A novel N-(pyrrolidinyl-2-methyl)glycine-based PNA with a strong preference for RNA over DNA

Article abstract of DOI:10.1002/1099-0690(200207)2002:14<2391::AID-EJOC2391>3.0.CO;2-D

A novel, N-(pyrrolidinyl-2-methyl)glycine-based (Pmg-based) PNA is introduced. The synthesis of the backbone was accomplished in good yield, starting from prolinol. Thymine (S)- and (R)-Pmg and adenine- and cytosine-derived (R)-Pmg monomers were prepared. Five different fragments - two with either the (R) or the (S) isomer of the thymine Pmg monomer, two oligomers with two consecutive (R)- or (S)-thymine Pmg units, and fully modified (R)-Pmg decamer - were assembled on a solid support. UV thermal melting experiments with complementary DNA and RNA were performed in order to determine the effects of conformational restriction, steric hindrance, and chirality on the duplex stability. It was found that the (R)-Pmg-containing PNAs bound better to DNA and RNA than those containing the (S)-Pmg isomer and that both (S)- and (R)-Pmg-containing PNAs preferentially bound to complementary RNA with a selectivity higher than that of 2-(aminoethyl)glycine (Aeg) PNA. However, even (R)-Pmg-containing oligomers showed poorer duplex stability than nonmodified Aeg-PNA, while no detectable binding either to the DNA or to the RNA was observed with the fully modified (R)-Pmg decamer. Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002.

Full text of DOI:10.1002/1099-0690(200207)2002:14<2391::AID-EJOC2391>3.0.CO;2-D

FULL PAPER
A Novel N-(Pyrrolidinyl-2-methyl)glycine-Based PNA with a Strong Preference
for RNA over DNA
Andis Slaitas[a] and Esther Yeheskiely*[a]
Keywords: Chirality / DNA recognition / Nitrogen heterocycles / Nucleic acids / RNA recognition / Solid-phase synthesis
A
novel, N-(pyrrolidinyl-2-methyl)glycine-based (Pmg-
duplex stability. It was found that the (R)-Pmg-containing
PNAs bound better to DNA and RNA than those containing
the (S)-Pmg isomer and that both (S)- and (R)-Pmg-con-
based) PNA is introduced. The synthesis of the backbone
was accomplished in good yield, starting from prolinol.
Thymine (S)- and (R)-Pmg and adenine- and cytosine-de- taining PNAs preferentially bound to complementary RNA
rived (R)-Pmg monomers were prepared. Five different frag- with a selectivity higher than that of 2-(aminoethyl)glycine
ments − two with either the (R) or the (S) isomer of the thym- (Aeg) PNA. However, even (R)-Pmg-containing oligomers
ine Pmg monomer, two oligomers with two consecutive (R)-
or (S)-thymine Pmg units, and fully modified (R)-Pmg de-
camer − were assembled on a solid support. UV thermal
melting experiments with complementary DNA and RNA
were performed in order to determine the effects of con-
formational restriction, steric hindrance, and chirality on the
showed poorer duplex stability than nonmodified Aeg-PNA,
while no detectable binding either to the DNA or to the RNA
was observed with the fully modified (R)-Pmg decamer.
( Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
2002)
Introduction
as in Aeg-PNA. Fragments would be constructed by intro-
duction of an amide bond between the carboxyl and the
pyrrolidineamino functions of two monomers. The second
amino group in the Pmg monomeric unit would be the at-
tachment point (carbonyl methylene) of the nucleobase.
Unlike aminoethylprolyl PNA[17] (Aep-PNA, II) and pyrro-
lidine PNA[18] (not shown), which are positively charged
(Figure 1), the Pmg backbone should remain uncharged at
physiological pH. Polycationic species are generally known
to form stable complexes with DNA or RNA due to charge
attraction; however, nonspecific binding may be a con-
sequence.[19]
PNA, an artificial nucleic acid, the backbone of which is
composed of repeating 2-(aminoethyl)glycine (Aeg) units,
forms stable complexes with complementary DNA and
RNA.[1Ϫ5] Previous publications have shown that PNAs
based on amino acids other than glycine possess similar
properties.[6Ϫ10] These reports have demonstrated that the
chirality of the backbone is an important factor determin-
ing the stability of the complexes. In addition, recent studies
have indicated that the introduction of conformational re-
striction into oligonucleotides,[11,12] PNA-DNA[13,14] chi-
meras, and modified PNAs[15,16] has a beneficial effect on
their hybridization properties. It is believed that suitable
conformational restriction in the nucleic acid backbone
should result in better positioning relative to the target
strand, thus giving higher affinity. We were interested in
investigating the hybridization properties of an uncharged
PNA modification that would combine both conformation
restriction and chirality, and so we designed and prepared
a new chiral PNA analogue, the backbone of which con-
tained both isomers of N-(2-pyrrolidinylmethyl)glycine
(Pmg) I.
Figure 1. Structures of Pmg- and Aep-PNAs
Study of the hybridization of Pmg-PNA should permit
the determination of the conformation restriction and chir-
ality effects on duplex formation in the absence of positive
charge. This may provide useful information for the devel-
opment of new PNA analogues.
As a part of an ongoing research program geared to the
design, synthesis, and investigation of potential antisense
agents and tools for biological studies,[20,21] we give here a
We reasoned that the pyrrolidine moiety would introduce
restraints and chirality, while the backbone would still con-
tain the requisite six covalent bonds in the monomeric unit,
[a]
Division of Organic and Bioorganic Chemistry, MBB, Scheele
Laboratory, Karolinska Institute,
17177 Stockholm, Sweden
Fax: (internat.) ϩ 46-8/311052
E-mail: esther.kuyl-yeheskiely@mbb.ki.se
Eur. J. Org. Chem. 2002, 2391Ϫ2399  WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/0714Ϫ2391 $ 20.00ϩ.50/0
2391

A. Slaitas, E. Yeheskiely
FULL PAPER
full report of the synthesis of and UV melting experiments
involving Pmg-containing PNAs.
Results and Discussion
The first step in the assembly of Pmg-containing PNA
fragments on a solid support was the preparation of
properly protected Pmg building blocks. Initially we chose
tert-butoxycarbonyl (Boc) as a temporary protecting group
for the pyrrolidine amino group, and benzoyl for the ex-
ocyclic amino functions of the nucleobases.
Thymine (S)-Pmg unit 1a was synthesized from commer-
cially available benzyl N-[(2S)-N-Boc-pyrrolidinylmethyl]-
glycinate[22] (2) (Scheme 1). Acylation of compound 2 at the
free secondary amino function was performed with thym-
ine-1-acetic acid (3) and the coupling agent 2-(benzotriazol-
1-yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate
(HBTU) in the presence of N-ethyldiisopropylamine
(DIEA) and 4-(dimethylamino)pyridine (DMAP). The isol-
ated yield of the fully protected (S)-Pmg monomer 4a was
80%. Basic hydrolysis of the benzyl ester 4a with aqueous
NaOH and transformation of the sodium salt into triethyl-
ammonium salt yielded monomer 1a in a form suitable for
solid-phase synthesis.
Scheme 2. Synthesis of adenine, cytosine and thymine (R)-Pmg
monomers; reagents: (a) Boc2O, 1  NaOH (aq.), dioxane; (b)
MsCl, Et3N, THF; (c) LiN3, DMF, 16 h, 64 °C; (d) Ph3P, H2O,
THF; (e) BrCH2COOAllyl, Et3N, THF; (f) 3, DIEA, HBTU,
DMF; (g) 1) 2  NaOH (aq.) dioxane, 2) KHSO4 (pH 2), 3) Et3N;
(h) 10, DIEA, HBTU, DMF; (i) 11, DIEA, HBTU, DMF (j)
Bu3SnH, [Pd(PPh3)2]Cl2, AcOH, CH2Cl2, 10 min
monomer 1b. Units 1c and 1d were obtained by Pd-cata-
lyzed deallylation[24] of compounds 12 and 13, respectively.
