An expeditious liquid-phase synthesis of cyclic peptide nucleic acids

Article abstract of DOI:10.1016/S0040-4039(00)00536-0

Suitably protected linear precursors of cyclic PNAs can be readily obtained by BOP/DiPEA coupling of the corresponding sub-monomers. Conversion of the linear PNA dimers into the pentafluorophenyl esters allows cyclization by intramolecular attack of the deprotected primary amino function under diluted conditions. After removal of the secondary amino protecting group(s), installation of the required nucleobase-acetyl function(s) affords cyclic PNAs. In addition, the latter compounds can be prepared following a direct coupling strategy. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00536-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 3991–3995
An expeditious liquid-phase synthesis of cyclic peptide
nucleic acids
Jeroen C. Verheijen, Gijsbert M. Grotenbreg, Ludo Hart de Ruyter, Pieter A. M. van der Klein,
Gijsbert A. van der Marel and Jacques H. van Boom ∗
Leiden Institute of Chemistry, Gorlaeus Laboratories, PO Box 9502, 2300 RA Leiden, The Netherlands
Received 17 February 2000; accepted 28 March 2000
Abstract
Suitably protected linear precursors of cyclic PNAs can be readily obtained by BOP/DiPEA coupling of the
corresponding sub-monomers. Conversion of the linear PNA dimers into the pentafluorophenyl esters allows
cyclization by intramolecular attack of the deprotected primary amino function under diluted conditions. After
removal of the secondary amino protecting group(s), installation of the required nucleobase-acetyl function(s)
affords cyclic PNAs. In addition, the latter compounds can be prepared following a direct coupling strategy. © 2000
Elsevier Science Ltd. All rights reserved.
Keywords: cyclization; peptide nucleic acid; coupling reactions; macrocycles.
Nearly two decades ago it was disclosed that both 30-50-cyclic diribonucleotides c[ApUp] and c[UpUp]
are inhibitors of the DNA dependent RNA polymerase in Escherichia coli.1 Later on, it was revealed
that c[GpGp] is an endogeneous regulator of the biological synthesis of cellulose in the Gram-negative
bacterium Acetobacter xylinum.2 X-Ray analysis showed that c[GpGp],3 as well as the closely related
cyclic dinucleotide c[dApdAp],4 adopts a conformation in which the 12-membered sugar-phosphate
backbone provides a rigid framework to hold the two nucleobases 6.8 Å apart in parallel planes.
The proposed intercalator potential3 of this class of cyclic compounds was an incentive in probing the
biological relevance of analogs in which the rigid cyclic framework in I (Fig. 1) is replaced by an isosteric
cyclic bis(aminoethylglycinamide) as in the peptide nucleic acid (PNA)5 analog II. It was expected that
this type of analog would be readily accessible by cyclization of linear PNA dimers. However, earlier
studies6,7 have noted that condensation of PNA monomers in solution is a surprisingly low yielding
process.
Recently, Di Giorgio et al. reported that liquid-phase synthesis of linear PNAs could be readily effected
via an indirect strategy using orthogonally N-protected aminoethylglycine monomeric synthons.8 In the
first instance, we explored the feasibility of converting an orthogonally (Z/Fmoc) protected linear PNA
∗
Corresponding author. Fax: +31 71 527 4307; e-mail: j.boom@chem.leidenuniv.nl (J. H. van Boom)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00536-0
tetl 16815

