Chemoenzymatic synthesis of a characteristic phosphorylated and glycosylated peptide fragment of the large subunit of mammalian RNA polymerase II

Article abstract of DOI:10.1021/ja970709e

The covalent modification of proteins by phosphorylation and addition of GlcNAc residues are important regulatory processes which mediate biological signal transduction. For instance, the cytosolic form of RNA polymerase II is heavily glycosylated but dur

Full text of DOI:10.1021/ja970709e

6702
J. Am. Chem. Soc. 1997, 119, 6702-6710
Chemoenzymatic Synthesis of a Characteristic Phosphorylated
and Glycosylated Peptide Fragment of the Large Subunit of
Mammalian RNA Polymerase II
Torsten Pohl and Herbert Waldmann*
Contribution from the Department of Organic Chemistry, UniVersity of Karlsruhe,
Richard-Willsta¨tter-Allee 2, D-76128 Karlsruhe, Germany
ReceiVed March 5, 1997X
Abstract: The covalent modification of proteins by phosphorylation and addition of GlcNAc residues are important
regulatory processes which mediate biological signal transduction. For instance, the cytosolic form of RNA polymerase
II is heavily glycosylated but during its transition from an initiating to an elongating complex the carbohydrates are
removed and the protein is phosphorylated. For the study of such biological phenomena, characteristic peptides
which embody both types of modifications may serve as efficient tools. However, their synthesis is complicated by
their pronounced acid and base lability as well as their multifunctionality. These properties make the application of
protecting groups necessary which can be removed under the mildest conditions. For the construction of such peptide
conjugates the enzyme labile PhAcOZ urethane blocking group was developed. This protecting group embodies (a)
a functional group (a phenylacetate) that is recognized by the biocatalyst (penicillin G acylase) and that is bound by
an enzyme labile linkage (an ester) to (b) a functional group (a p-hydroxybenzyl urethane) that undergoes a spontaneous
fragmentation upon cleavage of the enzyme-sensitive bond resulting in (c) the liberation of a carbamic acid derivative
which decarboxylates to give the desired peptide or peptide conjugate. When this enzymatic protecting group technique
was combined with classical chemical methods, a complex phosphoglycohexapeptide was built up, which embodies
two glycosylated, one phosphorylated, and one underivatized hydroxyamino acid. This peptide represents a
characteristic partial structure of the repeat sequence of the large subunit of RNA polymerase II which becomes
glycosylated or phosphorylated while the enzyme carries out its biological functions. The conditions under which
the enzymatic deprotections proceed are so mild that no undesired side reaction is observed (i.e., no rupture or
anomerization of the glycosidic bonds and no â-elimination of the phosphate or a carbohydrate occur). In addition,
the specificity of the biocatalyst guarantees that the peptide bonds and the other protecting groups present are not
attacked either.
Covalently modified proteins mediate the transduction of
physiological stimuli. The limited data available to date suggest
that the attachment and removal of N-acetylglucosamine is used
by cells as a regulatory mechanism which is not only similar
but often reciprocal to O-phosphorylation (i.e., phosphorylation
and glycosylation by O-GlcNAc show a so-called “yin-yang”
relationship).3,8 Thus, O-GlcNAc glycoproteins often are also
phosphoproteins in other sections of the cell cycle with the sites
of O-GlcNAc addition being virtually indistinguishable from
the sites of phosphorylation. An illustrative example is provided
by mammalian RNA polymerase II (Scheme 1).9
signals from the extracellular space across the plasma membrane
into the interior of cells and ultimately to the cell nucleus.1
Among the different types of covalent protein modification, in
particular, the dynamic phosphorylation and dephosphorylation
of hydroxyamino acids mediates the intracellular response to a
wide variety of extracellular stimuli.2 In addition, recently the
â-O-glycosidic attachment of single N-acetylglucosamine (O-
GlcNAc) residues to serine and threonine units of intracellular
proteins was discovered as a second type of modification.
Glycosylation with O-GlcNAc is analogous to phosphorylation
in many ways, in particular in terms of its dynamic character
and its ubiquitous occurrence.3 O-GlcNAc is found in all
eukaryotic cells on numerous proteins (e.g., on transcription
factors,4 oncogene products,5 and enzymes6 as well as on viral
proteins).7 The state of glycosylation of individual proteins is
often modified during the cell cycle or in response to specific
In its cytosolic form the C-terminal domain (CTD) of its large
subunit is heavily glycosylated, probably to facilitate the
transport of the enzyme into the cell nucleus. During the
intranuclear transition of RNA polymerase II from an initiating
(4) (a) Jackson, S. P.; Tjian, R. Cell 1988, 55, 125. (b) Lichtsteiner, S.;
Schibler, U. Cell 1989, 57, 1179. (c) Reason, A.; Dell, A.; Morris, H. R.;
Panico, M.; Treisman, R.; Marais, R.; Hart, G. W.; Haltiwanger, R. S.
Glycoconjugate J. 1991, 8, 211.
(5) (a) Privalsky, M. L. J. Virol. 1990, 64, 463. (b) Chou, T.-Y.; Hart,
G. W.; Dang, C. V. J. Biol. Chem. 1995, 270, 18961.
(6) (a) Datta, B.; Ray, M. K.; Chakrabarti, D.; Wylie, D. E.; Gupta, N.
K. J. Biol. Chem. 1989, 264, 20620. (b) Maikrantz, W.; Smith, D. M.;
Sladicka, M. M.; Schlegel, R. A. J. Cell. Sci. 1991, 98, 303.
(7) (a) Mullis, K. G.; Haltiwanger, R. S.; Hart, G. W.; Marchase, R. B.;
Engler, J. A. J. Virol. 1990, 64, 5317. (b) Whitford, M.; Faulkner, P. J.
Virol. 1992, 66, 3324. (c) Gonzalez, S. A.; Burrone, O. R. Virology 1992,
182, 8.
X Abstract published in AdVance ACS Abstracts, July 1, 1997.
(1) (a) Hunter, T.; Karin, M. Cell 1992, 70, 375. (b) Levitzki, A. Eur. J.
Biochem. 1994, 226, 1. (c) Casey, P. J. Science 1995, 268, 221. (d) Egan,
S. E.; Weinberg, R. A. Nature 1993, 365, 781. (e) Boguski, M. S.;
McCormick, F. Nature 1993, 366, 643.
(2) (a) Krebs, E. Angew. Chem. 1993, 105, 1173; Angew. Chem., Int.
Ed. Engl. 1993, 32, 1122. (b) Fischer, E. H. Angew. Chem. 1993, 105, 1181;
Angew. Chem., Int. Ed. Engl. 1993, 32, 1130. (c) Bridges, A. J.
Chemtracts: Org. Chem. 1995, 8, 73.
(3) (a) Hart, G. W.; Haltiwanger, R. S.; Holt, G. D; Kelly, W. G. Ann.
ReV. Biochem. 1989, 58, 841. (b) Hart, G. W.; Kelly, W. G.; Blomberg, M.
A.; Roquemore, E. P.; Dong, L.-Y.; Kreppel, L.; Chou, T.-Y.; Snow, D.;
Greis, K. Glyco- and Cellbiology, 44; Colloquium Mosbach 1993; Springer-
Verlag: Berlin, Heidelberg, 1994.
(8) Haltiwanger, R. S.; Kelly, W. G.; Roquemore, E. P.; Blomberg, M.
