Short synthesis of polyoxygenated macrocyclic rings using acetal linkages. Application to the preparation of a new lipidic polyamine

Article abstract of DOI:10.1021/jo0493020

A short preparation of polyoxygenated macrocycles can be carried out by combining the formation of an acetal linkage, to introduce long alkyl chains, with a ring closure metathesis. As an example, this methodology was used to synthesize a new polyamino li

Full text of DOI:10.1021/jo0493020

prone to toxicity, there is still a need for the development
of new models.
Sh or t Syn th esis of P olyoxygen a ted
Ma cr ocyclic Rin gs Usin g Aceta l Lin k a ges.
Ap p lica tion to th e P r ep a r a tion of a New
Lip id ic P olya m in e
Archeobacteria lipids5 are an attractive class of com-
pounds for drug and nucleic acid delivery. They consist
of regularly branched, and usually fully saturated, long
phytanyl chains attached via ether bonds to a glycerol
backbone(s). Archaeobacteria liposomes6 (archaeosomes)
demonstrate relatively higher stabilities to oxidative
stress, high temperature, alkaline pH, the action of
phospholipases, bile salts, and serum proteins. Further-
more, archaeosomes are safe and do not invoke any
noticeable toxicity.7 Consequently, several macrocyclic
models,8-12 with or without a phytanyl structure, have
been synthesized to overcome the difficulties encountered
in isolating gram quantities of naturally occurring ar-
chaebacterial lipids and to study structure/activity rela-
tionships.
Marie-Laure Miramon, Nathalie Mignet, and
J ean Herscovici*
Laboratoire de Pharmacologie Chimique et Ge´ne´tique U640
INSERM - UMR 8151 CNRS ENSCP 11 rue Pierre et Marie
Curie 75231 Paris Cedex
herscovi@ext.jussieu.fr
Received April 27, 2004
Abstr a ct: A short preparation of polyoxygenated macro-
cycles can be carried out by combining the formation of an
acetal linkage, to introduce long alkyl chains, with a ring
closure metathesis. As an example, this methodology was
used to synthesize a new polyamino lipid.
In an extension to our research in synthetic vectors
for DNA delivery,13 we became interested in these unique
features. We anticipated that liposomes made from
simpler macrocycles could provide, despite the absence
of phytanyl groups, some of the properties of archaeo-
somes that could be of great interest for gene and drug
delivery. Therefore, we designed a macrocycle that pos-
sesses the functionalities to interact with DNA by means
of cationic or polythiourea heads14 at one end and a
targeting moiety at the other (Figure 1).
Gene therapy is a rapidly evolving field of medicine,
in which DNA is the therapeutic agent.1,2 DNA delivery
systems mainly consist of either viral or synthetic vectors,
viral vectors being the most widely used. However, the
use of viruses is restricted by several drawbacks, such
as immunological responses, biological safety, and limita-
tion in the gene size. Synthetic vectors3,4 offer an alterna-
tive to viral transfection. Compared to modified viruses,
the major advantages of synthetic delivery systems are
their simplicity, ease of production, and relatively low
toxicity. However, because they are less efficient than
viral systems, are sensitive to blood components, and are
While searching for a methodology suitable for the
connection of a long alkyl chain to an alcohol, we came
across a report on the preparation of alkyl chloromethoxy
ethers.15 We have used these reactive species for the
synthesis of carbohydrate based cationic lipids.16 We thus
reasoned that the above strategy, combined with ring
closure metathesis (RCM), would provide a short and
F IGURE 1. DNA macrocyclic vectors.
F IGURE 2. Structure of polyoxygenated macrocycles.
10.1021/jo0493020 CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/10/2004
J . Org. Chem. 2004, 69, 6949-6952
6949

