Alternative procedures for the synthesis of methionine-containing peptide - oligonucleotide hybrids

Article abstract of DOI:10.1002/1099-0690(200007)2000:13<2495::AID-EJOC2495>3.0.CO;2-T

The synthesis of methionine-containing peptide-oligonucleotide hybrids has been found to be best accomplished by a stepwise solid-phase approach in which peptide assembly using the sulfoxide derivative of methionine is followed by elongation of the oligonucleotide chain using the phosphite triester methodology, ammonia deprotection, and reduction of the sulfoxide to thioether by reaction with N-methylmercaptoacetamide. Quantitative amino acid incorporation yields could not always be achieved when the order of assembly of the two moieties was reversed, i.e. by elongating the peptide chain on a resin-linked oligonucleotide in order to avoid exposure of the thioether function to oxidizing conditions.

Full text of DOI:10.1002/1099-0690(200007)2000:13<2495::AID-EJOC2495>3.0.CO;2-T

FULL PAPER
Alternative Procedures for the Synthesis of Methionine-Containing
Peptide؊Oligonucleotide Hybrids
[a]
[b]
[a]
[a]
´
´
Vicente Marchan, Chryslaine Rodrıguez-Tanty, Marta Estrada, Enrique Pedroso, and
Anna Grandas*[a]
Keywords: Methionine / Peptides / Nucleotides / Amino acids / Nucleopeptides
The synthesis of methionine-containing peptide−oligo-
nucleotide hybrids has been found to be best accomplished
methylmercaptoacetamide. Quantitative amino acid incor-
poration yields could not always be achieved when the order
by a stepwise solid-phase approach in which peptide assem- of assembly of the two moieties was reversed, i.e. by elon-
bly using the sulfoxide derivative of methionine is followed
gating the peptide chain on a resin-linked oligonucleotide in
by elongation of the oligonucleotide chain using the phos- order to avoid exposure of the thioether function to oxi-
phite triester methodology, ammonia deprotection, and re- dizing conditions.
duction of the sulfoxide to thioether by reaction with N-
Introduction
mained to find a suitable synthetic protocol for the me-
thionine-containing conjugates.
Many research groups have prepared various types of
peptideϪoligonucleotide hybrids, usually with the aim of
evaluating their possible therapeutic applications (for recent
examples, see refs.[1Ϫ16]). We have also used such hybrids as
models for studying the mode of action of the anticancer
drug cisplatin.[17] Among the questions that we would like
to address is whether coordination to two adjacent guanine
bases in DNA is preceded by coordination to the side
chains of amino acids, and how the metal might then be
transferred from proteins to DNA. Since the thioether
group and the imidazole ring of the methionine and his-
tidine residues, respectively, would seem to represent some
of the most likely platinum anchoring points, we wish to
report here on the preparation of peptideϪoligonucleotide
hybrids containing methionine, histidine, or both, with par-
ticular emphasis on the problems associated with the intro-
duction of the oxidizable amino acid methionine.
Nucleopeptides are commonly obtained in our labora-
tory using a stepwise solid-phase approach[6] in which the
assembly of the peptide moiety on a solid support is fol-
lowed by elongation of the oligonucleotide chain at the free
hydroxyl group of an amino acid side chain (serine, threo-
nine, tyrosine, or homoserine) using phosphoramidite build-
ing blocks. To prevent oxidation of the methionine thio-
ether to sulfoxide during the phosphite to phosphate oxida-
tion step, we decided to assemble our first synthetic target,
H-Ser(p5ЈdG)-Met-OH (1a), by coupling
a 2Ј-deoxy-
guanosine phosphate derivative to the dipeptideϪ
resin. The synthetic scheme is outlined in Figure 1.
The MSNT-mediated[19,20] coupling of the 2-chloro-
phenyl phosphate of 5Ј-O-[3Ј-O-dimethoxytrityl-(2N-di-
methylaminomethylene-2Ј-deoxyguanosine)][21] to Fmoc-
Ser-Met-O-HMFS-Leu-resin[6] was followed by removal of
the DMT group (an incorporation yield of 86% was estab-
lished through quantitative determination of the DMT cat-
ions formed upon acid treatment). Cleavage of the
peptideϪresin bond and removal of the Fmoc and 2-
chlorophenyl groups was accomplished by reaction with
TBAF, and then the guanine protecting group was removed
by treatment with ammonia. The only product isolated
after MPLC purification was the 5Ј-phosphate of 2Ј-deoxy-
guanosine, formation of which may only be explained by
assuming that a nucleosideϪserine bond had been formed
and subsequently cleaved by β-elimination during the de-
protection step. Replacement of the TBAF treatment by re-
action with either 0.03  TBAF in THF/H2O/py (8:8:1; 7 h,
room temp.) or 0.1  tetramethylguanidinium syn-2-nitro-
benzaldoximate in dioxane/H2O (1:1; 8 h, room temp.) led
to similar results. The serineϪnucleoside phosphodiester
linkage thus appears to be particularly labile in this mole-
Results and Discussion
Based on our previous experience in the preparation of
nucleopeptides,[6] we first decided to synthesize hybrids with
a covalent phosphodiester bond between the side-chain hy-
droxyl group of an amino acid and the 5Ј- or 3Ј-hydroxyl
group of a nucleoside. Histidine has already been success-
fully incorporated into nucleopeptides,[17,18] thus it re-
[a]
´
`
´
Departament de Quımica Organica, Facultat de Quımica,
Universitat de Barcelona,
´
`
Martı i Franques 1Ϫ11, E-08028 Barcelona, Spain
Fax: (internat.) ϩ 34-93/339-7878
E-mail: grandas@qo.ub.es
[b]
Present address: Centro Nacional de Investigaciones
´
Cientıficas,Avenida 25 no. 15202, Playa 12100, Cubanacan,
6990 La Habana, Cuba
Eur. J. Org. Chem. 2000, 2495Ϫ2500
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1434Ϫ193X/00/0713Ϫ2495 $ 17.50ϩ.50/0
2495

´
´
V. Marchan, C. Rodrıguez-Tanty, M. Estrada, E. Pedroso, A. Grandas
FULL PAPER
Figure 1. Assays carried out to optimize the synthesis of nucleopep-
tides 1a and 1b
cule, presumably because of the presence of the neigh-
bouring methionine residue.
