Synthesis and nucleic acid binding studies of novel pyrrolidinyl PNA carrying an N-amino-N-methylglycine spacer

Article abstract of DOI:10.1016/S0040-4039(03)00033-9

Two novel pyrrolidinyl peptide nucleic acids comprising alternating sequences of thymine-modified D- or L-proline and an N-amino-N-methylglycine spacer were synthesized using solid-phase methodology. UV and CD titrations together with a gel-binding shift assay revealed that neither of the homothymine PNA decamers bind to their complementary DNA or RNA. This was considered to be due to an unfavorable secondary structure which could not be alleviated by the presence of the positively charged protonated amine in the PNA backbone.

Full text of DOI:10.1016/S0040-4039(03)00033-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1663–1666
Synthesis and nucleic acid binding studies of novel pyrrolidinyl
PNA carrying an N-amino-N-methylglycine spacer
Tirayut Vilaivan,a,* Chaturong Suparpprom,a Preeyanut Duanglaor,a
Pongchai Harnyuttanakornb and Gordon Lowec
aOrganic Synthesis Research Unit, Department of Chemistry, Faculty of Science, Chulalongkorn University,
Phayathai Road, Patumwan, Bangkok 10330, Thailand
bDepartment of Biology, Faculty of Science, Chulalongkorn University, Phayathai Road, Patumwan, Bangkok 10330, Thailand
cDyson Perrins Laboratory, Department of Chemistry, University of Oxford, South Parks Road, Oxford OX1 3QY, UK
Received 4 November 2002; accepted 20 December 2002
Abstract—Two novel pyrrolidinyl peptide nucleic acids comprising alternating sequences of thymine-modified D- or L-proline and
an N-amino-N-methylglycine spacer were synthesized using solid-phase methodology. UV and CD titrations together with a
gel-binding shift assay revealed that neither of the homothymine PNA decamers bind to their complementary DNA or RNA. This
was considered to be due to an unfavorable secondary structure which could not be alleviated by the presence of the positively
charged protonated amine in the PNA backbone. © 2003 Elsevier Science Ltd. All rights reserved.
Peptide nucleic acid (PNA) is a DNA analogue with a
polyamide backbone which was shown to possess
strong nucleic acid binding properties.1 Recently there
has been much interest in conformationally rigid ana-
logues of PNA since in principle they should bind more
strongly with nucleic acids according to the concept of
pre-organization.2 We have recently synthesized and
studied the interactions of a conformationally con-
strained PNA based on an alternating sequence of
(2R,4R)-4-thyminylproline and b-amino acid spacers 1
with complementary DNA and RNA.3,4 It was evident
that the stereochemistry of the spacer amino acid is
critical for successful binding. Furthermore, we have
discovered that PNA 2 bearing the aminopyrrolidine-
RNA.4 From the limited data available, it was not
entirely clear whether a structural factor or the presence
of a positive charge on the PNA strand at physiological
pH, or both, were responsible for the unusual tight
binding properties of the PNA 2.
To investigate further the effect of structure and posi-
tive charge on the spacer, we designed a novel PNA 3
bearing a flexible N-amino-N-methylglycine spacer.
This is expected to decrease the binding affinity of 3
compared to 2 due to the higher entropy loss on
binding. On the other hand, the more flexible spacer
would allow 3 to adopt some conformations not avail-
able to 2, which may result in a PNA with different
binding characteristics. Indeed, related PNA systems
bearing a positively charged and conformationally
2R-carboxylic acid spacer (D-Apc) shows an unusual
preference for binding to its complementary DNA over
Keywords: nucleic acids; PNA; DNA; RNA; b-amino acids.
* Corresponding author. Tel.: +66-2-2187627; fax: +66-2-2187598; e-mail: vtirayut@chula.ac.th
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(03)00033-9

