selective receptor for Cbz-L-Ala-L-Ala-OH(-ΔG(ass) = 25 kJ mol^-1 at 293 K in CDCL₃).

Full text of DOI:10.1039/a903449h

A synthetic receptor for the Cbz-l-Ala-l-Ala-OH dipeptide sequence
Peter D. Henley and Jeremy D. Kilburn*
Department of Chemistry, University of Southampton, Southampton, UK SO17 1BJ. E-mail: jdk1@soton.ac.uk
Received (in Liverpool, UK) 28th April 1999, Accepted 26th May 1999
NHR
O
H
N
H
N
NHR
A novel bowl-shaped macrobicyclic receptor has been
prepared and is a particularly strong and selective receptor
for Cbz-l-Ala-l-Ala-OH (2DGass = 25 kJ mol21 at 293 K in
CDCl3).
H2N
N
NH2
i, ii
N
O
MeO2C
CO2Me
3
4
R = Z 81% overall
R = H 96%
iii
CO2Bn
CO2Bn
The development of synthetic receptors for biologically rele-
vant substrates such as peptides1 continues to be a fertile area of
research. The bacterial cell wall precursor peptide l-Lys-d-Ala-
d-Ala-OH represents a particularly interesting target because a
selective receptor2 for this peptide sequence might lead to
mimics for the vancomycin family of antibiotics.3 We have
described several systems designed to act as receptors for
peptides specifically with a free carboxylic acid terminus1a,1e,4
and in particular we have prepared a series of receptors 1 which
BocHN
BocHN
iv, v
vi
CO2R
NHBoc
OH
CO2H
5 63% overall
O
H
H
BocHN
RO2C
N
H
N
H
N
N
O
N
O
O
MeO2C
CO2Me
RS
RS
O
O
H
H
N
N
6
7
R = Bn 41%
R = H 98%
N
H
O
O
H
iii
N
H
O
O
H
N
N
CO2Bn
N
H
O
N
H
CO2R1
O
CO2R1
viii
CBS
O
H
H
O
N
H
H
N
N
CBS = carboxylate binding site
RS = rigid spacer
N
N
1
O
O
Ph
Ph
NH2
8
2
O
N
N
H
O
H
feature a specific binding site for the carboxylic acid terminus of
peptide guests at the base of a bowl-shaped cavity.4 Thus we
have described macrobicycle 2, which incorporates a diamido-
pyridine unit to serve as a carboxylic acid binding site and is an
effective receptor for certain dipeptides, in particular for Cbz-b-
Ala-d-Ala-OH, in chloroform solution.4b Despite the encourag-
ing binding selectivities found for 2, NMR and molecular
modelling studies revealed it to be a highly flexible molecule,
particularly in the portion derived from glutamic acid and
phenylalanine, which links the diamidopyridine unit and the rim
of the cavity. Thus 2 is far from preorganised for optimal
binding of peptide guests. Here we describe a new bowl-shaped
macrobicyclic receptor 13, with increased rigidity in compar-
ison to 2, which has proved to be a particularly strong and
selective receptor for Cbz-l-Ala-l-Ala-OH.
R2HN
NHR2
H
H
O
N
H
N
H
N
O
N
N
O
O
MeO2C
CO2Me
9
R1 = Bn, R2 = Boc 76%
viii
ix
10 R1 = OH, R2 = Boc
11 R1 = OC6F5, R2 = Boc
x
12 R1 = OC6F5, R2 = H2+ CF3CO2
–
xi
In order to produce a more rigid variant of our bowl-shaped
receptors, we chose to link the bisarylmethane-derived rim used
previously with a tyrosine derived aromatic unit, and a glutamic
acid residue so that ultimately the side-chain methyl ester could
be functionalised to give a water-soluble receptor.
O
Hd
N
N
H
O
O
Hc
N
H
N
O
Ho
Hp
Hq
Ht
Hr
The synthesis of macrobicycle 13 follows the strategy for the
synthesis of 2 and related macrobicycles, with a double
intramolecular cyclisation of a suitably activated acyclic
precursor 12 as the key step (Scheme 1). Assembly of the
cyclisation precursor then requires coupling together of the
various amino acid building blocks around the carboxylic acid
binding site, in this case diaminopyridine. Thus Cbz-l-glutamic
acid g-methyl ester was converted to the corresponding acid
fluoride, using cyanuric fluoride and pyridine,5 and coupled to
diaminopyridine, which had been pretreated with N,O-bis-
(trimethylsilyl)acetamide6 to give 3 in 81% yield. Removal of
the Cbz-protecting groups to yield the corresponding diamine 4
was readily achieved with Pd/C/H2. Acid 5 was prepared by a
Hs
H
Hb
Ha
O
N
H
N
N
O
N
N
O
O
MeO2C
CO2Me
13 6% overall from 10
Scheme 1 Reagents and conditions: i, N,O-bis(trimethylsilyl)acetamide; ii,
Cbz-l-Glu(OMe)-F, TBAF; iii, Pd, C, H2, MeOH; iv, Tf2O, pyridine,
CH2Cl2; v, Pd(OAc)2, Ph2P(CH2)3PPh2, KOAc, CO, DMSO, 65 °C, then 1
M HCl; vi, 5, EDC, HOBT, DMAP, DMF; vii, 8, EDC, HOBT, DMAP,
DMF; viii, Pd, C, H2, DMF; ix, F5C6OH, EDC, DMAP, DMF–THF (1:3
v/v); x, 10% TFA, CH2Cl2; xi, slow addition to a refluxing solution of
Pri2NEt in MeCN.