In
a second attempt to obtain a methionineϪ
deoxyguanosine hybrid (1b), the serine linking residue was
replaced by homoserine, which forms base-stable amino
acidϪnucleoside phosphodiester bonds.[21] Moreover, to re-
duce the final deprotection step to a single treatment with
ammonia/dioxane, the 2-cyanoethyl phosphate of 5Ј-O-[3Ј-
O-dimethoxytrityl-(2N-dimethylaminomethylene-2Ј-deoxy-
guanosine)] was synthesized and coupled to Phac-Hse-Met-
Figure 2. Synthesis scheme used for the preparation of hybrids 2,
3, and 4; for the synthesis of
5 (Phac-His-Gly-Met-linker-
dCATGGCT), tripeptide Phac-His(Dnp)-Gly-Met-OH was
coupled to H2N-linker-oligonucleotide-resin
O-HMFS-Leu-resin (Figure 1). Since the HPLC profile of linker (6-amino-1-hexanol) and the desired Nα-Fmoc-
the crude material obtained after removal of the DMT amino acid to it. The N-terminus was acylated to prevent
group and ammonia deprotection was very complex, we de- reaction with platinum(II) complexes in subsequent studies
cided to abandon this synthetic scheme.
(Figure 2). Nα-Fmoc- rather than Nα-Boc-amino acids were
Still with the idea of not submitting the methionine res- used, so as to avoid exposure of the nucleotide to acidic
idue to oxidation steps, our second choice was to synthesize conditions. A sarcosine residue was introduced between the
Phac-Met-linker-dG (2) by reversing the order in which the succinyl linker and the internal reference amino acid to
hybrid molecule was assembled, i.e. anchoring the 2Ј-deoxy- make the nucleosideϪresin linkage sufficiently stable to the
guanosine moiety on the solid support and then coupling a basic conditions required to remove the Fmoc group.[7,22]
Table 1. Assays carried out to determine the best amino acid coupling conditions
Entry
Coupling conditions[a]
Coupling yield[b]
1
2
3
4
5
6
7
8
9
10
15 equiv. Aa ϩ 15 equiv. DCC, DMF, 3 h
50%
15 equiv. Aa ϩ 15 equiv. DCC ϩ 15 equiv. HOBt, DMF, 3 h
30 equiv. Aa ϩ 15 equiv. DCC, DMF, 3 h
62%
62%
15 equiv. Aa ϩ 15 equiv. DCC, NMP, 3 h
42%
15 equiv. Aa ϩ 15 equiv. DCC, 0.5  LiCl in DMF, 3 h
15 equiv. Aa ϩ 15 equiv. DCC, 0.5  LiBF4 in DMF, 3 h
15 equiv. Aa ϩ 15 equiv. DCC ϩ 15 equiv. HOBt, 0.5  LiCl in DMF, 3 h
30 equiv. Aa ϩ 15 equiv. DCC, 0.5  LiCl in DMF, 3 h
15 equiv. Aa ϩ 15 equiv. TBTU ϩ 30 equiv. DIEA, 0.5  LiCl in DMF, 1 h
15 equiv. Aa ϩ 15 equiv. PyBOP ϩ 30 equiv. DIEA, 0.5  LiCl in DMF, 1 h
70%
54%
quant.
quant.
quant.
quant.
[a]
[b]
Aa ϭ Fmoc-Met-OH. Ϫ The extent of incorporation was determined from the relative areas of the peaks due to H-linker-dG and
H-Met-linker-dG in the HPLC trace of the crude material obtained after ammonia deprotection.
2496
Eur. J. Org. Chem. 2000, 2495Ϫ2500

Synthesis of Methionine-Containing PeptideϪOligonucleotide Hybrids
FULL PAPER
The synthesis proceeded smoothly up until the incor- TCA/DCM in the presence of Et3SiH as scavenger, and
poration of the linker, but problems were encountered be- some cleavage of the glycosidic linkage was also observed
yond removal of the monomethoxytrityl group. We had under these conditions. Literature reports on the compatib-
great difficulties in coupling Fmoc-methionine to the H2N- ility of the conditions required to deprotect His(Trt) and
linker-nucleotide-resin, and blocking of the N-terminus the stability of the glycosidic linkage are somewhat contra-
with Phac2O once the Fmoc group had been eliminated was dictory.[3,28]
not straightforward either (see below).
