Efficient conjugation of peptides to oligonuoleotides by 'native ligation'

Article abstract of DOI:10.1021/jo000214z

A new strategy has been developed for conjugation of peptides to oligonucleotides. The method is based on the 'native ligation' of an N-terminal thioester-functionalized peptide to a 5'-cysteinyl oligonucleotide. Two new reagents were synthesized for use

Full text of DOI:10.1021/jo000214z

4900
J . Org. Chem. 2000, 65, 4900-4908
Efficien t Con ju ga tion of P ep tid es to Oligon u cleotid es by “Na tive
Liga tion ”
Dmitry A. Stetsenko and Michael J . Gait*
Medical Research Council, Laboratory of Molecular Biology, Hills Road, Cambridge, CB2 2QH, U.K.
mgait@mrc-lmb.cam.ac.uk
Received February 15, 2000
A new strategy has been developed for conjugation of peptides to oligonucleotides. The method is
based on the “native ligation” of an N-terminal thioester-functionalized peptide to a 5′-cysteinyl
oligonucleotide. Two new reagents were synthesized for use in solid-phase peptide and oligonucle-
otide synthesis, respectively. Pentafluorophenyl S-benzylthiosuccinate was used in the final coupling
step in standard Fmoc-based solid-phase peptide assembly. Deprotection with trifluoracetic acid
generated in solution peptides substituted with an N-terminal S-benzylthiosuccinyl moiety. O-trans-
4-(N-R-Fmoc-S-tert-butylsulfenyl-L-cysteinyl)aminocyclohexyl O-2-cyanoethyl-N,N-diisopropylphos-
phoramidite was used in the final coupling step in standard phosphoramidite solid-phase
oligonucleotide assembly. Deprotection with aqueous ammonia solution generated in solution 5′-
S-tert-butylsulfenyl-L-cysteinyl functionalized oligonucleotides. Functionalized peptides and oligo-
nucleotides were used without purification in native ligation conjugation reactions in aqueous/
organic solution using tris-(2-carboxyethyl)phosphine to remove the tert-butylsulfenyl group in situ
and thiophenol as a conjugation enhancer. A range of peptide-oligonucleotide conjugates were
prepared by this route and purified by reversed-phase HPLC.
In tr od u ction
recent study showed that some peptide-oligonucleotide
conjugates are inactive as antisense agents because of
entrapment within endosomes.16 Systematic studies of
the sequence and structural requirements for good cell
penetration and compartmentalization in a range of cell
lines as well as correlation with biological activity have
not yet been reported for peptide-oligonucleotide conju-
gates. This is because such studies have been hampered
by the often cumbersome and inefficient methods re-
quired for the chemical synthesis of such bioconjugates.
Total stepwise solid-phase approaches on a single
support have posed serious difficulties because of incom-
patibilities in peptide and oligonucleotide protection and
assembly chemistries. Although some promising routes
have been suggested, so far only a restricted range of
peptide-oligonucleotide conjugates have been prepared
and the methods have been mostly unsuitable for the
synthesis of conjugates containing peptides of length and
type suitable for cell delivery studies.17-25 A post-as-
Oligonucleotides and their analogues are used widely
as sequence specific reagents to block gene expression of
RNA within cells.1,2 However, oligonucleotide efficacy in
cell culture is often limited by poor cellular uptake.3,4
Among the many molecules that have been reported to
enhance cell delivery of oligonucleotides are a number
of peptide carriers. Some of these peptides have been used
as additives to aid transfection.5-11 There are also several
reports of oligonucleotides conjugated covalently to pep-
tides for delivery to cell lines in culture.12-15 However, a
* To whom correspondence should be addressed. Tel: +44 1223
248011. Fax: +44 1223 402070.
(1) Antisense research and applications; Crooke, S. T., Lebleu, B.,
Ed.; CRC Press: Boca Raton, 1993.
(2) Chadwick, D. J .; Cardew, G. Oligonucleotides as therapeutic
agents; J ohn Wiley: Chichester, 1997.
(3) Crooke, S. T. J . Drug Targeting 1995, 3, 185-190.
(4) Stein, C. A.; Narayan, R. Perspect. Drug Discovery Design 1996,
4, 41-50.
(5) Morris, M. C.; Vidal, P.; Chaloin, L.; Heitz, F.; Divita, G. Nucl.
Acids Res. 1997, 25, 2730-2736.
(16) Antopolsky, M.; Azhayeva, E.; Tengvall, U.; Auriola, S.; J a¨a¨ske-
la¨inen, I.; Ro¨nkko¨, S.; Honkakoski, P.; Urtti, A.; Lo¨nnberg, H.; Azhayev,
A. Bioconjugate Chem. 1999, 10, 598-606.
(6) Dokka, S.; Toledo-Velasquez, D.; Shi, X.; Wang, L.; Rojanasukul,
Y. Pharmaceut. Res. 1997, 14, 1759-1764.
(7) Wyman, T. B.; Nicol, F.; Zelphati, O.; Scario, P. V.; Plank, C.;
Szoka, F. C. Biochemistry 1997, 36, 3008-3017.
(17) Haralambidis, J .; Duncan, L.; Angus, K.; Tregear, G. W.
Tetrahedron Lett. 1987, 28, 5199-5202.
(8) Wadhwa, M. S.; Collard, W. T.; Adami, R. C.; Mckenzie, D. L.;
Rice, K. G. Bioconjugate Chem. 1997, 8, 81-88.
(18) J uby, C. D.; Richardson, C. D.; Brousseau, R. Tetrahedron Lett.
1991, 32, 879-882.
(9) Pichon, C.; Freulon, I.; Midoux, P.; Mayer, R.; Monsigny, M.;
Roche, A.-M. Antisense and Nucl. Acid Drug. Dev. 1997, 7, 335-343.
(10) Bachmann, A. S.; Surovoy, A.; J ung, G.; Moelling, K. J . Mol.
Med. 1998, 76, 126-132.
(19) de La Torre, B. G.; Avino, A.; Tarrason, G.; Piulats, J .; Albericio,
F.; Eritja, R. Tetrahedron Lett. 1994, 35, 2733-2736.
(20) Truffert, J .-C.; Lorthoir, O.; Asseline, U.; Thuong, N. T.; Brack,
A. Tetrahedron Lett. 1994, 35, 2353-2356.
(11) Midoux, P.; Kichler, A.; Boutin, V.; Maurizot, J .-C.; M., M.
Bioconj. Chem. 1998, 9, 260-267.
(21) Bergmann, F.; Bannwarth, W. Tetrahedron Lett. 1995, 36,
1839-1842.
(12) Vive`s, E.; Lebleu, B. Tetrahedron Lett. 1997, 38, 1183-1186.
(13) Derossi, D.; J oliot, A. H.; Chassaing, G.; Prochiantz, A. J . Biol.
Chem. 1994, 269, 10444-10450.
(22) Basu, S.; Wickstrom, E. Tetrahedron Lett. 1995, 36, 4943-4946.
(23) Tetzlaff, C. N.; Schwope, I.; Bleczinski, C. F.; Steinberg, J . A.;
Richert, C. Tetrahedron Lett. 1998, 39, 4215-4218.
(24) De Napoli, L.; Messere, A.; Montesarchio, D.; Piccialli, G.;
Benedetti, E.; Bucci, E.; Rossi, F. Bioorg. Med. Chem. 1999, 7, 395-
400.