Chem. Commun., 1999, 1335–1336
1335

Table 1 Binding constants (Kass) and free energies of complexation
(2DGass) for the 1+1 complexes formed between macrobicycle 13 and
various substrates, in CDCl3 at 20 °C
but significant upfield shifts of both ArHq and ArHr (Dd =
0.1–0.2 ppm).
The association constants for the various guests studied
indicate that the receptor, while able to bind a range of
substrates, is a particularly strong receptor for Cbz-l-Ala-l-
Ala-OH (2DGass = 25.3 kJ mol21). The receptor is very
selective for this substrate as evidenced by the fact that Cbz-l-
Ala-l-Ala-OH is bound > 4 kJ mol21 more strongly than Cbz-
d-Ala-d-Ala-OH and ~ 7 kJ mol21 more strongly than Cbz-
Gly-l-Ala-OH. The latter result is particularly notable since the
difference in binding energies ( ~ 7 kJ mol21) is the con-
sequence of replacing a single proton with a methyl group at the
second residue of the dipeptide (Gly ? l-Ala), and is probably
a consequence of the methyl group establishing a stabilising
interaction with the bisarylmethane unit in the rim of the
macrobicycle—which is evidenced by the large upfield shift of
the signal for the methyl group (2Dd ≈ 0.7 ppm) on
complexation with macrobicycle 13.11
Entry
Substrate
Kassa/M21
2DGass/kJ mol21
1
2
3
4
5
6
7
8
Phenylacetic acid
Cbz-l-Ala-OH
Cbz-d-Ala-OH
Cbz-b-Ala-l-Ala-OH
Cbz-l-Ala-l-Ala-OH
Cbz-d-Ala-d-Ala-OH
Cbz-Gly-l-Ala-OH
Cbz-Gly-d-Ala-OH
2100 ± 100
5300 ± 500
800 ± 70
2500 ± 130
33000 ± 3200
4500 ± 300
1800 ± 150
3100 ± 370
18.6 ± 0.1
20.9 ± 0.3
16.3 ± 0.3
19.0 ± 0.1
25.3 ± 0.3
20.5 ± 0.2
18.3 ± 0.2
19.6 ± 0.3
a Errors were estimated from the quality of the fit of the experimental data
to the calculated, by carrying out several titration experiments and by
monitoring the shift of several protons (Ha, Ho, Hr) to obtain several
estimates of Kass and averaging the values obtained.
Thus the increased rigidity, and hence preorganisation,
introduced into the structure of macrobicycle 13 has provided a
stronger and considerably more selective receptor than the
previously described macrobicycle 2 and, whereas the latter was
a better receptor for b-alanyl-d-a-amino acid substrates,
macrobicycle 13 is selective for dipeptides derived from two a-
amino acids, and in particularly for the significant l-Ala-l-Ala-
OH sequence.
palladium catalysed carbonylation of the triflate derived from
Boc-l-tyrosine benzyl ester.7 In the presence of potassium
acetate8 and the bidentate dppp ligand,9 the carbonylation
worked well to give the desired acid 5 in 66% yield after
aqueous acid work-up. Acid 5 was then coupled to diamine 4
using 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydro-
chloride (EDC) and 1-hydroxybenzotriazole hydrate (HOBT),
and the resulting diester 6 was debenzylated to give the diacid
7, which in turn was coupled with the previously described
amine 84b to give the protected cyclisation precursor 9.
Debenzylation of 9 and formation of the bispentafluorophenyl
ester was followed by removal of the Boc protecting groups to
give the bisamine salt 12. Slow addition of 12 to a refluxing
solution of Pri2NEt in MeCN gave the desired macrobicycle 13,
but in only 6% yield overall from diacid 10, after purification by
column chromatography. The yield of the final cyclised product
was disappointing and in previous syntheses of related com-
pounds4 the analogous sequence typically yielded ~ 30% of
macrobicycle, but in the present case the low yield is probably
attributable to the greater rigidity of the precursor and of the
resulting macrobicycle.
We thank the EPSRC for a studentship for P. H. and Dr
Tobias Braxmeier and Dr Herbert Ro¨ttele (Karlsruhe Uni-
versity) for carrying out additional NMR experiments.
Notes and references
1 For recent examples of peptide receptors see (a) J. Dowden, P. D.
Edwards, S. S. Flack and J. D. Kilburn, Chem. Eur J., 1999, 5, 79; (b)
Md. A. Hossain and H.-J. Schneider, J. Am. Chem. Soc., 1998, 120,
11 208; (c) R. Breslow, Z. Yang, R. Ching, G. Trojandt and F. Odobel,
J. Am. Chem. Soc., 1998, 120, 3536; (d) D. W. P. M. Löwik, M. D.
Weingarten, M. Broekema, A. J. Brouwer, W. C. Still and R. M. J.