Amino acid coupling problems were similarly encoun-
The simplest standard activation conditions, even using tered when we undertook the synthesis of Phac-His-Gly-
a significant excess of the amino acid and coupling reagent, Met-linker-dCATGGCT (5). Only the use of the strongest
were not sufficiently effective for the incorporation of me- carboxyl activating agents, such as HATU,[29] was found to
thionine (Table 1, entries 1Ϫ3). Changing the solvent[23] did allow quantitative coupling of the methionine and glycine
not have a significant effect (entry 4), but the coupling yield derivatives, but incorporation of the histidine residue never
increased when LiCl[24] was added to the reaction mixture exceeded 20% yield. The target product (5) was finally ob-
(entry 5). LiCl proved to be more effective than LiBF4 tained by HATU-mediated coupling of the tripeptide Phac-
(entry 6). Carrying out the amino acid incorporation in the His(Dnp)-Gly-Met-OH to the linker-oligonucleotide-
presence of LiCl, using either standard or strong activating resin.[28]
reagents, allowed quantitative coupling of Fmoc-Met-OH
For amino acid or peptide couplings to H2N-linker-oligo-
(entries 7Ϫ10). It has been postulated that the unfavourable nucleotide-resins, some authors have reported low or non-
folding of resin-linked peptides may hinder the terminal reproducible yields.[9] Other authors have not described any
amine groups and render some steps of peptide syntheses particular problems in relation to the stepwise elongation of
particularly troublesome,[25] but it is hard to imagine why, a peptide on an oligonucleotideϪresin using Fmoc-amino
in the present case, the amine group of an H2N-linker-nu- acids,[3] whereas others have encountered more difficulties
cleotide-resin might not be accessible.
in the coupling of the first Fmoc-amino acid than in the
Phac-Met-linker-dG (2), Phac-His-linker-dG (3), and subsequent couplings.[7] The latter result[7] would confirm
Phac-His-Met-linker-dG (4) were finally obtained by coup- the hypothesis that coupling yields to phosphate-unprotec-
ling the appropriate amino acids [Fmoc-His(Dnp)-OH, ted oligonucleotide-resins may be superior to those of coup-
Fmoc-Met-OH, or both] using 15 equiv. of both the amino lings to phosphate-protected oligonucleotide-resins,[9] but
acid and PyBOP and 30 equiv. of DIEA in a 0.5  solution we have always had less difficulties in the first amino acid
of LiCl in DMF (1.5Ϫ3 h, Ar atmosphere). Blocking of the incorporation than in the subsequent ones. The cyanoethyl
N-termini was effected by reaction with Phac2O (15 equiv.), group is still present on the phosphate(s) in the first amino
DMAP (15 equiv.), and DIEA (30 equiv.) in a 0.5  LiCl acid coupling, but it is most probably lost during the piper-
solution in DMF (45 min., Ar atmosphere).
idine treatments that remove the Fmoc group.
We also tried to replace the coupling of Fmoc-His(Dnp)-
Taken together, the aforementioned findings indicated
OH and N-terminal phenylacetylation by a single coupling that it would not be easy to prepare peptideϪ
with Phac-His(Dnp)-OH to synthesize 3 and 4, which led oligonucleotide hybrids using this approach, hence we
to mixtures of - and -His-containing products. This result turned our attention back to our standard nucleopeptide
might have been anticipated, considering that an Nα-acyl- synthesis scheme.
amino acid had been activated rather than an Nα-urethane-
Methionine is generally incorporated into polymer-sup-
protected derivative,[26] and that histidine is very prone to ported peptide chains as the sulfoxide so as to avoid both
racemization.[27] McMinn and Greenberg[8] have suggested uncontrolled oxidation and, more importantly, alkylation
the use of strong activating reagents such as PyBOP or by reaction with carbocations formed under acidic depro-
HBTU to minimize histidine racemization, and did not ob- tection conditions.[30] Therefore, assembly of the
serve epimerization even when the imidazole-unprotected nucleopeptideϪresin using the methionine sulfoxide deriv-
Nα-acetylhistidine was activated with HBTU. In our hands, ative seemed the safest option, but some aspects of me-
the PyBOP-mediated coupling of Phac-His(Dnp)-OH (with thionine chemistry had to be considered first.
the electron-withdrawing 2,4-dinitrophenyl group on the
To evaluate the effect of the phosphite to phosphate oxi-
imidazole ring) afforded 56% of Phac--His--Met-linker- dizing conditions on a sulfoxide-protected methionine-con-
dG and 44% of Phac--His--Met-linker-dG, as determined taining peptide, an aliquot of the peptideϪresin Ac-
from the relative areas of the two product peaks in the Lys(Tfa)-Met(O)-Tyr-Pro-O-HMFS-Phe-resin (see below)
HPLC profile of the crude reaction mixture (Phac--His- was treated with the standard oxidation reagent, aqueous
linker-dG and Phac--His-linker-dG were co-eluted). The iodine. Not unexpectedly for a tyrosine-containing pep-
identities of these two products were verified by unequi- tide,[31] MALDI-TOF mass spectrometric analysis of the
vocal syntheses of the two diastereomeric hybrids using crude peptide obtained after deprotection with conc. aq.
either Fmoc--His(Dnp)-OH or Fmoc--His(Trt)-OH.
