Pyrrolidine nucleic acids: DNA/PNA oligomers with 2-hydroxy/aminomethyl-4-(thymin-1-yl)pyrrolidine-N-acetic acid

Article abstract of DOI:10.1021/ol0069655

(equation presented) Synthesis of pyrrolidine-based chiral positively charged DNA analogues is reported. The synthesis of (2S,4S) and (2R,4R) thymin-1-ylpyrrolidine-N-acetic acid, its site specific incorporation in PNA:DNA chimera and PNA, and the study o

Full text of DOI:10.1021/ol0069655

ORGANIC
LETTERS
2001
Vol. 3, No. 9
1269-1272
Pyrrolidine Nucleic Acids: DNA/PNA
Oligomers with 2-Hydroxy/Aminomethyl-
4-(thymin-1-yl)pyrrolidine-N-acetic acid
,†
Vaijayanti Kumar,* Pradeep S. Pallan, Meena, and Krishna N. Ganesh*
DiVision of Organic Chemistry (Synthesis), National Chemical Laboratory,
Pune 411008, India
Vkumar@dna.ncl.res.in
Received December 6, 2000 (Revised Manuscript Received March 27, 2001)
ABSTRACT
Synthesis of pyrrolidine-based chiral positively charged DNA analogues is reported. The synthesis of (2S,4S) and (2R,4R) thymin-1-ylpyrrolidine-
N-acetic acid, its site specific incorporation in PNA:DNA chimera and PNA, and the study of their binding properties with complementary
DNA/RNA sequences is presented.
Cat-anionic oligonucleotide analogues1 along with PNA:
DNA chimera2 are gaining increasing attention in the current
literature because of their favorable binding, aqueous solubil-
ity, and cellular permeation properties for potential thera-
peutic applications. The higher affinity of duplexes derived
from cat-anionic DNA and positively charged PNA3 is
probably due to electrostatic complexation leading to fast
on-rates of hybridization. Introduction of chirality and
conformational restraint in the achiral aminoethylglycyl PNA
with the intention of preorganizing the PNA structure
compatible for binding with target DNA/RNA and to
discriminate the parallel/antiparallel binding modes4 has been
partially successful. In our recent efforts,5 we simultaneously
introduced chirality, constrained flexibility, and positive
charge in the PNA backbone by linking the R-carbon of the
glycyl backbone to the â-carbon of the side chain (Figure 1,
path a) to result in an aminoethylglycyl PNA (aepPNA) that
formed highly stable PNA:DNA complexes. Extending the
logic of constrained backbone,4b,c,5 we herein report the
† E-mail: kng@ems.ncl.res.in.
(1) (a) Hashimoto, H.; Nelson, M. G.; Switzer, C. J. J. Am. Chem. Soc.
1993, 115, 7128-7134. (b) Barawkar, D. A.; Rajeev, K. G.; Kumar, V.
A.; Ganesh, K. N.; Nucleic Acids Res. 1996, 24, 1229-1237. (c) Barawkar,
D. A.; Kwok, Y.; Bruice, T. W.; Bruice, T. C. J. Am. Chem. Soc. 2000,
122, 5244-5250. (d) Prakash, T. P.; Manoharan, M.; Fraser, A.; Kawasaki,
A. M.; Lesnik, E. A.; Owens, S. R. Tetrahedron Lett. 2000, 41, 4855-
4859.
(2) Ulhmann, E.; Peyman, A.; Breipohl, G.; Will, D. W. Angew. Chem.,
Int. Ed. 1998, 37, 2796-2823. (b) Ganesh, K. N.; Nielsen, P. E. Curr.
Org. Chem. 2000, 4, 931-943.
(3) Gangamani, B. P.; Kumar, V. A.; Ganesh, K. N. Biophys. Biohem.
Res. Commun. 1997, 240, 778-782.
Figure 1. Structures of DNA, PNA, and their analogues. B )
adenine, cytosine, guanine, or thymine.