It had been reported that dipeptides of the type H-Pro-
Gly-OR were prone to cyclize under basic conditions.[25,26]
In order to investigate whether and to what extent this un-
desired side-reaction would occur during the synthesis of
Pmg-containing PNA fragments, the following experiment
was set up. Compound 14 was produced by acidolysis of
the fully protected thymine Boc-(S)-Pmg benzyl ester 4a
(Scheme 3). Such a system would be equivalent to a resin-
Scheme 1. Synthesis of thymine (S)-Pmg monomer; reagents: (a)
HBTU, DIEA, DMF, DMAP; (b) 1) 2  NaOH, dioxane, 2)
KHSO4 (pH 2.0), 3) Et3N
Unlike the (S) isomer, (R)-Pmg was not commercially bound thymine Pmg unit. This derivative was treated with
available, and so was prepared by the following route, start- a 0.4  DIEA solution in DMF (the concentration and the
ing from -prolinol (Scheme 2).
amount were in accordance with the solid-phase PNA syn-
The amino function of -prolinol was protected with the thesis coupling cycle described in Table 1). Compound 14
Boc group and the primary alcohol 5 was then mesylated. was fully transformed within 20 minutes into a new, less
Nucleophilic substitution of the mesylate 6 with lithium az- polar (by TLC) compound. Data obtained from 1H and 13
C
ide and subsequent reduction of azide 7 with triphenylphos- NMR spectroscopy were consistent with the ketopiperazine
phane in the presence of water afforded amine 8 in an over- 15. The fast cyclization of compound 14 in the DIEA solu-
all yield of 84% from -prolinol.[23]
tion implied that attachment of the first Pmg-based unit
Alkylation of amine 8 with allyl bromoacetate in the to the solid support through an ester bond was not to be
presence of triethylamine in THF gave the Boc-(R)-Pmg recommended. One way to circumvent the cyclization prob-
unit 9 in 66% isolated yield. Coupling of backbone 9 vari- lem was to anchor the first Pmg monomer to the resin
ously with thymine-1-acetic acid (3), N6-benzoyladenine-9- through an amide bond. Compound 16 was therefore pre-
acetic acid (10), or N4-benzoylcytosine-1-acetic acid (11) af- pared and exposed to the solution of DIEA in DMF as
forded the completely protected (R)-Pmg building blocks described above for compound 14.
4b, 12, and 13, respectively. Basic hydrolysis of ester 4b was
As expected, TLC and NMR analysis indicated no
performed as in the above synthesis of (S)-Pmg, yielding formation of compound 15 even after 20 hours.
2392
Eur. J. Org. Chem. 2002, 2391Ϫ2399

A Novel N-(Pyrrolidinyl-2-methyl)glycine-Based PNA
FULL PAPER
Scheme 3. Reagents: (a) 50% TFA/CH2Cl2; (b) 15 equiv. DIEA as
0.4  solution in DMF; (c) BzlNH2, HBTU, DIEA, DMF
Scheme 4. Coupling in solution; reagents: (a) HBTU, DIEA, DMF;
(b) PyBroP, DIEA, DMF
Table 1. Steps involved in the solid phase synthesis by the mixed protection group strategy
Step
Function
Solvents and reagents[a]
Time [min]
1.
2.
3.
4.
Fmoc deprotection of resin-bound glycine
Single coupling
MMT deprotection during chain elongation
Double coupling
22% piperidine in NMP (v/v)[b]
2 ϫ 3.5
17.5
3
MMT Aeg-PNA monomers[c], HBTU[c], DIEA[d]
1% TFA in dichloroethane (v/v)[e]
Boc Pmg-PNA monomer 1a or 1b[c], HBTU[c],
DIEA[d] (double coupling)
2 ϫ 17.5
5.
6.
Boc deprotection
Lysine termination
50% TFA in dichloromethane (v/v)[f]
2 ϫ 15
2 ϫ 17.5
Fmoc--Lys(Boc)-OH[c], HBTU[c], DIEA[d] (double coupling)
[a] Synthesis was performed on 41 mg support (1 µmol). [b] 3.5 mL. [c] 0.3  solution in NMP. [d] 0.4  solution in NMP. [e] 6 mL. [f] 10 mL.
This finding not only implied that linking of Pmg-based further progress of the reaction could be detected over
monomers to the resin through an amide bond would be 2.5 h. The minor product was possibly the symmetric an-
advantageous over attachment through an ester bond but hydride of compound 1a.
also that condensation of a chain-bound Pmg monomer
should not result in any significant loss of the above build- thymine (S)- or (R)-Pmg units in the middle of the strand
ing block as the result of cyclization. (either at the fifth, or at both the fifth and the sixth posi-
Initially, four PNA fragments containing one or two
Prior to the synthesis of the Pmg-containing PNA frag- tion, counting from the C-terminal) were synthesized on a
ments on solid support, experiments were performed in or- solid support by mixed protecting group strategies (Table 1
der to identify suitable coupling conditions for the intro- and Scheme 5). The solid support was preloaded with
duction of the amide bond between two Pmg units. Two Fmoc-protected glycine, incorporated into the PNA strands
coupling reagents Ϫ HBTU and bromotris(pyrrolidinyl)- to prevent ketopiperazine formation (as described above).
phosphonium hexafluorophosphate (PyBroP) Ϫ were se- The Aeg-PNA part was assembled with MMT/acyl-pro-
lected for condensation of 1a and 16 (Scheme 4). HBTU is tected monomers.[29] The PNA chain was elongated by
commonly used both in solution and solid-phase peptide HBTU-mediated coupling of the appropriate monomers,
and PNA synthesis, while PyBroP has been reported to followed by removal of the MMT group with 1% TFA in
give better yields for the coupling of hindered amino ac- dichloroethane. The thymine Pmg building blocks 1a or 1b
ids.[27,28]
In two parallel experiments, equal amounts of com- ecuted twice prior to the acidolysis of the Boc group.
pounds 1a and 16 in DMF were treated with 1.5 equivalents Thus, a 50% TFA solution in dichloromethane was
were incorporated under the same coupling conditions, ex-
of the coupling agent, and the formation of dimer 17 was manually applied to the disconnected synthesis vessel for
monitored by HPLC. In the case of HBTU, no further the removal of the Boc group from monomers 1a or 1b.
change in the composition of the reaction mixture could be
In addition, to prevent acyl shift[30] and to increase the
observed after 55 min One major product was formed, and solubility, all fragments were terminated with two consecut-
1
this was isolated (yield 93%) and characterized by H and ive lysines.[1,31]
13C NMR spectroscopy and by mass spectrometry. It was
identified as the desired dimer 17 (Scheme 4).
An essential point in the incorporation of the thymine
Pmg units was the replacement of the resin neutralization
On the other hand, analysis of the PyBroP-mediated step after Boc deprotection by an extensive wash. This
coupling after 55 min showed two new compounds in a 1:4 change allowed the coupling yields of Pmg monomers to be
ratio. The major of these was identified as the dimer 17. No improved from 71% to 99%. The average coupling yields of
Eur. J. Org. Chem. 2002, 2391Ϫ2399
2393

A. Slaitas, E. Yeheskiely
FULL PAPER
points, the melting temperatures of PNA-DNA and PNA-
RNA duplexes were used (Table 2, entries 3Ϫ4). PNA de-
camer IV, with a single thymine (S)-Pmg unit, gave Tm
values of 21 °C and 39 °C when hybridized to complement-
ary DNA and RNA, respectively (Table 2, entries 5, 6).
Fragment V, containing two consecutive thymine (S)-Pmg
units, showed no detectable transition with DNA and a Tm
of 27 °C with target RNA (entries 9, 10). The above data
for fragments IV and V imply a ∆Tm of 19 °C per single
(S)-Pmg modification, calculated from the Tm of unmodi-
fied PNA duplex with DNA, and ∆Tm values of 14 °C and
13 °C for fragments IV and V, respectively, calculated from
unmodified PNA duplex with RNA (see entries 5 vs. 3, 6
and 10 vs. 4).
PNA decamer VI, with a single thymine (R)-Pmg unit,
gave Tm values of 24 °C and 44 °C when hybridized to com-
plementary DNA and RNA, respectively (Table 2, entries
7, 8). Fragment VII, containing two consecutive thymine
(R)-Pmg units, gave Tm values of 5 °C and 36 °C with com-
plementary DNA and RNA, respectively (Table 2, entries
11, 12). The above data for fragments VI and VII imply a
decrease of 17 °C per (R)-Pmg modification with DNA and
9 °C with RNA, relative to Aeg-PNA III.