NH3/dioxane (1:1) (overnight, room temp.) showed the
The use of the NIm-trityl-protected histidine derivative presence of iodinated products. In a further experiment,
might be an alternative to imidazole protection with the both HPLC and MALDI-TOF mass spectrometric analysis
Dnp group, but, in our hands, only about 10% of the Trt after ammonia deprotection showed that the peptide had
group could be removed after nine treatments with 10% remained intact after treatment with 1  tBuOOH for 1 h,
Eur. J. Org. Chem. 2000, 2495Ϫ2500
2497

´
´
V. Marchan, C. Rodrıguez-Tanty, M. Estrada, E. Pedroso, A. Grandas
FULL PAPER
hence we selected this reagent for the oxidation of phos-
phites to phosphates.
Sulfoxides have long been known to be capable of oxidiz-
ing PIII compounds (phosphanes, phosphite triesters) to PV
derivatives,[32] with their concomitant reduction to thio-
ethers. It was thus conceivable that, during each nucleotide
incorporation cycle, some methionine sulfoxides might be
reduced, although we could expect these side chains to be
re-oxidized by the tBuOOH treatment. Even though there
seems to be a consensus of opinion that anhydrous
tBuOOH solutions do not oxidize thioethers to sulfox-
ides,[33] it has been reported that methionines are oxidized
to sulfoxides by tBuOOH when anchored to a solid ma-
trix.[34] Our experience is that, at least in solution, methio-
nine is quickly oxidized by 1  tBuOOH solutions, and thus
we expect such oxidation to also take place on the solid
support.
Of the many different methods described for the reduc-
tion of sulfoxides to thioethers following peptide
assembly,[35Ϫ39] reduction with N-methylmercaptoacetam-
ide[35] seemed the safest method for the oligonucleotide
moiety. In a preliminary assay, oligonucleotide 5ЈGCGAT-
ATCG3Ј was treated with a large excess (1000 equivalents,
overnight, 37 °C) of N-methylmercaptoacetamide in water
and was found to remain totally unaffected.
We then undertook the stepwise solid-phase synthesis of
a methionine-containing nucleopeptide, specifically Ac-Lys-
Met-Tyr(p3ЈdACTAGT)-Pro-OH (6) (Figure 3).
The tyrosineϪnucleoside phosphodiester linkage has pre-
viously been shown to be stable to treatment with 1:1 NH3/
dioxane at 55 °C,[40] hence the trifluoroacetyl group and the
standard benzoyl and isobutyryl groups were chosen for the
permanent protection of the lysine side chain and the exo-
cyclic amines of the nucleobases, respectively. Since the
tyrosineϪnucleoside phosphodiester bond has been re-
ported to be labile in the presence of tetrabutylammonium
fluoride,[41] the nucleopeptide was linked to the resin
through a 9-fluorenylmethyl ester group, which allowed
cleavage of the nucleopeptideϪHMFS handle[42] bond
simply by treating with ammonia.[6] Ac-Lys(Tfa)-Met(O)-
Tyr-Pro-O-HMFS-Phe-resin was assembled according to
standard procedures,[6] with the sole exception that, to pre-
vent loss of the peptide chains through diketopiperazine
formation, Boc-Met(O)-OH was coupled to the triflu-
oroacetate salt of Tyr-Pro-resin by activation with PyAOP
in the presence of DIEA.[43] The oligonucleotide chain was
elongated at the side chain of the tyrosine residue, and the
crude nucleopeptide was obtained after deprotection with
ammonia and reduction of the sulfoxide with N-methylmer-
captoacetamide. The target product, 6, was obtained after
MPLC purification in 23% yield, which is within the same
range as achieved for other nucleopeptides constructed
using the same methodology.
Figure 3. Stepwise solid-phase synthesis of nucleopeptide 6; * step
(iv) was suppressed to incorporate Boc-Met(O)-OH, which was ac-
tivated with PyAOP and coupled to the trifluoroacetate salt of the
dipeptideϪresin in the presence of DIEA
oxidation to incorporate each nucleotide unit. In our opin-
ion, this synthetic strategy is more reliable than the meth-
odology in which the peptide is either coupled to or as-
sembled in a stepwise manner on oligonucleotideϪresins,
since high and reproducible yields are difficult to achieve in
these cases.
Work is currently in progress aimed at extending this
stepwise solid-phase synthesis methodology to the introduc-
tion of other trifunctional amino acids into nucleopeptides
or other peptideϪoligonucleotide conjugates.