Liskamp, Angew. Chem., 1998, 37, 1846; (e) M. Davies, M. Bonnat, F.
Guillier, J. D. Kilburn and M. Bradley, J. Org. Chem., 1998, 63,
8696.
2 The in vitro antimicrobial activity, against selected Gram-positive
bacteria, of a calixarene derived receptor for the dipeptide sequence d-
Ala-d-Ala-OH has been described: A. Casnati, M. Fabbi, N. Pelizzi, A,
Pochini, F. Sansone, R. Ungaro, E. Di Modugno and G. Tarzia, Bioorg.
Med. Chem. Lett., 1996, 6, 2699.
3 H. R. Perkins, Biochem. J., 1969, 111, 195; D. H. Williams, Nat. Prod.
Rep., 1996, 13, 469; D. H. Williams, M. S. Searle, M. S. Westwell, J. P.
Mackay, P. Groves and D. A. Beauregard, Chemtracts: Org. Chem.,
1994, 7, 133.
4 (a) G. J. Pernia, J. D. Kilburn, J. W. Essex, R. J. Mortishire-Smith and
M. Rowley, J. Am. Chem. Soc., 1996, 118, 10 220; (b) C. P. Waymark,
J. D. Kilburn and I. Gillies, Tetrahedron Lett., 1995, 36, 3051.
5 L. A. Carpino, D. Sadat-Aalaee, H. G. Chao and R. H. DeSelms, J. Am.
Chem. Soc., 1990, 112, 9651.
6 S. Rajeswari, R. J. Jones and M. P. Cava, Tetrahedron Lett., 1987, 28,
5099.
7 J. W. Tilley, R. Sarabu, R. Wagner and K. Mulkerins, J. Org. Chem.,
1990, 55, 906; X. Creary, B. Benage and K. Hilton, J. Org. Chem., 1983,
48, 2887.
8 S. Cacchi and A. Lupi, Tetrahedron Lett., 1992, 33, 3939.
9 R. E. Dolle, S. J. Schmidt and L. I. Kruse, J. Chem. Soc., Chem.
Commun., 1987, 904.
10 The binding constants were calculated by fitting the data to a 1+1
binding isotherm using NMRTit HG software, kindly provided by
Professor C. A. Hunter, University of Sheffield. See: A. P. Bisson, C. A.
Hunter, J. C. Morales and K. Young, Chem. Eur. J., 1998, 4, 845.
11 In view of the selectivity of macrobicycle 13 for Cbz-l-Ala-l-Ala-OH
over Cbz-d-Ala-d-Ala-OH (Table 1, entries 5 and 6), and for Cbz-l-
Ala-OH over Cbz-d-Ala-OH (entries 2 and 3), the preference for Cbz-
Gly-d-Ala-OH over Cbz-Gly-l-Ala-OH (entries 7 and 8) is somewhat
surprising. The origin of this selectivity is unclear at this stage and will
need to be examined in further studies.
1
Macrocycle 13 gave a well-resolved H NMR spectrum in
CDCl3 which could be fully assigned with the help of 2D NMR
experiments. Binding studies on macrobicycle 13 were there-
fore carried out with a number of Cbz-protected alanine derived
amino acids and dipeptides in CDCl3, using a standard 1H NMR
titration experiment, monitoring the shift of various signals, and
analysing the resultant binding curves (Table 1).10 In each
binding experiment a 1:1 binding stoichiometry has been
assumed which was generally supported by the good fit of the
measured data to the theoretical model, on analysis. In each
titration experiment significant downfield shifts of the signal for
NHa were observed (Dd = 0.3–0.8 ppm), consistent with a
strong association between the carboxylic acid and the amido-
pyridine moiety. Addition of corresponding methyl esters to a
1
solution of 13 led to no significant changes in the H NMR
spectrum for 13, further confirming the importance of the
carboxylic acid–amidopyridine interaction in the observed
binding. The addition of all carboxylic acid substrates, with the
exception of phenylacetic acid (Table 1, entries 2–8), also led to
large downfield shifts of the signals for NHc (Dd = 0.4–0.9
ppm) and addition of the dipeptide substrates (entries 5–8) led
to large downfield shifts of NHd (Dd
= 0.1–0.5 ppm)
implicating these NHs in hydrogen bonding with the respective
substrates. (The signal for NHb was obscured by the aromatic
region and could not be accurately monitored.) Significant
shifts in the signals for the aromatic protons were also observed.
In particular, addition of substrates incorporating an l-Ala-OH
moiety (entries 2, 4, 5 and 7) led to an upfield shift of ArHp (Dd
= 0.1–0.2 ppm) and a downfield shift of ArHt (Dd = 0.1– 0.3
ppm), whereas addition of substrates incorporating d-Ala-OH
(entries 3, 6 and 8) gave much smaller shifts of ArHp and ArHt,
Communication 9/03449H
1336
Chem. Commun., 1999, 1335–1336