Scheme 5. Solid-phase synthesis of Pmg-PNA by a mixed pro-
tecting group strategy; reagents: (a) 22% piperidine/NMP (v/v); (b)
MMT-Aeg-PNA building block, DIEA, HBTU; (c) 1% TFA/
dichloroethane (v/v); (d) Boc-Pmg-PNA building block 1a or 1b,
DIEA, HBTU; (e) TFA/DCM 1:1 (v/v); (f) Fmoc--Lys(Boc)-OH,
DIEA, HBTU; (g) 95% aq. TFA; (h) NH3/MeOH saturated,
16 h, 55 °C
In comparison with the oligomers containing the (S) iso-
mer, the (R)-Pmg PNAs showed higher affinities for both
the Pmg building blocks were determined by comparison of DNA and RNA. The above Tm values also indicated that
the absorbance of the released MMT cation of the Aeg- Pmg-PNAs (i.e., fragments IVϪVII) bound complementary
PNA monomers coupled prior to and after the thymine RNA with higher selectivity than Aeg-PNA. This effect was
Pmg unit(s). The assembled PNA fragments (Table 2, more pronounced with the (R)-Pmg-containing fragments VI
IVϪVII) were detached from the solid support and the nu- and VII. For example, the ∆Tm values (RNA vs. DNA) were
cleobases deprotected by ammonolysis.[32] The oligomers 20 °C and 31 °C for decamers VI and VII, respectively, com-
were then purified by HPLC and analyzed by mass spectro- pared to 13 °C for unmodified PNA III. The Tm values of
metry.
decamer VII with DNA (entry 2, Table 2) and with comple-
The abilities of the Pmg-containing PNA fragments to mentary RNA were approximately the same (38 °C and 36
form duplexes with complementary DNA and RNA were °C, respectively). The ability to bind RNA, however, was
evaluated by UV thermal melting experiments. As reference substantially higher, the ∆Tm (RNA vs. DNA) values being 3
Table 2. UV thermal melting data; all PNA sequences were of the following structure (N Ǟ C): H-(Lys)2-(PNA)10-Gly-NH2; lowercase
letters in PNA sequences denote Aeg-PNA (a, c, t), bold uppercase denote Pmg-PNA (TX) where the superscript letter indicates the isomer
Entry
PNA #
Sequences
Tm (°C)
∆Tm(RNA vs. DNA)
(°C)
∆Tm/mod. (°C)
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
d(TCACTTCCAT):DNA
d(TCACTTCCAT):RNA
tcacttccat:DNA
35.0
38.0
40.0
53.0
21.0
39.0
24.0
44.0
Ϫ[c]
27.0
5.0
3.0
13.0
18.0
20.0
Ϫ
III
III
IV
tcacttccat:RNA
tcactTSccat:DNA
Ϫ19.0[a]
Ϫ14.0[b]
Ϫ16.0[a]
Ϫ9.0[b]
Ϫ[c]
IV
tcactTSccat:RNA
VI
VI
tcactTRccat:DNA
tcactTRccat:RNA
V
V
VII
VII
VIII
VIII
tcacTSTSccat:DNA
tcacTSTSccat:RNA
tcacTRTRccat:DNA
tcacTRTRccat:RNA
(TCACTTCCAT)R:DNA
(TCACTTCCAT)R:RNA
Ϫ13.0[b]
Ϫ17.5[a]
Ϫ8.5[b]
Ϫ[c]
36.0
Ϫ[c]
Ϫ[c]
31.0
Ϫ
Ϫ[c]
[a]
[b]
[c]
Compared to the Tm of PNA:DNA duplex. Compared to the Tm of PNA:RNA duplex. No sigmoidal transition was observed.
2394
Eur. J. Org. Chem. 2002, 2391Ϫ2399

A Novel N-(Pyrrolidinyl-2-methyl)glycine-Based PNA
FULL PAPER
°C and 31 °C for the RNA and compound VII, respectively.
All UV thermal melting experiments were performed at
least three times and the deviation of the measured Tm
values did not exceed 0.5 °C.
It had previously been reported that incorporation of
modifications (single or multiple) in the nucleic acid strand
could cause severe distortion of the duplex structure.[18]
This so-called chimeric effect can be avoided by examining
the hybridization of fully modified sequences.
Since (R)-Pmg-containing PNA fragments hybridized to
DNA and RNA better than those containing the (S)-Pmg,
a fully modified (R)-Pmg decamer VIII was assembled. In
order to be able to perform a fully automated synthesis of
fragment VIII, we followed the Fmoc strategy with building
blocks 18aϪc, prepared from the previously synthesized
Boc-monomers 1bϪd (See Scheme 6). Acidolysis of the Boc
groups from compounds 1bϪd and subsequent treatment
of the resulting intermediates with N-(9-fluorenylmethoxy-
carbonyloxy)succinimide (FmocOSu) and DIEA in THF
provided the desired building blocks 18aϪc in excellent
yields after workup and purification.
Figure 2. Model compounds for testing the stability of the N-
benzoyl group in 22% piperidine
Scheme 7. Synthesis of all-(R)-Pmg PNA decamer IX; reagents:
(a) 22% piperidine/NMP (v/v); (b) Fmoc-Pmg-PNA building block
18aϪc, DIEA, HBTU, 2ϫ17.5 min; (c) Fmoc--Lys(Boc)-OH,
DIEA, HBTU, 2ϫ17.5 min; (d) 95% aqueous TFA; (e) NH3/
MeOH saturated, 16 h, 55 °C
tecting group cleavage afforded crude all-(R)-Pmg PNA,
which was further purified by RP-HPLC and characterized
by mass spectrometry.
Scheme 6. Synthesis of Fmoc (R)-Pmg monomers; reagents: (a)
TFA/DCM/MeOH/H2O 16:16:1:1 (v/v/v/v); (b) FmocOSu, DIEA,
THF
Compound VIII was then subjected to the same UV ther-
mal melting experiments as described above for the oligo-
mers IIIϪVII. Unfortunately, no thermal melting transition
was observed either with complementary RNA or DNA.
This inability to form stable duplexes may be due to unsuit-
able positioning of the nucleobases in the fully modified
PNA strand VIII.
Next, since the exocyclic amino groups of the adenine
and cytosine nucleobases were protected with benzoyl
groups, the stability of this protecting group in 22% piperid-
ine solution in NMP was tested by TLC experiments on
model compounds 19 and 20 (Figure 2). No cleavage prod-
ucts were detected in either case, and the starting com-
pounds were intact after 3.5 h.
With the Fmoc (R)-Pmg monomers available and the nu-
cleobase protecting group compatibility tested, an auto-
mated solid-phase synthesis of the fully modified (R)-Pmg-
PNA decamer VIII was performed. Chain elongation and
Conclusion
Both thymine (S)- and (R)-Pmg building blocks were syn-
Fmoc deprotection were performed as described in Table 3 thesized in good yields. Four PNA decamers containing one
and Scheme 7. or two thymine (S)- or (R)-Pmg units were assembled and
All-Pmg-PNA fragment VIII was assembled starting their hybridization properties determined. The duplexes
from resin-linked glycine and terminated with two lysines. formed by (R)-Pmg-containing PNAs both with DNA and
Detachment from the solid support and nucleobase pro- with RNA were more stable than those formed by thymine
Table 3. Steps involved in solid-phase synthesis of PNA VIII
Step[a]
Function
Solvents and reagents
Time (min)
1.
2.
3.
Fmoc deprotection
Coupling
Lysine termination
22% piperidine in NMP (v/v)[b]
2 ϫ 3.5
2 ϫ 17.5
2 ϫ 17.5
Fmoc Pmg-PNA monomers 18aϪc[c], HBTU[c], DIEA[d], (double coupling)
Fmoc--Lys(Boc)-OH[c], HBTU[c], DIEA[d], (double coupling)
[a]
[b]
[c]
[d]
Synthesis was performed on 41 mg support (1 µmol). 3.5 mL. 0.3  solution in NMP. 0.4  solution in NMP.
Eur. J. Org. Chem. 2002, 2391Ϫ2399
2395

A. Slaitas, E. Yeheskiely
FULL PAPER
(5.1 mL), followed by the addition of HBTU (0.90 g, 2.36 mmol).