A.; Dong, L.-Y. D.; Kreppel, L.; Chou, T.-Y.; Greis, K.; Hart, G. W.
Biochem. Soc. Trans. 1992, 20, 264.
(9) Kelly, W. G.; Dahmus, M. E.; Hart, G. W. J. Biol. Chem. 1993,
268, 10416.
S0002-7863(97)00709-9 CCC: $14.00 © 1997 American Chemical Society

Chemoenzymatic Synthesis of a Peptide Fragment
J. Am. Chem. Soc., Vol. 119, No. 29, 1997 6703
Scheme 1. Modification of the Cytosolic and the
Nucleosolic Forms of RNA Polymerase II by Glycosylation
with O-GlcNAc and by Phosphorylation
Scheme 2. Acid and Base Lability of Phospho- and
Glycopeptides
of phosphoglycopeptides that embody both modifications, these
difficulties are potentiated. Consequently, many of the classical
chemical protecting groups can not be applied in the construction
of phosphoglycopeptides, and to date their synthesis has not
been described. Enzymatic protecting group techniques12 offer
viable alternatives to the established classical chemical methods.
Thus, on the one hand enzymatic transformations can be carried
out under characteristically mild reaction conditions (pH 6-8,
room temperature). In addition, enzymes often combine a high
specificity for the structures they recognize and the reactions
they catalyze with a broad substrate tolerance. These properties
have opened up new and advantageous routes to sensitive and
multifunctional lipo-,13 glyco-,14 phospho-,15 and nucleopep-
tides.16 In this paper, we report that by means of a combination
of enzyme labile and classical protecting groups phosphogly-
copeptides can be built up efficiently.17
to an elongating transcription complex the O-GlcNAc glycosides
are removed and the protein is rapidly and extensively phos-
phorylated.
The fact that only the unphosphorylated form of RNA
polymerase II contains O-GlcNAc suggests that phosphorylation
and glycosylation are separate events and that the CTD of RNA
polymerase II appears to exist in three distinct states: unmodi-
fied, phosphorylated, and glycosylated with O-GlcNAc. But
in succession of conversion a partly glycosylated and partly
phosphorylated intermediate has to appear. Thus, O-GlcNAc
may either mask or tag phosphorylation sites. Therefore, the
covalent modifications with both O-GlcNAc and phosphate
fulfill an decisive role in the expression of genes which are
transcribed under control of RNA polymerase II.
To study the biological phenomena associated with such
covalently modified protein conjugates appropriate tools are
required. The application of characteristic partial structures of
the parent proteins which embody both types of modifications
and also underivatized hydroxyamino acids may open up new
opportunities for bioorganic investigations (e.g., for the genera-
tion of monoclonal antibodies (O-GlcNAc modified peptides
are highly antigenic))3 or for inhibition and microinjection
studies. Furthermore, the precise location of the glycosylation
sites may be determined by comparison of glycosylated peptides
obtained from natural O-GlcNAc-containing glycopeptides with
synthetic analogs. However, the synthesis of phospho-10 or
glycopeptides11 alone is severely complicated by their pro-
nounced acid and base sensitivity as well as their multifunc-
tionality. Thus, under weakly basic conditions (pH > 8) the
phosphates and the glycosides may be lost by â-elimination
reactions,10,11 and at low pH an anomerization or a rupture of
the O-glycosides may occur11 (Scheme 2).
Results and Discussion
As a biologically relevant target compound, the hexapeptide
2 which carries one unmodified, one phosphorylated, and two
glycosylated hydroxyamino acids (Scheme 3) was chosen.
Compound 2 represents a characteristic partial structure of the
repeat sequence 1 of the large subunit of RNA polymerase II
which becomes glycosylated or phosphorylated while the
enzyme carries out its biological functions (Vide supra).
For the synthesis of 2, the strategy delineated in Scheme 4
was chosen. The target compound was divided into a perma-
nently masked N- and a C-terminal amino acid (4 and 7) and
two temporarily protected dipeptides 5 and 6. After the amino
blocking groups were removed, the peptide chain was to be
completed by stepwise N-terminal chain elongation. The
protecting group strategy employed the use of O-acetates for
the GlcNAc residues of the glycopeptide fragments 6 and 7,
since O-acetates can be cleaved from glycosylated peptides
under weakly basic conditions without â-elimination or race-
mization of the amino acids.18 In addition, this allowed for the
(12) For reviews, see: (a) Pohl, T.; Na¨gele, E.; Waldmann, H. Catal.
Today 1994, 22, 407. (b) Drauz, K.; Waldmann, H. Enzymes in Organic
Synthesis - A ComprehensiVe Handbook; VCH: Weinheim, 1995. (c)
Waldmann, H.; Sebastian, D. Chem. ReV. 1994, 94, 911.
(13) (a) Waldmann, H.; Na¨gele, E. Angew. Chem. 1995, 107, 2425;
Angew. Chem., Int. Ed. Engl. 1995, 34, 2259. (b) Schelhaas, M.; Glomsda,
S.; Ha¨usler, M.; Jakubke, H.-D. Angew. Chem. 1996, 108, 82; Angew.
Chem., Int. Ed. Engl. 1996, 35, 106. (c) Waldmann, H.; Na¨gele, E.;
Schelhaas, M. Bioorg. Med. Chem. 1996, 5, 75.
(14) Braun, D.; Waldmann, H.; Kunz, H. Bioorg. Med. Chem. 1993, 1,
197.
(15) (a) Waldmann, H.; Heuser, A.; Schulze, S. Tetrahedron Lett. 1996,
37, 8725. (b) Sebastian, D.; Waldmann, H. Tetrahedron Lett. 1997, 38,
2927.
These properties make necessary the application of a variety
of protecting groups that can be removed selectively under the
mildest and preferably neutral conditions. In the construction
(10) (a) Perich, J. W. Peptides and Protein Phosphorylation; Kemp, B.
E., Ed.; CRC Press: Boca Raton, FL, 1990. (b) Shapiro, G.; Buechler, D.
Tetrahedron Lett. 1994, 35, 5421.
(11) (a) Kunz, H. Angew. Chem. 1987, 99, 297; Angew. Chem., Int. Ed.
Engl. 1987, 26, 294. (b) Garg, H.; Jeanloz, R. W. AdV. Carbohydr. Chem.
Biochem. 1985, 43, 135.
(16) Gabold, S.; Waldmann, H. Submitted for publication.
(17) Part of this work was published in preliminary form, see: Pohl, T.;
Waldmann, H. Angew. Chem. 1996, 108, 1829; Angew. Chem., Int. Ed.
Engl. 1996, 35, 1720.
(18) Schultheiss-Reimann, P.; Kunz, H. Angew. Chem. 1983, 95, 64;
Angew. Chem., Int. Ed. Engl. 1983, 22, 62.

6704 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Pohl and Waldmann
Scheme 3. Schematic Representation of the Structure of
RNA Polymerase II and the Structure of the Target
Compound 2
upon amino acid activation. (4) It had to be removable with a
biocatalyst which does not attack the other functional groups
present (i.e., the acetates, the allyl and the tert-butyl esters, the
formed peptide bonds, and the phosphate and the glycosidic
bonds).
In developing such a blocking group, we have drawn from
our previous work on enzymatic protecting group techniques.
