A peptide nucleic acid embedding a pseudopeptide nuclear localization sequence in the backbone behaves as a peptide mimic

Article abstract of DOI:10.1002/ejoc.201000123

In this paper, the synthesis of a short modified peptide nucleic acid (PNA), obtained by using several different L-amino acids as synthons, is shown. The synthesis was performed by a submonomeric strategy, obtaining a model trimeric PNA containing embedded amino acid derived side chains in its backbone that mimic the peptide sequence PKKKRKV, which is a nuclear localization signal (NLS) widely used for translocating cargo molecules into cell nuclei. Fluorescence experiments demonstrated that this modified PNA, and not a standard unmodified PNA having the same nucleobase sequence, was able to penetrate Rhabdomyosarcoma cell nuclei, exactly behaving as the NLS standard peptide.

Full text of DOI:10.1002/ejoc.201000123

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201000123
A Peptide Nucleic Acid Embedding a Pseudopeptide Nuclear Localization
Sequence in the Backbone Behaves as a Peptide Mimic
Stefano Sforza,*[a] Tullia Tedeschi,[a] Alessandro Calabretta,[a] Roberto Corradini,[a]
Consuelo Camerin,[b] Roberto Tonelli,[c] Andrea Pession,[b] and Rosangela Marchelli[a]
Keywords: Peptide nucleic acids / Peptidomimetics / Amino acids / Cell recognition
In this paper, the synthesis of a short modified peptide nu- is a nuclear localization signal (NLS) widely used for translo-
cleic acid (PNA), obtained by using several different
L
-amino
cating cargo molecules into cell nuclei. Fluorescence experi-
ments demonstrated that this modified PNA, and not a stan-
dard unmodified PNA having the same nucleobase se-
quence, was able to penetrate Rhabdomyosarcoma cell
nuclei, exactly behaving as the NLS standard peptide.
acids as synthons, is shown. The synthesis was performed by
a submonomeric strategy, obtaining a model trimeric PNA
containing embedded amino acid derived side chains in its
backbone that mimic the peptide sequence PKKKRKV, which
serting amino acid derived side chains in the 2- and 5-posi-
tions (Figure 1). In this way, a real “peptide” and “nucleic
acid” could be obtained, fully exploiting the intrinsic
double nature of the molecule.
Introduction
Peptide nucleic acids (PNAs, Figure 1) are DNA ana-
logues, first introduced by Nielsen and co-workers in
1991,[1] which demonstrate superior ability in specific DNA
binding through standard Watson–Crick hydrogen bonds.
PNAs are able to specifically bind DNA and RNA to give
PNA–DNA and PNA–RNA duplexes that are usually more
stable than DNA–DNA and DNA–RNA duplexes.[2]
A
particularly fascinating, but seldom explored, aspect of
PNAs is their “double nature”: PNAs are not only able to
recognize a complementary DNA sequence through the nu-
cleobases (a property typical of nucleic acids), but they also
possess a pseudopeptidic backbone and therefore might po-
tentially display properties similar to those performed by
proteins. Indeed, almost all the papers published on PNAs
have so far explored their “nucleic acid” characteristics (re-
cognition of complementary sequences) and almost no one
has concentrated on their “peptide” characteristics. The
reason for this asymmetry is easily understood by consider-
ing the structure of standard PNAs: the backbone can be
considered as a long sequence of pseudoglycylglycine dipep-
tides; therefore, a mimic of a polyGly protein. In order to
express all the “proteic potential” of PNAs, the backbone
should be modified to become a real peptide mimic, by in-
Figure 1. A standard PNA compared with a 2S,5S-modified PNA:
both are able to perform DNA recognition, but the backbone of
the latter also has the potential to mimic a peptide sequence.
The first steps toward the exploration of the full poten-
tial of PNAs as “peptide mimics” have been taken by our
group in the last years by synthesizing 2- and 5-substituted
chiral PNAs.[3] Seminal studies have been devoted to estab-
lishing the basic rules by which the configurations of the
single stereogenic centers, independently from the type of
side chains, influence the overall PNA conformation and in
turn how this conformation affects the DNA binding abil-
ity, both in terms of specificity and affinity.[4] In particular,
it has been found that the insertion of a stereogenic center
[a] Department of Organic and Industrial Chemistry
University of Parma
Viale G.P. Usberti 17a, University Campus, 43100 Parma, Italy
Fax: +39-0521-905472
E-mail: stefano.sforza@unipr.it
[b] Department of Pediatrics
University of Bologna
Via Massarenti 11, 40138 Bologna, Italy
[c] Department of Pharmacology, University of Bologna,
Via Irnerio 48, 40126 Bologna, Italy
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000123.
Eur. J. Org. Chem. 2010, 2441–2444
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2441