10.1021/ol0069655 CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/11/2001

linking of the R′-carbon of the ethylenediamine backbone
segment to the â-carbon of the side chain, generating the
isomeric cationic, chiral PNA (Figure 1, path b). Further,
among the several DNA backbone modifications, the ideal
amide linkage6 (3′-CH2CONH-5′) presumably organizes the
sugar into a high N-pucker, resulting in a stable A-type
duplex formation. Introduction of a positively charged
pyrrolidine ring carrying the HN+-CH2-CO-NH-5′ linkage
for replacement of the sugar phosphate backbone in DNA
(Figure 1) may therefore be expected to have cumulative
advantages, leading to novel pyrrolidine PNA:DNA chimeric
antisense constructs. The monomeric unit 2-hydroxy/ami-
nomethyl-4-thymin-1-ylpyrrolidine-N-acetic acid I containing
a positively charged tertiary nitrogen has dual utility. It is
compatible for introduction in both PNA and DNA (in the
form of dimeric unit 5c) depending upon the nature of the
C2-substituent (hydroxymethyl or aminomethyl) on the
pyrrolidine ring. In this Letter, we report the synthesis and
site-specific incorporation of I (R ) NH2) and 5c into PNA/
DNA oligomers and the evaluation of their binding properties
with target DNA sequences.
Scheme 2a
a (a) DMTCl, pyridine, 70%; (b) H2 Pd/C, MeOH-ethyl acetate,
98%; (c) BrCH2COOEt, CH3CN, KF/Celite, 67%; (d) aqueous
NaOH, MeOH, 90%; (e) TBTU, HOBt, 3′-O-benzoyl-5′-amino-
5′-deoxythymidine, 42%; (f) tetrazole, P(OCH2CH2CN){N(iPr)2}2,
CH3CN, 80%.
bromoacetate to give the DMT ester 4a. Hydrolysis of the
ester with methanolic NaOH to give the acid 4b and
condensation with 3′-O-benzoyl-5′-amino-5′-deoxythymidine
in the presence of TBTU-HOBt yielded the dimer 5a.
Alkaline deprotection of the 3′-O-benzoyl in 5a to 5b,
followed by 3′-O-phosphitylation, gave the protected chi-
meric dimer phosphoramidite 5c for site-specific incorpora-
tion into oligonucleotides by automated solid-phase synthesis.
All the synthesized compounds were characterized by H
and 13C NMR and mass spectral analysis.7
Synthesis of 2-Aminomethyl Monomer I for PNA. The
monomer 7b needed for insertion into PNA oligomers was
synthesized from 3a (Scheme 3) by sequential tosylation of
Synthesis of Dimeric Block for Chimeric DNA:PNA.
N3-Benzoylthymine was alkylated at N1 with N-protected
proline methyl esters5 1a (2S,4R) or 1b (2R,4S) (Scheme 1)
1
Scheme 1a
Scheme 3a
a (a) N3-Benzoylthymine, DEAD, PPh3, benzene, 68%; (b)
LiBH4, THF, 92%.
via Mitsunobu reaction at C4 to yield 2a and 2b, respectively.
Subsequent reduction with excess LiBH4 afforded the alcohol
3a/3b with simultaneous removal of the N-3 benzoyl of
thymine. The pyrrolidine derivative 3b (2R,4R) with stere-
ochemistry equivalent to that of the natural nucleoside was
used to synthesize chimeric dimer block 5c suitable for solid-
phase DNA synthesis (Scheme 2). The primary hydroxyl
function in 3b was protected as the dimethoxytrityl ether
followed by deprotection of the ring nitrogen by hydrogena-
tion over Pd-C and subsequent alkylation with ethyl
a (a) TsCl, pyridine, 66%; (b) NaN3, DMF, 70 °C, 98%; (c)
Raney-Ni, H2, 95%; (d) Boc N3/Et3N, DMSO, 50 °C, 90%; (e)
Pd/C, H2, 98%; (f) BrCH2COOEt, CH3CN, KF/Celite, 60 °C, 56%;
(g) aqueous NaOH, MeOH, 95%.
the primary hydroxyl group, azidation with NaN3, and
conversion to amine by hydrogenation with Ra-Ni followed
by protection as a tert-butylcarbamate to give 6. The
deprotection of the ring nitrogen by hydrogenation over
Pd-C and subsequent alkylation with ethyl bromoacetate
gave 7a (2S,4S). The monomer7 7b was obtained by
hydrolysis with 1 M NaOH (aqueous methanol) and used
for solid-phase peptide synthesis using Boc chemistry.