3992
Fig. 1.
dimer into a cyclic PNA dimer. Accordingly, condensation (see Scheme 1) of Z-protected PNA backbone9
1 with Fmoc-protected PNA building block 2 under the agency of Castro’s reagent gave fully protected
dimer 3. Linear dimer 3 could now be transformed into the cyclic intermediate 7 by the following four
step procedure. Removal of the allyl group in 3 via palladium-catalyzed hydrostannylation was followed
by EDC coupling with pentafluorophenol to give 5. Deblocking of the N-Boc protective group in 5 with
50% TFA in CH2Cl2 afforded primary amine 6 as its TFA salt. Following a well-established procedure
for the synthesis of cyclic peptides,10 slow addition of a solution of 6 to a diluted solution of Hünig’s
base in acetonitrile gave orthogonally protected cyclic dimeric PNA precursor 7 in an overall yield of
57% (based on 3). Conversion of 7 into cyclic PNA dimer 12 could be accomplished by executing the
following sequence of reactions. The Fmoc protective group in 7 was removed with a 20% piperidine
solution (→8) and the first nucleobase was installed by coupling thymin-1-ylacetic acid11 to the free
secondary amine in 8 to give 9. Hydrogenolysis of the benzyloxycarbonyl group and condensation of the
resulting 10 with N-benzoyl-adenin-9-ylacetic acid12 furnished, after ammonolysis, target compound 12
in a yield of 41% over the last five steps.
The successful transformation of linear dimer 6 into cyclic PNA precursor 7 implied that replacement
of one of the urethane protective groups in 6 by a nucleobase-acetyl moiety would be an attractive
alternative. To this end, we first prepared dimer 15a by BOP-mediated condensation of guanine monomer
13a with Fmoc-protected sub-monomer 14a (Scheme 2). Transformation of 15a into the corresponding
pentafluorophenyl ester, followed by acidolysis of the Boc group, gave 16a. Cyclization of the linear
PNA derivative 16a, as described for the conversion of 6 into 7, proceeded smoothly to furnish 17
in a yield of 62%. Removal of the Fmoc protective group followed by coupling of 18 with N-iso-
butyryl-guanin-9-yl-acetic acid13 allowed direct isolation (34% yield) of crystalline 19 from the crude
reaction mixture. Deprotection of the exocylic amines in 19 was effected with methanolic ammonia to
give the homogeneous cyclic bis-guaninyl PNA dimer 20. In a similar fashion, thymine monomer 13b
was condensed with building block 14b to give linear dimer 15b. Saponification of the latter dimer
and subsequent formation of the pentafluorophenyl ester was followed by deprotection of the primary
amine to give 16b. Also in this case, cyclization proceeded smoothly to give 9. At this stage, we were
interested in determining whether the presence of a nucleobase in monomer 14 instead of 13 would
influence the outcome of the cyclization. It was established that condensation of carboxylic acid 13c with
N-aminoethyl-N-thymin-1-yl-glycine 14c14 gave, after further processing of 15c as mentioned earlier, the
activated linear dimer 16c. Subjection of this compound to DiPEA under diluted conditions gave cyclic
dimer 9 as the only detectable product. In this respect, it should be noted that the cyclization products

3993
Scheme 1. Reagents and conditions: (i) BOP, DiPEA, DMF, 77%; (ii) Cl2Pd(PPh3)2, AcOH, Bu3SnH, CH2Cl2, 83%; (iii)
pentafluorophenol, EDC, DMF, 98%; (iv) 50% TFA, CH2Cl2, quant.; (v) DiPEA, dilution in ACN, 70%; (vi) 20% piperidine,
DMF, 85%; (vii) T-CH2COOH, BOP, DiPEA, DMF, 64%; (viii) H2, Pd/C, iPrOH:DMF (1:1), quant.; (ix) A(Bz)-CH2COOH,
BOP, DiPEA, DMF, 83%; (x) NH3/MeOH, 90%
1
from 16b and 16c were identical in all aspects (i.e. H NMR, 13C NMR and mass spectrometry) with
earlier prepared 9 (cf. Scheme 1).
Scheme 2. Reagents and conditions: (i) BOP, DiPEA, DMF, 75% (15a), 76% (15b), 66% (15c); (ii) (1) Cl2Pd(PPh3)2, AcOH,
Bu3SnH, CH2Cl2, 76%; (2) pentafluorophenol, EDC, DMF, 70%; (3) 50% TFA in CH2Cl2, quant. (16a); (iii) (1) NaOH (1N),
MeOH:H2O (1:1, v/v); (2) HCl; (3) pentafluorophenol, EDC, DMF; (4) 50% TFA, CH2Cl2, 90% (16b), 60% (16c); (iv) DiPEA,
dilution in ACN, 62% (17), 67% (9 from 16b), 60% (9 from 16c); (v) 20% piperidine, DMF, 85%; (vi) G(iBu)-CH2COOH,
BOP, DiPEA, 34%; (vii) NH3/MeOH, 96%
The results obtained thus far clearly show that the cyclization of linear pentafluorophenyl esters is not
influenced by the nature of the substituents at the secondary amine (i.e. R1 and R2 in 16). We surmised
that the poor yields, observed in the introduction of an amide bond between PNA monomers in a direct
coupling strategy, could be improved using pentafluorophenyl active esters. In order to validate this