In particular, we had devised a viable strategy for the develop-
ment of enzyme labile urethane protecting groups. It consists
in the use of a urethane that embodies (a) a functional group
that is recognized by the biocatalyst and that is bound by an
enzyme labile linkage to (b) a functional group that undergoes
a spontaneous fragmentation upon cleavage of the enzyme-
sensitive bond resulting in (c) the liberation of a carbamic acid
derivative which decarboxylates to give the desired peptide or
peptide conjugate 11 (Scheme 5).
Using this principle, the p-acetoxybenzyloxycarbonyl (AcOZ)
group was developed whose removal could be initiated by the
acetyl esterase- or lipase-catalyzed cleavage of an acetic acid
ester.13a This urethane, however, could not be employed in the
construction of 2, since the carbohydrates were already masked
as acetic acid esters which are attacked by acetyl esterase12,19
and lipases.12
Scheme 4. Retrosynthetic Analysis of the Target
Phosphoglycopeptide 2
However, the above-mentioned principle for the development
of enzyme labile urethanes is general and not restricted to a
particular biocatalyst. Therefore, it could be adopted to the
problem at hand by switching to a different enzyme and a
different functional group recognized by it. To this end, we
developed the p-(phenylacetoxy)benzyloxycarbonyl (PhAcOZ)
group. It embodies a phenylacetic acid ester which is recognized
by penicillin G acylase.20 This biocatalyst operates under mild
conditions, is highly specific for the phenylacetyl group and
does not attack peptide-, phosphate- or glycosyl bonds at all.
As in the case of the AcOZ group a para-hydroxybenzylurethane
was employed as functional group that undergoes a spontaneous
fragmentation reaction to a quinone methide and thereby
liberates the desired amine (Scheme 5).
The PhAcOZ-protected dipeptide esters 17 were built up from
amino acid esters 16 and PhAcOZ-protected amino acids 15 in
the presence of N,N′-dicyclohexylcarbodiimide (DCC)/1-hy-
droxybenzotriazole (HOBt) as condensing reagents (Scheme 6).
Neither by HPLC nor by NMR spectroscopy could a racem-
ization of the activated amino acid in the formed peptides be
detected. The introduction of the PhAcOZ group into the amino
acids was achieved by means of p-(phenylacetoxy)benzyloxy-
carbonyl chloride 13 (PhAcOZCl). Compound 13 was obtained
in high yield via acylation of p-hydroxybenzaldehyde, reduction
of the aldehyde group in 12 to give the benzyl alcohol, and
conversion to the chloroformate 13 (Scheme 6). The acylation
of amino acids with 13 under the usual Schotten-Baumann
conditions at pH 10-12 yielded the desired PhAcOZ amino
acids only in low yield. Presumably the urethane which may
be regarded as a phenyl ester (i.e., an activated ester) is cleaved
under these alkaline conditions. However, upon treatment of
N,O-bissilylated amino acids21 14 with the chloroformate 13,
the selectively protected amino acids 15 were obtained in
satisfactory yields.
use of the tert-butyl ester as the permament C-terminal blocking
function which had to be removed only after assembly of the
entire peptide chain. Under the conditions necessary to split
off this ester, acetyl-masked glycosides remain unattacked11 and
phosphopeptides10 are not destroyed under acidic conditions
either. For the masking of the phosphate and the permanent
protection of the N-terminus in 4, we chose allyl-based blocking
groups which can be removed from phospho-10 and glycopep-
tides11 without undesired side reactions, too. Similarly, allyl
esters were introduced as temporary carboxy blocking groups
into the dipeptide intermediates 5 and 6.
As a temporary N-terminal protecting group, an enzyme labile
blocking function was needed. This protecting group had to
meet several demands: (1) It had to be orthogonally stable to
the base labile acetates, the acid labile tert-butyl ester, and the
noble metal sensitive allyl groups. (2) It had to be removable
under mild conditions to prevent â-elimination or anomerization.
(3) It had to have a urethane structure to avoid racemization
In order to determine if the PhAcOZ group can selectively
be removed enzymatically, in initial experiments, the peptide
allyl esters 17a and 17b were treated with penicillin acylase at
(19) (a) Waldmann, H.; Heuser, A.; Reidel, A. Synlett 1994, 65. (b)
Waldmann, H.; Heuser, A. Bioorg. Med. Chem. 1994, 2, 477.
(20) (a) Pohl, T.; Waldmann, H. Tetrahedron Lett. 1995, 36, 2963 and
references given therein. (b) Cole, M. Biochem. J. 1969, 115, 741, 747.
(21) Bolin, D. R.; Sytwu, I. I.; Humiec, F.; Meienhofer, J. Int. J. Pept.
Protein Res. 1989, 33, 353 and references therein.

Chemoenzymatic Synthesis of a Peptide Fragment
J. Am. Chem. Soc., Vol. 119, No. 29, 1997 6705
Scheme 5. Principle for the Development of Urethane
Protecting Groups which Can Be Removed by an
Enzyme-Initiated Fragmentation
Scheme 6. Synthesis and N-Terminal Deprotection of
PhAcOZ-Protected Peptides
pH 7.5. In the course of the ensuing reaction, the phenylacetates
were smoothly hydrolyzed to give the phenolates corresponding
to 9 (Scheme 5). These intermediates underwent a spontaneous
fragmentation and liberated the deprotected esters 18, pheny-
lacetic acid, and the quinone methide 10. However, due to
competing side reactions, the yield of 18a and 18b was only
low. On one hand, the dipeptide esters cyclized to the
diketopiperazines 19 under the reaction conditions, a well-known
side reaction in peptide chemistry displayed by N-terminally
unmasked dipeptide esters. On the other hand, the amino group
of the peptides competed with the solvent for the electrophile
10 to give the N-alkylated dipeptide esters 20. Diketopiperazine
formation was overcome by using the tert-butyl esters which
are not prone to this cyclization instead of the allyl esters. The
formation of the N-alkylated peptides 20 was efficiently
suppressed by performing the enzymatic reactions in the
presence of suitable nucleophiles,13a,22 which trap the quinone
methide 11 faster than the amines 18. Thus, if the deprotection
of 17c was carried out in the presence of an excess of NaHSO3
at pH 7.5, no undesired addition product was observed and the
dipeptide tert-butyl ester 18c was isolated in 88% yield (Scheme
7). Alternatively, KI which was the nucleophile of choice in
the enzymatic removal of the AcOZ protecting group could be
employed, too, however with inferior results. To guarantee that
the PhAcOZ-protected substrates are soluble under the reaction
conditions and thereby accessible to the biocatalyst, the
enzymatic transformations were most advantageously carried
out in the presence of 30 vol % of methanol as cosolvent.
Under these optimized reaction conditions, a fast enzymatic
reaction was assured and no undesired side reactions were
(22) Le Corre´, G.; Guibe´-Jampel, E.; Wakselman, M. Tetrahedron 1978,
34, 3105.

6706 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Pohl and Waldmann
Scheme 8. Synthesis of the Diglycotripeptide 27a
Scheme 7. Optimized Reaction Conditions for the
Enzymatic Removal of the PhAcOZ Group
observed. The selectivity of penicillin acylase for phenylacetic
acid derivatives guarantees that exclusively the PhAcOZ group
is cleaved and that the C-terminal esters and the peptide bonds
are left intact.
Two advantageous features of the PhAcOZ-protecting func-
tion have to be emphasized: The enzyme used recognizes the
same structural element in all cases. The variable peptide part
of the substrates is remote from the site of the biocatalyst’s
attack. Thus, possible sterically or electronically unfavorable
interactions of the protein with the substrates caused by bulky
amino acid side chains are guaranteed not to limit the substrate
tolerance of the enzyme. Furthermore, this enzymatic protecting
group technique can be applied for the construction of peptides
and analogs thereof containing, for instance, unnatural amino
acids including D-configurated amino acids.