DNA:PNA Chimeras. The dimer block phosphoramidite
5c was incorporated into oligonucleotide sequences 9-13
at the desired positions by automated DNA synthesis to yield
the chimeric oligomers. The purity of all modified oligomers
(4) (a) Haaima, G.; Lohse, A.; Buchardt, O.; Nielsen, P. E. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 1939-1941. (b) Gangamani, B. P.; Kumar, V. A.;
Ganesh, K. N. Tetrahedron 1996, 52, 15017-15030. (c) Gangamani, B.
P.; Kumar, V. A.; Ganesh, K. N. Tetrahedron 1999, 55, 177-192. (d)
Sforza, S.; Corradini, R.; Ghirardi, S.; Dossena, A.; Marchelli, R. Eur. J.
Org. Chem. 2000, 2905-2913.
(5) D’Costa, M.; Kumar, V. A.; Ganesh, K. N. Org. Lett. 1999, 1, 1513-
1516.
(6) (a) De Mesmaeker, A.; Altmann, K.-H.; Waldner, A.; Wendeborn,
S. Curr. Opin. Struct. Biol. 1995, 5, 343-355. (b) Robins, M. J.;
Doboszwski, B.; Nilsson, B. L.; Peterson, M. A. Nucleoside Nucleotide
Nucl. Acids 2000, 19, 69-86.
1270
Org. Lett., Vol. 3, No. 9, 2001

Table 1. UV-Tm of DNA Duplexes and Triplexesa
entry
complex
Tm (°C)
∆Tm (°C)
1
2
3
4
5
6
7
8
9
8:14
9:14
10:14
11:15
12:15
13:15
16:15
8*17:18
9*17:18
10*17:18
45
42
38
51
45
50
57
29
29
25
-3
-7
-6
-12
-7
0
-4
10
a Buffers: (a) duplexes, 10 mM sodium phosphate, pH 7.0, 100 mM
NaCl, 0.1 mM EDTA; (b) triplexes, 10 mM sodium phosphate, 200 mM
NaCl, pH 7.0. All Tm values are accurate to (0.5 °C and measured in four
melting experiments.
other in the center (Table 1, entry 5) was destabilized to a
greater extent (<12 °C), while introduction of a second
modified unit at the 3′ terminus as in 13:15 effected a lesser
destabilization of 7 °C (Table 1, entry 6). The dimer 5c,
though configurationally equivalent to natural DNA, perhaps
induces structural deviations in the duplexes of oligomers
9-13 resulting in their destabilization as compared to the
control, unmodified duplexes. The polypyrimidine oligomers
8, 9, or 10 also form a third (Hoogsteen) strand of a triplex
with the duplex 17:18 (Figure 2B). It was found that the
stability of the triplex 9*17:18 with a chimeric oligomer
having the pyrrolidine dimer 5c at the 3′-end was similar to
that of the corresponding unmodified triplex 8*17:18. The
triplex with a third strand carrying the dimer in the middle
(10*17:18) exhibited a 4 °C destabilization (Table 1, entries
8, 9, and 10). This indicates that the modification is
accommodated better at the 3′ terminus compared to the
modification in the middle of the sequence.
Figure 2. First-derivative normalized UV-Tm plots: A, PNA:
DNA chimeric duplexes; B, triplexes derived from duplexes 17:18
and PNA:DNA chimeric third strands 8-10; C, PNA:DNA/RNA
triplexes (a) 19:20, (b) 19:21, (c) 19:22, and (d) 21:poly rA and
(e) 22:poly rA.
was established by RP-HPLC, and the compounds were
characterized by mass spectra. The chimeric ODNs 9-10
and 11-13 were individually hybridized with the comple-
mentary DNA strands 14 and 15, respectively, to obtain
duplexes for UV-Tm measurements (Figure 2A), while
duplexes 8:14 and 16:15 served as the control.