Shortly after the addition of HBTU, a precipitate formed. DMAP
(0.29 g, 2.36 mmol) was added, and the reaction mixture was al-
lowed to react until a clear solution was obtained (48 h room tem-
perature), at which point TLC analysis indicated complete conver-
sion of starting materials. The solvent was removed under reduced
pressure, and the residue was dissolved in ethyl acetate and washed
with saturated NaHCO3, water, and acetate buffer (pH 4.0). The
organic layer was then dried with anhydrous Na2SO4, filtered, and
concentrated under reduced pressure. Crude product 4a was puri-
(S)-Pmg-containing PNAs, but less stable than those ob-
tained with unmodified Aeg-PNA. Both the (R)- and the
(S)-Pmg-containing PNAs formed tighter duplexes with
complementary RNA than with counterpart DNA. In addi-
tion, both the (S)- and the (R)-Pmg PNA exhibited sub-
stantially stronger preferences for RNA, compared to all-
Aeg-PNA and native DNA. This enhanced selectivity, espe-
cially of the (R)-Pmg-containing fragments, towards RNA
could be utilized in several possible ways. For example, a
more stable RNA-preferring nucleic acid analogue might be fied by flash chromatography (FC) with ethyl acetate as eluent.
Fractions containing the product were pooled and concentrated un-
der reduced pressure. Ester 4a was obtained as a white foam, in a
yield of 81.0% (0.98 g). The benzyl ester 4a (0.90 g, 1.75 mmol) was
next dissolved in dioxane (5 mL) and water (1 mL) and hydrolyzed
with 3 mL of 2  aqueous NaOH. After 10 min the reaction mix-
ture was neutralized to pH 7 (aq. KHSO4), and the dioxane was
removed under reduced pressure. The residue was diluted with ethyl
acetate (25 mL), acidified to pH 2 (aq. KHSO4), and extracted.
The organic phase was washed with water, dried over Na2SO4, and
concentrated. The obtained intermediate carboxylic acid was fur-
ther converted into its triethylammonium salt by redissolving it in
ethyl acetate and adding triethylamine, followed by FC purification
(gradient 5 to 10% MeOH/CH2Cl2 ϩ 1% Et3N). Fractions con-
taining the product were pooled, concentrated to dryness, co-evap-
orated with 5 mL CH2Cl2, and briefly dried in vacuo. The residue
was redissolved in CH2Cl2, triturated from dry diethyl ether, and
dried in vacuo. Compound 1a was obtained as a white powder
created by increasing the Aeg-PNA population around (R)-
Pmg based units. This analogue might be used to label or
identify specific RNA in cell extracts, or gels and might
have other applications in studies in which discrimination
between RNA over DNA was desired.
Experimental Section
General Remarks: Solvents were purchased from Merck Eurolab.
Dry solvents were obtained by use of appropriate molecular sieves,
except for THF. The water content in dry solvents did not exceed
50 ppm as determined by KarlϪFisher titration. THF was distilled
from sodium benzophenone ketyl. -Prolinol was purchased from
Aldrich. HBTU and PyBroP were purchased from Novabiochem.
TLC was performed on analytical Merck silica plates with F254
indicator. TLCs were viewed either under 254 nm UV or by stain-
ing with o-toluidine reagent or ninhydrin where appropriate. Solid-
phase syntheses were performed on Pharmacia Gene Assembler
Special with custom-written cycles. NMP for solid-phase synthesis
1
(0.77 g, 82.1%). H NMR (CDCl3, selected signals): δ ϭ 9.49 (br.
s, 1 H), 7.08 ϩ 7.06 (s, 1 H), 4.85Ϫ4.25 (m, 3 H), 4.10Ϫ3.90 (m, 3
H), 3.90Ϫ3.70 (m, 1 H), 3.40Ϫ3.13 (m, 3 H), 3.07 (q, J ϭ 7.3 Hz,
6 H), 1.85 (s, 3 H), 1.80Ϫ1.62 (br. s, 2 H), 1.44 ϩ 1.42 (s, 9 H),
1.26 (t, 9 H, J ϭ 7.3 Hz) ppm. 13C NMR: δ ϭ 173.5, 168.5, 165.0,
155.2 ϩ 154.7, 151.7, 142.1, 110.6, 80.0 ϩ 79.6, 56.0 ϩ 55.7, 51.4,
49.9, 48.3, 47.0 ϩ 46.6, 45.8, 29.4, 23.9 ϩ 23.0, 12.8, 8.9 ppm. MS:
m/z ϭ 447.25 [M Ϫ Et3N ϩ Na]ϩ. Accurate mass calculated for
C19H28N4NaO7 m/z ϭ 447.1856, observed 447.1873.
˚
was purchased from Applied Biosystems and dried over 4 A mo-
lecular sieves. The solid phase was a low-loading (24.4 µmol/g),
highly cross-linked polystyrene, loaded with Fmoc-glycine through
a 4-(hydroxymethyl)-benzoic acid linker. Reversed-phase HPLC
was performed on a Jasco HPLC system equipped with a Hypersil
ODS 5 µm, 150 ϫ 4.6 mm (Column 1) or a Phenomenex Jupiter 5
µm, 250 ϫ 4.60 (Column 2) or 5 µm, 250 ϫ 10.0 mm (Column 3).
Buffer A was 0.1% TFA in water (pH 2.0), buffer B 90% aqueous
acetonitrile with 0.1% TFA. The following gradients were used.
Gradient A: 0Ϫ40% buffer B in 40 min; gradient B: 0Ϫ25% buffer
B in 40 min 25Ϫ40% B in 20 min; gradient C: 0Ϫ50% buffer B in
50 min All runs were performed at 60 °C. Fractions containing the
PNA were lyophilized. Pure Pmg PNAs were stored as frozen (Ϫ
78 °C) solutions in deionized water. Complementary DNA was
purchased from TAG Copenhagen A/S (Denmark), RNA from
Dharmacon Research Inc. (USA). DNA and RNA sequences were
purified by both cation exchange and reversed-phase HPLC,
yielding chromatographically pure materials. Nominal mass spectra
were recorded on Lasermat 2000 (Finnigan MAT) MALDI TOF
(as a service at the Protein Analysis Centrum, Karolinska Institute)
and/or Micromass LCT mass spectrometer equipped with an ES
probe. Accurate masses were obtained on Micromass LCT mass
spectrometer in positive ion mode with leucine enkephalin as an
internal lock mass standard. NMR spectra were recorded on a
Bruker Avance DRX 400 spectrometer at 400.13 and 100.61 MHz
for 1H and 13C, respectively. Thermal melting experiments were
performed on a Varian Cary 300 spectrophotometer with pro-
grammable temperature block.
(2R)-N-Boc-Pyrrolidinylmethanol
(5):
-Prolinol
(1.05 g,
10.4 mmol) was dissolved in a mixture of dioxane, water, and 2 
aqueous NaOH (20.8 mL, 15.6 mL, and 5.2 mL, respectively). The
solution was cooled in an ice/water bath for 15 min, followed by
addition of di-tert-butyl dicarbonate (Boc2O, 2.5 g, 11.4 mmol).
After the reagent had dissolved, the ice bath was removed and the
reaction was allowed to proceed at ambient temperature until the
starting material was consumed (1 h). The organic solvent was re-
moved under reduced pressure, and ethyl acetate (50 mL) was ad-
ded to the crude compound 5 in water. The pH was then adjusted
to 7 and the title compound was extracted. The organic phase was
washed with water (1 ϫ 50 mL), dried with Na2SO4, and concen-
trated under reduced pressure, yielding 2.08 g (99.4%) of the com-
pound 5 as a colorless oil, which solidified in vacuo overnight. The
protected prolinol 5 was chromatographically pure (TLC, NMR)
1
and was used for the next step without additional purification. H
NMR (CDCl3): δ ϭ 4.78 (br. s, 1 H), 3.95 (m, 1 H), 3.60 (m, 1 H),
3.42 (m, 1 H), 3.31 (m, 1 H), 1.99 (m, 1 H), 1.80 (m, 2 H), 1.55
(m, 1 H), 1.46 (s, 9 H) ppm. 13C NMR: δ ϭ 157.5, 80.6, 68.0, 65.3,
60.6, 47.9 ϩ 47.2, 28.9, 24.5 ppm.