SCHEME 1. Retr osyn th etic An a lysis of th e 47-Mem ber Ma cr ocycle
SCHEME 2. P r ep a r a tion of th e 47-Mem ber Ma cr ocyclea
a
(i) a MeONa, MeOH, (ii) TMSCl, paraformaldehyde 100%, (iii) TrCl, 85%, (iv) 2, DIPEA, tBu4NI, THF 60 °C 82%, (v) LiAlH4, reflux
THF 91%, (vi) NaH, tBu4NI, allyl iodide, 90%, (vii) RuCl2(dCHPh)(PCy3)2, CH2Cl2 reflux 100%.
simple route to macrocycles. Furthermore, the calculation
of the octanol/water partition coefficients (logP) indicated
a gain of eight logs between the saturated (16.12) and
the oxygenated ring (8.36). These data showed that the
polyoxygenated ring will decrease the highly hydrophobic
nature of polythiourea lipids and will therefore lead to
an easier lipid/DNA complex formulation.14
In this note, we describe the preparation of two models
(Figure 2), a 47-member macrocycle and a tricyclic 101-
atom ring containing two symmetrical benzene groups,
in order to avoid folding that has been described for
bolaform amphiphile.17-20 Second, the transformation of
the 47-member macrocycle to a polyamino lipid will be
depicted.
Therefore, we began our synthesis with 2, which is
available in two steps from ω-pentadecalactone (Scheme
2). Protection of diethanolamine produced the trityl
derivative 4 in 85% yield. Condensation of 4 with the
(1) Rubanyi, G. M. Mol. Aspects Med. 2001, 22, 113.
(2) Nabel, G. J . Proc. Nat. Acad. Sci. 1999, 96, 324.
(3) Verma, I. M.; Somia, N. Nature 1997, 389, 239.
(4) Miller, A. D. Angew. Chem., Int. Ed. Engl. 1998, 37, 7, 1768.
(5) (a) De Rosa, M. Thin Solid Films 1996, 284-285, 13-17. (b)
Gliozzi, A.; Relini, A.; Chong, P. L.-G. J . Membrane Sci. 2002, 206,
131.
(6) Patel, G. B. Sprott, G. D. Crit. Rev. Biotechnol. 1999, 19, 317.
(7) Patel, G. B; Omri, A.; Deschatelets, L.; Sprott, G. D. J . Liposome
Res. 2002, 12, 353.
(8) Patwardhan, A. P.; Thompson, D. H. Org. Lett. 1999, 1, 241.
(9) Menger, F. M.; Chen, X. Y. Tetrahedron Lett. 1996, 37, 323.
(10) (a) Eguchi, T.; Terachi, T.; Kakinuma, K. J . Chem. Soc., Chem.
Commun. 1996, 365. (b) Eguchi, T.; Ibaragi, K.; Kakinuma, K. J . Org.
Chem. 1998, 63, 2689. (c) Arakawa, K.; Kano, H.; Eguchi, T.; Nish-
iyama, Y.; Kakinuma, K. Bull. Chem. Soc. J pn. 1999, 72, 1575.
(11) Arakawa, K.; Eguchi, T.; Kakinuma, K. J . Org. Chem. 1998,
63, 4741.
(12) Eguchi, T.; Arakawa, K.; Terachi, T.; Kakinuma, K. J . Org.
Chem. 1997, 62, 1924.
(13) Carriere, M.; Tranchant, I.; Niore, P.-A.; Byk, G.; Mignet, N.;
Escriou, V.; Scherman, D.; Herscovici, J . J . Liposome Res. 2002, 12,
95-106.
(14) Herscovici, J .; Scherman, D.; Tranchant, I.; Mignet, N.;
Girard, C. Patent WO02092558; WO2002FR01626, FR20010006330,
US20010297482P, 14/05/2001. Girard, C.; Tranchant, I.; Gorteau, V.;
Potey, L.; Herscovici, J . J . Comb. Chem. 2004, 6, 275.
Syn th esis of th e 47-Mem ber Ma cr ocycle
Our retrosynthetic analysis for the macrocycle A is
shown in Scheme 1. Macrocyclization would occur upon
a ring-closing metathesis21 (RCM) of precursor B, which
may be prepared by allylation of diol C. The construction
of C would be easily achieved by the reduction of methyl
ester D. Accordingly, disconnection of D gives the known
compounds diethanolamine and methyl 15-hydroxy-pen-
tadecanoate.
6950 J . Org. Chem., Vol. 69, No. 20, 2004