Cationic and Chiral PNA. The monomer (2S,4S)-2-(N-
Boc-aminomethyl)-4-thymin-1-ylpyrrolidine-N-acetic acid 7b
was used to prepare the homopyrimidine PNA oligomers 21
and 22 by standard PNA synthesis protocols8 along with the
control PNA 20. The purity of the oligomers was rechecked
by RP HPLC on a C18 column and characterized by
MALDI-TOF mass spectrometry. No precipitation was
(7) 5b. 1H NMR in CDCl3 8.6 (NHCO), 7.62 (s, 1H, H6 Thy), 7.4-7.2
(m, 9H, DMT + 1H, NH Thy, + 1H, H6 Thy), 6.8-6.45 (d, 4H, DMT),
6.04 (t, 1H, H1′), 5.05 (bs, 1H, C4H), 4.25 (bs, 1H H3′), 3.85 (m, 1H,
H4′), 3.75 (s, 6H, OMe), 3.7-3.5 (m, 2H, CH2ODMT), 3.5-3.15 (m, 4H,
H5, H5′, C5H, C5′H), 3.0-2.6 (m, 2H, CH2N), 2.4-2.2 (m, 3H, H2′, H2′′,
C3H), 2.1-1.9 (m, 1H, C3′H), 1.82 (s, 3H, Me Thy), 1.67 (s, 3H, 5-MeThy).
13C NMR CDCl3 + pyridine-d5 δ 169.74 (CO amide), 163.86, 163.73 (2
× C4 Thy), 157.59 (C2 Thy), 148.72, 148.52 (2 × C6 Thy), 109.83 (C5),
85.26 (C1′), 84.47 (C4′), 70.73 (C3′), 62.78 (amide CH2), 62.06 (C4), 58.27
(C6), 55.71 (C5), 54.03 (OMe), 51.62 (C2), 40.54 (C5), 38.88 (C2′), 34.42
(C3), 11.46 (Thy Me). FAB mass 831 (M + Na), calculated mass 808.3.
The duplex with polypyrimidine ODN 9 carrying the
amide-linked pyrrolidine-sugar dimer at the 3′ terminus
exhibited a 3 °C destabilization compared to the control
duplex 8:14 (Table 1, entries 1 and 2) while the destabiliza-
tion increased when the dimer was at the center as in 10:14
(-7 °C, entry 3). The duplex 11:15 with a single modification
at the center in a mixed pyrimidine-purine sequence also
showed a similar destabilization of 6 °C compared to the
relevant control duplex 16:15 (Table 1, entries 4 and 7). The
duplex 12:15 containing two modified units adjacent to each
1
5c. 31P in CDCl3, 149.50, 148.42. 7b. H NMR in D2O δ 7.5 (s, 1H, Thy
H6), 4.8 (m, 1H, H4), 4.1-3.55 (m, 2H, H5, H5′), 3.6-3.35 (m, 5H, CH2-
NH, CH2CO, H2), 2.85-2.6 (m 1H, H3), 2.4-1.9 (m, 1H, H3′), 1.9 (s,
3H, Thy CH3), 1.49 (s, 9H, t-Boc). 13C in D2O δ 170.95 (CO acid), 167.53
(CO ThyC4), 159.22 (CO Boc), 144.19 (CO ThyC2), 144.19 (C6 Thy),
111.48 (C5 Thy), 82.55 (C-Boc), 68.29 (C4 Pro), 60.31 (CH2COO), 59.24
(C2 Pro), 56.04 (CH2NH), 38.44 (C5 Pro), 32.7 (C3 Pro), 28.55 (Me Boc),
12.33 (Me Thy). FAB mass 383 (M++ H), calculated mass 382.
(8) Peptide Nucleic Acids protocols and applications; Nielson, P. E.,
Egholm, M., Eds.; Horizon Scientific Press: Norfolk, England, 1999.
Org. Lett., Vol. 3, No. 9, 2001
1271

observed in the samples of modified PNAs 21 and 22 even
after prolonged storage.