Experimental Section
H-Ser(p5’dG)-Met-OH (1a): DipeptideϪresin Fmoc-Ser-Met-O-
HMFS-Leu-resin was assembled according to the general proced-
ure described in ref.[6] (using the HMFS handle). The 2-chloro-
phenyl phosphate of 5Ј-O-[3Ј-O-dimethoxytrityl-(2N-dimethylami-
nomethylene-2Ј-deoxyguanosine)] was obtained as described in
ref.[20] (pp. 94Ϫ96) and coupled to the side chain of serine in the
presence of MSNT (molar ratio: nucleotide/MSNT/peptideϪresin
8:40:1; pyridine, 5 h, room temp.). The DMT group was sub-
sequently removed by reaction with 3% trichloroacetic acid (86%
nucleotide coupling yield). Final deprotection was carried out by
In summary, we have prepared nucleopeptides containing
a methionine residue for the first time. Sulfoxide protection
to prevent alkylation of the methionine side chain has al-
lowed the target molecule to be safely assembled on a solid
matrix using the phosphite triester approach and tBuOOH treating the nucleopeptideϪresin first with 0.05  TBAF in anhy-
2498 Eur. J. Org. Chem. 2000, 2495Ϫ2500

Synthesis of Methionine-Containing PeptideϪOligonucleotide Hybrids
drous THF (30 min., room temp.) and then with conc. aq. NH3/ with 1  tBuOOH. The MMT group was subsequently removed
FULL PAPER
dioxane (1:1; overnight, room temp.).
under standard detritylation conditions and the amine groups on
the resin were neutralized by treatment with 2% DIEA in dichloro-
methane prior to amino acid incorporation. The extent of amino
acid incorporation was monitored by HPLC analysis of the crude
material obtained following ammonia treatment of an aliquot of
Bis(2-cyanoethyloxy)(diisopropylamino)phosphane: This compound
was obtained by the reaction of chloro(2-cyanoethyloxy)(diisopro-
pylamino)phosphane with dry 2-cyanoethanol and DIEA (molar
ratio 1:2:1.5). It was purified by silica gel column chromatography
(hexanes/tBuOMe, 1:1 and 1:3).
amino-acid-linker-nucleotide-resin (A: 0.01
 NH4OAc, B:
CH3CN/H2O, 1:1, linear gradient from 10 to 50% of B over
40 min., 1 mL/min): tR dG 7.0 min, H-linker-dG 9.2 min, H-Met-
linker-dG 20.1 min, Phac-Met-linker-dG 30.0 min. Once as-
sembled, the target hybrid was obtained after overnight treatment
with conc. aq. NH3/dioxane (1:1; room temp.) and purification by
MPLC (A: 0.05  NH4OAc, B: CH3CN/H2O, 1:1, gradient from 0
to 40% of B for hybrids 2 and 4, and from 0 to 30% of B for hybrid
3). 2, 3, and 4 were characterized by (a) nucleoside composition
after enzymatic digestion with snake venom phosphodiesterase and
alkaline phosphatase, and (b) mass spectrometry.
2: (a) dG; (b) m/z ϭ 694.8 (MALDI-TOF, negative mode), calcd.
for C29H42N7O9PS: 695.69.
3: (a) dG; (b) m/z ϭ 701.2 (MALDI-TOF, negative mode), calcd.
for C30H40N9O9P: 701.64.
4: (a) dG; (b) m/z ϭ 831.1 (electrospray, negative mode), calcd. for
C35H49N10O10PS: 832.83.
2-Cyanoethyl Phosphate of 5Ј-O-[3Ј-O-dimethoxytrityl-(2N-di-
methylaminomethylene-2Ј-deoxyguanosine): This compound was
prepared was prepared by reaction of 3Ј-O-dimethoxytrityl-(2N-di-
methylaminomethylene-2Ј-deoxyguanosine) with bis(2-cyanoethyl-
oxy)diisopropylaminophosphane and tetrazole followed by: (i)
tBuOOH oxidation, and (ii) treatment with Et3N/py (1:1; 3 h, room
temp.) to remove one cyanoethyl group.
Phac-Hse(p5’dG)-Met-OH (1b): Nucleopeptide 1b was assembled
essentially as described for 1a, by coupling the 2-cyanoethyl phos-
phate of 5Ј-O-[3Ј-O-dimethoxytrityl-(2N-dimethylaminomethylene-
2Ј-deoxyguanosine)] to the dipeptideϪresin Phac-Hse-Met-O-
HMFS-Leu-resin. Nucleotide coupling yield: 61%. Basic deprotec-
tion treatment: conc. aq. NH3/dioxane (1:1; overnight, room
temp.).
Fmoc-His(Dnp)-OH: The Boc group was removed from commer-
cially available Boc--His(Dnp)-OH by reaction with 4  HCl/diox-
ane for 1 h at room temp. A yellow solid [HCl·H-His(Dnp)-OH]
was obtained after removal of the solvent under reduced pressure.
A solution of Fmoc-Cl (1.5 equiv.) in dioxane was then added to
a cooled solution of the HCl·H-His(Dnp)-OH thus obtained in
10% aq. Na2CO3, and the mixture was stirred overnight in an ice
bath and for a further 24 h at room temp. Amino protection was
then completed by treatment with an additional 0.5 equiv. of Fmoc-
Cl. The dioxane was evaporated under reduced pressure, and the
concentrated mixture was poured onto a mixture of ice and H2O
and extracted with diethyl ether. The aqueous layer was cooled,
acidified to pH 2 with conc. HCl, and extracted with ethyl acetate.