(14) Pichon, C.; Arar, K.; Stewart, A. J .; Dodon, M. D.; Gazzolo, L.;
Courtoy, P. J .; Mayer, R.; Monsigny, M.; Roche, A.-C. Mol. Pharmacol.
1997, 51, 431-438.
(15) Chaloin, L.; Vidal, P.; Lory, P.; Me´ry, J .; Lautredou, N.; Divita,
G.; Heitz, F. Biochem. Biophys. Res. Comm. 1998, 243, 601-608.
(25) Schwope, I.; Bleczinski, C. F.; Richert, C. J . Org. Chem. 1999,
64, 4749-4761.
10.1021/jo000214z CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/15/2000

Conjugation of Peptides to Oligonucleotides by “Native Ligation”
J . Org. Chem., Vol. 65, No. 16, 2000 4901
sembly conjugation strategy has found much more gen-
eral applicability. Here, peptide and oligonucleotide
moieties are assembled separately on their own solid
supports. Each biomolecule is designed to carry a reactive
functionality that is released upon full deprotection and
cleavage from the support. Peptide and oligonucleotide
parts are then joined in aqueous solution through these
reactive functionalities by a selective reaction. To date,
only a limited number of such reactions has been
described, for example the formation of a disulfide
bond,12,26,27 reaction of a cysteine peptide with a male-
imido oligonucleotide,28-30 or a bromoacetyl peptide with
a thiol-functionalized oligonucleotide.31
These established routes suffer from certain disadvan-
tages. For example, a disulfide bond is unstable to
reducing agents that may be present under many assay
conditions or within cells. The maleimide-thiol route of
conjugation requires a functionalization step on the
oligonucleotide part after release from the solid support.
Two of the routes present limitations in peptide sequence
(requirement for a cysteine). Finally, post-assembly join-
ing routes may sometimes lead to inefficient conjugation,
due to secondary structure or poor solubility of certain
peptide components in aqueous solution, and this may
be relieved in some cases by the addition of denaturants.
However, a major advantage of post-assembly conjuga-
tion is that all peptide side chains and nucleobases are
deprotected before conjugation, and thus there is no
problem of compatibility of assembly and deprotection
chemistries. An interesting variation of the solution
phase strategy is where both peptide and oligonucleotide
fragments are released from their respective supports
still carrying side-chain and nucleobase protection, puri-
fied and conjugated under nonaqueous peptide-bond-
forming conditions.32,33 Although encouraging, this route
is not yet validated for a wide-range of peptides carrying
side-chain protecting groups that require deprotection
following conjugation.
linker.35,36 After cleavage from the support, the conjugate
was deprotected and analyzed. By this route, we were
able to obtain efficient conjugation for small peptides,
especially if 2-cyanoethyl phosphate protecting groups
were removed from the oligonucleotide prior to conjuga-
tion and if the spacer distance between oligonucleotide
and amino group was 6 carbon atoms or more. However,
longer hydrophobic or highly basic peptide fragments
were poorly conjugated under these conditions.36
We decided to search for alternative conjugation chem-
istries that might proceed both with high yield and
selectivity for preparation of peptide conjugates of length
and variety useful in cellular uptake studies. Such
chemistries should be adaptable to a range of solvent
conditions in order to maintain solubility of peptides and
oligonucleotides that vary substantially in hydrophobicity
and charge. Such solubility may depend not only on the
intrinsic sequence but also on the level of chemical
protection that remains on each biomolecule and the
nature of any analogue incorporated. We also wished any
functionalization method to be compatible with mild
Fmoc-based peptide assembly chemistry and a wide
range of side-chain protecting groups. Further, we wished
the functionalities to be introduced on to each biomolecule
by techniques that could be carried out simply and on-
column, without the need for extended manipulation
following release from the solid supports.
In the search for suitable conjugation chemistry, a
method of protein synthesis came to our attention that
involves “native ligation” of two largely unprotected
peptide fragments, one containing a C-terminal thioester
and the other an N-terminal cysteine.37 This technique
has been applied recently to numerous interesting protein
syntheses.38,39 This conjugation method is carried out in
aqueous solution but has the advantage that efficient
coupling is also achieved in the presence of denaturing
agents and organic solvent additives and thus is suitable
in principle for both hydrophobic and hydrophilic pep-
tides. As we began our studies, the only reported pep-
tide-oligonucleotide conjugation procedure based on a
ligation approach was the coupling of a 3′-amino oligo-
nucleotide to a peptide thioester oligonucleotide aligned
on a DNA template.40 However, the requirement for three
separate oligonucleotide chains, one of which must carry
a 3′-amino group, makes this synthesis route highly
impractical and is further restricted to 3′-peptide conju-
gated oligonucleotides. Very recently, the native ligation
of a 5′-thioester functionalized RNA with a Cys-Gly
dipeptide and with a 13-mer cysteine-containing peptide
was reported.41
We have reported recently a solid-phase peptide frag-
ment coupling strategy, which was precedented by a
previous report of a tripeptide conjugation by Grandas
et al.34 In our studies, oligonucleotide and peptide
moieties were assembled separately on their own sup-
ports but only the peptide fragment was removed from
its support and purified as a peptide that now carries
only NR-Fmoc protection. We then explored the solid-
phase conjugation of such peptide fragments under
various peptide bond-forming conditions. Here, the C-
terminus of the peptide fragment was reacted with
support-bound oligonucleotides that had been 5′-func-
tionalized on-column with a primary amino group
We have now developed a novel and general method
of peptide-oligonucleotide conjugation based on the
principle of “native ligation” which is template free. The
method involves conjugation of a peptide fragment as an
(26) Eritja, R.; Pons, A.; Escarceller, M.; Giralt, E.; Albericio, F.
Tetrahedron 1991, 47, 4113-4120.
(27) Bongartz, J . P.; Aubertin, A. M.; Milhaud, P. G.; Lebleu, B. Nucl.
Acids Res. 1994, 22, 4681-4688.
(28) Tung, C.-H.; Rudolph, M. J .; Stein, S. Bioconjugate Chem. 1991,
2, 464-465.
(35) Peyrottes, S.; Mestre, B.; Burlina, F.; Gait, M. J . Tetrahedron
1998, 54, 12513-12522.
(29) Harrison, J . G.; Balasubramanian, S. Nucl. Acids Res. 1998,
26, 3136-3145.
(36) Peyrottes, S.; Mestre, B.; Burlina, F.; Gait, M. J . Nucleosides
Nucleotides 1999, 18, 1443-1448.
(30) Soukchareun, S.; Haralambidis, J .; Tregear, G. Bioconjugate
Chem. 1998, 9, 466-475.
(37) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H.
Science 1994, 266, 776-779.
(31) Arar, K.; Aubertin, A.-M.; Roche, A.-C.; Monsigny, M.; Mayer,
M. Bioconjugate Chem. 1995, 6, 573-577.
(32) McMinn, D. L.; Greenberg, M. M. Bioorg. Med. Chem. Lett.
1999, 9, 547-550.
(38) Wilken, J .; Kent, S. B. H. Curr. Opin. Biotech. 1998, 9, 412-
426.
(39) Canne, L. E.; Botti, P.; Simon, R. J .; Chem, Y.; Dennis, E. A.;
Kent, S. B. H. J . Am. Chem. Soc. 1999, 121, 8720-8727.
(40) Bruick, R. K.; Dawson, P. E.; Kent, S. B. H.; Usman, N.; J oyce,
G. F. Chem. Biol. 1996, 3, 49-56.
(41) McPherson, M.; Wright, M. C.; Lohse, P. A. Synlett 1999, S1,
978-980.