(2R)-2-Azidomethyl-N-Boc-pyrrolidine (7): Triethylamine (4.25 mL)
was added to a solution of alcohol 5 (2.04 g, 10.1 mmol) in dry
THF (103 mL), and the reaction flask was cooled in an ice/water
N-Boc-(S)-Pmg Thymine Monomer 1a: DIEA (1.23 mL,
7.07 mmol) and thymine-1-acetic acid (3) (0.44 g, 2.36 mmol) were bath for 15 min Mesyl chloride (1.56 mL, 20.2 mmol) was added in
added to a solution of amine 2 (0.55 g, 1.57 mmol) in dry DMF
a single portion, resulting in instantaneous formation of a white
2396
Eur. J. Org. Chem. 2002, 2391Ϫ2399

A Novel N-(Pyrrolidinyl-2-methyl)glycine-Based PNA
FULL PAPER
precipitate. After 1 h, the mixture was diluted with dichlorome-
thane (120 mL), washed with saturated aq. NaHCO3 (100 mL),
dried over MgSO4, and concentrated under reduced pressure. The
intermediate mesylate 6 was obtained as a colorless oil, and was
used without further purification. Thus, mesylate 6 was dissolved
in dry DMF (100 mL), and lithium azide (3.0 g, 60.6 mmol) was
added. The reaction mixture was stirred and kept at 64 °C for 16 h,
and then diluted with diethyl ether (250 mL), washed with water
(1ϫ500 mL, 3ϫ250 mL) and saturated NaCl (50 mL), dried with
MgSO4, and concentrated under reduced pressure. Crude azide 7
was purified by silica FC, with isocratic eluent Ϫ n-hexane/ethyl
acetate 3:2. Fractions containing the product were pooled and con-
centrated under reduced pressure, yielding azide 7 (1.93 g, 84.4%)
ϩ 45.4, 28.5, 28.8, 23.5 ϩ 22.6, 12.3, 8.5 ppm. MS: m/z ϭ 526.25
[M ϩ H]ϩ, 424.14 [M Ϫ Et3N ϩ H]ϩ. Accurate mass calculated
for C19H28N4O7Na m/z ϭ 447.1856, observed 447.1851.
N-Boc-(R)-Pmg Adenine Monomer 1c: Compound 12 was prepared
as described above for compound 4a, starting from backbone 9
with the slight modifications that DMAP was not used and that
HBTU and N6-benzoyladenine-9 acetic acid were 1.1 equivalents
relative to backbone 9. Ester 12 was obtained as a white foam
(0.272 g, 56.4%). Pd-assisted deallylation proceeded as follows: es-
ter 12 (0.257 g, 0.465 mmol) in dry dichloromethane (3.5 mL) was
treated with Bu3SnH (185 µL, 0.7 mmol) and [Pd(PPh3)2]Cl2
(10 mg, 0.014 mmol) in the presence of acetic acid (40 µL,
0.70 mmol). The solution was stirred for 10 min, followed by addi-
tion of Et3N (174 µL, 1.25 mmol). The reaction solvents were evap-
orated to near dryness, and the residue was redissolved in dichloro-
methane and precipitated from light petroleum ether. Crude prod-
uct 1c was purified by FC with a gradient of methanol in dichloro-
methane (0 to 10% MeOH) containing 1% Et3N. Solvent removal
under reduced pressure and trituration from dry diethyl ether
1
as colorless liquid. H NMR (CDCl3): δ ϭ 3.94 (m, 1 H), 3.58 ϩ
3.37 (m, 4 H), 1.97 ϩ 1.80 (m, 4 H), 1.47 (s, 9 H). 13C δ ϭ 154.9,
80.0, 56.9, 54.1 ϩ 53.1 47.5 ϩ 47.0, 28.9, 29.8 ϩ 28.1, 24.3 ϩ
23.4 ppm.
(2R)-2-Aminomethyl-N-Boc-pyrrolidine (8): Azide
7
(1.91 g,
8.44 mmol) in anhydrous THF (70 mL) was reduced with triphenyl-
phosphane (4.43 g, 16.9 mmol) and water (0.312 mL, 17.3 mmol). yielded pure N-Boc-(R)-Pmg adenine 1c as a white powder (0.253 g,
1
The reaction mixture was heated to reflux until all the starting mat-
erial had been consumed. The organic solvent was removed under
reduced pressure. Diethyl ether (180 mL) was added to the re-
maining oil, and the pH of the aqueous phase was lowered to 1.75
with 2  HCl with vigorous stirring. The aqueous layer was separ-
ated and washed with diethyl ether (2 ϫ 90 mL), and the pH was
then adjusted to 13.0 (2  NaOH). After extraction with dichloro-
methane (6ϫ100 mL), the combined organic phases were dried
over Na2SO4 and concentrated under reduced pressure. Amine 8
was obtained in quantitative yield as a colorless liquid (1.71 g) and
85.2%). H NMR (CDCl3): δ ϭ 8.70 (d, J ϭ 13.0 Hz, 1 H), 8.21
(d, J ϭ 11.7 Hz, 1 H), 8.05 (dd, J ϭ 1.4, 7.2 Hz, 2 H), 7.56 (m, 1
H), 7.48 (m, 2 H), 3.04 (q, J ϭ 7.4 Hz, 6 H), 1.44 mi ϩ 1.40 ma
(s, 9 H), 1.25 (t, 9 H, J ϭ 7.4 Hz) ppm. 13C NMR (selected signals):
δ ϭ 173.4, 167.6, 165.4, 154.7, 152.7, 149.8, 145.3, 134.2, 133.1,
129.2, 128.4, 122.6, 79.6, 66.3, 55.7, 52.0, 50.2, 47.0, 45.6, 44.6,
34.6, 30.7, 30.1, 27.5, 23.9, 15.7, 9.0 ppm. Accurate mass calculated
for C26H32N7O6 m/z ϭ 538.2414 and C26H32N7NaO6 m/z ϭ
560.2234, observed 538.2412 and 560.2221.
N-Boc-(R)-Pmg Cytosine Monomer 1d: Compound 1d was pre-
pared as described above for compound 1c, with backbone 9
(0.208 g, 0.697 mmol) and N4-benzoylcytosine-1-acetic acid
(0.220 g, 0.805 mmol). The yield of ester 13 was 0.147 g (36.2%).
After the removal of the allyl ester, compound 1d was obtained as
a white power in a yield of 93.8% (0.135 g). 1H NMR (CDCl3,
selected signals): δ ϭ 7.96 (d, J ϭ 7.4 Hz, 2 H), 7.75 (m, 1 H), 7.59
1
used without additional purification. H NMR (MeOD): δ ϭ 3.78
(m, 1 H), 3.33 (m, 2 H), 2.83 (dd, J ϭ 12.7, 4.4 Hz, 1 H), 2.59 (m,
1 H), 1.87 (m, 4 H), 1.49 (s, 9 H) ppm. 13C NMR: δ ϭ 157.2 ϩ
157.0, 81.4 ϩ 81.1, 61.1, 48.5 ϩ 48.1, 45.7, 29.6, 29.2 ϩ 29.0, 24.9
ϩ 24.2 ppm.
Allyl [N-(2R)-Boc-Pyrrolidin-2-yl)methyl]glycinate (9): Amine 8
(1.69 g, 8.44 mmol) was dissolved in a mixture of dry THF (45 mL) (m, 2 H), 7.50 (m, 2 H), 3.09 (m, 2 H), 1.44 (s, 9 H) ppm. 13C
and triethylamine (1.18 mL, 8.44 mmol) and alkylated with allyl
bromoacetate (1.51 g, 8.44 mmol) at ambient temperature. The re-
action was allowed to continue overnight. The formed triethylam-
NMR: δ ϭ 173.1, 168.4, 167.3, 163.4, 156.2, 155.2, 151.4, 133.4,
129.2, 128.4, 128.3, 97.3, 79.8, 55.8, 53.0, 51.1, 47.1, 45.9, 29.3,
28.9, 23.9, 23.0, 8.9, 7.9 ppm. Accurate mass calculated for
monium salt was removed by filtration, and the organic solvent C25H32N5O7 m/z ϭ 514.2302 and C25H31N5NaO7 m/z ϭ 536.2121,
was evaporated. The residue was then dissolved in ethyl acetate,
washed with sat. NaHCO3 and water, dried over Na2SO4, and con-
centrated. Pure compound 9 was obtained by silica FC with iso-
cratic eluent Ϫ dichloromethane/ethyl acetate 1:1 ϩ 0.2% Et3N (v/
v/v). The yield of compound 9 (colorless oil) was 65.8% (1.65 g).
1H NMR (CDCl3): δ ϭ 5.91 (m, 1 H), 5.28 (dd, J ϭ 10.3, 17.2 Hz,
2 H), 4.63 (d, J ϭ 5.7 Hz, 2 H), 3.91 ϩ 3.80 (m, 1 H), 3.46 ϩ 3.32
(m, 4 H), 2.79 ϩ 2.60 (m, 2 H), 1.94 ϩ 1.80 (m, 4 H), 1.47 (br. s,
9 H) ppm. 13C NMR: δ ϭ 172.5, 155.3 ϩ 155.0, 132.3, 119.0, 79.7,
65.7, 57.4, 53.0, 51.4, 47.2 ϩ 46.8, 30.0 ϩ 29.4, 28.9, 24.2 ϩ 23.4
ppm. Accurate mass calculated for C15H27N2O4 m/z ϭ 299.1971,
observed 299.1979.
observed 514.2291 and 536.2124.