For the selective deprotection of the intermediates 5 and 6,
orthogonal stability of the PhAcOZ group to the allyl ester was
required (Scheme 4). This was demonstrated by selective
removal of the allyl ester from the dipeptide PhAcOZ-Pro-
ThrOAll 17d by treatment with (PPh3)4Pd(0) in the presence
of morpholine as accepting nucleophile.23 Thereby, the desired
peptide carboxylic acid PhAcOZ-Pro-ThrOH was obtained
without attack on the N-terminal urethane group in 93% yield
(see the Experimental Section).
The PhAcOZ group thus proved to be an enzyme labile
urethane blocking group which is orthogonally stable to the other
protecting groups chosen and which ensures deprotection under
the mildest conditions. With this synthetic tool in hand, the
construction of the acid and base labile complex phosphogly-
copeptide 2 following the strategy detailed above (Scheme 4)
was approached.24 The fully protected C-terminal tripeptide 27
was synthesized via the building blocks 6 and 7. To this end,
the PhAcOZ-protected serine glycoside 22 was built up in 54%
yield by glycosylation of PhAcOZ serine 15e with the oxazoline
2125 (Scheme 8). The C-terminally unmasked serine derivative
22 was then condensed with proline allyl ester 23 in high yield
to give 6. In addition, it was converted into the tert-butyl ester
7 by employing the tert-butyl isourea 24.26 The glycosylated
dipeptide allyl ester 6 was then selectively deprotected at the
C-terminus by palladium(0)-mediated allyl transfer to morpho-
line as the accepting nucleophile.27 From the PhAcOZ-masked
glycosylated amino acid ester 7, the N-terminal protecting group
was removed by treatment with penicillin G acylase under the
reaction conditions described above to give the N-terminally
deprotected serine derivative 26 in high yield. In the course of
the enzymatic transformation, no undesired side reaction oc-
curred. Neither an anomerization or a rupture of the glycosidic
bond nor an attack on the O-acetates could be observed. The
selectively deprotected fragments 25 and 26 were then con-
densed to give the fully protected diglycotripeptide 27.
a Conditions: (A) EDC/HOBt (2 equiv/2 equiv); (B) penicillin G
acylase, 0.07 M phosphate buffer [30 vol % methanol, NaHSO3 (500
equiv), pH 7.5, 25 °C]; (C) [(C6H5)3P]4Pd, morpholine (1.1 equiv).
In order to elongate the peptide chain in the N-terminal
direction, 27 was deprotected by means of the enzyme-initiated
removal of the PhAcOZ-protected urethane under the mildest
conditions (Scheme 9). Thereby, the selectively unmasked
diglycotripeptide 28 was obtained in high yield and again
without disturbing side reactions. It was then condensed in high
yield with the dipeptide carboxylic acid 29 to give the
diglycopeptide 30, which contains two glycosylated and one
unmodified hydroxyamino acid. Compound 29 was obtained
by cleavage of the respective allyl ester (Vide supra; see the
Experimental Section).
From the glycosylated pentapeptide, the N-terminal PhAcOZ
urethane was cleaved off by employing once more penicillin G
acylase. The enzyme-initiated fragmentation proceeded smoothly,
and the desired N-terminally unmasked glycopeptide 31 was
formed in high yield. We stress that the acylase is capable of
removing the PhAcOZ group from an N-terminal proline. If
(23) Friedrich-Bochnitschek, S.; Waldmann, H.; Kunz, H. J. Org. Chem.
1989, 54, 751.
(24) For a synthesis of a glycosylated analog to 2, see: Meinjohanns,
E.; Vargas-Berenguel, A.; Meldal, M.; Paulsen, H.; Bock, K. J. Chem. Soc.,
Perkin Trans. 1 1995, 2165.
(25) Arsequell, G.; Krippner, L.; Dwek, R. A.; Wong, S. Y. C. J. Chem.
Soc., Chem. Commun. 1994, 2383.
(26) Mathias, L. Synthesis 1979, 561.
(27) Waldmann, H. Liebigs Ann. Chem. 1988, 1179 and references given
therein.

Chemoenzymatic Synthesis of a Peptide Fragment
J. Am. Chem. Soc., Vol. 119, No. 29, 1997 6707
Scheme 9. Synthesis of the Protected
Phosphoglycohexapeptide 3a
Scheme 10. Synthesis of the Phosphoglycohexapeptide 2a
a Conditions: (A) [(C6H5)3P]4Pd, formic acid/n-butylamine (10
equiv/6 equiv), room temperature (rt); (B) TFA, rt; (C) hydrazine
hydrate (3000 equiv), methanol, rt.
phosphoglycopeptide 3. The serine phosphate 32 was obtained
from the respective heptyl ester 5 (Scheme 4) by lipase-mediated
saponification of the ester group.15b Thereby, a second enzy-
matic protecting group technique could advantageously be
employed for the construction of the RNA polymerase derived
target compound.
a Conditions: (A) penicillin G acylase, 0.07 M phosphate buffer (30
vol % methanol), NaHSO3 (500 equiv), pH 7.5, 25 °C; (B) EDC/HOBt
(2 equiv/2 equiv).
Finally, all blocking functions were removed from the fully
masked phosphoglycohexapeptide 3 (Scheme 10). To this end,
first the N-terminal Aloc group and the allyl phosphate
protecting functions were cleaved by treatment with a palladium-
(0) catalyst and formic acid/n-butylamine as recommended by
Noyori et al.28 These weakly acidic conditions were chosen to
avoid any base-mediated â-elimination, in particular, of the
phosphate but also of the carbohydrates. Thus, the partially
deprotected hexapeptide 33 was obtained in high yield without
undesired side reaction. Next, the C-terminal tert-butyl ester
was cleaved selectively by treatment of 33 with trifluoroacetic
acid. Under these conditions an anomerization or rupture of
the glycosides did not occur. Finally, all O-acetates were
removed from 34 by means of hydrazine-mediated transesteri-
fication of the acetic acid esters with methanol.18
the phenylacetamide is employed instead, the respective peptides
are not attacked by penicillin acylase.20,27 This finding high-
lights a particularly important advantage of the 4-acyloxy-
benzyloxy-type blocking functions. The biocatalyst always has
to recognize and attack the same functional group (here a
phenylacetate), which is distant to the variable part of the
substrates to be deprotected (here the peptide chain). Therefore,
the introduction of sterically demanding or otherwise unfavor-
able structures (here a cyclic N-alkylamino acid) only has a
subordinate (if any) influence on the desired enzymatic trans-
formation.
The assembly of the peptide chain was completed by coupling
of the selectively deprotected diglycopentapeptide 31 with the
phosphorylated serine derivative 32 to give the complex
(28) Hayakawa, Y.; Kato, H.; Uchiyama, M.; Kajino, H.; Noyori, R. J.
Org. Chem. 1986, 51, 2402.

6708 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Pohl and Waldmann
MHz, CDCl3) δ 7.43-7.22 (m, 7H), 7.11 (d, J ) 8.5 Hz, 2H), 5.28 (s,
2H), 3.88 (s, 2H); 13C NMR (62.8 MHz, CDCl3) δ 169.8, 151.4, 150.7,
133.2, 130.9, 130.3, 129.3, 128.8, 127.5, 122.0, 72.6, 41.4; MS m/e
calcd for (M+) C16H13O4Cl 304.050, found 304.051.