(33) Kahl, J . D.; Greenberg, M. M. J . Org. Chem. 1999, 64, 507-
510.
(34) Grandas, A.; Robles, J .; Pedroso, E. Nucleosides Nucleotides
1995, 14, 825-828.

4902 J . Org. Chem., Vol. 65, No. 16, 2000
Stetsenko and Gait
Sch em e 1a
Sch em e 2a
a
Key: (a) BnSH, DMAP, pyridine; (b) DCC, pentafluorophenol,
CH2Cl2.
a
Key: (a) either 2 (4.5 equiv), HOAt (1 equiv), DMF, 4 h, rt, or
N-terminal thioester to an oligonucleotide functionalized
at the 5′-end with a cysteine moiety.
1 (5 equiv), HATU (4.5 equiv), DIEA (10 equiv), DMF, 1 h, rt; (b)
TFA-BnSH-PhOH-H2O (90:5:2.5:2.5 v/v/v/v), 1-6 h, rt.
Resu lts
to note that no evidence was found for loss of the S-benzyl
group from the thiosuccinate during the acidic deprotec-
tion step.
For protein synthesis by native ligation, it is common
for C-terminal thioesters of peptide fragments to be
prepared. Such C-terminal thioesters have been obtained
primarily by Boc/benzyl peptide assembly chemistry on
specially designed solid supports either through alkyla-
tion of a thio acid intermediate42 or more recently by a
direct route.43 Use of the milder Fmoc/tert-butyl chem-
istry for peptide assembly has been hampered by the
instability of polymer-anchored thioesters to repetitive
piperidine treatment required for N-terminal Fmoc re-
moval. The use of less nucleophilic secondary amines for
N-deprotection as well as an HOBt amine scavenger has
improved yields somewhat.44 But, in general, post-as-
sembly introduction of a thioester is necessary.
No suitable route for introduction of a C-terminal
thioester following peptide assembly by the Fmoc/tert-
butyl method has been reported until very recently.45
Therefore to enable initial studies of peptide-oligonucle-
otide conjugation via native ligation, it was simpler to
develop a method for synthesis of N-terminal thioesters
of peptide fragments. For this purpose we designed a new
reagent, pentafluorophenyl S-benzylthiosuccinate 2, which
is prepared in two steps. First, reaction of succinic
anhydride with benzyl mercaptan produced S-benzylth-
iosuccinic acid in 80% yield, which was converted into
the corresponding pentafluorophenyl ester in 92% yield
(Scheme 1).
The thiosuccinate 2 may be incorporated into a peptide
fragment following standard Fmoc/tert-butyl peptide
assembly on a commercially available peptide synthesizer
(Scheme 2). In order that the C-terminus of the peptide
is protected from possible interference with native liga-
tion, we have utilized a PEG-polystyrene support con-
taining a standard Rink amide linker or a PAL linker.
Thus N-terminal benzyl thioester peptide is released into
solution from the solid support as a C-terminal amide
during side-chain deprotection by treatment with TFA/
phenol/benzyl mercaptan/water (92.5:2.5:2.5:2.5 v/v). A
range of peptides containing N-terminal S-benzylthio-
succinate was assembled and deprotected. Each peptide
product was assayed for purity by reversed-phase HPLC
as well as for the correct mass of the major product by
MALDI-TOF mass spectrometry (Table 1). It is important
One advantage of preparation of N-terminal thioesters
is that the thioester group is spaced away from the
terminal amino acid. This is helpful in that C-terminal
thioesters of peptides containing sterically hindered
amino acids at the C-terminus, such as threonine, iso-
leucine, valine or proline, are known to couple only very
slowly in native ligation reactions.43 A second advantage
is that there is no possibility of peptide racemization
during conjugation reactions.
To obtain oligonucleotides containing a cysteine moiety,
we synthesized a novel phosphoramidite reagent 5 in two
simple steps (Scheme 3). Thus, trans-4-aminocyclohex-
anol was acylated by reaction with the pentafluorophenyl
ester of N-R-Fmoc-S-tert-butylsulfenylcysteine in 82%
yield. The corresponding phosphoramidite was prepared
by subsequent standard phosphitylation chemistry and
purified by column chromatography in 66% yield. Even
higher yields (95%) were obtained in preparation of the
corresponding S-trityl-protected cysteine phosphoramid-
ite 6.
Whereas many commercially available phosphoramid-
ite reagents used for 5′- functionalization of oligonucle-
otides that utilize short linear alkyl chains are oils,
phosphoramidite 5 containing the trans-4-aminocyclo-
hexanol moiety, a secondary alcohol, is a solid and is
much more stable. Phosphoramidite 5 was used in the
final step of coupling following standard oligonucleotide
assembly on a controlled pore glass solid-support by the
phosphoramidite method (1 µM nucleoside bound to the
support) using a commercial DNA/RNA synthesizer. A
number of oligonucleotides were prepared in this way,
deprotected at the nucleobases and released into solution
by treatment with concentrated aqueous ammonia solu-
tion at 55 °C. The S-tert-butylsulfenyl protecting group
is maintained but the NR-Fmoc group is removed from
the cysteine moiety during ammonia treatment. The
cysteine-modified oligonucleotides were characterized by
ion exchange and/or reversed phase HPLC and by
MALDI-TOF mass spectrometry (Table 2). Average cou-
pling yields of phosphoramidites 5 and 6 were >97% as
judged by HPLC analysis following deprotection.
As a first trial, we then evaluated the conjugation in
aqueous solution of a peptide containing an N-terminal
thioester to an oligonucleotide containing a 5′-cysteine
(Scheme 4). We chose in each case a peptide and an
oligonucleotide fragment, containing their respective
functionalities, that had been assembled in high yield as
judged by HPLC analysis, and that could be used
therefore without the need for HPLC purification. We
(42) Canne, L. E.; Walker, S. M.; Kent, S. H. B. Tetrahedron Lett.
1995, 36, 1217-1220.
(43) Hackeng, T. M.; Griggin, J . H.; Dawson, P. E. Proc. Natl. Acad.
Sci. U.S.A. 1999, 96, 10068-10073.
(44) Li, X.; Kawakami, T.; Aimoto, S. Tetrahedron Lett. 1998, 39,
8669-8672.
(45) Ingenito, R.; Bianchi, E.; Fattori, D.; Pessi, A. J . Am. Chem.
Soc. 1999, 121, 11369-11374.

Conjugation of Peptides to Oligonucleotides by “Native Ligation”
J . Org. Chem., Vol. 65, No. 16, 2000 4903
Ta ble 1. Syn th esis of P ep tid e N-Ter m in a l Th ioester sa
HPLC retention
time, min
purity of crude
yield,e
%
peptide sequenceb
PTSQSRGDPTGPKE P1
Sar-Leu-Gly-Ile-Gly P 2
ALPPLERLTL P 3
GALFLGFLGAAGSTMGAWSQPKSKRKV P 4
PQIK(Tfa)IWFPNRRK(Tfa)PFK(Tfa)K(Tfa) P 5
GRK(Tfa)K(Tfa)RRQRRR
DRVIEVVQGAYRAIRNIPRRIRQG
QAKKKKLDK
MALDI-TOF MSc
product,d
%
1660.2 (1660.8)
657.9 (657.8)
13.6
17.6
19.6
21.6
22.4
15.1
20.0
11.9
98.5
92.2
74.7
48.3
72.0
76.5
28.6
90.3
86.6f
89.9f
43.3
25.0
44.2
38.1
17.8
68.9
1325.7 (1326.7)
2969.1 (2969.6)
2715.0 (2715.9)
1791.6 (1792.9)
3040.6 (3039.7)
1289.9 (1290.8)
a
b
Abbreviation used: Tfa ) trifluoroacetyl. All the peptides contain an N-terminal S-benzyl thiosuccinyl group and are C-terminal
amides. c Values in parentheses are calculated. Integrated from HPLC traces. e Calculated from resin mass and loading indicated by
d
supplier. f Not purified.