The combined organic extracts were washed with 0.1  HCl and
H2O, dried over MgSO4, and concentrated. The desired product
was obtained following silica gel column chromatography eluting
with mixtures of dichloromethane and ethyl acetate (100/0 Ǟ 0/
100) containing 1% acetic acid; 30% yield. Electrospray-MS
(positive mode): m/z ϭ 543.8 [M ϩ H]ϩ; calcd. for C27H21N5O8:
543.46.
Phac-His(Dnp)-Gly-Met-OH: Solid-phase assembly using standard
Boc methodology on Boc-Met-O-PAM-resin (PAM ϭ phenylacet-
amidomethyl), followed by phenylacetylation of the N-terminus
and cleavage of the peptideϪresin bond by reaction with anhydrous
HF (in the presence of 10% anisole) afforded the target tripeptide.
No purification was required. Electrospray-MS (positive mode):
m/z ϭ 628.0 [M ϩ H]ϩ; calcd. for C27H29N7O9S: 627.59.
Phac-His-Gly-Met-linker-5ЈdCATGGCT (5): The oligonucleotide
was assembled on DMT-T-succinyl-sarcosil-Ala-resin using the
phosphite triester approach. The linker and tripeptide Phac-
His(Dnp)Gly-Met-OH were subsequently incorporated as de-
scribed above. Final deprotection was achieved by treatment with
conc. aq. NH3/dioxane (1:1; overnight, room temp.). 5 was purified
by MPLC (gradient from 0 to 30% of B, the same solvent system
as used for hybrids 2Ϫ4). Nucleoside composition after enzymatic
digestion: dC 2.09, dG 1.98, dT 2.04, dA 0.88. MALDI-TOF MS:
m/z ϭ 2717.6 (negative mode); calcd. for C95H126N31O48P7S:
2718.91.
Phac-His(Dnp)-OH: This derivative was prepared following essen-
tially the same methodology as described for the synthesis of
Fmoc-His(Dnp)-OH, but using the O-phenylacetyl derivative of N-
hydroxysuccinimide (Phac-OSu) to block the amine group. FAB-
MS (positive mode): m/z ϭ 440.1 [M ϩ H]ϩ, 462.1 [M ϩ Na]ϩ;
calcd. for C20H17N5O8: 439.36.
Ac-Lys-Met-Tyr(p3ЈdACTAGT)-Pro-OH (6): Nucleopeptide 6 was
assembled according to the general procedure described in ref.[6]
with minor modifications to prevent diketopiperazine formation
(see text). PeptideϪresin: Ac-Lys(Tfa)-Met(O)-Tyr-Pro-O-HMFS-
Leu-resin. After oligonucleotide elongation using the phosphite
triester methodology and removal of the 5Ј-DMT group, treatment
of the nucleopeptideϪresin with conc. aq. NH3/dioxane (1:1) over-
night at 55 °C afforded Ac-Lys-Met(O)-Tyr(p3ЈdACTAGT)-Pro-
OH. The target nucleopeptide, 6, was then obtained by reduction
of the sulfoxide (73 OD of hybrid, 1.0 m in H2O; Ar atmosphere)
with 20 equiv. of N-methylmercaptoacetamide for 96 h at 37 °C
followed by MPLC purification (gradient from 0 to 30% of B, the
same solvent system as used above). Nucleoside composition after
enzymatic digestion: dC 1.14, dG 0.97, dT 2.09, dA 1.80; amino
acid analysis after acid hydrolysis (6  HCl in the presence of
phenol and dithiothreitol, 1 h, 160 °C): Lys 1.05, Met 0.93, Tyr
Phac-OSu: Reaction of equimolar amounts of phenylacetic acid,
DCC, and N-hydroxysuccinimide in dichloromethane (overnight,
room temp.), followed by the addition of diethyl ether to precipitate
N,NЈ-dicyclohexylurea and aqueous work-up, afforded the active
ester. CI-MS (NH3): m/z
ϭ 251 [M ϩ
NH4]ϩ; calcd. for
C12H11NO4: 233.21.
Phac-Met-linker-5ЈdG (2), Phac-His-linker-5ЈdG (3), and Phac-His-
Met-linker-5ЈdG (4): DCC-mediated coupling of Boc-sarcosine and
the 3Ј-O-succinate of 5Ј-O-dimethoxytrityl-(2N-dimethylaminome-
thylene)-2Ј-deoxyguanosine to Fmoc-alanine-resin (0.21 mmol/g) 0.97, Pro 1.05. MALDI-TOF MS (reflectron): m/z ϭ 2430.4; elec-
was followed by incorporation of MMT-NH-(CH2)6-O-P- trospray-MS: m/z ϭ 607.1 [M Ϫ 4H]4Ϫ, 809.8 [M Ϫ 3H]3Ϫ, 1215.0
(OCNE)NiPr2 in the presence of tetrazole, capping, and oxidation
[M Ϫ 2H]2Ϫ; found 2432.3; calcd. for C86H115N27O34P6S: 2432.74.