S. Sforza et al.
SHORT COMMUNICATION
influences the preference of the PNA strand to adopt a
right-handed or a left-handed helix, and as a general rule,
PNAs preferring the right-handed helix will bind to DNA
(which is right-handed) more effectively. The correct config-
urations for having right-handed PNAs in 2,5-disubstituted
PNAs have been found to be 2R,5S and 2S,5S. Thus, a
backbone including -amino acid derived side chains at the
2- and 5-positions (Figure 1) has the potential to be a good
mimic of natural peptides, formed by all -amino acids, and
at the same time preserving the correct handedness for
DNA binding.
Although the introduction of substituents in the PNA
backbone has been used to introduce functional groups and
charges (thus modulating the properties of the PNAs),[5–7]
the use of side chains mimicking an entire complex peptide
consensus sequence with a specific biological function has
never been reported, and this represents a challenging task
both conceptually and synthetically.
In this paper we explore the ability of PNAs to act as a
peptide mimic, reporting the design and the synthesis of a
modified PNA containing an embedded mimic of a nuclear
localization signal (NLS) in its backbone. NLS are short,
predominantly basic, peptide sequences identified as nu-
clear import signals, which are translocated across the nu-
clear envelope by interaction with specific carrier proteins.[8]
The sequence chosen to be embedded into the PNA back-
bone was PKKKRKV, a hydrophilic sequence derived from
the simian virus 40 large T-antigen. This peptide has al-
ready been used, covalently linked to the molecule of inter-
est, to carry antigen oligonucleotides[9] and antigen
PNAs[10–12] in different cell nuclei. The modified PNA was
Figure 2. The NLS peptide PKKKRKV (1), a standard PNA tri-
mer ATG (2), the modified PNA trimer ATG containing the em-
bedded NLS sequence (3). A = adenine, T = thymine, G = guanine,
Gly = glycine, Rho = rhodamine, AEEA = 2-aminoethoxyethoxya-
cetyl spacer.
synthesized and tested for its ability to enter into cell nuclei 1 and PNA trimer 2 were purified by RP-HPLC and char-
and then assayed by fluorescence microscopy. The modified acterized by ESI mass spectrometry (see the Supporting In-
PNA, but not a homologous standard achiral PNA, exactly formation). A Boc submonomer strategy was used for
behaved as the NLS peptide, that is, it entered inside the modified PNA 3 to better preserve the optical purity at the
nuclei of RH30 cancer cells.
2-position.[13] The submonomers were prepared, as re-
ported in Scheme 1, starting from the corresponding com-
mercial Boc--amino acids suitably protected on the side
chains. Boc--Lys-(2-Cl-Z)-OH, to be used as a synthon for
inserting the lysine side chains into the 5-position, was
Results and Discussion
The structure of the NLS peptide (1) chosen to be em- transformed into the corresponding Weinreb amide and
bedded into the PNA backbone is represented in Figure 2. then reduced to the aldehyde. Boc--Lys-(2-Cl-Z)-OH, Boc-
A model PNA consisting of only three residues was chosen -Arg-(Tos)-OH, and Boc--Val were transformed into the
as the substrate into which the peptidic structure would be corresponding methyl esters and Boc deprotected by using
embedded (2, Figure 2). The fusion of the two structures HCl/MeOH. The PNA backbones were then synthesized by
led us to design modified PNA 3 (Figure 2) containing em- reductive amination of the -amino acid methyl esters with
bedded side chains derived from natural -amino acids. The the Boc--Lys-(2-Cl-Z) aldehyde. The methyl esters were hy-
proline residue was linked as an amino acid residue at the drolyzed and the secondary amines were protected by the
N-terminus, for it bears a secondary amine and could not introduction of a Fmoc group (Scheme 1, details in the
be included in a PNA structure. All compounds were syn- Supporting Information). All compounds were charac-
1
thesized with a rhodamine linked at the N-terminus, with terized by H and 13C NMR spectroscopy and ESI mass
an aminoethoxyethoxyacetyl (AEEA) spacer in between, so spectrometry.
that it could be visualized in the interaction and cell pene-
tration studies, and with a glycine residue at the C-terminus, ral submonomeric units were inserted by manual coupling
for easiness of synthesis on the solid phase. with the HATU/DIPEA [O-(7-azabenzotriazol-1-yl)-
According to previously reported procedures,[13] the chi-
The syntheses of NLS peptide 1 and standard PNA tri- N,N,NЈ,NЈ-tetramethyluronium hexafluorophosphate/N,N-
mer 2 were performed on solid phase by following standard diisopropylethylamine] protocol on a MBHA-PS (4-methyl-
Boc protocols. After cleavage from the resins, NLS peptide benzhydrylamine polystyrene) resin, and after Fmoc depro-
2442
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2441–2444