Sch em e 3a
oligonucleotide had reacted to give a major product with
later elution time. Following HPLC purification, the
product was analyzed by MALDI-TOF mass spectrometry
and found to have the expected mass for the desired
peptide-oligonucleotide conjugate. The isolated yield of
conjugate based on the amount of unpurified oligonucle-
otide was 65% (Table 3).
These first two successful examples of peptide-oligo-
nucleotide native ligation involved use of a peptide P 1
that contained an N-terminal proline. One possible
concern of the use of the S-benzylthiosuccinate linker was
that under basic conditions a cyclization side-reaction
could take place, involving attack of a deprotonated
N-terminal amido group of the peptide on the thioester
carbonyl group. Such a side reaction cannot take place
in the case of N-terminal proline, since the amido group
is secondary and does not contain an acidic hydrogen
atom. To determine whether such a side reaction can take
place we chose a particularly challenging conjugation of
a long 27-mer peptide P 4 that contains an N-terminal
glycine, an unhindered amino acid that would be the most
likely to be prone to cyclization. Peptide P 4 contains both
a long hydrophobic section as well as a C-terminal basic
region. Thus, peptide P 4 (2.5 µmol) and oligonucleotide
ODN2 (0.5 µmol) were incubated at pH 7.5 in the absence
of denaturing agent, but in the presence of TCEP and
thiophenol. After 48 h at room temperature, HPLC
examination showed that no conjugation product was
obtained. Instead, MALDI-TOF mass spectra of the crude
reaction mixture showed that most of the peptide P 4 had
been converted into two products, one consistent in mass
with N-terminal cyclization and loss of the thioester
moiety and the other consistent with the free carboxylic
acid hydrolysis product (data not shown).
To circumvent the cyclization reaction and also to
investigate alternative parameters for conjugation, we
chose two further peptide conjugation examples. A pep-
tide containing an N-terminal sarcosine (N-methyl gly-
cine) that cannot undergo cyclization, peptide P 2, was
conjugated to oligonucleotide ODN2 and also to ODN3.
In the conjugation to ODN2, acetonitrile was used as
cosolvent (conditions C), whereas DMF was cosolvent in
conjugation to ODN 3 (conditions D). The concentrations
of peptide and oligonucleotide were 10-fold lower in each
case. In addition, since we noticed that the rate of
reduction by TCEP may be pH-dependent, we carried out
the ODN2 conjugation reaction at pH 6.5 (conditions C)
for 48 h and in the case of ODN3 a prereduction for 3 h
by TCEP at pH 6.5 was followed by alteration of the pH
to 7.5 and addition of the peptide component for a
reaction time of 48 h (conditions D). In these cases the
major products of the reactions were the expected con-
a
Key: (a) trans-4-aminocyclohexanol hydrochloride, Et3N, DMF;
(b) 2-cyanoethyl-N,N-diisopropylamino chlorophosphine, DIEA,
CH2Cl2 (5); (c) 2-cyanoethyl N,N,N′,N′-tetraisopropyl phosphor-
diamidite, diisopropylammonium tetrazolide, CH2Cl2 (6).
found that it was possible to avoid the necessity for prior
removal of the S-tert-butylsulfenyl protecting group from
the peptide fragment by in situ reduction with the water
soluble reducing agent tris(2-carboxyethyl)phosphine
(TCEP)46 during the native ligation reaction. To facilitate
thiol exchange and to improve reactivity, excess thiophe-
nol was also added to the conjugation reaction.47 The
conjugation was carried out for 24 h at room temperature
between a 14-mer peptide P1 (2.5 µmol) and a 5-mer
oligonucleotide of thymidine ODN1 (0.5 µmol) in 1 mL
of 0.1 M phosphate buffer (pH 7.5) in the presence of 7
M urea (conditions A, Table 3and Figure 2). The products
were separated by reversed-phase HPLC, which showed
predominantly a single product of conjugate with a
slightly later elution time compared to the starting
oligonucleotide (Figure 1a). Following HPLC purification,
the product (75% isolated yield starting from unpurified
oligonucleotide) was analyzed by MALDI-TOF mass
spectrometry and found to have the expected mass for
the desired peptide-oligonucleotide conjugate.
As a second example, the reaction products of conjuga-
tion of excess 14-mer peptide N-terminal benzyl thioester
P 1 with a 5′-cysteinyl 15-mer mixed sequence phosphodi-
ester oligodeoxyribonucleotide ODN2 were separated by
reversed phase HPLC (Figure 1b). The conditions of
conjugation were identical except that the 7 M urea was
omitted (conditions B). This showed that all the cysteinyl
(46) Burns, J . A.; Butler, J . C.; Moran, J .; Whitesides, G. M. J . Org.
Chem. 1991, 56, 2648-2650.
(47) Dawson, P. E.; Churchill, M. J .; Ghadri, M. R.; Kent, S. B. H.
J . Am. Chem. Soc. 1997, 119, 4325-4329.

4904 J . Org. Chem., Vol. 65, No. 16, 2000
Ta ble 2. Syn th esis of 5′-Cystein yloligon u cleotid esa
Stetsenko and Gait
HPLC retention
time, min
purity of crude
isolated
oligonucleotide sequenceb
H-Cys(StBu)-TTT TT
H-Cys(Trt)-TTT TT
H-Cys-TTT TT
H-Cys(StBu)-CTC CCA GGC TCA AAT
H-Cys(StBu)-GCT CCC AGG CTC AAA
H-Cys(StBu)-AGC TCC CAG GCT CAA
H-Cys-GCT CCC AGG CTC AAA
H-Cys(Trt)-CTC CCA GGC TCA AAT
H-Cys(Trt)-GCT CCC AGG CTC AAA
H-Cys(Trt)-AGC TCC CAG GCT CAA
MALDI-TOF MSc
product,d
%
yield,e %
1827.1 (1827.4)
1983.2 (1982.6)
1738.2 (1739.4)
4868.0 (4866.4)
4888.1 (4888.9)
4888.0 (4888.9)
4803.1 (4804.0)
5017.1 (5018.0)
5043.7 (5043.0)
5043.9 (5043.0)
16.6
22.1
14.8
15.0
14.4
17.0
13.7
21.5
20.8
20.7
87.1
76.6
82.2
82.6
89.7
86.5
100
91.2
89.3
90.5
76.3
72.5
ND
61.6
66.8
60.3
ND
72.2
68.4
70.1
a
b
Abbreviations used: StBu ) tert-butylsulfenyl, Trt ) triphenylmethyl, ND ) not determined. All the oligonucleotides contain 5′-
terminal S-protected or unprotected trans-N-(L-cysteinyl)aminocyclohexyl phosphate group (denoted as Cys). c Values in parentheses are
d
e
calculated. Integrated from HPLC traces. A260 units obtained from 1 µmol synthesis.