Eur. J. Org. Chem. 2000, 2495Ϫ2500
2499

´
´
V. Marchan, C. Rodrıguez-Tanty, M. Estrada, E. Pedroso, A. Grandas
FULL PAPER
done, Nu ϭ nucleoside, Phac ϭ phenylacetyl, PS ϭ polysty-
rene-co-1%-divinylbenzene resin functionalized as p-methyl-
benzhydrylamine, PyBOP ϭ benzotriazole-1-yl-oxy-tris(pyrrol-
idino)phosphonium hexafluorophosphate, R ϭ resin (solid
matrix), TBAF ϭ tetrabutylammonium fluoride, TCA ϭ tri-
chloroacetic acid, Tfa ϭ trifluoroacetyl, TFA ϭ trifluoroacetic
acid, TMG ϭ tetramethylguanidine.
Acknowledgments
This work was supported by funds from the Spanish MEC (grant
PB97Ϫ0941-C02Ϫ01) and the Generalitat de Catalunya
(SGR98Ϫ0001 and Centre de Referencia de Biotecnologia). The
authors thank Dr. Irene Fernandez and Dr. Marta Vilaseca of the
Servei d’Espectrometria de Masses and acknowledge the use of the
facilities of the Servei de RMN d’Alt Camp (Serveis Cientıfico-
`
´
[20]
B. S. Sproat, M. J. Gait, in Oligonucleotide Synthesis: A Prac-
tical Approach (Ed.: M. J. Gait), IRL Press, Oxford, 1985, pp.
´
104Ϫ106.
`
Tecnics, Universitat de Barcelona).
[21]
´
M. Beltran, M. Maseda, J. Robles, E. Pedroso, A. Grandas,
Lett. Pept. Sci. 1997, 4, 147Ϫ155.
[22]
[23]
[1]
T. Brown, C. E. Pritchard, G. Turner, S. A. Salisbury, J. Chem.
J. D. Kahl, M. M. Greenberg, J. Org. Chem. 1999, 64,
Soc., Chem. Commun. 1989, 891Ϫ893.
507Ϫ510.
[2]
C. J. Bagley, K. M. Otteson, B. L. May, S. N. McCurdy, L.
Pierce, F. J. Ballard, J. C. Wallace, Int. J. Peptide Prot. Res.
1990, 36, 356Ϫ361.
I. Schwope, C. F. Bleczinski, C. Richert, J. Org. Chem. 1999,
64, 4749Ϫ4761.
[3]
L. De Napoli, A. Messere, D. Montesarchio, G. Piccialli, E.
[24]
[25]
A. Thaler, D. Seebach, F. Cardinaux, Helv. Chim. Acta 1991,
74, 628Ϫ643.
Benedetti, E. Bucci, F. Rossi, Bioorg. Med. Chem. 1999, 7,
395Ϫ400.
[4]
P. Lloyd-Williams, F. Albericio, E. Giralt, in Chemical Ap-
proaches to the Synthesis of Peptides and Proteins, CRC Press,
Boca Raton, 1997, pp. 65Ϫ67.
D. S. Kemp, in The Peptides: Analysis, Synthesis, and Biology,
vol. 1 (Major Methods of Peptide Bond Formation) (Eds.: E.
Gross, J. Meienhofer), Academic Press, New York, 1979, pp.
331, 339Ϫ340.
N. L. Benoiton, in The Peptides: Analysis, Synthesis, and Bio-
logy, vol. 5 (Special Methods in Peptide Synthesis, Part B)
(Eds.: E. Gross, J. Meienhofer), Academic Press, New York,
1980, pp. 169Ϫ175.
A. Sakakura, Y. Hayakawa, H. Harada, M. Hirose, R. Noyori,
Tetrahedron Lett. 1999, 40, 4359Ϫ4362.
[5]
S. Flohr, V. Jungmann, H. Waldmann, Chem. Eur. J. 1999, 5,
[26]
[27]
669Ϫ681.
[6]
´
´
´
J. Robles, M. Beltran, V. Marchan, Y. Perez, I. Travesset, E.
Pedroso, A. Grandas, Tetrahedron 1999, 55, 13251Ϫ13263.
C. N. Tetzlaff, I. Schwope, C. F. Bleczinski, J. A. Stenberg, C.
Richert, Tetrahedron Lett. 1998, 39, 4215Ϫ4218.
D. L. McMinn, M. M. Greenberg, J. Am. Chem. Soc. 1998,
120, 3289Ϫ3294.
[7]
[8]
[9]
S. Peyrottes, B. Mestre, F. Burlina, M. J. Gait, Tetrahedron
[28]
[29]
[30]
N. N. Polushin, B.-C. Chen, L. W. Anderson, J. S. Cohen, J.
Org. Chem. 1993, 58, 4606Ϫ4613.
1998, 4, 12513Ϫ12522.
[10]
[11]
[12]
[13]
´
D. Gottschling, H. Seliger, G. Tarrason, J. Piulats, R. Eritja,
L. A. Carpino, A. El-Faham, C. Minor, F. Albericio, J. Chem.
Soc., Chem. Commun. 1994, 201Ϫ203.