1664
T. Vilai6an et al. / Tetrahedron Letters 44 (2003) 1663–1666
Scheme 1. Reagents and conditions: (i) excess 40% aq. MeNH2, KI (1 equiv.) (91%); (ii) NaNO2, 3 M H2SO4 0°C (87%); (iii) Zn,
50% HOAc 0°C (60%); (iv) FmocCl, DIEA; CH2Cl2 (55%); (v) TFA; CH2Cl2; (vi) PfpOCOCF3, DIEA; CH2Cl2 (68%, two steps).
flexible aminoethyl spacer showed good binding to
RNA and DNA.5 Since different stereochemistries on
the thyminylproline unit can have a dramatic effect on
the binding characteristics of the PNA,5,6 the PNA 4
was also included in our target molecules.
Binding of the PNA 3-T10 and 4-T10 to polydeoxy-
adenylic acid (complementary DNA) and polyadenylic
acid (complementary RNA) were studied by UV titra-
tion (20°C, 10 mM sodium phosphate buffer, pH 7.0).
If the PNA and DNA can form a stable hybrid at or
above the temperature at which the titration was car-
ried out, a non-linear increment in absorbance at 260
nm would be expected due to hypochromism, with a
discontinuity at the equivalence point. According to
such criteria, no binding between 3-T10 or 4-T10 and
complementary DNA was observed as the absorbance
showed a near-perfect linearity with the observed and
calculated absorbance in good agreement. Likewise, the
UV-titration experiment indicated that 3-T10 and 4-T10
did not bind to their complementary RNA. The lack of
binding was also supported by polyacrylamide gel elec-
trophoresis of 3-T10 in the presence of fluorescent-
labeled (dA)10, the only fluorescent band observed
migrating at the same rate as the unhybridized fluores-
cent-labeled (dA)10.
The
Fmoc-protected
N-amino-N-methylglycine
monomer 5 was synthesized from readily available
starting materials as shown in Scheme 1.7 The pen-
tafluorophenyl-activated
Fmoc-protected
thymine
building blocks 6 and 7 were synthesized as previously
described.3,8 The target sequence for both 3 and 4 was
H-T10-LysNH2, i.e. B=thymine. The lysinamide was
included at the C-termini of both PNA for comparison
with previous PNA in this series.9 The PNA syntheses
were carried out following our previously reported pro-
tocol.10 The acid-sensitive Rink-amide Novasyn TGR
resin was used as the solid support. The lysine,
nucleoamino acids and spacer were introduced as their
pentafluorophenyl esters (4 equiv.) in the presence of
1-hydroxy-7-azabenzotriazole (HOAt) in DMF (120
min, single coupling). UV monitoring of the dibenzoful-
vene–piperidine adduct suggested that the efficiency of
each coupling step was greater than 95%, nevertheless,
capping (Ac2O/DIEA) was routinely performed at
every step. The overall coupling efficiency was deter-
mined to be 48% for (3-T10) and 58% for (4-T10). After
the synthesis of the entire sequence was completed, the
PNA’s were released from the resin by treatment with
trifluoroacetic acid followed by ether precipitation and
were purified by reverse phase HPLC. The purified
PNAs were homogeneous by reverse phase HPLC (3-
T10, tR=28.5 min; 4-T10, tR=25.5 min, column: C-18, 4
m particle size, acetonitrile:water gradient 10:90–90:10
over 20 min, hold time 5 min, flow rate 0.5 mL/min)
and showed the expected molecular weight (MALDI
TOF, 3-T10 calcd for M+, 3216.42, found 3218.06;
4-T10, calcd for M+, 3216.42 found 3216.40).
Finally, CD experiments confirmed the lack of binding
between 3-T10/poly(dA), 3-T10/poly(rA), 4-T10/poly(dA)
and 4-T10/poly(rA). Interestingly, the single stranded
PNA 3-T10 showed strong CD signals with an ellipticity
maximum at 280 nm and a minimum at 225 nm while
4-T10 showed a CD spectrum which is almost the
mirror image of 3-T10 (Fig. 1).11 This suggests the
presence of a stable secondary structure in both PNA
as found in polymers of b-amino acids.12 Nevertheless,
in all cases the CD spectra of a 1:1 mixture of the PNA
and poly(rA) or poly(dA) were identical to the sum of
each component even at a temperature as low as 10°C
(Fig. 2). In addition, a plot of ellipticity at different
ratios of PNA:DNA in CD-titration experiments did
not show any discontinuity, confirming the lack of
binding.

T. Vilai6an et al. / Tetrahedron Letters 44 (2003) 1663–1666
1665
Figure 1. CD spectra of single stranded PNA 3-T10 and 4-T10. Conditions: concentration of PNA strand=1 mM, 10 mM sodium
phosphate pH 7.0, 20°C.
Figure 2. CD spectra of single-stranded poly(dA), 3-T10 and their 1:1 mixture compared with the calculated spectra. Conditions:
concentration of PNA=DNA=10 mM nucleotides, 10 mM sodium phosphate pH 7.0, 20°C.
The binding experiments indicate that the PNA 3-T10 and
4-T10 were unable to form stable hybrids with poly(dA)
or poly(rA). This is in sharp contrast to the PNA 2-T10
which forms a very stable 1:1 PNA–DNA hybrid with
a Tm of 80°C (10 mM sodium phosphate buffer, pH 7.0)
and a less stable hybrid with RNA (Tm 32°C).4 The
results suggested that the presence of a positively charged
protonated amine on the PNA strand is probably not the
major contributor to the unusual nucleic acid binding
properties of PNA 2. Consequently, the presence of the
structurally rigid aminopyrrolidine carboxylic acid
spacer appeared to be the more important factor. The
fact that single stranded PNA 3-T10 and 4-T10 show
strong CD spectra (Fig. 1) indicates that the PNA’s adopt
secondary structures which are not favorable for binding
to complementary DNA or RNA, the need for struc-
tural-reorganization rendering the binding process ther-
modynamically unfavorable.

