Solid-phase synthesis using (Allyloxy)carbonyl(Alloc) chemistry of a putative heptapeptide intermediate in vancomycin biosynthesis containing m-chloro-3-hydroxytyrosine

Article abstract of DOI:10.1002/1522-2675(20000906)83:9<2572::AID-HLCA2572>3.0.CO;2-X

A convenient method for the solid-phase synthesis of putative linear heptapeptide intermediates in vancomycin biosynthesis is described, in particular, the heptapeptide D-Leu-Cyt-L-Asn-Hpg-Hpg-Cyt'-Dhpg (Cyt = (2R,3R)-m-chloro-3-hydroxytyrosine, Hpg = (R)-2-(p-hydroxyphenyl)glycine, Cyt' = (2S,3R)-m-chloro-3-hydroxytyrosine and Dhpg = (S)-2-(3,5-dihydroxyphenyl)glycine). The synthesis was performed on chlorotrityl resin and employed the (allyloxy)carbonyl protecting group for temporary N(α) protection during peptide-chain assembly.

Full text of DOI:10.1002/1522-2675(20000906)83:9<2572::AID-HLCA2572>3.0.CO;2-X

2572
Helvetica Chimica Acta ± Vol. 83 (2000)
Solid-Phase Synthesis Using (Allyloxy)carbonyl(Alloc) Chemistry of a
Putative Heptapeptide Intermediate in Vancomycin Biosynthesis Containing
m-Chloro-3-hydroxytyrosine
Preliminary Communication
by Ernst Freund, Francesca Vitali, Anthony Linden and John A. Robinson*
Institute of Organic Chemistry, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich
Dedicated to Professor Albert Eschenmoser on the occasion of his 75th birthday
A convenient method for the solid-phase synthesis of putative linear heptapeptide intermediates in
vancomycin biosynthesis is described, in particular, the heptapeptide d-Leu-Cyt-l-Asn-Hpg-Hpg-Cyt'-Dhpg
(Cyt ˆ (2R,3R)-m-chloro-3-hydroxytyrosine, Hpg ˆ (R)-2-(p-hydroxyphenyl)glycine, Cyt' ˆ (2S,3R)-m-chloro-
3-hydroxytyrosine and Dhpg ˆ (S)-2-(3,5-dihydroxyphenyl)glycine). The synthesis was performed on chloro-
trityl resin and employed the (allyloxy)carbonyl protecting group for temporary N(a) protection during
peptide-chain assembly.
1. Introduction. ± Recent studies of glycopeptide antibiotic biosynthesis have led to
the cloning of glycosyl transferase genes responsible for the addition of monosacchar-
ides to the vancomycin heptapeptide aglycon [1], from both vancomycin-and
chloroeremomycin-producing strains of Amycolatopsis orientalis. Also, 72 kb of
chromosomal DNA surrounding the glycosyl transferase genes from the chloroer-
emomycin producer has been sequenced [2]. This revealed 39 putative open reading
frames (orfs), including many encoding enzymes likely to be involved in the
biosynthesis of the antibiotic. Putative functions were assigned to these biosynthetic
genes based on sequence comparisons to enzymes of known function. The genes
encoding the peptide synthetase, required for assembly of the heptapeptide core, were
easily identified, and more recently, biochemical studies of the recombinant peptide
synthetase produced in E. coli were reported [3]. Another gene was shown to encode
an N-methyltransferase, responsible for methylating the N-terminal leucine residue in
vancomycin [4].
Of special interest here are the enzymes that catalyze the oxidative cross-linking of
the phenolic rings during vancomycin biosynthesis. Three closely linked P450-like orfs
were identified in the chloroeremomycin cluster [2], and a similar arrangement of three
P450 genes (oxyA, oxyB and oxyC) has been found in the balhimycin producer
Amycolatopsis mediterranei [5]. Insertional inactivation of these oxygenase gene(s)
resulted in a balhimycin mutant which was unable to produce the antibiotic, but,
instead, produced the linear heptapeptide 1, which is possibly the direct product of the
balhimycin peptide synthetase [5][6]. Hence, the P450-like orfs appear to be
responsible for the oxidative phenol-coupling reactions.
Presently, the biosynthetic route to glycopeptides of the vancomycin family is still
largely unknown. In particular, the timing of the phenol-coupling reactions, and the