G. Barany, R. B. Merrifield, in The Peptides: Analysis, Syn-
thesis, Biology, vol. 2 (Special Methods in Peptide Synthesis,
Part A) (Eds.: E. Gross, J. Meienhofer), Academic Press, New
York, 1980, pp. 223Ϫ229.
Bioconjugate Chem. 1998, 9, 831Ϫ837.
J. G. Harrison, S. Balasubramanian, Nucleic Acids Res. 1998,
26, 3136Ϫ3145.
M. A. Farrow, F. Aboul-ela, D. Owen, A. Karpeisky, L.
Beigelman, M. J. Gait, Biochemistry 1998, 37, 3096Ϫ3108.
C. Guibourdenche, D. Seebach, F. Natt, Helv. Chim. Acta 1997,
80, 1Ϫ13.
[31]
[32]
[33]
[34]
´
H. Lamthanh, C. Roumestand, C. Deprun, A. Menez, Int. J.
Peptide Protein Res. 1993, 41, 85Ϫ95.
[14]
[15]
´
E. Vives, B. Lebleu, Tetrahedron Lett. 1997, 38, 1183Ϫ1186.
E. H. Amonoo-Neizer, S. K. Ray, R. A. Shaw, B. C. Smith, J.
R. K. Bruick, P. E. Dawson, S. B. H. Kent, N. Usman, G. F.
Chem. Soc. 1965, 4296Ϫ4300.
Joyce, Chem. Biol. 1996, 3, 49Ϫ56.
D. M. Andrews, J. Kitchin, P. W. Seale, Int. J. Peptide Prot.
Res. 1991, 38, 469Ϫ475.
Novabiochem Catalog & Peptide Synthesis Handbook, Technical
Notes, 1999.
R. A. Houghten, C. H. Li, Anal. Biochem. 1979, 98, 36Ϫ46.
S. Futaki, T. Taike, T. Yagami, T. Akita, K. Kitagawa, Tetra-
hedron Lett. 1989, 30, 4411Ϫ4412.
[16]
[17]
[18]
[19]
C.-H. Tung, K. J. Breslauer, S. Stein, Bioconjugate Chem. 1996,
7, 529Ϫ531.
´
M. Beltran, G. B. Onoa, E. Pedroso, V. Moreno, A. Grandas,
J. Biol. Inorg. Chem. 1999, 4, 701Ϫ707.
[35]
[36]
´
M. Beltran, E. Pedroso, A. Grandas, Tetrahedron Lett. 1998,
39, 4115Ϫ4118.
[37]
[38]
[39]
Abbreviations: Aa ϭ amino acid, Boc ϭ tert-butyloxycarbonyl,
W. Beck, G. Jung, Lett. Pept. Sci. 1994, 1, 31Ϫ37.
M. W. Pennington, M. Byrnes, Peptide Res. 1995, 8, 39Ϫ43.
Bz ϭ benzoyl, CNE ϭ 2-cyanoethyl, DCC ϭ N,NЈ-dicyclohex-
ylcarbodiimide, DIEA
ϭ
N-ethyl-N,N-diisopropylamine,
´
E. Nicolas, M. Vilaseca, E. Giralt, Tetrahedron 1995, 51,
DMAP ϭ 4-dimethylaminopyridine, Dmf ϭ dimethylamino-
methylene, DMF ϭ N,N-dimethylformamide, DMT ϭ 4,4Ј-di-
methoxytrityl, Dnp ϭ 2,4-dinitrophenyl, Fmoc ϭ 9-fluorenyl-
methoxycarbonyl, HATU ϭ N-[(dimethylamine)-1H-1,2,3-tri-
azole[4,5-b]pyridine-1-yl-methylene]-N-methylmethanaminium
5701Ϫ5710.
[40]
[41]
[42]
[43]
´
J. Robles, M. Maseda, M. Beltran, M. Concernau, E. Pedroso,
A. Grandas, Bioconjugate Chem. 1997, 8, 785Ϫ788.
E. Kuyl-Yeheskiely, C. M. Tromp, A. W. M. Lefeber, G. A. van
der Marel, J. H. van Boom, Tetrahedron 1988, 44, 6515Ϫ6523.
F. Rabanal, E. Giralt, F. Albericio, Tetrahedron Lett. 1992,
39, 1775Ϫ1778.
hexafluorophosphate, HBTU
ϭ 2-(1H-benzotriazole-1-yl)-
1,1,3,3-tetramethyluronium hexafluorophosphate, HMFS ϭ
2-(9-oxymethylfluorenyl)succinyl, HOBt ϭ 1-hydroxybenzotri-
´
azole, Hse
ϭ
homoserine, iBu
ϭ
isobutyryl, MMT
ϭ
M. Gairı, P. Lloyd-Williams, F. Albericio, E. Giralt, Tetrahe-
4-methoxytrityl, MPLC ϭ medium-pressure liquid chromato-
graphy, MSNT ϭ 1-(mesitylene-2-sulfonyl)-3-nitro-1H-1,2,4-
triazole, NMI ϭ N-methylimidazole, NMP ϭ N-methylpyrroli-
dron Lett. 1990, 31, 7363Ϫ7366.
Received January 24, 2000
[O00043]
2500
Eur. J. Org. Chem. 2000, 2495Ϫ2500