Bolaforms with Fourteen Galactose Units: A Proposed Site-Directed Cohesion of Cancer Cells

Article abstract of DOI:10.1021/ol030135k

(Matrix presented) The multistep synthesis of a calixarene joined to a second calixarene via a long spacer is described. Since each calixarene bears multiple galactose-based units (known to bind strongly to rat hepatoma cells), there existed the possibility of cross-linking the cancer cells into a network. The compounds did not serve this purpose, a fact potentially correctable by adjusting or rigidifying the spacer. Formation of a cancer net around a solid tumor remains a viable approach to retarding growth and/or inhibiting metastasis.

Full text of DOI:10.1021/ol030135k

ORGANIC
LETTERS
2
004
Vol. 6, No. 2
61-264
Bolaforms with Fourteen Galactose
Units: A Proposed Site-Directed
Cohesion of Cancer Cells
2
Fredric M. Menger,* Jianwei Bian, Elena Sizova, Deborah E. Martinson, and
Victor A. Seredyuk
Department of Chemistry, Emory UniVersity, Atlanta, Georgia 30322
menger@emory.edu
Received November 17, 2003
ABSTRACT
The multistep synthesis of a calixarene joined to a second calixarene via a long spacer is described. Since each calixarene bears multiple
galactose-based units (known to bind strongly to rat hepatoma cells), there existed the possibility of cross-linking the cancer cells into a
network. The compounds did not serve this purpose, a fact potentially correctable by adjusting or rigidifying the spacer. Formation of a
“cancer net” around a solid tumor remains a viable approach to retarding growth and/or inhibiting metastasis.
Consider the possibility of an anti-cancer drug that functions
by interconnecting a layer of cancer cells. Formation of such
a “cancer-net” around a solid tumor might, conceivably,
retard growth and/or inhibit metastasis. Since the therapy
depends on chemistry occurring at the periphery of the tumor,
the common problem of delivering a drug to an entire cell
mass, often via an impaired blood supply, is mitigated. To
be effective, of course, a cancer-net drug must selectively
adhere to pathological cells in preference to healthy cells.
Interconnecting cells into networks is not a new concept.
mannose substituents that promote the aggregation of
leukemia cells. Polymers can also be used to coat cells so
4
that adhesion is sterically blocked.
5
The work of Y. Aoyama et al. gave us the impetus to
pursue nonpolymeric cancer-net compounds. They found that
rat hepatoma cells bind a calixarene having two galactoses
on each of its four aromatic rings. Spleen cells lacking
specific galactoside receptors do not bind this calixerene.
Nor do rat hepatoma cells bind the calixarene substituted
with glucose instead of galactose.
Thus, it possible, that molecules with two galacto-
calixarene macrocycles, covalently joined by a long spacer,
might “glycotarget” a group of cancer cells and thereby
1
W. Meier, for example, showed that certain cells, surface-
modified with a biotin-terminated polymer, are subsequently
agglutinated by addition of streptavidin. The polymer thus
2
mimics natural “cell adhesion molecules”, i.e., proteins that
(3) Gestwicki, J. E.; Strong, L. E.; Cairo, C. W.; Boehm, F. J.; Kiessling,
unite cells via intermembrane ligand-receptor interactions.
L. L. Chem. Biol. 2002, 9, 163.
(4) Elbert, D. L.; Hubbell, J. A. Chem. Biol. 1998, 5, 177.
3
Kiessling et al. synthesized polymer chains with up to 142
(5) (a) Aoyama, Y.; Tanaka, Y.; Sugahara, S. J. Am. Chem. Soc. 1989,
(
1) Meier, W. Langmuir 2000, 16, 1457.
111, 5397. (b) Fujimoto, K.; Miyata, T.; Aoyama, Y. J. Am. Chem. Soc.
2000, 122, 3558.
(2) Dembo, M.; Bell, G. I. Curr. Top. Membr. Transp. 1987, 29, 71.
1
0.1021/ol030135k CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/24/2003