SCHEME 3. Retr osyn th etic An a lysis of th e 101-Mem ber Ma cr ocycle P r ecu r sor
SCHEME 4. P r ep a r a tion of th e 101 Atom s Ma cr ocycle P r ecu r sor 16a
a
(i) I2 PPh3 imidazole 95%, (ii) APTS dihydropyran 85%, (iii) 2, diiosopropylethylamine, tetrabutylammonium iodide reflux, (iv) LiAlH4
THF (80% two steps), (v) NaH, allyl iodide 66%, (vi) Dowex H+ 95%, (vii) NaH, 9 0°-5 °C then 60 °C, one night 32%.
chloromethoxy ether 2 at 60 °C afforded the diester 5
(82%), which was reduced with LiAlH4 to give the diol 6
(91%). Allylation of 6 with allyl iodide afforded the
bis-(2-(15-allyloxy-pentadecyloxymethoxy)-ethyl)-trityl-
amine (7) in 90% yield. Heating 7 for 2 days in the
presence of Grubb’s catalyst produced the 47-member
macrocycle 8 quantitatively.22,23 NMR data was consistent
with the 1,3,9,11,27,32-hexaoxa-6-aza-cycloheptapenta-
cont-29-ene structure for 8. Examination of the DEPT
spectrum showed only CH olefinic carbons (δ 130.2 ppm),
which were correlated to the proton signal at δ 5.8 ppm.
chloromethoxy ether 2 followed by LiAlH4 reduction
generated 13 (80% yield, two steps). Allylation of alcohol
13 provided 14. Removal of the THP group using a Dowex
resin gave the benzylic alcohol 15. Condensation of the
diiodide 9 with 15 afforded the bis(allyl) ether 16 in 32%
yield.
To improve this modest yield, another approach was
investigated (Scheme 5). Acetalation of 15 with the
chloromethoxy ether 2 and reduction of the resulting
ester 17 afforded the bis-acetal 18 in 80% yield (two
steps). Reaction of 18 with methanesulfonyl chloride gave
the mesyl 19. Condensation of the diol 4 with 19 in the
presence of HMPA produced the macrocycle’s precursor
20 in 56% yield. To complete the synthesis, 20 was
refluxed in the presence of Grubb’s catalyst to quantita-
tively yield the macrocycle 21 as a white solid.
P r ep a r a tion of th e 101-Mem ber Ma cr ocycle
Having completed the synthesis of 8 we investigated
the preparation of the 101-member macrocycle. The
retrosynthetic analysis showed that the precursor of
macrocycle F could be prepared by the condensation
between G and H. Therefore, the diodide 9 was prepared
by the reaction of the diol 6 with I2/PPh3 in the presence
of imidazole (Scheme 4). Tetrahydropyranylation of 10
led to the monoprotected derivative 11. Coupling 11 with
(20) Cornell, B. A.; Braach-Maksvytis, V. L. B.; King, L. G.; Osman,
P. D. J .; Raguse, B.; Wieczorek, L.; Pace, R. J . Nature 1997, 387, 580.
(21) For the preparation of macrocycle using RCM see: (a) Pat-
wardhan, A. P.; Thompson, D. H. Org. Lett. 1999, 1, 241-244. (b) Lee,
C. W.; Grubbs, R. H. J . Org. Chem. 2001, 66, 7155-7158 and ref 10b.
(22) Maynard, H. D.; Grubbs, R. H. Macromolecules 1999, 32, 6917.
(23) Schwab, P.; Grubbs, R. H.; Ziller, J . W. J . Am. Chem. Soc. 1996,
118, 100.
(15) Shipov, A. G.; Savost’yanova, I. A.; Baukov, Y. I Zh. Obshch.
Khim. 1984, 54, 2649.
(16) Herscovici, J .; Egron, M. J .; Quenot, A.; Leclercq, F.; Leforestier,
N.; Mignet, N.; Wetzer, B.; Scherman, D. Org. Lett. 2001, 1893.
(17) Yamauchi, K.; Moriya, A.; Kinoshita, M. Biochim. Biophys. Acta
1989, 1003, 151.
(24) The azido sulfonate 22 was prepared from 11-bromo undecanol
in a two step sequence entailing (i) substitution of bromine with sodium
azide. (ii) mesylation of the hydroxyl group with methanesulfonyl
chloride.
(18) Raguse, B.; Culshaw, P. N.; Prashar, J . K.; Raval, K. Tetrahe-
dron Lett. 2000, 41, 2971.
(19) Moss, R. A.; Li, J .-M. J . Am. Chem. Soc. 1992, 114, 9227.
J . Org. Chem, Vol. 69, No. 20, 2004 6951