In this final deprotection sequence, the order in which the
different blocking groups were removed had to be chosen
carefully. Under the basic conditions necessary to split off the
carbohydrate ester protecting groups, the diallyl phosphate would
most probably have been lost by â-elimination. Therefore, the
N-terminal Aloc group and the allyl phosphates were cleaved
first. The resulting phosphoric acid monoester is deprotonated
under basic conditions and, therefore, no longer prone to base-
induced elimination. The second step had to be the cleavage
of the tert-butyl ester under acidic conditions in the presence
of peracetylated GlcNAc residues. Glycosidic bonds to acety-
lated carbohydrates are significantly more acid stable than, e.g.,
bonds to benzylated or deprotected sugars.11 Thus, by perform-
ing the C-terminal deprotection as the second step a possible
anomerization or cleavage of the O-glycosides was prevented.
Synthesis of PhAcOZ-Protected Amino Acids 15. To a vigorously
stirred suspension of the amino acid (7.5 mmol) in 25 mL of CH2Cl2
was added trimethylsilyl chloride (TMS-Cl) (1.63 g, 15 mmol) in one
portion. The mixture was then refluxed for 1 h and cooled to 0 °C,
and diisopropylethylamine (1.68 g, 13 mmol) and chloroformate 13
(1.5 g, 5 mmol) were added. The reaction mixture was stirred at 0 °C
for 30 min and at room temperature for 2 h. The solvent was removed
in Vacuo, and the residue was taken up in 25 mL of Et2O and 50 mL
of a NaHCO3 solution (0.5 N). The layers were separated, and the
aqueous layer was extracted with 10 mL of Et2O. The combined
organic layers were washed with 10 mL of H2O. The pH of the
combined aqueous layers was adjusted to pH 2 with HCl (0.5 N), and
the aqueous solution was extracted with ethyl acetate. The combined
organic layers were then dried with MgSO4 and concentrated to give
the desired amino acids.
Conclusion
We have developed an efficient chemoenzymatic strategy for
the synthesis of differently functionalized complex peptide
conjugates. Its usefulness was demonstrated by the first
construction of the sensitive phosphoglycohexapeptide 2 which
represents a characteristic partial structure of the C-terminal
repeating unit of the large subunit of mammalian RNA
polymerase II. In all enzymatic and nonenzymatic transforma-
tions described above, no undesired side reactions were
observed. Thus, the peptide backbone and the glycosidic bonds
of the carbohydrates as well as the phosphate remained intact.
In addition, an anomerization of the glycosidic bonds which
may occur under acidic conditions was not observed. Also a
base-mediated â-elimination of the phosphate or the carbohy-
drates could not be detected. When enzymatic and classical
chemical protecting group techniques are combined in the way
described above or in similar ways, further sensitive O-
phosphorylated and O-glycosylated peptide conjugates will be
accessible. The use of these compounds may open up new
avenues of research in bioorganic chemistry and biology.
N-(4-(Phenylacetoxy)benzyloxycarbonyl)-L-alanine (15a). Data
for 15a: yield 75%; colorless solid, mp 91 °C; H NMR (250 MHz,
1
CDCl3) δ 10.27 (br, 1H), 7.41-7.21 (m, 7H), 7.04 (d, J ) 8.5 Hz,
2H), 5.43 (d, J ) 7 Hz, 1H), 5.08 (s, 2H), 4.40 (m, 1H), 3.83 (s, 2H),
1.41 (d, J ) 7 Hz, 3H); 13C NMR (62.8 MHz, CDCl3) δ 177.5, 170.1,
155.7, 150.5, 133.8, 133.3, 129.3, 129.2, 128.7, 127.4, 121.6, 66.4,
20
49.5, 41.0, 18.3; [R]D 6.7 (c ) 1, CH2Cl2); MS m/e calcd for (M+)
C19H19NO6 357.121, found 357.118. Anal. Calcd: C, 63.84; H, 5.36;
N, 3.92. Found: C, 63.82; H, 5.34; N, 3.88.
N-(4-(Phenylacetoxy)benzyloxycarbonyl)-O-(2-acetamido-3,4,6-
tri-O-acetyl-2-deoxy-â-D-glucopyranosyl)-L-serine (22). To a mixture
of peracetylated GlcNAc24 (480 mg, 1.23 mmol), 4A molecular sieves,
and CH2Cl2 (8 mL) was added BF3‚Et2O (0.4 mL, 3.8 mmol) dropwise
at 0 °C (freshly distilled under argon). After 24 h of stirring at room
temperature, the formation of the oxazoline 21 was complete as
monitored by TLC (CHCl3/MeOH 10:1). Then, Et3N (0.16 mL, 1.15
mmol) was added at 0 °C, the reaction mixture was stirred for 10 min,
and a solution of PhAcOZ-Ser (15e, 465 mg, 1.25 mmol) in 12 mL of
CH2Cl2/CH3CN (2:1) was added. After 4-5 days at room temperature,
the reaction was complete (periodic monitoring by TLC (CH3Cl/MeOH/
AcOH 80:10:1)). The mixture was neutralized at 0 °C with Et3N,
diluted with CH2Cl2, and filtered through Celite. The filtrate was
concentrated in Vacuo, and the residue was purified by column
chromatography (silica gel, CHCl3/MeOH 20:1 to 5:1) to yield 467
Experimental Section
General Procedures. 1H and 13C NMR spectra were recorded on
a Bruker AC 250, AM 400, and DRX-500 spectrometer. Mass spectra
were measured on a Finnigan MAT MS 70 spectrometer. Analytical
chromatography was performed on E. Merck silica gel 60 F254 plates.
Flash chromatography was performed on Baker silica gel (40-64 µm).
Tetrahydrofuran was distilled from potassium metal, dichloromethane
was from lithium aluminium hydride, and toluene was from sodium.
Penicillin G acylase was obtained in immobilized form on Eupergit C
from Boehringer Mannheim.
1
mg (54%) of 22 as a colorless solid: mp 87 °C; H NMR (250 MHz,
CD3OD) δ 7.43-7.23 (m, 7H), 7.06 (d, J ) 8.6 Hz, 2H), 5.21 (dd, J2,3
) J3,4 ) 10 Hz, 1H), 5.07 (s, 2H), 4.94 (dd, J3,4 ) J4,5 ) 10 Hz, 1H),
4.64 (d, J1,2 ) 8 Hz, 1H), 4.24 (m, 2H), 4.11 (m, 2H), 3.89 (s, 2H),
3.88-3.69 (m, 3H), 2.03 (s, 3H), 2.00 (s, 3H), 1.96 (s, 3H), 1.83 (s,
3H); 13C NMR (62.8 MHz, CD3OD) δ 174.6, 172.9, 171.4, 170.9, 170.8,
170.3, 157.0, 151.0, 134.9, 134.2, 129.5, 129.3, 128.7, 127.3, 121.7,
101.0, 72.9, 71.9, 69.1, 70.0, 66.0, 62.2, 55.8, 54.3, 40.8, 21.9, 19.7,
19.7, 19.6; [R]D20 -17.7 (c ) 2, CH3OH); FAB MS m/e calcd for (M+
+ 1) C33H38N2O15 703.235, found 703.237.