1666
T. Vilai6an et al. / Tetrahedron Letters 44 (2003) 1663–1666
In conclusion, we have successfully synthesized two
novel pyrrolidinyl PNAs carrying the N-amino-N-
methylglycine spacer. The T10 sequence of both PNAs
did not hybridize with poly(dA) and poly(rA) under the
conditions studied. The lack of binding provides valu-
able new information on the structural requirements for
the PNA–nucleic acids interaction which is important
in designing further PNA systems with improved bind-
ing properties.
2. Kumar, V. A. Eur. J. Org. Chem. 2002, 2021–2032.
3. Vilaivan, T.; Suparpprom, C.; Harnyuttanakorn, P.;
Lowe, G. Tetrahedron Lett. 2001, 42, 5533–5536.
4. Vilaivan, T.; Lowe, G. J. Am. Chem. Soc. 2002, 124,
9326–9327.
5. (a) Vilaivan, T.; Khongdeesameor, C.; Harnyuttanakorn,
P.; Westwell, M. S.; Lowe, G. Bioorg. Med. Chem. Lett.
2000, 10, 2541–2545; (b) D’Costa, M.; Kumar, V. A.;
Ganesh, K. N. Org. Lett. 1999, 1, 1513–1516; (c)
D’Costa, M.; Kumar, V. A.; Ganesh, K. N. Org. Lett.
2001, 3, 1281–1284.
Acknowledgements
6. (a) Lowe, G.; Vilaivan, T.; Westwell, M. S. Bioorg. Chem.
1997, 25, 321–329; (b) Puschl, A.; Boesen, T.; Zuccarello,
G.; Dahl, O.; Pitsch, S.; Nielsen, P. E. J. Org. Chem.
2001, 66, 707–712.
This work is supported by a grant from the Thai
Government Research Fund (The Bureau of The Bud-
get Office of the Prime Minister, 2002) and the Thai-
land Research Fund (RSA/04/2545) (to T.V.). C.S. is a
research fellow under the Royal Golden Jubilee Ph.D.
project. P.D. is supported by the Thai Government’s
Promotion and Development of Science and Technol-
ogy Talents Project (DPST) and Thailand’s Junior Sci-
ence Talents Project (JSTP). We also thank Dr.
Carolyn A. Carr (Synaptica Research, UK) for obtain-
ing the MALDI-TOF mass spectra.
7. 5: white crystalline solid, mp 122.4–125.2°C; anal. calcd.
for C24H17F5N2O4 C, 58.54, H, 3.48, N, 5.69; found C,
58.51, H, 3.46, N, 5.69%; 1H NMR (200 MHz, CDCl3)
lH 2.82 (s, 3H), 4.03 (br m, 2H), 4.22 (t, 1H, J=6.8 Hz),
4.46 (d, 2H, J=6.8 Hz), 6.60 (br m, 1H), 7.25–7.42 (m,
4H), 7.57 (d, 2H, J=7.3 Hz), 7.75 (d, 2H, J=7.3 Hz); 13
C
NMR (50 MHz, CDCl3) lC 44.2, 47.1, 57.7, 67.2, 120.0,
125.0, 127.1, 127.8, 135.4–140.5, 141.3, 143.6, 155.6,
166.4; HRMS (ESI-TOF) calcd for C24H17F5N2O4+Na+,
515.1006; found 515.1000.
8. (a) Lowe, G.; Vilaivan, T. J. Chem. Soc., Perkin Trans. 1
1997, 539–546; (b) Lowe, G.; Vilaivan, T. J. Chem. Soc.,
Perkin Trans. 1 1997, 547–554.
References
9. Egholm, M.; Buchardt, O.; Nielsen, P. E.; Berg, R. H. J.
1. Recent reviews: Uhlman, E.; Peyman, A.; Breipohl, G.;
Will, D. W. Angew. Chem., Int. Ed. 1998, 37, 2796–2823;
Nielsen, P. E.; Haaima, G. Chem. Soc. Rev. 1997, 73–78;
Nielsen, P. E. Pure Appl. Chem. 1998, 70, 105–110;
Nielsen, P. E. Acc. Chem. Res. 1999, 32, 624–630;
Falkiewicz, B. Acta Biochimica Polonica 1999, 46, 509–
529; Ganesh, K. N.; Nielsen, P. E. Curr. Org. Chem.
2000, 4, 931.
Am. Chem. Soc. 1992, 114, 1895–1897.
10. Lowe, G.; Vilaivan, T. J. Chem. Soc., Perkin Trans. 1
1997, 555–560.
11. PNA 3-T10 and 4-T10 are not exactly mirror images
because they contain L-lysinamide at the C-termini.
12. Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. Chem.
Rev. 2001, 101, 3219–3232.