SCHEME 5. P r ep a r a tion of th e 101-Mem ber
Ma cr ocyclea
SCHEME 6. P r ep a r a tion of th e Ma cr ocyclic
P olya m in o Lip id a
a
(i) 2, diisopropylethylamine tetrabutylammonium iodide, (ii)
LiAlH4 80% (two steps), (iii) methanesulfonyl chloride DMAP RT
one night 90%, (iv) 4, NaH THF HMPA 0-5 °C 2 h then RT,
overnight, 56%. (v) benzylidene-bis(tricyclohexylphosphine)dich-
lororuthenium, CH2Cl2, 48 h, reflux 100%.
a
(i)BH3-THF, 0 °C, 70%, (ii) NaH, THF reflux 40% (iii PS-
PPh3 H2O, THF 100%, (iv) succinic anhydride, DMAP, CH2Cl2,
pyridine 72%, (v) Pd(C) MeOH 100%, (vi) DIPEA, BOP, CH2Cl2,
46%.
To demonstrate the interest of macrocyclic polycationic
vectors, the preparation of polyamine 27 was undertaken
(Scheme 6). Hydroboration of 8 with borane-tetrahydro-
furan complex gave the 6-trityl-1,3,9,11,27,32-hexaoxa-
6-aza-cycloheptapenta-contan-27-ol (22) in 70% yield as
a yellow oil. Alkylation of 22 with the sulfonate 2324
afforded the azide 24. Treatment of 24 with a polystyrene
grafted triphenylphosphine yields the amine 25 quanti-
tatively. Reaction of 25 with succinic anhydride pro-
ceeded smoothly to yield the succinate 26 (72%). Hydro-
genolysis of 26 afforded the secondary amine 27 quantita-
tively. Finally, spermine condensation with 27 was
performed using BOP. Purification of the resulting
product on Sephadex LH-20 afforded the lipopolyamine
28 (yield 46%).
In conclusion, the work presented here describes
methodologies for the preparation of polyoxygenated
macrocycles and their transformation into a macrocyclic
cationic lipid. Efforts toward the introduction of a target-
ing moiety25 and evaluation of their transfecting proper-
ties are underway.
(25) In an attempt to prepare a lipid possessing an interacting head
and a targeting moiety, lipids 29 and 30 were synthesized. However,
condensation with spermine has thus far been unsuccessful.
Ack n ow led gm en t. This work was financially sup-
ported by the CNRS, Aventis-Gencell, the MENRT and
the Re´gion Ile de France (SESAME). M.L.M. thanks
Aventis-Gencell for a postdoctoral fellowship.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and NMR data for compounds 2 to 28. This material
is available free of charge via the Internet http://pubs.acs.org.
J O0493020
6952 J . Org. Chem., Vol. 69, No. 20, 2004