Sch em e 4a
a
Key: (a) peptide thioester (5 equiv), 100 mM TCEP, excess PhSH under various solvent and buffer conditions (see the Experimental
Section), rt, 24 h.
Ta ble 3. Syn th esis of P ep tid e-N-to-5′-oligon u cleotid e Con ju ga tes
HPLC retention
time, min
yield,c
%
oligonucleotide sequencea
TTT TT ODN1
GCT CCC AGG CTC AAA ODN2
GCT CCC AGG CTC AAA ODN2
GCT CCC AGG CTC AAA ODN2
AGC TCC CAG GCT CAA ODN3
AGC TCC CAG GCT CAA ODN3
peptide sequencea
method
MALDI-TOF MSb
PTSQSRGDPTGPKE P 1
PTSQSRGDPTGPKE P 1
Sar-Leu-Gly-Ile-Gly P 2
ALPPLERLTL P 3
Sar-Leu-Gly-Ile-Gly P 2
ALPPLERLTL P 3
A
B
C
C
D
D
3276.7 (3275.6)
6340.2 (6340.5)
5349.0 (5349.1)
6043.4 (6044.5)
5348.5 (5349.1)
6043.9 (6044.5)
15.8
15.6
17.6
20.1
17.4
19.8
75
65
26d
22
43e
51
a
All the peptide-oligonucleotide conjugates are linked from the N-terminus to the 5′-end via (N-succinyl-L-cysteinyl)-trans-
aminocyclohexyl phosphate group. Values in parentheses are calculated. c Calculated from A260 units of starting crude oligonucleotide.
b
Additionally, 17% of mixed disulfide with thiophenol was isolated. e 14% of mixed disulfide with thiophenol was isolated.
d
jugates as seen from HPLC, which were isolated in 26
and 43% yields respectively (Table 3). There was also in
each case a side product of the mixed disulfide of
conjugate with thiophenol. A hydrophobic 10-mer peptide
P 3, which has a N-terminal alanine and which has the
sequence corresponding to a nuclear export signal was
also conjugated to ODN2 and ODN3 under the same
conditions as above. Once again the major product by
HPLC was the desired conjugates which were isolated
in 22 and 51% yields, respectively. No cyclized product
was detected in conjugations with peptide P 3. In each
case the product conjugate showed a single peak after
HPLC purification and the correct MALDI-TOF mass
(Table 3).
Discu ssion
Efficient and convenient procedures for conjugation of
oligonucleotides to peptides have hitherto proved elusive.
The native ligation strategy we have developed has many
attractions in this respect. First, the reagents used for
functionalization of peptide and oligonucleotide moieties
are stable solids that are prepared in each case in two
steps from readily accessible materials. Second, the
reagents are used in standard protocols of solid-phase

Conjugation of Peptides to Oligonucleotides by “Native Ligation”
J . Org. Chem., Vol. 65, No. 16, 2000 4905
F igu r e 1. (a) Analytical reversed-phase HPLC of crude (A) H-PQIK(Tfa)IWFPNRRK(Tfa)PFK(Tfa)K(Tfa)-NH2 and (B) P 5 thioester
(Table 1, entry 5), as obtained after acidolytic cleavage. HPLC conditions as listed in the Experimental Section, General Procedures.
(b) Analytical HPLC of (cysteinylamino)cyclohexyl-modified oligonucleotides: (A) H-Cys-GCT CCC AGG CTC AAA ODN2 (Table
2, entry 7), (B) H-Cys(StBu) ODN2 (Table 2, entry 5), and (C) H-Cys(Trt)-TTT TT ODN1 (Table 2, entry 2). HPLC conditions (a),
as listed in the Experimental Section.
F igu r e 2. (a) Reversed-phase HPLC of ODN1-P 1 conjugate (Table 3, entry 1), reaction mixture after 24 h, (conditions b). Inset:
purified conjugate (conditions a). (b) Reversed-phase HPLC of ODN2-P 1 conjugate (Table 3, entry 2), reaction mixture after 24
h (conditions b). Inset: purified conjugate (conditions b). Note that the most prominent peak eluting at around 20 min is thiophenol.
synthesis and couple to their respective peptide or
oligonucleotide partners in high yield. Third, assuming
the peptide and oligonucleotide fragments are themselves
assembled efficiently, the functionalized oligonucleotides
and peptides may be used in unpurified form for conjuga-
tion without the need for further manipulation following
deprotection and release from their respective solid
supports.
mercaptan that was recently suggested,43 or addition of
another reducing agent after ligation is complete. A side-
reaction resulting from intramolecular cyclization of the
succinyl functionality on the peptide was observed in
place of conjugation in the case of N-terminal glycine,
but no cyclization was detected in the case of a peptide
containing N-terminal alanine. Use of N-terminal proline
or sarcosine peptides obviates any possibility of cycliza-
tion. We have also learnt that in cases where conjugation
may be slow, prior TCEP reduction of the oligonucleotide
component at pH 6.5 to remove the tert-butylsulfenyl
protecting group may also be helpful.
We have shown some examples of conjugation reactions
using peptides and oligonucleotides of mixed composition
and of length and type appropriate for biological studies.
In these examples, we have begun to learn about the
requirements for efficient conjugation. First, the rate of
conjugation appears to be dependent on the concentra-
tions of the two components. Reactions carried out with
2.5 mM peptide P 1 and 0.5 mM oligonucleotide gave high
yields within a 24 h period either in the presence or in
the absence of a denaturing agent (7 M urea). By
contrast, reactions carried out at 10-fold lower concentra-
tions (e.g., with peptides P 2 and P 3) still showed some
starting oligonucleotide even after 48 h and some mixed
disulfide side-products were also in evidence in some
cases in addition to the desired product. Nevertheless the
conjugate was the major product in all examples shown.
We are currently exploring the possibility of addition of
a more reducing thiol during ligation, such as benzyl
Further experimentation will be necessary to optimize
the solvent composition and reaction conditions for
conjugation, especially in the case of longer peptides that
may have secondary structure, but in the case of hydro-
phobic peptide segments the addition of DMF has helped
to maintain solubility. But there is already a sufficient
number of examples to suggest that in principle a very
wide range of peptides should be able to be conjugated
to oligonucleotides by the native ligation technique and
there should be very few restrictions to peptide sequence.
It should be noted that internal cysteine residues in a
peptide do not interfere with native ligation reactions.37
Our ligation method has advantages over that described
recently for RNA-peptide conjugation by McPherson et

4906 J . Org. Chem., Vol. 65, No. 16, 2000
Stetsenko and Gait
al.41 Our method is in principle suitable for conjugation
of peptides containing thioesters at the C- or N-terminus
and utilizes chemically synthesized 5′-cysteine-substi-
tuted oligonucleotides obtained by standard phosphora-
midite methods. By contrast, the method of McPherson
et al is restricted to N-terminal cysteine-containing
peptides and to transcribed RNA which is functionalized
with a 5′-thiophosphate and then alkylated to produce a
5′-thioester. Further, reagent 2 could be used to prepare
thioesters from other amino-functionalized biomolecules.