4-(Phenylacetoxy)benzyl Chloroformate (13). A solution of
4-(phenylacetoxy)benzaldehyde (12) (4.3 g, 18 mmol) in 75 mL of
2-propanol was treated with sodium borohydride (0.225 g, 6 mmol)
suspended in 25 mL of 2-propanol. The reaction mixture was
periodically monitored by TLC. After the conversion was complete,
the reaction was quenched by addition of HCl (0.5 N), the pH was
adjusted to 7 with Na2CO3 solution, and the 2-propanol was removed
under reduced pressure. After extraction with ethyl acetate, the organic
layer was dried (MgSO4) and concentrated under reduced pressure.
Purification of the residue by flash chromatography (silica gel; ethyl
acetate/hexane 1:1) yielded 3 g (70%) of 4-(phenylacetoxy)benzyl
Peptide Couplings to Give PhAcOZ Protected Peptides. (A)
N-(4-(Phenylacetoxy)benzyloxycarbonyl)-O-(2-acetamido-3,4,6-tri-
O-acetyl-2-deoxy-â-D-glucopyranosyl)-L-serine, tert-Butyl Ester (7).
A solution of peptide 22 (71 mg, 0.1 mmol) and N,N′-diisopropyl-O-
tert-butylisourea (24) (100 mg, 0.5 mmol) in 20 mL of CH2Cl2 was
stirred for 3 days at room temperature. After filtration to remove
precipitated urea, the CH2Cl2 solution was extracted with phosphate
buffer (pH 5), the organic layer was dried (MgSO4), and CH2Cl2 was
removed under reduced pressure. Purification of the crude residue by
flash chromatography (silica gel; ethyl acetate/hexane 4:1) yielded 43
1
alcohol as a white solid: mp 46 °C; H NMR (250 MHz, CDCl3) δ
7.42-7.23 (m, 7H), 7.03 (d, J ) 8.5 Hz, 2H), 4.68 (s, 2H), 3.89 (s,
2H), 1.81 (br, 1H); 13C NMR (62.8 MHz, CDCl3) δ 169.9, 149.9, 138.3,
133.2, 129.1, 128.5, 127.8, 127.2, 121.4, 64.5, 41.4. Anal. Calcd: C,
74.35; H, 5.83. Found: C, 74.34; H, 5.69.
A solution of 4-(phenylacetoxy)benzyl alcohol (1.24 g, 5 mmol) in
10 mL of toluene was added at 0 °C to a solution of phosgene in 5.3
mL of toluene (1.93 M). After 2 h at 0 °C, the mixture was stirred for
1 h at room temperature and the toluene was removed under reduced
pressure to yield 1.5 g (99%) of 13 as a colorless oil: 1H NMR (250
1
mg (56%) of 7 as a colorless solid: mp 93 °C; H NMR (250 MHz,
CD3OD) δ 7.43-7.24 (m, 7H), 7.06 (d, J ) 8.5 Hz, 2H), 5.19 (dd, J2,3
) J3,4 ) 10 Hz, 1H), 5.06 (s, 2H), 4.94 (dd, J3,4 ) J4,5 ) 10 Hz, 1H),
4.66 (d, J1,2 ) 8 Hz, 1H), 4.39-4.07 (m, 4H), 3.89 (s, 2H), 3.88-3.67
(m, 3H), 2.03 (s, 3H), 2.01 (s, 3H), 1.96 (s, 3H), 1.85 (s, 3H), 1.47 (s,
9H); 13C NMR (62.8 MHz, CD3OD) δ 171.8, 170.9, 170.5, 170.2, 169.6,
168.5, 156.2, 150.2, 133.8, 133.0, 128.9, 128.4, 128.2, 127.0, 121.2,
99.9, 82.2, 72.2, 71.3, 68.4, 68.6, 65.8, 61.8, 54.5, 53.8, 40.8, 27.3,

Chemoenzymatic Synthesis of a Peptide Fragment
J. Am. Chem. Soc., Vol. 119, No. 29, 1997 6709
20
22.3, 20.5, 19.9, 19.9; [R]D -13.6 (c ) 1, CH3OH). Anal. Calcd:
tri-O-acetyl-2-deoxy-â-D-glucopyranosyl)-L-serine, tert-butyl ester
(26): yield 78%, colorless oil; 1H NMR (250 MHz, CDCl3) δ 6.46 (d,
J ) 8 Hz, 1H), 5.28 (dd, J2,3 ) J3,4 ) 10 Hz, 1H), 5.08 (dd, J3,4 ) J4,5
) 10 Hz, 1H), 4.78 (d, J1,2 ) 8 Hz, 1H), 4.31-4.06 (m, 3H), 3.93-
3.68 (m, 3H), 3.58 (t, J ) 4 Hz, 1H) 2.08 (s, 3H), 2.02 (s, 3H), 2.01
(s, 3H), 1.97 (s, 3H), 1.46 (s, 9H); 13C NMR (62.8 MHz, CDCl3) δ
171.7, 170.9, 170.7, 170.6, 169.4, 100.8, 81.9, 72.6, 71.8, 68.6, 71.3,
C, 58.55; H, 6.11; N, 3.69. Found: C, 58.37; H, 6.12; N, 3.44.
(B) Peptides 17a-d, 6, 27, 30, and 3. To a solution of PhAcOZ-
protected amino acid, dipeptide, or serine derivative 32 (0.25 mmol)
and the relevant amino acid ester or peptide tert-butyl ester in 5 mL of
CH2Cl2 was added N-hydroxybenzotriazole (67.5 mg, 0.5 mmol) and
at 0 °C the carbodiimide DCC or N-ethyl-N′-((dimethylamino)propyl)-
carbodiimide hydrochloride (EDC) (0.5 mmol). After 2 h of stirring
at this temperature, the mixture was left overnight at room temperature,
the urea was filtered, and the solution was extracted with 3 mL of H2O
(pH 4). The organic layer was dried (MgSO4) and concentrated under
reduced pressure. The crude residue was purified by flash chroma-
tography (silica gel; CHCl3/CH3OH). Data for N-(4-(phenylacetoxy)-
benzyloxycarbonyl)-L-valyl-L-alanine, allyl ester (17a): yield 88%,
colorless solid, mp 119 °C; 1H NMR (250 MHz, CDCl3) δ 7.41-7.27
(m, 7H), 7.04 (d, J ) 8 Hz, 2H), 6.52 (d, J ) 7 Hz, 1H), 5.99-5.83
(m, 1H), 5.46 (d, J ) 8 Hz, 1H), 5.31 (dd, Jtrans ) 17 Hz, Jgem ) 1.5
Hz, 1H), 5.24 (dd, Jcis ) 10 Hz, Jgem ) 1.5 Hz, 1H), 5.07 (s, 2H),
4.68-4.54 (m, 3H), 4.05 (m, 1H), 3.87 (s, 2H), 2.11 (m, 1H), 1.39 (d,
J ) 6 Hz), 0.96 (d, J ) 7 Hz, 3H), 0.89 (d, J ) 7 Hz, 3H); 13C NMR
(62.8 MHz, CDCl3) δ 172.3, 170.7, 169.9, 156.2, 150.5, 133.9, 133.3,
131.4, 129.3, 129.2, 128.7, 127.3, 121.5, 118.8, 66.3, 66.0, 60.1, 48.0,
20
62.1, 55.0, 54.6, 27.9, 23.5, 23.3, 20.7, 20.6; [R]D -29.8 (c ) 1,
CH2OH); FAB MS m/e calcd for (M+ + 1) C21H34N2O11 491.224, found
491.232.