Scheme 1
interconnect them (not unlike the cross-linking of polymer
chains). Compound 1 in Scheme 1 sprung from this idea.
Several features of 1 need explanation. (a) The compounds
are polyvalent at both termini; multiple simultaneous interac-
6
tions are known to enhance associative processes in biology.
Previously described scaffolds bearing polyvalent glycocon-
7
8
9
jugates include cyclodextrins, proteins, dendrimers, gold
nanoparticles, vesicles,
1
0
11,12
13
self-assembled monolayers,
self-aggregates,14 and cluster compounds.15 (b) Spacers
separating our calixarenes vary in length from 16 atoms (1a),
to 21 atoms (1b), to 29 atoms (1c).
The variations were prompted by the fact that we did not
know if undesirable “looping” (i.e., intracellular binding
competitive with the transcellular connectivity) would be
minimized with certain spacer lengths. Both polar and
nonpolar spacers were investigated. (c) In addition to 14
galactose units, bis-calixarenes 1 each possess 8 undecyl
Figure 1. MALDI-MS of 1a after using (A) Cs
CO in the 6-to-7 transformation.
2 3 2
CO and (B) K -
3
compounds. First and foremost, we faced the need to
monoderivatize one of eight equivalent hydroxyls in calix-
[
4]resorcarene to form 2. By using 1.3 equiv of bromo-ester,
it was possible to obtain 2 in 30% yield along with, roughly,
0% starting calixarene and 20% material with 2-4 alkyl
groups. Chromatography with gradient elution (0-20% CH
CN in CHCl on silica) allowed isolation of 2. Acetylation
of the remaining seven hydroxyls with Ac O followed by
5
groups in accordance with Aoyama’s cell-binding monomer.
It is likely, but not yet established, that the hydrocarbon
5
16
chains (assumed to be all-cis ) contribute hydrophobically
to docking at the cancer cells’ outer membrane surface.
The nine-step synthesis of 1 (Scheme 2) encountered
special problems related to the handling of polyfunctional
3
-
3
2
deprotection of the carboxyl (leaving the acetates untouched)
gave 4. Compound 4 allowed a DCC-promoted coupling with
a diamine “spacer” in moderate yields. Under our reaction
conditions, the diamines preferentially linked the calixarenes
(as opposed to deacylating them) to give 5. Coupling yields
were low when the hydroxyls were not acetylated. Removal
of the acetate groups in the presence of the amide groups
was surprisingly touchy: since neither aqueous HCl nor
(6) Mammen, M.; Choi, S.-K.; Whitesides, G. M. Angew. Chem., Int.
Ed. 1998, 37, 2755.
(7) Fulton, D. A.; Stoddart, J. F. Bioconjugate Chem. 2001, 12, 655.
(8) Lemieux, G. A.; Bertozzi, C. R. Chem. Biol. 2001, 8, 265.
(9) Zanini, D.; Roy, R. J. Am. Chem. Soc. 1997, 119, 2088.
(10) Hern a´ iz, M. J.; de la Fuente, J. M.; Barrientos, A. G.; Penad e´ s, S.
Angew. Chem., Int. Ed. 2002, 41, 1554.
11) Ma, Z.; Li, J.; Liu, M.; Cao, J.; Zou, Z.; Tu, J.; Jiang, L. J. Am.
Chem. Soc. 1998, 120, 12678.
12) Chenevier, P.; Grandjean, C.; Loing, E.; Malingue, F.; Angyalosi,
G.; Gras-Masse, H.; Roux, D.; Melnyk, O.; Bourel-Bonnet, L. Chem. Comm.
002, 20, 2446.
13) Horan, N.; Yan, L.; Isobe, H.; Whitesides, G. M.; Kahne, D. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 11782.
14) Akiyoshi, K.; Deguchi, S.; Moriguchi, N.; Yamaguchi, S.; Sunamoto,
J. Macromolecules 1993, 26, 3062.
15) Grandjean, C.; Angyalosi, G.; Loing, E.; Adriaenssens, E.; Melnyk,
O.; Pancre, V.; Auriault, C.; Gras-Masse, H. Chem. BioChem. 2001, 2, 747.
(
(
2 3
NaOH performed well, we resorted to aqueous K CO to
secure 6 in high yield.
2
(
In synthesizing their monomeric calixarene, Aoyama et
5
(
al. octa-alkylated the parent calixarene with bromo-aceto-
(
(16) Fujimoto, T.; Shimizu, C.; Hayashida, O.; Aoyama, Y. J. Am. Chem.
Soc. 1997, 119, 6676.
262
Org. Lett., Vol. 6, No. 2, 2004