3994
assumption, we prepared the linear bis-thyminyl15 PNA dimer 22 in an excellent yield by condensing
the pentafluorophenyl ester 2116 with primary amine 14c (Scheme 3). Hydrolysis of the methyl ester
in 22 with sodium hydroxide, followed by neutralization with Dowex-H+, afforded dimeric carboxylic
acid 23. Subjection of 23 to the same sequence of reactions as described earlier for the cyclization
of PNA precursors gave access to the bis-thyminyl cyclic PNA 25. HPLC–MS analysis of the crude
reaction mixture revealed that the only detectable side-products formed in the cyclization reaction were a
hydrolyzed linear dimer and a cyclized tetramer. The latter macrocycle is formed through intermolecular
coupling of two dimers, followed by cyclization of the resulting tetrameric pentafluorophenyl ester.
The latter finding indicates that the methodology described here may be successfully extended to the
construction of larger ring-systems.
Scheme 3. Reagents and conditions: (i) DiPEA, ACN, 90%; (ii) (1) NaOH (1N), MeOH:H2O (1:1, v/v); (2) Dowex-H+; (iii)
pentafluorophenol, EDC, DMF, 51% (three steps); (iv) (1) 50% TFA in CH2Cl2; (2) DiPEA, dilution in ACN, 68% (isolated
yield)
In conclusion, we have shown that the liquid-phase synthesis of cyclic PNA dimers may be readily
effected by three different routes, each of which has its own merit in the construction of cyclic
bis(aminoethylglycinamide) frameworks carrying a variety of functional groups. Adaptation of this
methodology also opens the way to a liquid-phase synthesis of cyclic oligo-PNAs. The biological
relevance of cyclic PNA dimers as well as extension of the reported strategy to the solid-phase synthesis
of cyclic PNAs is currently under investigation.
Acknowledgements
This work was supported by the Council for Chemical Sciences of the Netherlands Organization for
Scientific Research (CW-NWO). We thank Dr. M. Overhand for stimulating discussions.
References
1. Hsu, C.-Y. J.; Dennis, D. Nucleic Acids Res. 1982, 10, 5637–5647.
2. Ross, P.; Weinhouse, H.; Aloni, Y.; Michaeli, D.; Weinberger-Ohana, P.; Mayer, R.; Braun, S.; de Vroom, E.; van der Marel,
G. A.; van Boom, J. H.; Benziman, M. Nature 1987, 325, 279–281.
3. Liaw, Y.-C.; Gao, Y.-C.; Robinson, H.; Sheldrick, G. M.; Sliedregt, L. A. J. M.; van der Marel, G. A.; van Boom, J. H.;
Wang, A. H.-J. FEBS Lett. 1990, 264, 223–227.
4. Frederick, C. A.; Coll, M.; van der Marel, G. A.; van Boom, J. H.; Wang, A. H.-J. Biochemistry 1988, 27, 8350–8361.
5. Nielsen, P. E.; Egholm, M.; Berg, R. M.; Buchardt, O. Science 1991, 254, 1497–1500.
6. Farèse, A.; Patino, N.; Condom, R.; Dalleu, S.; Guedj, R. Tetrahedron Lett. 1996, 37, 1413–1416.
7. Farèse, A.; Pairot, S.; Patino, N.; Ravily, V.; Condom, R.; Aumelas, A.; Guedj, R. Nucleosides and Nucleotides 1997, 16,
1893–1906.

3995
8. Di Giorgio, C.; Pairot, S.; Schwergold, C.; Patino, N.; Condom, R.; Farèse-Di Giorgio, A.; Guedj, R. Tetrahedron 1999,
55, 1937–1958.
9. Atherton, E.; Law, H. D.; Moore, S.; Elliott, D. F.; Wade, R. J. Chem. Soc. (C) 1971, 3393–3396.
10. Seebach, D.; Overhand, M.; Kühnle, F. N. M.; Martinoni, B. Helv. Chim. Acta 1996, 79, 913–941.
11. Rabinowitz, J. L.; Gurin, S. J. Am. Chem. Soc. 1953, 75, 5758–5759.
12. Efimtseva, E. V.; Mikhailov, S. N.; Jasko, M. V.; Malakhov, D. V.; Semizarov, D. G.; Fomicheva, M. V.; Kern, E. R.
Nucleosides and Nucleotides 1995, 14, 373–375.
13. Will, D. W.; Breipohl, G.; Langner, D.; Knolle, J.; Uhlmann, E. Tetrahedron 1995, 51, 12 069–12 082.
14. van der Laan, A. C.; Meeuwenoord, N. J.; Oosting, R. S; Brands, R.; Kuyl-Yeheskiely, E.; van Boom, J. H. Recl. Trav.
Chim. Pays-Bas 1995, 114, 295–297.
15. Similarly, the corresponding linear bis-guaninyl dimer could be obtained in a yield of 48%.
16. Clivio, P.; Guillaume, D.; Adeline, M.-T.; Hamon, J.; Riche, C.; Fourrey, J. L. J. Am. Chem. Soc. 1998, 120, 1157–1166.