N-Fmoc-(R)-Pmg Cytosine Monomer 18c: Compound 1d (0.535 g,
0.87 mmol) was dissolved in CH2Cl2 (8 mL), MeOH (0.5 mL), and
H2O (0.5 mL). TFA (8 mL) was added. Once TLC analysis indic-
ated complete acidolysis of the Boc group, the mixture was diluted
with toluene (20 mL) and concentrated under reduced pressure.
The resulting oily residue was co-evaporated with toluene
(2ϫ20 mL) and then with dry THF (2ϫ10 mL). The remaining yel-
low oil was dissolved in dry THF (7 mL), and DIEA (0.662 mL,
1.92 mmol) was added, resulting in the formation of a dense, white
precipitate. Subsequently, dry DMF (1 mL) and FmocOSu
(0.324 g, 0.96 mmol) were added. The reaction vessel was sealed
and placed in an ultrasonic bath until a clear solution resulted.
After 16 hours, TLC analysis indicated complete conversion of the
starting material into the product. The solvent was removed under
reduced pressure, and the residue was dissolved in ethyl acetate
(30 mL), washed with aq. KHSO4 (1ϫ10 mL, pH 1.46) and water
N-Boc-(R)-Pmg Thymine Monomer 1b: Compound 1b was pre-
pared from backbone 9 (0.211 g, 0.706 mmol) as described above
for monomer 1a. Ester 4b (0.330 g, 80.5%) was obtained as a white
foam. Ester 4b (0.300 g, 0.646 mmol) was then subjected to base
hydrolysis as described above for compound 1a, yielding 1b as a
white powder (0.286 g, 84.2%). 1H NMR (CDCl3): δ ϭ 9.64 ϩ 9.45 (3ϫ10 mL), dried over Na2SO4, and concentrated to yield a white
(s, 1 H), 7.10 (s, 1 H), 3.83 (m, 1 H), 3.10 (m, 3 H), 1.45 (s, 9 H) foam. The product was then transformed into a fine powder by
ppm. 13C NMR: δ ϭ 172.7 ϩ 172.3, 168.1, 164.6, 154.9, 151.3,
141.7, 110.3, 79.7 ϩ 79.3, 55.6 ϩ 55.3, 51.0 ϩ 50.0, 47.9, 46.7, 46.3 0.493 g (89.2%). H NMR (CDCl3, selected signals): δ ϭ 7.95 (d,
redissolving it in CH2Cl2 and triturating from n-hexane. Yield
1
Eur. J. Org. Chem. 2002, 2391Ϫ2399 2397

A. Slaitas, E. Yeheskiely
FULL PAPER
J ϭ 7.2 Hz, 2 H), 7.86Ϫ7.73 (m, 3 H), 7.65Ϫ7.57 (m, 3 H), m/z ϭ 660.66 [M ϩ H]ϩ, 682.70 [M ϩ Na]ϩ, 698.64 [M ϩ K]ϩ.
7.57Ϫ7.46 (m, 3 H), 7.42Ϫ7.34 (m, 2 H), 7.33Ϫ7.25 (m, 2 H), Accurate mass calculated for C36H34N7O6 m/z ϭ 660.2571, ob-
7.02Ϫ4.80 (m, 3 H), 4.77Ϫ4.52 (m, 2 H), 4.50Ϫ4.28 (m, 2 H),
served 660.2568. A Fmoc-deprotected sample for NMR was pre-
4.11Ϫ3.74 (m, 1 H), 3.74Ϫ3.56 (m, 1 H), 2.10Ϫ1.55 (m, 4 H) ppm. pared as described above for compound 18c. 1H NMR (D2O): δ ϭ
13C NMR: δ ϭ 171.3, 168.0, 164.1, 157.5, 156.1 ϩ 155.9, 151.2,
144.3, 141.7, 133.6, 133.1, 128.8, 128.2, 127.9, 127.5, 127.2, 125.1,
124.8, 120.0, 97.4, 78.5, 78.3, 67.5, 66.4, 57.0 ϩ 55.9, 50.6, 28.7 ϩ
8.79 (s, 1 H), 8.58 (s, 1 H), 8.00 (d, J ϭ 7.2 Hz, 2 H), 7.70 (t, J ϭ
7.5 Hz, 1 H), 7.58 (t, J ϭ 7.7 Hz, 2 H), 5.40 (s, 2 H), 4.45 (AB, 2
H, J ϭ 19.2), 3.98 (dd, J ϭ 9.3, 14.8 Hz, 1 H), 3.87 (ddd, J ϭ 12.5,
28.3, 23.6 ϩ 22.4 ppm. MS: m/z ϭ 636.61 [M ϩ H]ϩ, 658.61 [M 3.2, 9.7 Hz, 1 H), 3.57 (dd, J ϭ 3.5, 14.9 Hz, 1 H), 3.39Ϫ3.21 (m,
ϩ
Na]ϩ, 674.59 [M
ϩ
K]ϩ. Accurate mass calculated for
2 H), 2.24Ϫ2.12 (m, 1 H), 2.11Ϫ1.94 (m, 2 H), 1.72 (ddt, J ϭ 13.3,
C35H34N5O7 m/z ϭ 636.2458, observed 636.2477. C35H33N5O7 8.9, 8.9 Hz, 1 H) ppm. 13C NMR: δ ϭ 173.3, 169.9, 150.1, 147.9,
(635.6826): calcd. C 66.13, H 5.23, N 11.02; found C 65.91, H 5.32, 147.1, 134.3, 132.2, 129.3, 128.7, 121.2, 59.5, 50.9, 49.9, 45.8, 45.7,
1
N 10.94. The H spectrum of the above compound 18c was com-
plex and poorly resolved. In order to obtain a better resolved spec-
trum, the Fmoc group of compound 18c (11 mg) was removed by
β-elimination with 22% piperidine solution in dry DMF (1 mL).
After 7 minutes the pH of the solution was reduced to 2 (aqueous
TFA). The deprotected compound was then purified by RP-HPLC
(Column 3, gradient: 0Ϫ75% buffer B in 60 min at 30 °C). The
appropriate fractions were pooled and lyophilized to yield the pure
27.6, 22.5 ppm. Accurate mass of deprotected monomer calculated
for C21H24N7O4 m/z ϭ 438.1890, observed 438.1875.
Reference Aeg-PNA H-Lys-Lys-tcacttccat-Gly-NH2 (III): Solid-
phase synthesis was performed by a mixed Fmoc/MMT protecting
group strategy on 41.0 mg (1 µmol scale) of nonswelling, highly
cross-linked aminomethyl-polystyrene, preloaded with Fmoc-gly-
cine attached through a 4-(hydroxymethyl)benzoic acid (HMBA)
linker (loading: 24.4 µmol/g, as determined spectrophotometrically
by the amount of released dibenzofulvene). The assembly of the
PNA was performed after cleavage of the Fmoc group from resin-
bound glycine with 22% piperidine in NMP. The procedure for one
elongation/deprotection cycle consisted of (1) wash: NMP (5 mL);
(2) coupling: MMT-Aeg-PNA monomer (15 equiv.), HBTU (16
equiv.), DIEA (32 equiv.) in NMP (550 µL), 17.5 min; (3) wash:
NMP (5 mL); (4) MMT deprotection: 1% TFA/dichloroethane
(6 mL), 3 min; (5) wash: dichloroethane (2.5 mL) 1 min The MMT-
Aeg-PNA monomers and HBTU were delivered as 0.3  solutions
in NMP. DIEA was delivered as a 0.4  solution. The PNA de-
camer was terminated with two consecutive -lysines by use of
commercially available Fmoc--Lys(Boc)-OH. The procedure for
the lysine incorporation/deprotection cycle consisted of (1) wash:
NMP (5 mL); (2) double coupling: lysine monomer (15 equiv.),
1
compound as a hygroscopic trifluoroacetate salt. H NMR (D2O):
δ ϭ 7.96 mi ϩ 7.90 ma (d, J ϭ 7.2 Hz, 1 H), 7.81Ϫ7.79 (m, 2 H),
7.61Ϫ7.57 (m, 1 H), 7.48Ϫ7.44 (m, 2 H), 7.38Ϫ7.35 (m, 1 H), 4.73
(AB, 2 H, J ϭ 16.6), 4.26 (AB, 2 H, J ϭ 19.2), 3.87 (dd, J ϭ 9.6,
14.9 Hz, 1 H), 3.77 (ddd, J ϭ 2.9, 9.5, 12.4 Hz, 1 H), 3.46 (dd, J ϭ
3.1, 14.9 Hz, 1 H), 3.40Ϫ3.22 (m, 1 H), 3.22Ϫ3.13 (m, 1 H),
2.14Ϫ2.04 (m, 1 H), 2.03Ϫ1.83 (m, 2 H), 1.62 (ddt, J ϭ 13.2, 8.9,
8.8 Hz, 1 H) ppm. 13C NMR: δ ϭ 173.3, 170.6, 170.2, 164.1, 157.6,
151.9, 134.0, 132.9, 129.3, 128.4, 98.9, 59.6, 51.9, 51.0, 50.0, 46.0,
45.7, 27.6, 22.6 ppm. Accurate mass of the deprotected monomer
calculated for C20H24N5O5 m/z ϭ 414.1777, observed 414.1776.