PNA Behaves as a Peptide Mimic
Scheme 1. Synthesis of the submonomers. Reagents and conditions:
(i) N-methyl-N-methoxyamine·HCl (0.97 equiv.), HBTU [O-
(benzotriazol-1-yl)-N,N,NЈ,NЈ-tetramethyluronium
hexafluoro-
phosphate], DIPEA, DMF, η: 90%; (ii) LiAlH4 (1  in THF,
4.6 equiv.), THF, η: 75%; (iii) SOCl2 (1  in CH3OH), η: quantita-
tive; (iv) L–AA–OMe (1 equiv.), Boc-Lys-H (1 equiv.), DIPEA
(1 equiv.), NaBH3CN (3 equiv.), CH3COOH (1.1 equiv.), CH3OH,
η: 30–35%; (v) NaOH (10 equiv.), THF/H2O = 1:1, then HCl up
to pH 5.5, η: 87–90%; (vi) BTSA [N,O-bis(trimethylsilyl)acetamide]
(2 equiv.), DIPEA (1.5 equiv.), FmocCl (2 equiv.), CH2Cl2, η: 30–
35%.
Scheme 2. Solid-phase synthesis of the modified PNA containing
the embedded NLS sequence by the Boc submonomeric strategy.
Reagents and conditions: (i) neat TFA (added of 5% m-cresol);
(ii) HATU/DIEA; (iii) piperidine/NMP 20%; (iv) carboxymethyl-
Z-adenine, carboxymethyl-O-benzylguanine, or carboxymethylthy-
mine, DIC/DhBtOH.
tection, the nucleobases residues were introduced by a
double coupling with DIC/DhBTOH (N,NЈ-diisopropyl-
carbodiimide/3,4-dihydro-3-hydroxy-4-oxo-1,2,3-benzotri-
azine; Scheme 2). After cleavage from the resin, the PNA
trimer was purified by RP-HPLC and characterized by ESI
mass spectrometry (see Supporting Information).
The ability of the different compounds to be internalized penetration are absent, or inefficient under the conditions
in the cell nuclei was assessed in vivo: RH30 cells were incu- used for the experiment. NLS peptide 1, on the other side,
bated with 10 µmol/L of peptide 1, standard PNA 2, or shows an uptake consistent with a receptor-mediated pro-
modified PNA 3 and analyzed after 3 h. Fluorescence mi- cess, as already known.[9–12] Likewise, the concentration of
croscopy images showed that after 3 h in the RH30 cell line, PNA 3 in cell nuclei is a strong indication that this molecule
high intracellular levels with a main nuclear localization is able to follow the same pathway as the NLS peptide. As
could be observed for NLS peptide 1 (Figure 3), as ex- a matter of fact, the uptake in the nuclear region and the
pected, and for modified PNA 3 (Figure 4), whereas stan- homogeneous diffusion is indeed consistent with a receptor-
dard PNA 2 was not detected inside the nuclei (Figure 5), mediated internalization process. Because the importin-me-
confirming the previous results.[12,14]
diated nuclear transport mechanism has been demonstrated
The lack of nuclear fluorescence when standard PNA 2 for NLS peptide 1, it can be assumed that modified PNA 3
was used demonstrates that these probes are not recognized may be able to interact with this receptor protein under the
by any carrier protein, and alternative pathways for cell conditions used.
Eur. J. Org. Chem. 2010, 2441–2444
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2443