Helvetica Chimica Acta ± Vol. 83 (2000)
2573

2574
Helvetica Chimica Acta ± Vol. 83 (2000)
roles and preferred substrates of the P450-like enzymes implicated in catalyzing these
steps remain to be defined. To address these questions, access is required to putative
heptapeptide substrates such as 1. In earlier work [7], we reported a convenient solid-
phase synthesis of the related heptapeptide 3. We now show that the solid-phase
methods developed earlier may also be used to allow convenient access to
heptapeptides such as 1 and 2.
2. Results and Discussion. ± The main concern in planning a solid-phase synthesis of
1 was the sensitivity of several of the constituent amino acids to epimerization. The
widely used solid-phase Fmoc- or Boc-chemistries, which require basic or acidic
conditions at each step of chain elongation, appeared to be unsuitable for the synthesis.
For this reason, we favoured the use of Alloc1) chemistry [8], and could show in earlier
work that 3 can be prepared in good yield by this methodology [7]. To apply the same
approach here, we first set out to synthesize the suitably protected building blocks 4 ± 10
(11 is available commercially).
2.1. Synthesis of Amino-Acid Building Blocks 4 ± 10. Compound 4 was prepared
from (S)-3,5-dihydroxybenzoic acid following a procedure of Evans et al. [9], as shown
in Scheme 1. Homologation of acid 12 to 13 was achieved by the Arndt-Eistert reaction
according to a published procedure [10]. After coupling to the chiral auxiliary (S)-4-
benzyloxazolidin-2-one, the asymmetric aziridination was performed by the method of
Evans et al. [9]. After reduction of the azide group to an amine, protection of the amino
group, and removal of the chiral auxiliary, the product 4 was obtained in good yield, in
1
crystalline form, with > 98% optical purity (determined by H-NMR after coupling 4
to (S)-and ( R)-1-phenylethylamine).
1
)
Abbreviations: Alloc: allyloxycarbonyl, Cyt: m-chloro-3-hydroxytyrosine, Dhpg: 2-(3,5-dihydroxyphe-
nyl)glycine, HATU: 2-(1H-7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate,
HOAt: 1-hydroxy-7-azabenzotriazole, Hpg: 2-(4-hydroxyphenyl)glycine, Mtt: 4-methyltrityl, NMM: N-
methylmorpholine.

Helvetica Chimica Acta ± Vol. 83 (2000)
2575
Scheme 1
i) BnCl, K2CO3, DMF, r.t. ii) KOH, EtOH/H2O, r.t.; 97% over 2 steps. iii) SOCl2. iv) CH2N2, Et3N, THF 08;
81% over 2 steps. v) MeOH, silver benzoate, Et3N. vi) 1m KOH, MeOH/H2O 9 :1; 79% over 2 steps. vii) Et3N,
pivaloyl chloride, THF, then (S)-4-benzyloxazolidin-2-one, BuLi; 89%. viii) K salt of 1,1,1,3,3,3-hexamethyldi-
silazane (KHMDS), trisyl azide (trisyl ˆ 2,4,6-triisopropylbenzenesulfonyl), THF, À 788, then AcOH; 81%. ix)
SnCl2, dioxane/H2O. x) Allyl chloroformate, dioxane/aq. NaHCO3; 77% over 2 steps. xi) LiOH, THF/H2O 3 :1; 88%.
Scheme 2
i) 1.1 equiv. BuLi, 1.1 equiv. Ti(NEt2)3Cl, 1.2 equiv. aldehyde, À 788; 76%. ii) Ac2O, pyridine. iii) 3 equiv. TFA,
H2O/MeCN. iv) Allyl chloroformate, NaHCO3; 60% over three steps. v) K2CO3, MeOH/water; 91%.
The synthesis of 5 and 6 was achieved using the method of Schöllkopf and
coworkers [11][12]. The titanium derivative of the Schöllkopf bis-lactim ether reacted
smoothly with both required aldehydes (Scheme 2), to afford almost exclusively the
required stereoisomers 16 and 17. These adducts were then converted to 5 and 6 by a
slight adaptation of published procedures [11][12].
The amino-acid derivative 7 was prepared from d-(2-(4-hydroxyphenyl)glycine, by
first alkylating the phenol group with BnBr, following the method of Wünsch et al. [13].
Then, the Alloc group was introduced by standard methods. The enantiomeric purity of
the product (> 97%) was analyzed by 1H-NMR after coupling the product to optically
pure 1-phenylethylamine. Compound 8 was prepared from commercially available
Fmoc-Asn(Mtt)-OH, by exchanging the Fmoc group for Alloc.
The synthesis of 9 and 10 was performed by the route described by Rao et al. [14].
The azide intermediates 20 and 21 were crystalline (Scheme 3). This allowed
recrystallization to improve the optical purity to close to 100% (determined by
1H-NMR with a chiral solvating agent). Also, both the relative and absolute
configuration of 20 was confirmed by single-crystal X-ray diffraction, as was (only)
the relative configuration of 21 (see Fig.). The azides 20 and 21 were then transformed
in straightforward steps into 9 and 10, respectively (Scheme 3).