N-Fmoc-(R)-Pmg Thymine Monomer 18a: Compound 18a was pre-
pared as described for 18c, starting from 1b (0.752 g, 1.43 mmol).
Yield 0.692 g (88.5%) as a white powder. H NMR (CDCl3): δ ϭ
1
10.15 ϩ 10.10 (br. s, 1 H), 7.85Ϫ7.70 (m, 2 H), 7.70Ϫ7.50 (m, 2 HBTU (16 equiv.), DIEA (32 equiv.) in NMP (550 µL), 17.5 min
H), 7.50Ϫ7.35 (m, 2 H), 7.35Ϫ7.20 (m, 2 H), 7.05Ϫ7.00 (s, 1 H), NMP wash (2.5 mL), lysine monomer (15 equiv.), HBTU (16
4.85Ϫ4.55 (m, 1 H), 4.50Ϫ4.35 (m, 2 H), 4.35Ϫ4.15 (m, 2 H), equiv.), DIEA (32 equiv.) in NMP (550 µL), 17.5 min; (3) wash:
4.15Ϫ3.95 (m, 1 H), 3.85Ϫ3.60 (m, 1 H), 3.55Ϫ3.10 (m, 3 H), 1.83 NMP (5 mL); (4) Fmoc deprotection: 22% piperidine/NMP
ϩ 1.82 (s, 3 H) ppm. 13C NMR: δ ϭ 171.8, 168.6, 165.6, 156.0,
151.8, 144.3 ϩ 144.2, 142.3, 141.7, 128.2 ϩ 127.5, 125.4 ϩ 125.0,
(3.5 mL), 3.5 min NMP (2.5 mL), 22% piperidine/NMP (3.5 mL),
3.5 min Prior to the cleavage from the solid support, the Nε-Boc
120.4, 111.0, 73.3, 67.9 ϩ 66.6, 57.3, 56.2, 51.5, 50.0 ϩ 49.7, 49.0, groups of the lysines were deprotected with 95% aqueous TFA,
48.3, 47.6, 46.8, 29.3 ϩ 28.8, 25.9, 24.1 ϩ 23.9, 22.9, 12.6 ppm.
followed by neutralization with 10% DIEA/CH2Cl2, washing with
MS: m/z ϭ 547.57 [M ϩ H]ϩ. Accurate mass calculated for additional CH2Cl2 and diethyl ether, and then drying in vacuo. Re-
C29H30N4O7Na m/z ϭ 569.2012, observed 569.2008. C29H30N4O7 lease of the PNA fragment from the support was achieved by am-
(546.5851): calcd. C 63.73, H 5.53, N 10.25; found C 63.55, H 5.67,
N 9.99. A Fmoc-deprotected sample for NMR was prepared as
described above for compound 18c. H NMR (D2O): δ ϭ 7.29 ϩ
monolysis of the HMBA linker with saturated anhydrous ammo-
nia/methanol at 55 °C/16 hours. Crude PNA III was purified by
reversed-phase HPLC. The appropriate fractions were pooled and
1
7.24 (s, 1 H), 4.56 (AB, 2 H, J ϭ 16.9), 4.26 (AB, 2 H, J ϭ 20.2), lyophilized. MS (MALDI-TOF): calculated for C120H165N56O35
3.87 (dd, J ϭ 15.0, 9.2 Hz, 1 H), 3.75Ϫ3.73 (m, 1 H), 3.35Ϫ3.22
(m, 1 H), 3.42 (dd, J ϭ 15.0, 3.1 Hz, 1 H), 2.15Ϫ2.05 (m, 1 H),
2.02Ϫ1.82 (m, 2 H), 1.76 (s, 3 H), 1.61 (ddt, J ϭ 13.1, 8.7, 8.8 Hz,
1 H) ppm. 13C NMR: δ ϭ 143.4, 55.7, 51.6, 50.3, 48.9, 47.1, 28.6,
23.8, 11.7 ppm. Accurate mass of deprotected monomer calculated
for C14H21N4O5 m/z ϭ 325.1512, observed 325.1523.
m/z ϭ 2950.3, observed 2951.4 [M ϩ H]ϩ, 2973.3 [M ϩ Na]ϩ.
PNA-Containing one Thymine (S)-Pmg Unit (IV): The solid-phase
synthesis of PNA IV was performed by a mixed Fmoc/MMT/Boc
strategy. The Aeg part of decamer IV was assembled as described
above for PNA III. Attachment of thymine (S)-Pmg monomer 1a
was executed as follows: the Pmg unit was double-coupled (i.e.,
after the deprotection of MMT group from the last Aeg-PNA unit,
building block 1a was attached by the same procedure as described
for the elongation of the Aeg-part). In this case, however, after an
N-Fmoc-(R)-Pmg Adenine Monomer 18b: Compound 18b was pre-
pared as described for 18c, starting from 1c (0.752 g, 1.43 mmol).
Yield 95.0% as a white powder. 1H NMR (MeOD, selected signals):
δ ϭ 7.46 (d, J ϭ 8.0 Hz, 2 H), 7.23Ϫ7.08 (m, 2 H), 7.05Ϫ6.86 (m, additional washing step, the coupling cycle was repeated. Next, the
5 H), 6.82Ϫ6.58 (m, 4 H), 3.70Ϫ3.50 (m, 2 H), 3.26Ϫ3.00 (m, 1
synthesis vessel was disconnected from the synthesizer and the re-
H), 2.85Ϫ2.75 (m, 1 H), 2.65Ϫ2.50 (m, 1 H) ppm. 13C NMR: δ ϭ moval of the Boc group was accomplished manually by application
171.3, 168.4, 167.0, 152.1, 149.7, 145.7, 144.2, 141.6, 134.0, 132.9, of a TFA solution (TFA/DCM, 1:1 v/v, 2 ϫ 5 mL, 2 ϫ 15 min)
128.8, 128.4, 127.9, 127.5, 127.2, 125.1, 122.8, 120.0, 78.5, 78.3,
into and through the vessel by syringe. The synthesis vessel was
67.4 ϩ 66.5, 57.0 ϩ 56.0, 51.2, 44.3, 28.8 ϩ 28.3, 23.4 ppm. MS: then reconnected to the synthesizer, on which the support-bound
2398
Eur. J. Org. Chem. 2002, 2391Ϫ2399

A Novel N-(Pyrrolidinyl-2-methyl)glycine-Based PNA
FULL PAPER
N. J. Peffer, J. C. Hanvey, J. E. Bisi, S. A. Thomson, C. F.
Hassman, S. A. Noble, L. E. Babiss, Proc. Natl. Acad. Sci. U.
S. A. 1993, 90, 10648Ϫ10652.
K. K. Jensen, H. Orum, P. E. Nielsen, B. Norden, Biochemistry
1997, 36, 5072Ϫ5077.
A. Lomakin, M. D. Frank-Kamenetskii, J. Mol. Biol. 1998,
276, 57Ϫ70.
E. Uhlmann, A. Peyman, G. Breipohl, D. W. Will, Angew.
Chem. Int. Ed. 1998, 37, 2796Ϫ2823.
A. C. Laan, Van der, I. Amsterdam, Van, G. I. Tesser, J. H.
Boom, Van, E. Kuyl-Yeheskiely, Nucleosides Nucleotides 1998,
17, 219Ϫ231.