in prolyl PNA4b,c and aepPNA.5 As originally envisaged by
us, the possibility of four diastereomers from the two chiral
centers in the pyrrolidene ring, in conjunction with different
combinations of ring substitution, leads to a number of
regio- and stereocentric modifications. From our work and
that of others, interesting and unforeseen aspects of prolyl/
pyrrolidyl-PNA:DNA/RNA binding selectivity are now
emerging. While cationic aepPNA in both (2S,4S) and
(2R,4S) modifications exhibited high binding affinity to
cDNA,5 the corresponding (2R,4R) analogue was reported
to bind preferentially to cRNA.9b In the present modification
series, the (2R,4R) PNA analogues displayed improved
binding affinity toward the complementary DNA and RNA.9c
The (2S,4S) geometry of the pyrrolidine ring in PNA reported
here does not seem to be structurally compatible in the PNA
backbone, although it was very effective in forming DNA:
aepPNA2 triplexes.5
The (2R,4R) stereochemistry used for the pyrrolidine-
sugar thymine dimer unit was expected to be compatible with
DNA geometry, the nucleobase being cis to the aminomethyl
segment of the backbone, as in DNA sugar. However, a
positively charged pyrrolidine ring nitrogen might adapt a
ring pucker, resulting in a nucleobase orientation detrimental
to duplex formation with the complementary DNA. Further
studies on the effective stereochemical preferences exerted
by these units on the DNA/RNA binding properties are
necessary.
In conclusion, the preliminary results reported here add
to a growing stereooligomeric library of pyrrolidine PNAs
that offer new insights into understanding PNA-DNA/RNA
complexation. Such studies may eventually allow for rational
design of potential antisense molecules that display a high
discrimination in recognition of DNA/RNA. The binding
results obtained so far from mixed backbone, homopyrimi-
dine/purine sequences may have limitations in extrapolation
to mixed sequence complexes. Pure modified backbone
oligomers may bind DNA/RNA differently than the mixed
backbone oligomers. The presence of a positive charge on
pyrrolidine PNA may also favor cell permeation as seen in
the case of cationic peptide-PNA conjugates.10 Future work
in these directions is in progress to delineate the binding
preferences of different stereochemically divergent pyrroli-
dine PNA analogues to complementary DNA/RNA.
Homopyrimidine PNA sequences are known to form
PNA2:DNA triplexes with a homopurine DNA such as 19.
The PNA oligomers 20-22 were independently hybridized
with the complementary DNA oligomer 19 and poly rA, with
a PNA:DNA/RNA stoichiometry of 2:1 for Tm studies
(Figure 2C). The DNA:PNA complex 19:21 having the
modified unit 7b at the C-terminus had a Tm 6° lower than
that of the control duplex 19:20 (Table 2, entries 1 and 2).
Table 2. UV-Tm of PNA:DNA/poly rAa
entry
complex
Tm (°C)
1
2
3
4
5
19:20
19:21
19:22
poly rA:21
poly rA:22
46
40
70
27
*All Tm values are accurate to (0.5 °C and measured in four melting
experiments. Buffers: entry 1-3, 10 mM sodium phosphate, pH 7.3; entries
4-5, as above with 100 mM NaCl.
The simultaneous presence of the modified unit in the center
as well as at the C-terminus in the complex 19:22 largely
affected the stability, and the absence of any temperature-
dependent UV absorbance change suggested no complexation
with the target DNA (curve c, Figure 2C).
The hybridization efficiency of modified PNAs with RNA
was also examined by complexation with poly rA. The PNA
21 with a single modification had a Tm of 70 °C whereas
the Tm of PNA 22 was much lower (27 °C). Under the buffer
conditions employed, the control poly rA:20 complex did
not melt.
While this Letter was in preparation, three interesting
reports concerning pyrrolidine PNA appeared in the litera-
ture.9 These are related to our ideas on constrained flexibility
(9) (a) Hickman, D. T.; King, P. M.; Cooper. M. A.; Slater, J. M.;
Micklefield Chem. Commun. 2000, 2251-2252. (b) Vilaivan, T.; Khongdee-
sameor, C.; Harnyuttanakorn, P.; Westwell, M. S.; Lowe, G. BioMed. Chem.
Lett. 2000, 10, 2541-2545. (c) Puschi, A.; Tedeschi, T.; Nielsen, P. E.
Org. Lett. 2000, 2, 4161-4163.
(10) Simmons, C. G.; Pitts, A. E.; Mayfield, L. D.; Shay, J. W.; Corey,
D. R. BioMed. Chem. Lett. 1997, 7, 3001-3006.
Acknowledgment. P.S.P and Meena thank CSIR, New
Delhi, for the award of research fellowships.
OL0069655
1272
Org. Lett., Vol. 3, No. 9, 2001