2576
Helvetica Chimica Acta ± Vol. 83 (2000)
Scheme 3
i) SnCl2, dioxane/H2O. ii) Allyl chloroformate, NaHCO3; 83% over the last two steps. iii) LiOH ´ H2O, THF/
H2O; 91%.
2.2. Peptide Synthesis. The solid-phase synthesis of peptide 1 was performed as
described in [7] on 2-chlorotrityl-chloride resin. The first amino acid, 4, was coupled to
the resin in CH2Cl2 with NMM as base, without significant epimerisation. The peptide
chain was then extended using for each coupling HATU/HOAt (4 equiv.) and the
required amino-acid derivative, 5 ± 11. For Alloc removal, Pd(PPh3)4 and PhSiH3 were
used, as outlined in Scheme 4. The assembly of the heptapeptide was monitored after
each amino-acid coupling and Alloc-removal step, by cleaving a small sample of inter-
mediate from the resin and analysis by reversed-phase HPLC. This revealed a single
component at the dipeptide stage, but, after coupling of 7, two components were detected
in a 4 :1 ratio, presumably due to partial epimerization during coupling of this amino
acid. The same ratio of two heptapeptide epimers was also observed at the end of the
synthesis. Upon completion of the synthesis, the side-chain-protected peptide was cleaved
from the solid support using a mixture of CH2Cl2/CF3CH2OH/AcOH (see Scheme 4).
The major component of the protected heptapeptide product was then treated with
CF3COOH/thioanisole/(i-Pr)3SiH to remove all protecting groups, and the product 1
was purified by reversed-phase HPLC in an overall yield (not optimized) of ca. 13%.
The heptapeptide 1 was characterized by MS and NMR. The electrospray MS
showed a main molecular ion at m/z 1135, as well as peaks at higher mass (m/z 1136,
1
1137, 1138, 1139 and 1140), as expected for the isotopes of Cl. The H-NMR spectra
were recorded in H2O (H2O :D2O 9 :1)/CD3OD 5 :2 (at 300 K), which gave sharp
peaks also for the amide resonances. Analysis of 2D-NOESY, TOCSY and DQF-
COSY spectra allowed a complete assignment of all non-exchangeable resonances2),
1H-NMR Assignment for the heptapeptide 1 in (H2O/D2O 9 :1)/CD3OD 5 :2 at 300 K (pH* ca. 5): 8.49
(weak d, NH, Asn3); 8.40 (d, NH, Hpg4); 8.30 (d, NH, Dhpg7); 8.29 (d, NH, Hpg5); 8.24 (d, NH, Cyt6); 7.45
(s, NH(d) (trans), Asn3); 7.23 (d, HÀC(2), HÀC(6), Hpg4); 7.23 (d, HÀC(2), Cyt2); 7.15 (d, HÀC(2),
Cyt6); 6.99 (dd, HÀC(6), Cyt2); 6.98 (d, HÀC(2), HÀC(6), Hpg5); 6.92 (d, HÀC(5), Cyt2); 6.90 (dd,
HÀC(6), Cyt6); 6.83 (d, HÀC(3), HÀC(5), Hpg4); 6.79 (s, NH(d) (cis), Asn3); 6.76 (d, HÀC(3), HÀC(5),
Hpg5); 6.70 (d, HÀC(5), Cyt6); 6.51 (d, HÀC(2), HÀC(6), Dhpg7); 6.30 (t, HÀC(4), Dhpg7); 5.51 (d,
HÀC(a), Hpg4); 5.38 (d, HÀC(a), Hpg5); 5.18 (d, HÀC(b), Cyt6); 5.04 (d, HÀC(a), Dhpg7); 4.90 (d,
HÀC(b), Cyt2); 4.73 (m, HÀC(a), Asn3); 4.72 (m, HÀC(a), Cyt6); 4.45 (d, HÀC(a), Cyt2); 3.86 (t,
HÀC(a), Leu1); 2.77 (dd, 1 HÀC(b), Asn3); 2.71 (dd, 1 HÀC(b), Asn3); 1.38 (m, HÀC(g), HÀC(b),
Leu1); 0.78 (d, 1 HÀC(d), Leu1); 0.75 (d, 1 HÀC(d), Leu1). The assignments were made by 2D-TOCSY,
DQF-COSY and NOESY experiments at 600 MHz. The connectivity within each spin system was
2
)
confirmed by intra-residue J couplings and NOEs, and the amino acid sequence by inter-residue da (i, i ‡ 1)
N
NOEs. The NH resonances for Leu1 and Cyt2 were not observed due to exchange.