S. Sforza, G. Haaima, R. Marchelli, P. E. Nielsen, Eur. J. Org.
Chem. 1999, 1, 197Ϫ204.
L. Zhang, J. Min, L. Zhang, Bioorg. Med. Chem. Lett. 1999,
9, 2903Ϫ2908.
M. Hoffmann, A. M. Brückner, T. Hupp, B. Engels, U. Dieder-
ichsen, Helv. Chim. Acta 2000, 83, 2580Ϫ2593.
S. Sforza, R. Corradini, S. Ghirardi, A. Dossena, R. Marchelli,
Eur. J. Org. Chem. 2000, 16, 2905Ϫ2913.
A. A. Koshkin, S. K. Singh, P. Nielsen, V. K. Rajwanshi, R.
Kumar, M. Meldgaard, C. E. Olsen, J. Wengel, Tetrahedron
1998, 54, 3607Ϫ3630.
J. Wengel, Acc. Chem. Res. 1999, 32, 301Ϫ310.
J. C. Verheijen, A-M.M. M. Roon, A. C. Laan, Van der, G. A.
Marel, Van der, J. H. Boom, Van, Nucleosides Nucleotides 1999,
18, 493Ϫ508.
[2]
oligomer was additionally washed with dichloroethane (4 mL), ace-
tonitrile (2 mL), and NMP (2 mL). The automated synthesis of the
remaining Aeg-PNA was then resumed and performed as described
above. Comparison of the absorbance of the released MMT cation
from the fourth and sixth Aeg PNA monomers (i.e., before and
after the thymine (S)-Pmg unit) indicated an average coupling effi-
ciency of 89.0%. Deprotection, cleavage from the resin, and puri-
fication was performed as described above for the synthesis of PNA
III (Rt ϭ 24.0 min Column 1, gradient A). MS (MALDI-TOF):
calculated for C123H169N56O35 m/z ϭ 2992.0, observed 2991.7 [M
ϩ H]ϩ.
[3]
[4]
[5]
[6]
[7]
[8]
PNA-Containing Two Thymine S-Pmg Units (V): Compound V was
prepared as described above for compound IV, with the exception
that two thymine (S)-Pmg units were incorporated. Average coup-
ling efficiency for (S)-Pmg unit was 90.1% (Rt 23.8 min Column 1,
gradient B). MS (MALDI-TOF): calculated for C126H173N56O35
m/z ϭ 3030.4, observed 3032.5 [M ϩ H]ϩ.
[9]
[10]
[11]
PNA-Containing One Thymine (R)-Pmg Unit (VI): Compound VI
was prepared as described for PNA IV, with the exception that
thymine (R)-Pmg monomer 1b was used. Average coupling effici-
ency for the thymine (R)-Pmg unit was 72.0% (Rt 12.2 min column
2, gradient A). MS (MALDI-TOF): calculated for C123H169N56O35
m/z ϭ 2992.0, observed 2992.2 [M ϩ H]ϩ.
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
V. Kumar, P. S. Pallan, Meena, K. N. Ganesh, Org. Lett. 2001,
PNA-Containing Two Thymine (R)-Pmg Units (VII): Compound
VII was prepared as described for PNA IV, with the exception that
two thymine (R)-Pmg units were incorporated. Average coupling
efficiency for the Pmg unit 98.8% (Rt 17.0 min Column 2, gradient
B). MS (MALDI-TOF): calculated for C126H173N56O35 m/z ϭ
3030.4, observed 3032.5 [M ϩ H]ϩ, 3054.4 [M ϩ Na]ϩ, 3070.4 [M
ϩ K]ϩ.
3, 1269Ϫ1272.
G. Lowe, T. Vilaivan, J. Chem. Soc., Perkin Trans. 1 1997, 1,
555Ϫ560.
G. Lowe, T. Vilaivan, M. Westwell, Bioorg. Chem. 1997, 25,
321Ϫ329.
M. D’Costa, V. A. Kumar, K. N. Ganesh, Org. Lett. 1999, 1,
1513Ϫ1516.
A. Püschl, T. Tedeschi, P. E. Nielsen, Org. Lett. 2000, 2,
4161Ϫ4163.
H. Deng, V. A. Bloomfield, J. M. Benevides, G. J. Thomas, Jr.,
Nucleic Acids Res. 2000, 17, 3379Ϫ3385.
A. Slaitas, E. Yeheskiely, Nucleos. Nucleot. Nucl. 2001, 20,
1377Ϫ1379.
All (R)-Pmg PNA (VIII): Solid-phase synthesis was performed by
the Fmoc strategy. The chain elongation/deprotection cycle was
identical to that used for lysine incorporation and described above
for the synthesis of PNA III, with the exception that monomers
18aϪc were used. Deprotection, cleavage from the resin, and puri-
fication were performed as described above for synthesis of PNA
III. Average coupling efficiency could not be determined due to the
setup of the synthesizer, which was MMT dedicated (Rt 27.2 min
column 2, gradient C). MS (ES-TOF): calculated for
A. C. Laan, Van der, R. Strömberg, J. H. Boom, Van, E. Kuyl-
Yeheskiely, V. A. Efimov, O. G. Chakhmakhcheva, Tetrahedron
Lett. 1996, 37, 7857Ϫ7860.
[22]
[23]
[24]
Neosystem Laboratoire, France.
M. J. Soth, J. S. Nowick, J. Org. Chem. 1999, 64, 276Ϫ281.
O. Dangles, F. Guibe, G. Balavoine, S. Lavielle, A. Marquet, J.
Org. Chem. 1987, 52, 4984Ϫ4993.
H. N. Rydon, P. W. G. Smith, J. Chem. Soc. 1954, 3,
3642Ϫ3648.
B. R. De Costa, X. He, J. T. M. Linders, C. Dominguez, Z. Q.
Gu, W. Williams, W. D. Bowen, J. Med. Chem. 1993, 36,
2311Ϫ2320.
E. Frerot, J. Coste, A. Pantaloni, M. N. Dufour, P. Jouin, Tetra-
hedron 1991, 47, 259Ϫ70.
C150H205N56O35 m/z ϭ 3350.6, observed 1676.6 [M ϩ 2 H]2ϩ
1118.4 [M ϩ 3 H]3ϩ, 838.9 [M ϩ 4H]4ϩ
,
.
[25]
[26]
Thermal Melting Experiments: Melting temperatures were meas-
ured with 1:1 molar mixtures of PNA and the corresponding target
DNA or RNA (amounts determined by the UV absorption and
ε260), each at a concentration of 3 m in 10 m phosphate buffer,
pH 7, containing 100 m NaCl, and 0.1 m EDTA. Prior to the
measurement of the melting profiles, the solutions were heated to
90 °C for 5 minutes at a fast rate, then cooled to 0.5 °C over 90
min and equilibrated at this temperature for 45 min Subsequently,
the absorbance A260 was recorded vs. temperature for both dissoci-
ation and annealing at the rates 0.5 °C/min and 0.2 °C/min, respect-
ively.
[27]
[28]
[29]
J. Coste, E. Frerot, P. Jouin, B. Castro, Tetrahedron Lett. 1991,
32, 1967Ϫ1970.
A. C. Laan, Van der, N. J. Meeuwenoord, E. Kuyl-Yeheskiely,
R. S. Oosting, R. Brands, J. H. Boom, Van, Recl. Trav. Chim.
Pays-Bas. 1995, 114, 295Ϫ297.
L. Christensen, R. Fitzpatrick, B. Gildea, K. H. Petersen, H.
F. Hansen, T. Koch, M. Egholm, O. Buchardt, P. E. Nielsen,
J. Coull, R. H. Berg, J. Pept. Sci. 1995, 1, 175Ϫ183.
G. Dieci, R. Corradini, S. Sforza, R. Marchelli, S. Ottonello,
J. Biol. Chem. 2001, 276, 5720Ϫ5725.
[30]
[31]
[32]
Acknowledgments
We wish to thank the Swedish Natural Science Research Council
(NFR) for financial support.
R. Vinayak, A. C. Laan, Van der, R. Brill, K. Otteson, A.
Andrus, E. Kuyl-Yeheskiely, J. H. Boom, Van, Nucleosides
Nucleotides 1997, 16, 1653Ϫ1656.
[1]
P. E. Nielsen, M. Egholm, R. M. Berg, O. Buchardt, Science
Received February 6, 2002
1991, 254, 1497Ϫ1500.
[O02075]
Eur. J. Org. Chem. 2002, 2391Ϫ2399
2399