Pd(0)-Mediated Cleavage of the Allyl Esters. Peptides 25 and
29. To a solution of the dipeptide allyl ester (0.23 mmol) and 2 mol%
(Ph3P)4Pd in 20 mL of THF under an argon atmosphere was added
dropwise a solution of morpholine (21 mg, 0.24 mmol) in 5 mL of
THF. The mixture was stirred for 30 min, the solvent was removed
under reduced pressure and the residue purified by flash chromatography
(silica gel; ethyl acetate/hexane 3:1 then CHCl3/CH3OH 10:1 to 5:1).
Data for N-(4-(phenylacetoxy)benzyloxycarbonyl)-O-(2-acetamido-
3,4,6-tri-O-acetyl-2-deoxy-â-D-glucopyranosyl)-L-seryl-L-proline
(25): yield 89%, colorless solid, mp 138 °C; 1H NMR (250 MHz, CD3-
OD) δ 7.42-7.22 (m, 7H), 7.03 (d, J ) 8.5 Hz, 2H), 5.22 (dd, J2,3
)
J3,4 ) 10 Hz, 1H), 5.08 (s, 2H), 5.04 (dd, J3,4 ) J4,5 ) 10 Hz, 1H),
4.78-4.45 (m, 2H, 1H), 4.29 (m, 2H), 4.06 (m, 1H) 4.04-3.38 (m,
8H), 2.20 (m, 1H), 2.13-1.92 (m, 3H), 2.01 (s, 3H), 1.98 (s, 3H), 1.96
(s, 3H), 1.87 (s, 3H); 13C NMR (62.8 MHz, CD3OD) δ 173.7, 172.4,
171.9, 171.3, 170.6, 170.0, 169.5, 158.1, 152.1, 135.9, 135.2, 130.5,
130.2, 129.7, 128.3, 122.7, 101.7, 74.3, 73.0, 70.1, 69.5, 67.2, 63.2,
20
41.3, 31.3, 19.1, 18.2, 17.7; [R]D -3.5 (c ) 1, CH2Cl2). Anal.
Calcd: C, 65.29; H, 6.50; N, 5.64. Found: C, 65.19; H, 6.66; N, 5.82.
Data for N-(4-(phenylacetoxy)benzyloxycarbonyl)-L-prolyl-L-threonyl-
O-(2-acetamido-3,4,6-tri-O-acetyl-2-deoxy-â-D-glucopyranosyl)-L-seryl-
L-prolyl-O-(2′-acetamido-3′,4′,6′-tri-O-acetyl-2′-deoxy-â-D-glucopyra-
nosyl)-L-serine, tert-butyl ester (30): yield 71%, colorless solid, mp
112-116 °C; 1H NMR (500 MHz, CD3OD) δ 7.45-7.26 (m, 7H), 7.06
(d, J ) 8.5 Hz, 2H), 5.19 (dd, J2,3 ) J3,4 ) J2′,3′ ) J3′,4′ ) 9 Hz, 2H),
5.09 (s, 2H), 5.01 (dd, J3,4 ) J4,5 ) J3′,4′ ) J4′,5′ ) 9 Hz, 2H), 4.73 (d,
J1,2 ) 8.5 Hz, 1H), 4.69 (d, J1′,2′ ) 8.5 Hz, 1H), 4.66 (t, J ) 6.5 Hz,
1H), 4.57 (d, J ) 6.5 Hz, 1H), 4.56-4.44 (m, 2H), 4.38-4.20 (m,
5H), 4.17-4.04 (m, 2H), 4.01-3.48 (m, 13H), 2.33-2.22 (m, 2H),
2.16-1.87 (m, 30H), 1.48 (s, 9H), 1.19, 1.06 (d, J ) 6 Hz, 3H, 2
rotamers); 13C NMR (125.7 MHz, CD3OD) δ 173.8, 173.7, 173.6, 173.3,
172.9, 172.3, 172.2, 172.0, 171.8, 171.7, 171.2, 171.1, 170.5, 169.7,
157.2, 152.0, 135.6, 135.0, 130.4, 130.0, 129.6, 128.3, 122.7, 101.4,
101.1, 83.5, 74.2, 74.1, 73.1, 72.9, 70.0, 69.6, 71.7, 71.5, 68.8, 68.6,
63.1, 62.8, 61.7, 61.6, 59.6, 55.3, 55.1, 54.4, 54.2, 48.2, 47.7, 41.8,
30.6, 30.5, 28.3, 26.0, 25.5, 23.1, 23.0, 20.8, 20.8, 20.7, 20.7, 20.6,
20.6, 20.2; [R]D20 -54.7 (c ) 1, CH3OH). Anal. Calcd: C, 55.53; H,
6.24; N, 6.67. Found: C, 55.34; H, 6.20; N, 6.77. Data for
N-allyloxycarbonyl-O-(diallylphosphato)-L-seryl-L-prolyl-L-threonyl-O-
(2-acetamido-3,4,6-tri-O-acetyl-2-deoxy-â-D-glucopyranosyl)-L-seryl-
L-prolyl-O-(2′-acetamido-3′,4′,6′-tri-O-acetyl-2′-deoxy-â-L-glucopyra-
nosyl)-L-serine, tert-butyl ester (3): yield 62%, colorless amorphous
solid; 1H NMR (400 MHz, CD3OD) δ 6.06-5.88 (m, 3H), 5.43-5.08
(m, 8H), 5.01 (dd, J3,4 ) J4,5 ) J3′,4′ ) J4′,5′ ) 9 Hz, 2H), 4.78 (d, J1,2
) 8.5 Hz, 1H), 4.73 (d, J1′,2′ ) 8.5 Hz, 1H), 4.71 (t, J ) 6 Hz, 1H),
4.64-4.43 (m, 10H), 4.41-4.26 (m, 5H), 4.23-4.07 (m, 5H), 4.02-
3.77 (m, 8H), 3.72-3.61 (m, 2H), 2.40-2.19 (m, 2H), 2.16-1.85 (m,
30H), 1.46 (s, 9H), 1.19, 1.14 (d, J ) 6 Hz, 3H, 2 rotamers); 13C NMR
(125.7 MHz, CD3OD) δ 173.8, 173.6, 173.3, 172.3, 172.2, 171.9, 171.8,
171.6, 171.5, 171.1, 171.1, 170.0, 169.8, 169.6, 155.7, 133.2, 132.8,
132.7, 118.1, 117.9, 117.6, 100.7, 100.2, 83.7, 74.7, 74.5, 73.1, 72.7,
71.0, 70.9, 69.9, 69.8, 68.9, 68.1, 66.6, 66.5, 66.3, 62.7, 62.6, 61.0,
60.6, 60.1, 54.4, 54.3, 54.0, 53.9, 52.1, 47.9, 47.8, 30.1, 29.9, 28.0,
20
62.2, 55.4, 53.9, 49.3, 41.9, 30.5, 25.8, 23.0, 20.7, 20.7, 20.6; [R]D
-65.8 (c ) 1, CH3OH); FAB MS m/e calcd for (M+ + 1) C38H45N3O16
800.287, found 800.281.
Final Removal of All Protecting Groups. (A) O-Phosphato-L-
seryl-L-prolyl-L-threonyl-O-(2-acetamido-3,4,6-tri-O-acetyl-2-deoxy-
â-D-glucopyranosyl)-L-seryl-L-prolyl-O-(2′-acetamido-3′,4′,6′-tri-O-
acetyl-2′-deoxy-â-D-glucopyranosyl)-L-serine, tert-Butyl Ester (33).