Helvetica Chimica Acta ± Vol. 83 (2000)
2577
Figure. X-Ray crystal structures of the intermediates 20 and 21 (Scheme 3)

2578
Helvetica Chimica Acta ± Vol. 83 (2000)
Scheme 4
i) For coupling 4 (Dhpg), 5 (Cyt), 7 (Hpg), 8 (Asn), 9 (Cyt) and 11 (each 4 equiv.), HATU and HOAt (each
4 equiv.), NMM, DMF, and for Alloc deprotection Pd(PPh3)4 (0.5 equiv.) and PhSiH3 (30 equiv.) in CH2Cl2. ii)
CF3CH2OH/AcOH/CH2Cl2 1:1:3. iii) TFA/thioanisole 3 :1 ‡ 5% (i-Pr)3SiH, r.t., 2 h., 13% yield over whole
synthesis.
and also provided confirmation of the constitution of the molecule. No long-range
NOEs were observed in NOESY spectra, indicating that the peptide is not folded under
these conditions. Finally, the assignments made from 1D-1H and 2D-NOESY, TOCSY
and DQF-COSY-NMR spectra recorded in (D6)DMSO were essentially identical to
those reported earlier [6] for 1 isolated from a mutant of the balhimycin producer.
In summary, a convenient method is decribed for the solid-phase synthesis, using
Alloc chemistry, of putative intermediates in the biosynthesis of vancomycin, in
particular, the heptapeptide 1 containing, amongst other amino acids, 2-(3,5-
dihydroxyphenyl)glycine, 2-(4-hydroxyphenyl)glycine and m-chloro-3-hydroxytyro-
sine. The method should allow rapid access also to a variety of related heptapeptides
that may prove useful in studies of the later steps in glycopeptide antibiotic
biosynthesis.

Helvetica Chimica Acta ± Vol. 83 (2000)
2579
Crystallographic data (excluding structure factors) for the structures of 20 and 21 reported in this paper
have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication Nos.
CCDC-148683 and CCDC-148684, respectively. Copies of the data can be obtained, free of charge, on
application to the CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax: ‡ 44-(0)1223-336033 or e-mail:
deposit@ccdc.cam.ac.uk).
The authors thank the Swiss National Science Foundation for financial support.
REFERENCES
[1] P. J. Solenberg, P. Matsushima, D. R. Stack, S. C. Wilkie, R. C. Thompson, R. H. Baltz, Chem. Biol. 1997, 4,
195.
[2] A. M. A. van Wageningen, P. N. Kirkpatrick, D. H. Williams, B. R. Harris, J. K. Kershaw, N. J. Lennard, M.
Jones, S. J. M. Jones, P. J. Solenberg, Chem. Biol. 1997, 5, 155.
[3] J. W. Trauger, C. T. Walsh, Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 3112.
[4] D. P. OꢀBrien, P. N. Kirkpatrick, S. W. OꢀBrien, T. Staroske, T. I. Richardson, D. A. Evans, A. Hopkinson,
J. B. Spenser, D. H. Williams, Chem. Commun. 2000, 103.
[5] S. Pelzer, R. Sussmuth, D. Heckmann, J. Recktenwald, P. Huber, G. Jung, W. Wohlleben, Antimicrob.
Agents Chemother. 1999, 43, 1565.
[6] R. D. Sussmuth, S. Pelzer, G. Nicholson, T. Walk, W. Wohlleben, G. Jung, Angew. Chem., Int. Ed. 1999, 38,
1976.
[7] E. Freund, J. A. Robinson, Chem. Commun. 1999, 2509.
Â
[8] F. Guibe, Tetrahedron 1998, 54, 2967 ± 3042.
[9] D. A. Evans, T. C. Britton, J. A. Ellman, R. L. Dorow, J. Am. Chem. Soc. 1990, 112, 4011.
[10] N. Katagiri, J. Nakano, K. Tetsuzo, J. Chem. Soc., Perkin Trans I 1981, 2710.
[11] T. Beulshausen, U. Groth, U. Schöllkopf, Liebigs Ann. Chem. 1991, 1207.
[12] U. Schöllkopf, T. Beulshausen, Liebigs Ann. Chem. 1989, 223.
[13] E. Wünsch, G. Fries, A. Zwick, Chem. Ber. 1958, 91, 542.
[14] A. V. Rama Rao, T. K. Chakraborty, K. L. Reddy, A. S. Rao, Tetrahedron Lett. 1994, 35, 5043 ± 5046.
Received August 21, 2000