In the future, we plan to investigate the conjugation
of peptides to 3′-cysteine-functionalized oligonucleotides,
to oligonucleotides bound to solid supports, as well as the
use in conjugation reactions of C-terminal peptide
thioesters prepared by new Fmoc synthetic routes.45,48 We
also hope to extend the conjugation chemistry to other
oligonucleotide analogues which are more commonly used
for antisense experiments, such as oligodeoxynucleotide
phosphorothioates and 2′-O-alkyl oligoribonucleotides,
that are more resistant to attack by nucleases. There
should be no impediment in principle to such conjuga-
tions using native ligation technology. In applications of
such conjugates in cellular uptake studies, the presence
of a free thiol in the cysteine residue remaining at the
join between peptide and oligonucleotide parts could be
useful for attachment of a fluorescent or other reporter
group. In conclusion, we believe the method should be
generally applicable for the development of procedures
for efficient peptide-oligonucleotide conjugation.
(25 × 300 mm) and 215 nm UV detection. Oligonucleotide and
conjugate analytical and semipreparative HPLC was carried
out using a Phenomenex RP-C18 column (4.6 × 250 mm) and
dual wavelength (218 and 254 nm) UV detection.
Gen er a l Meth od s for Solid -P h a se P ep tid e Syn th esis.
Solid-phase peptide synthesis was carried out by the solid-
phase Fmoc-tert-butyl procedures49 on a PE Biosystems Pio-
neer peptide synthesis system on a 0.1 mmol scale using
HATU/DIEA in situ activation protocols50 and either a Rink
NovaGel HL or a PAL-PEG-PS solid support. After removal
of the last N-R-Fmoc group, pentafluorophenyl S-benzyl thio-
succinate 1 was manually coupled to the last amino acid (4.5
equiv of 1, 1 equiv of HOAt in 2 mL of DMF) for 4 h at room
temperature. Alternatively, S-benzyl thiosuccinic acid can be
coupled using the same HATU protocol as for standard peptide
chain assembly. Then the resin was washed with DMF (5 × 5
mL), MeOH (3 × 5 mL), and diethyl ether (2 × 5 mL), and
dried. Cleavage from the support and deprotection of the side
chain protecting groups was carried out in TFA-benzyl
mercaptan-phenol-water (90:5:2.5:2.5 v/v/v/v) mixture for
1-6 h at room temperature depending on NG-2,2,4,6,7-pen-
tamethyldihydrobenzofuran-5-sulfonyl (Pbf) arginine content.
The filtrate was flushed by a stream of nitrogen to remove
most of the TFA and the peptide thioesters were precipitated
by addition of cold (-20 °C) diethyl ether, washed thrice with
diethyl ether, and dried in vacuo. The solid was redissolved
in 0.1% TFA with a variable proportion of acetonitrile (10-
30%) to solubilize hydrophobic peptide thioesters, to ca. 3-5
mg mL-1 concentration, and subjected to HPLC analysis using
a gradient of acetonitrile in 0.1% aqueous TFA from 10 to 90%
in 30 min. If needed, peptide thioesters were purified by
preparative HPLC. Appropriate fractions were pooled, lyoph-
ilized, and analyzed by MALDI-TOF MS. Analytical data of
the synthesized peptide thioesters are given in Table 1.
Exp er im en ta l Section
Gen er a l Meth od s for Solid -P h a se Oligon u cleotid e
Syn th esis. Assembly of oligonucleotides was carried out by
the standard 2-cyanoethyl phosphoramidite method51 on long
chain alkylamine controlled pore glass (LCAA-CPG) support
(Glen Research via Cambio). All oligonucleotides were syn-
thesized on a 1 µmol scale using an ABI 380B DNA/RNA
Synthesizer. Protected 2′-deoxyribonucleoside phosphoramid-
ites were obtained from Cruachem (Scotland). After the last
dimethoxytrityl group removal, modified phosphoramidite 5
or 6 (0.15 M solution in dry acetonitrile) was coupled to the
support-bound oligonucleotide using an extended coupling
protocol (10 min) to ensure complete coupling. After the usual
iodine-water oxidation, the support was manually flushed with
20% piperidine solution in DMF for 10 min to remove Fmoc
groups, washed with 10 mL of DMF, 10 mL of acetonitrile and
briefly air-dried. The glass beads were then treated with 0.5
mL of 30% aqueous ammonia solution at room temperature
for 2 h to cleave the oligonucleotide from the support, washed
with an additional 0.5 mL of concentrated ammonia solution,
and the filtrate was then transferred to a screw-capped
polypropylene tube and heated at 55 °C for 16 h to completely
deprotect oligonucleotides at the nucleobase and phosphate
residues. After cooling and evaporating most of the solution
under a stream of nitrogen, 1 mL of deionized water was added
and the solution was evaporated on a SpeedVac vacuum
concentrator to dryness, product redissolved in deionized
water, and the quantity of the oligonucleotide was then
assessed by checking the absorbance at 260 nm on a Perkin-
Elmer Lambda 2 UV-vis spectrophotometer. Purity of oligo-
nucleotides was established by analytical HPLC using gradi-
ents of acetonitrile 0-2%, 5 min, 2-40% 20 min, 40-100% 25
min in either 0.1 M ammonium acetate buffer, pH 7.0 (a), or
0.1M triethylammonium acetate buffer, pH 7.0 (b). Molecular
Gen er a l P r oced u r es. Unless otherwise noted, materials
were obtained from commercial suppliers and used without
further purification. Rink amide NovaGel HL and Fmoc-PAL-
PEG-PS resins were purchased from Novabiochem and PE
Biosystems, respectively. Fmoc-amino acids and derivatives
were purchased from Novabiochem, Millipore and PE Biosys-
tems. DMF (Fisher) was distilled in vacuo and used shortly
after. Dichloromethane and acetonitrile (Fisher) were refluxed
over CaH2 followed by distillation. Basic solvents were pur-
chased from Fisher. Standard peptide synthesis reagents were
purchased from PE Biosystems (HATU, HOAt, DIEA) and
Romil (UK) (piperidine, trifluoroacetic acid). All other fine
chemicals were supplied by Aldrich and Fluka. Thin-layer
chromatography was carried out on Merck 60 F254 aluminum-
coated silica gel plates. TLC elution systems: (a) EtOAc-CH2-
Cl2-AcOH (10:85:5 v/v/v) (system A); (b) EtOAc-hexane (1:1
v/v) (system B); (c) MeOH-CHCl3-AcOH (2:93:5 v/v/v) (system
C); (d) EtOAc-hexane-NEt3 (35:60:5 v/v/v) (system D). Spots
were visualized by UV irradiation with a 254 nm lamp.
Column chromatography was carried out using Macherey-
Nagel 60 230-400 mesh silica gel. All 1H and 31P NMR spectra
were recorded at 300 MHz. MALDI-TOF mass spectra were
recorded on
a Voyager-DE workstation (PE Biosystems).
Matrixes used for preparing MALDI-TOF samples were the
following: (a) R-cyano-4-hydroxycinnamic acid, 10 mg mL-1
in acetonitrile-3% aqueous TFA (1:1 v/v) for all peptides, (b)
2,6-dihydroxyacetophenone, 20 mg mL-1, and diammonium
hydrogen citrate, 40 mg mL-1, in 50% aqueous methanol for
all oligonucleotides, conjugates and phosphoramidites, and (c)
2,5-dihydroxybenzoic acid, 10 mg mL-1 in methanol for all
other low molecular weight compounds. Elemental analyses
were carried out by The Chemical Laboratory, University of
Cambridge, UK. Analytical HPLC of peptides was carried out
with dual wavelenth (215 and 230 nm) UV detection using a
Vydac RP-C8 column (4.6 × 250 mm). Preparative peptide
chromatography was carried out using a Vydac RP-C8 column
(49) Atherton, E.; Sheppard, R. C. Solid-phase peptide synthesis: A
practical approach; Oxford University Press: Oxford, 1989.