Scheme 2
nitrile followed by reduction to an octa-amine. The octa-
2 3
instead of Cs CO in the 6-to-7 step, the spectrum (Figure
amine was subsequently reacted with a sugar lactone. This
strategy was unsuccessful with our compounds due to the
difficulty of reducing the fourteen nitriles in the presence of
the two spacer amides. Hence, we shifted to a nonreductive
1B) displayed an abundance of lower homologues.
Our cell experiments used epithelial-like 4-dimethyl-amino
azobenzene-transformed rat liver cells (Riken RLC-16,
Japan) cultured in Williams’ Medium E containing 1%
L-glutamine, 1% penicillin/streptomycin, and 10% v/v fetal
bovine serum. The surface-adhered cells were washed three
times with Dulbecco’s phosphate-buffered saline prior to
being trypsinized via a quick rinse with 0.25% trypsin/EDTA
followed by a 30 min incubation at 37 °C also with trypsin/
EDTA. Fetal bovine serum in medium or buffer was then
added to retard the trypsin. Repeated pipeting broke the
pathway: intermediate 6 was treated with Br(CH
a “14-site” alkylation that proceeded to completion only in
the presence of Cs CO . Removal of the Boc groups from 7
2 3
) NHBoc,
2
3
with HCl produced polyamine 8 that could now be reacted
with the sugar lactone to give 1. Figure 1A, the spectrum
+
with the correct [MW + Na] of 8024, was critical in
2 3
affirming the proposed structure of 1. When K CO was used
Org. Lett., Vol. 6, No. 2, 2004
263

suspension into single cells. Samples (1.0 mL) placed in wells
of a 24-well polystyrene microtiter plate contained light to
heavy cell concentrations (50-1000 K per well) and one of
intercellular cohesion to compete with intracellular looping.
17
Rigid spacers would have, in retrospect, favored the former.
Despite our apparent setback, the concept of a cancer-net
drug, in addition to the design and synthesis of polyglyco-
sylated bolaforms that address the concept (if not actually
manifesting it), clears the way for future work in the area.
-
4
-2
our compounds (1.5 × 10 to 5 × 10 mM). Exposure of
the cells to 1, with and without gentle shaking, or with and
2+
without the presence of Ca , was allowed to proceed for
-30 min at room temperature. Subsequent inspection of
5
the cell suspensions by light microscope and light scattering
gave no convincing evidence for aggregation. Since the work
of Aoyama et al. with simple analogues of 1 affirmed the
Acknowledgment. This work was supported by the
National Institutes of Health. We thank Prof. Aoyama for
his assistance in acquiring the cancer cells.
5
presence of binding to the identical cell line, lack of cell
flocculation may have more to do with spacer length than
with binding equilibria. Spacer lengths must, of course, allow
Supporting Information Available: Synthetic procedures
for the compounds in Scheme 1. This material is available
free of charge via the Internet at http://pubs.acs.org.
(17) Schwab, P. F. H.; Levin, M. D.; Michl, J. Chem. ReV. 1999, 99,
1
863.
OL030135K
264
Org. Lett., Vol. 6, No. 2, 2004