To a solution of peptide 3 (5 mg, 3.26 µmol) and 2 mol % Pd(Ph3P)4
in 2.5 mL of THF were added under argon atmosphere a solution of
formic acid (1.5 mg, 33 µmol) in 0.5 mL of THF and a solution of
n-butylamine (1.4 mg, 20 µmol) in 0.5 mL of THF. After 30 min of
stirring at room temperature, at which time no traces of 3 could be
detected by TLC, the solvent was removed under reduced pressure and
the crude residue was purified by flash chromatography (silica gel;
ethyl acetate/hexane 4:1 followed by acetone/hexane 3:1 and CHCl3/
CH3OH 5:1), yielding 4 mg (89%) of 33 as a colorless amorphous
solid: 1H NMR (500 MHz, CD3OD) δ 5.21 (dd, J2,3 ) J3,4 ) 9 Hz,
1H), 5.19 (dd, J2′,3′ ) J3′,4′ ) 9 Hz, 1H), 5.01 (dd, J3,4 ) J4,5 ) J3′,4′
)
J4′,5′ ) 9 Hz, 2H), 4.78 (d, J1,2 ) 8.5 Hz, 1H), 4.72 (d, J1′,2′ ) 8.5 Hz,
1H), 4.70 (m, 1H), 4.59-4.47 (m, 3H), 4.40-4.28 (m, 4H), 4.27-
4.10 (m, 5H), 4.02-3.78 (m, 8H), 3.72-3.61 (m, 4H), 2.40-2.19 (m,
2H), 2.16-1.88 (m, 30H), 1.48 (s, 9H), 1.21, 1.16 (d, J ) 6 Hz, 3H,
20
2 rotamers); [R]D -14.5 (c ) 0.4, CH3OH); FAB MS m/e calcd for
(M+) C55H85N8O30P (1369.5), found 1328 (M-Ac), 1273 (M + H -
H2PO4), 1231 (1273 - Ac), 1202 (M - serylphosphate), 1160 (1202
- Ac).
(B) O-Phosphato-L-seryl-L-prolyl-L-threonyl-O-(2-acetamido-
3,4,6-tri-O-acetyl-2-deoxy-â-D-glucopyranosyl)-L-seryl-L-prolyl-O-
(2′-acetamido-3′,4′,6′-tri-O-acetyl-2′-deoxy-â-D-glucopyranosyl)-L-
serine (34). A solution of peptide 33 (4 mg, 3 µmol) in 0.8 mL (10
mmol) of trifluoracetic acid was stirred for 4 h at room temperature.
At this time, 3 mL of toluene were added and the solvent was removed
under reduced pressure. For further purification, the residue was
washed with hexane and then with CH2Cl2, yielding 2.8 mg (73%) of
34 as a colorless amorphous solid: 1H NMR (500 MHz, CD3OD) δ
20
25.2, 25.0, 23.1, 23.0, 20.5, 20.5, 20.5, 20.4, 20.4, 20.4, 20.2; [R]D
-65.4 (c ) 1, CH2OH); FAB MS m/e calcd for (M+ + 1) C65H97N8O32P
1533.6025, found 1533.768.
Enzymatic Removal of the PhAcOZ Group from Peptides and
Glycopeptides. Peptides 18c, 26, 28, and 31. A solution of the
respective PhAcOZ-protected peptide (0.05 mmol) in a mixture of 33
mL of CH3OH, 13 mL of a NaHSO3 solution (40%), and 68 mL of a
Na2HPO4 buffer (0.07 M) was treated with penicillin G acylase (300
units) at pH 7.5 for 24 h at room temperature. The immobilized enzyme
was filtered, and the CH3OH was removed under reduced pressure.
The reaction mixture was extracted three times with CHCl3, and the
combined organic layers were dried (MgSO4) and concentrated to
dryness. The crude residue was purified by flash chromatography (silica
gel, CHCl3/CH3OH and 1 vol % Et3N). Data for O-(2-acetamido-3,4,6-
5.23 (dd, J2,3 ) J3,4 ) J2′,3′ ) J3′,4′ ) 9 Hz, 2H), 5.00 (dd, J3,4 ) J4,5
J3′,4′ ) J4′,5′ ) 9 Hz, 2H), 4.78 (d, J1,2 ) 8.5 Hz, 1H), 4.75 (d, J1′,2′
)
)
8.5 Hz, 1H), 4.73-4.64 (m, 2H), 4.61-4.53 (m, 1H), 4.47 (t, J ) 6
Hz, 1H), 4.43-4.27 (m, 5H), 4.27-4.10 (m, 3H), 4.08-3.77 (m, 9H),
3.75-3.61 (m, 4H), 2.41-2.18 (m, 2H), 2.16-1.87 (m, 30H), 1.19,
1.14 (d, J ) 6 Hz, 3H, 2 rotamers); [R]D20 - 10.8 (c ) 0.28, CH3OH);
FAB MS m/e calcd for (M+) C51H77N8O30P (1313.4), found 1238 (M
- H2PO4 + Na), 1215 (M - H2PO4), 1196 (1238 - Ac), 1145 (M -
serylphosphate).

6710 J. Am. Chem. Soc., Vol. 119, No. 29, 1997
Pohl and Waldmann
(C) O-Phosphato-L-seryl-L-prolyl-L-threonyl-O-(2-acetamido-2-
deoxy-â-D-glucopyranosyl)-L-seryl-L-prolyl-O-(2′-acetamido-2′-deoxy-
â-D-glucopyranosyl)-L-serine (2). A solution of peptide 34 (2.8 mg,
2.13 µmol) in 1.5 mL of CH3OH was treated with hydrazine hydrate
(6 mmol) for 3 h at room temperature. After cooling the reaction
mixture to 0 °C, 3 mL of acetone was added, and after 1 h, the solvent
was removed under reduced pressure. This procedure was repeated
10 times; then, the residue was dissolved in CH3OH, acetone was added,
and the mixture was evaporated to dryness, yielding 1.2 mg (55%) of
2 as a colorless amorphous solid: 1H NMR (500 MHz, CD3OD) δ
4.88-4.79 (m, 2H), 4.50 (d, J1,2 ) 8.4 Hz, 1H), 4.48-4.18 (m, 5H),
4.17-3.98 (m, 3H), 3.97-3.81 (m, 6H), 3.79-3.62 (m, 10H), 3.57 (t,
J ) 4.7 Hz, 2H), 3.53-3.41 (m, 8H), 2.41-2.19 (m, 2H), 2.17-1.89
FAB MS m/e calcd for (M+) C39H65N8O24P (1060.5), found 1061.5
(M + H), 1040 (M - Ac + Na), 921 (M + H - Ac - H2PO4).
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft and the Fonds der Chemischen Indus-
trie. We thank Dr. Dagmar Sebastian for supplying compound
32.
Supporting Information Available: 1H and 13C NMR data
for compounds 12, 15b-e, 17b-d, 6, 27, 18c, 28, and 29 as
1
well as copies of the H and 13C NMR spectra of compounds
12, 13, 15b-d, 18c, 22, 25, 26, 28, 29, 31, 3, 33, 34, and 2 (33
pages). See any current masthead page for ordering and Internet
access instructions.
20
(m, 12H), 1.19 (d, J ) 6 Hz, 3H); [R]D -5.5 (c ) 0.1, CH3OH);
JA970709E