(50) Carpino, L. A.; El-Faham, A.; Minor, C. A.; Albericio, F. J .
Chem. Soc., Chem. Commun. 1994, 201-202.
(51) Brown, T.; Brown, D. J . S. In Oligonucleotides and Analogues:
A Practical Approach; Eckstein, F., Ed.; OUP: Oxford, 1991; pp 1-24.
(48) Shin, Y.; Wianans, K. A.; Backes, B. J .; Kent, S. B. H.; Ellman,
J . A.; Bertozzi, C. R. J . Am. Chem. Soc. 1999, 121, 11684-11689.

Conjugation of Peptides to Oligonucleotides by “Native Ligation”
J . Org. Chem., Vol. 65, No. 16, 2000 4907
masses were confirmed by MALDI-TOF mass spectrometry
(positive ion mode). Analytical data are summarized in Table
2.
mL of anhydrous DMF was added triethylamine (0.45 mL, 3.1
mmol), and the resulting solution was stirred at room tem-
perature for 3 h until TLC showed complete reaction. The
reaction mixture was evaporated to dryness and the white
residue transferred to a sintered glass filter, washed succes-
sively with small amounts of DMF, ethanol, and diethyl ether,
and dried in vacuo. Yield of white powder: 0.87 g (82%). TLC
(C): Rf 0.35. 1H NMR (DMSO-d6): δ 1.18 (m, 4H), 1.27 (s, 9H),
1.69 (m, 2H), 1.77 (m, 2H), 2.95 (m, 2H), 3.45 (m, 2H), 4.24
(m, 4H), 7.30 (t, 2H, J ) 7.4 Hz), 7.41 (t, 2H, J ) 7.4 Hz), 7.61
(d, 1H, J ) 8.5 Hz), 7.72 (d, 2H, J ) 7.3 Hz), 7.86 (m, 3H).
MALDI-TOF MS: [M + H]+ 528.9 (529.7 calcd), [M + Na]+
550.8 (551.7 calcd), [M + K]+ 566.6 (567.7 calcd). Anal. Calcd
for C28H36N2O4S2: C, 63.61; H, 6.86; N, 5.30. Found: C, 63.85;
H, 6.84; N, 5.28.
Gen er a l Meth od s for P ep tid e-Oligon u cleotid e Con -
ju ga tion by Na tive Liga tion . The following oligonucleotide
sequences were used for model conjugation studies: TTT TT
(ODN1), GCT CCC AGG CTC AAA (ODN2), and AGC TCC
CAG GCT CAA (ODN3). Three representative peptide N-
terminal thioesters were chosen for conjugation: HIV-1 Tat
protein C-terminal 14-mer PTSQSRGDPTGPKE (P 1), a hy-
drophobic pentapeptide Sar-Leu-Gly-Ile-Gly (P 2), and HIV-1
Rev protein nuclear export signal (NES) decamer ALPPLER-
LTL (P 3). Four sets of conditions in generally 1 mL reaction
volumes were explored: (a) 0.5 mM ODN, 5 equiv of peptide,
0.1 M TCEP, 2% PhSH (v/v), 7 M urea, 0.1 M sodium
phosphate, pH 7.5, 24 h, room temperature (conditions A); (b)
0.5 mM ODN, 5 equiv of peptide, 0.1 M TCEP, 2% PhSH (v/
v), 25% DMF (v/v) in 0.1 M sodium phosphate, pH 7.5, 24 h,
room temperature (conditions B); (c) 0.05 mM ODN, 10 equiv
of peptide, 0.1 M TCEP, titrated to pH 6.5 by addition of 20%
sodium hydroxide solution, 2% PhSH (v/v), 25% acetonitrile,
48 h, room temperature (conditions C); (d) 0.1 mM ODN,
pretreated with 0.2 M TCEP, pH 6.5, for 3 h at room
temperature, then 10 equiv, peptide in an equal volume of 50%
aqueous DMF was added together with PhSH (2% v/v final)
and kept at room temperature for 48 h (conditions D). Reaction
mixtures were analyzed by reversed phase HPLC and products
were separated by semipreparative RP-HPLC. Fractions con-
taining products were lyophilized twice from deionized water
and analyzed by MALDI-TOF. Analytical data for conjugates
synthesized are given in Table 3.
N-r-F m oc-S-t r it yl-L-cyst ein e 4-h yd r oxy-tr a n s-cyclo-
h exyla m id e (4). To a slurry of trans-4-aminocyclohexanol
hydrochloride (0.32 g, 2.1 mmol) and N-R-Fmoc-S-trityl-L-
cysteine pentafluorophenyl ester (1.5 g, 2 mmol) in 25 mL of
anhydrous DMF was added triethylamine (0.31 mL, 2.2 mmol),
and the resulting solution was stirred at room temperature
for 3 h until TLC showed complete reaction. The reaction
mixture was evaporated to dryness, redissolved in ethyl
acetate and washed successively with ice-cold 5 wt % citric
acid solution, water, 5% sodium bicarbonate solution, and
brine, dried over sodium sulfate, and evaporated to a light
brown foam. The residue was chromatographed on a silica gel
column eluted by 5-15% ethyl acetate in hexane + 0.5%
triethylamine. Appropriate fractions were pooled and evapo-
rated to give 1.31 g (95%) of the title product as a white foam.
1
TLC (C): Rf 0.42. H NMR (DMSO-d6): δ 1.17 (m, 4H), 1.65
S-Ben zyl Th iosu ccin ic Acid (1). Benzyl mercaptan (2.595
mL, 22 mmol) was added under nitrogen to a stirred solution
of succinic anhydride (2.00 g, 20 mmol) and 4-(dimethylamino)-
pyridine (122.2 mg, 1 mmol) in 25 mL of anhydrous acetoni-
trile/pyridine (9:1 v/v). Stirring was continued at room tem-
(m, 2H), 1.75 (m, 2H), 2.29 (m, 2H), 3.36 (m, 2H+water), 4.01
(q, 1H, J ) 7.0 Hz), 4.22 (m, 3H), 4.50 (d, 1H, J ) 4.3 Hz),
7.28 (m, 17H), 7.39 (t, 2H, J ) 7.4 Hz), 7.57 (d, 1H, J ) 8.7
Hz), 7.66 (d, 1H, J ) 7.7 Hz), 7.72 (d, 2H, J ) 7.3 Hz), 7.87 (d,
2H, J ) 7.5 Hz). MALDI-TOF MS: [M + Na]+ 704.1 (705.9
perature for 3 h and the mixture was evaporated to near
dryness. The product was dissolved in 30 mL of aqueous
sodium bicarbonate solution, pH 8.5, and extracted twice with
10 mL of diethyl ether. The aqueous phase was then cooled in
an ice bath and acidified with 5 N hydrochloric acid to pH 2.