S. Sforza et al.
SHORT COMMUNICATION
tide sequence encrypted in the backbone of the modified
PNA can behave as a fully functional peptide mimic, shed-
ding light on a new dimension concerning the properties of
PNAs, which may find numerous applications in bioorganic
chemistry and molecular biology. For example, a PNA with
the embedded NLS sequence could be used to efficiently
translocate the PNA in the cell nuclei after injection in the
blood stream, avoiding degradation from serum proteases,
as it may happen when the NLS sequence is simply linked
as a normal peptide at the C-terminus. In more general
terms, a peptide nucleic acid possessing the functionalities
of the proteins might be potentially used to develop PNAs
endowed with catalytic functions, fully exploiting their in-
trinsic nucleic acid and peptide natures. More experiments
are needed to obtain long PNAs enclosing these modified
monomers at one terminus to verify their binding properties
towards complementary nucleic acids.
Figure 3. Fluorescence microscopy analysis (left) of uptake of
RH30 cells treated for 3 h with NLS peptide 1. Optical images are
also reported (right).
Supporting Information (see footnote on the first page of this arti-
cle): Experimental details, LC/ESI-MS and LC/UV profiles of com-
pounds 1–3.
Acknowledgments
This work was partially supported by grants from the Ministero
dellЈUniversità e della Ricerca (MIUR) (PRIN2005, Grant No.
2005038704, and PRIN2007, Grant No. 2007F9TWKE).
[1] P. E. Nielsen, M. Egholm, R. H. Berg, O. Buchardt, Science
1991, 254, 1497–1500.
[2] M. Egholm, O. Buchardt, L. Christensen, C. Behrens, S. M.
Freier, D. A. Driver, R. H. Berg, S. K. Kim, B. Norden, P. E.
Nielsen, Nature 1993, 365, 556–568.
Figure 4. Fluorescence microscopy analysis (left) of uptake of
RH30 cells treated for 3 h with modified PNA 3. Optical images
are also reported (right).
[3] T. Tedeschi, S. Sforza, R. Corradini, R. Marchelli, Tetrahedron
Lett. 2005, 46, 8395–8399.
[4] S. Sforza, T. Tedeschi, R. Corradini, R. Marchelli, Eur. J. Org.
Chem. 2007, 5879–5885.
[5] V. Menchise, G. De Simone, T. Tedeschi, R. Corradini, S.
Sforza, R. Marchelli, D. Capasso, M. Savino, C. Pedone, Proc.
Natl. Acad. Sci. USA 2003, 100, 12021–12026.
[6] A. Dragulescu-Andrasi, S. Rapireddy, G. He, B. Bhattacharya,
J. J. Hyldig-Nielsen, G. Zon, D. H. Ly, J. Am. Chem. Soc. 2006,
128, 16104–16112.
[7] V. Chenna, S. Rapireddy, B. Sahu, C. Ausin, E. Pedroso, D. H.
Ly, ChemBioChem 2008, 9, 2388–2391.
[8] R. Cartier, R. Reszka, Gene Ther. 2002, 9, 157–167.
[9] D. Lochmann, E. Jauk, A. Zimmer, Eur. J. Pharm. Biopharm.
2004, 58, 237–251.
[10] N. Bendifallah, F. W. Rasmussen, V. Zachar, P. Ebbesen, P. E.
Nielsen, U. Koppelhus, Bioconjugate Chem. 2006, 17, 750–758.
[11] R. Tonelli, S. Purgato, C. Camerin, R. Fronza, F. Bologna, S.
Alboresi, M. Franzoni, R. Corradini, S. Sforza, A. Faccini,
J. M. Shohet, R. Marchelli, A. Pession, Mol. Cancer Ther.
2005, 4, 779–786.
Figure 5. Fluorescence microscopy analysis (left) of uptake of
RH30 cells treated for 3 h with unmodified PNA 2. Optical images
are also reported (right).
[12] G. Cutrona, E. M. Carpaneto, M. Ulivi, S. Roncella, O. Landt,
M. Ferrarini, L. C. Boffa, Nat. Biotechnol. 2000, 18, 300–303.
[13] S. Sforza, T. Tedeschi, R. Corradini, D. Ciavardelli, A. Dos-
sena, R. Marchelli, Eur. J. Org. Chem. 2003, 1056–1063.
[14] A. Pession, R. Tonelli, R. Fronza, E. Sciamanna, R. Corradini,
S. Sforza, T. Tedeschi, R. Marchelli, L. Montanaro, C. Camerin,
M. Franzoni, G. Paolucci, Int. J. Oncol. 2004, 24, 265–272.
Received: January 29, 2010
Conclusions
In conclusion, we have demonstrated the chemical feasi-
bility of synthesizing peptide nucleic acids containing em-
bedded amino acid side chains in their backbone that mimic
a complex peptide sequence. The preliminary data here on
the NLS-mimicking PNA demonstrate that the pseudopep-
Published Online: March 24, 2010
2444
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2441–2444