The white precipitate was filtered off, washed with cold 0.1M
HCl solution, followed by ice-cold water, and dried in a vacuum
desiccator over phosphorus pentoxide overnight. Yield of white
powder 3.6 g (80%). TLC (A): Rf 0.76. 1H NMR (CDCl3): δ
2.74 (t, 2H, J ) 6.6 Hz), 2.92 (t, 2H, J ) 6.8 Hz), 4.16 (s, 2H),
7.27 (m, 5H). MALDI-TOF MS: [M + Na]+ 247.6 (247.3 calcd),
[M + K]+ 263.4 (263.2 calcd). Anal. Calcd for C11H12O3S: C,
58.91; H, 5.39. Found: C, 58.69; H, 5.41.
calcd), [M
43H42N2O4S: C, 75.63; H, 6.20; N, 4.10. Found: C, 75.33; H,
+
K]+ 720.0 (721.8 calcd). Anal. Calcd for
C
6.10; N, 4.11.
O-tr a n s-4-(N-r-F m oc-S-ter t-bu tylsu lfen yl-L-cystein yl)-
a m in ocycloh exyl O-2-Cya n oeth yl-N,N-d iisop r op ylp h os-
p h or a m id ite (5). To a chilled (ice bath) solution of 3 (0.83 g,
1.578 mmol) in 15 mL of anhydrous dichloromethane contain-
ing 3 equiv (0.79 mL) of diisopropylethylamine was added
dropwise via syringe and under nitrogen 2-cyanoethoxy-N,N-
diisopropylaminochlorophosphine (1.5 equiv, 0.529 mL). After
1 h of stirring on ice, the mixture was allowed to warm
gradually to room temperature and stirring was continued for
2 h. The reaction mixture was quenched with 0.1 mL of
methanol and evaporated to dryness. The product was taken
up in ethyl acetate, washed with saturated sodium bicarbonate
solution and brine, dried over anhydrous sodium sulfate and
evaporated to a small volume. The product was chromato-
graphed on a silica gel column eluted with 15-40% ethyl
acetate in hexane + 2% triethylamine. Appropriate fractions
were pooled and evaporated to dryness. Yield of title product:
P en ta flu or op h en yl S-Ben zyl Th iosu ccin a te (2). A solu-
tion of dicyclohexyl carbodiimide (2.27 g, 11 mmol) in 15 mL
of dichloromethane was added dropwise to a stirred and cooled
(ice bath) solution of S-benzyl thiosuccinic acid (2.24 g, 10
mmol) and pentafluorophenol (2.12 g, 11.5 mmol) in 25 mL of
dichloromethane. The reaction mixture was stirred for 0.5 h
in an ice bath and then allowed to warm slowly to room
temperature, stirred for 4 h and left overnight in a refrigerator.
Dicyclohexylurea precipitate was filtered off and the remaining
solution concentrated in vacuo, redissolved in a minimal
volume of ethyl acetate, the solution filtered again to remove
particulates, and hexane was added. After standing overnight
in a freezer, crystals were filtered off, washed with cold ethyl
acetate and hexane mixture (1:9 v/v) and dried in vacuo
overnight. Yield of white needles 3.44 g (88%). After evapora-
tion of mother liquor and further treatment with hexane, an
additional 0.18 g of the title compound was obtained. Total
yield of two crops 3.62 g (92%). TLC (B): Rf 0.84. 1H NMR
(DMSO-d6): δ 3.08 (s, 4H), 4.16 (s, 2H), 7.26 (m, 5H). MALDI-
TOF MS: [M + H]+ 393.4 (391.3 calcd). Anal. Calcd for
1
0.76 g (66%). TLC (D): Rf 0.53. H NMR (CD3CN): δ 1.16 (d,
6H, J ) 0.9 Hz), 1.18 (d, 6H, J ) 1.0 Hz), 1.27 (m, 1H), 1.32
(s, 9H), 1.45 (m, 1H), 1.87 (m, 2H), 1.95 (q, 2H, J ) 2.5 Hz),
2.01 (m, 1H), 2.64 (t, 2H, J ) 5.9 Hz), 2.95 (m, 1H), 3.12 (m,
1H), 3.62 (m, 3H), 3.75 (m, 2H), 4.32 (m, 3H), 6.07 (d, 1H, J )
8.6 Hz), 6.59 (d, 1H, J ) 7.8 Hz), 7.34 (t, 2H, J ) 7.5 Hz), 7.43
(t, 2H, J ) 7.4 Hz), 7.68 (d, 2H, J ) 7.3 Hz), 7.85 (d, 2H, J )
7.5 Hz). 31P NMR (CD3CN): δ 146.51 ppm. MALDI-TOF MS:
[M + H]+ 729.4 (730.0 calcd).
O-tr a n s-4-(N-r-F m oc-S-tr ityl-L-cystein yl)a m in ocyclo-
h exyl O-2-Cya n oeth yl-N,N-d iisop r op ylp h osp h or a m id ite
(6). To a solution of 4 (0.24 g, 0.34 mmol) in 10 mL of
anhydrous dichloromethane containing 75 mg (1.5 equiv) of
diisopropylammonium tetrazolide was added 2-cyanoethoxy-
N,N,N′,N′-tetraisopropyl phosphorodiamidite (0.13 mL, 0.39
mmol), and the mixture was stirred for 6 h at room temper-
ature, until TLC revealed complete reaction. Dichloromethane
was then removed by evaporation and the product was taken
up in ethyl acetate, washed with 5% sodium bicarbonate
C
17H11F5O3S: C, 52.31; H, 2.84. Found: C, 51.50; H, 2.85.
N-r-F m oc-S-ter t-bu tylsu lfen yl-L-cystein e 4-Hyd r oxy-
tr a n s-cycloh exyla m id e (3). To a slurry of trans-4-aminocy-
clohexanol hydrochloride (0.3 g, 2 mmol), N-R-Fmoc-S-tert-
butylsulfenyl-L-cysteine pentafluorophenyl ester (1.2 g, 2
mmol), and 1-hydroxybenzotriazole (2 mmol, 270.3 mg) in 20

4908 J . Org. Chem., Vol. 65, No. 16, 2000
Stetsenko and Gait
solution and brine, dried over anhydrous sodium sulfate and
evaporated to a small volume. The product was chromato-
graphed on a silica gel column eluted with 10-35% ethyl
acetate in hexane + 2% triethylamine. Appropriate fractions
were pooled and evaporated to dryness. Yield of title product
Ack n ow led gm en t. We thank Richard Grenfell and
Donna Williams for oligonucleotide assemblies and
Robert C. Sheppard and David Owen for valuable advice
on peptide synthesis.
1
1.30 g (95%). TLC (D): Rf 0.78. H NMR (DMSO-d6): δ 1.11
(d, 6H, J ) 1.6 Hz), 1.13 (d, 6H, J ) 1.4 Hz), 1.19 (m, 8H),
1.35 (m, 2H), 1.69 (m, 2H), 1.86 (m, 2H), 2.28 (m, 2H), 2.76 (t,
2H, J ) 5.8 Hz), 3.48 (m, 4H), 3.67 (m, 3H), 4.02 (q, 1H, J )
7.1 Hz), 4.22 (m, 3H), 7.29 (m, 17H), 7.39 (t, 2H, J ) 7.5 Hz),
7.59 (d, 1H, J ) 8.6 Hz), 7.72 (d, 2H, J ) 6.9 Hz), 7.88 (d, 2H,
J ) 7.5 Hz). 31P NMR (CD3CN) δ 146.50 ppm. MALDI-TOF
MS: [M + H]+ 922.0 (922.2 calcd).
Su p p or tin g In for m a tion Ava ila ble: NMR spectra of
intermediates and examples of HPLC traces and MALDI-TOF
spectra of peptide thioesters, cysteine oligonucleotides and
conjugates. This material is available free of charge via the
Internet at http://pubs.acs.org.
J O000214Z