S. B. A. Halkes et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1567–1570
1569
in the series of 7, 5, and gallic acid. However, these dif-
ferences can be totally accounted for by the number of
galloyl moieties per molecule and the relative size of the
backbone to which these galloyls are attached.16
Netherlands Organization for International Coopera-
tion in Higher Education (NUFFIC) for a Huygens
fellowship. The Royal Netherlands Academy of Arts
and Sciences (KNAW) is gratefully acknowledged for a
fellowship to R.J.P.
Effects on cross-linking of collagen were assessed
according to the method of Heijmen et al.17 (Table 1).
Compounds were dissolved in 85% glycerol and incu-
bated for 48 h at room temperature with purified skin
collagen matrices. The degree of collagen cross-linking
was determined by the hydrothermal shrinkage tem-
perature (Ts). The Ts denotes the change in molecular
conformation of collagen from triple helix to random
coil and will be increased when cross-linking has taken
place.18 The strongest increase in Ts was observed for
the collagen matrices pretreated with tannic acid. For 5,
elevation of the Ts was less pronounced but not sig-
nificantly different from tannic acid. In comparison, the
cross-linking capacity of 7 and 9 lagged behind, the Ts
of collagen matrices pretreated with 9 not even being
distinct from the controls without tannins/dendrimers.
These results fit within the concept for tannic acid–col-
lagen interactions as proposed by Haslam.19 In this
model, pentagalloylglucose as the prototype molecule
representative for tannic acid, has in its most favored
conformation a disc-like shape with a thickness of
approximately 0.7 nm and a diameter of approximately
2.1 nm. These dimensions are ideally suited to infiltrate
in the gaps between the tropocollagen molecules that
compose the collagen fibrils. Hydrophobic effects are
thought to initiate the association with hydrogen bond-
ing acting to reinforce the structure. The decline in col-
lagen cross-linking capacity going from the first- to
third-generation dendrimers is consistent with the disc
model. Bis-galloyl compound 5 is more likely to adopt a
flat conformation than the more three-dimensional 7
and 9. Alternatively, spatial contraints may simply not
allow the accommodation of the significantly larger
structures 7 and 9, as compared to 5 or tannic acid.
Furthermore the difference in hydrophobicity between
the galloyl dendrimers cannot be ruled out as a deter-
mining factor either considering the importance of
hydrophobicity in the initiation of the association.
References and Notes
1. (a) Haslam, E. In Plant Polyphenols—Vegetable Tannins
Revisited; Cambridge University Press: Cambridge, 1989. (b)
Nishizawa, M.; Yamagishi, T.; Nonaka, G.-I.; Nishioka, I.
J. Chem. Soc., Perkin Trans. 1 1982, 2963.
2. Halkes, S. B. A.; Van den Berg, A. J. J.; Hoekstra, M. J.;
Du Pont, J. S.; Kreis, R. W. Wounds 2001, 13, 144.
3. Hoekstra, M. J.; Dutrieux, R. P.; Kreis, R. W. Tannin
Ointment Alternated with Silversulfadiazine 1% Cream Versus
Tannin Ointment Alternated with Silversulfadiazine 1% Liquid.
Pre-clinical investigation report, no: 92871008. Burns Research
Institute: Beverwijk, The Netherlands, 1992.
4. Kreis, R. W.; Vloemans, A. F. P. M. Tannin Ointment.
Clinical Pilot Trial Report; Burns Research Institute: Bev-
erwijk, The Netherlands, 1992.
5. Delahaye, P.; Verzele, M. J. Chromatogr. 1983, 265, 363.
6. Halkes, S. B. A.; Van den Berg, A. J. J.; Hoekstra, M. J.;
Du Pont, J. S.; Kreis, R. W. Burns 2001, 27, 299.
7. (a) Mulders, S. J. E.; Brouwer, A. J.; Liskamp, R. M. J.
Tetrahedron Lett. 1997, 38, 3085. (b) Mulders, S. J. E.;
Brouwer, A. J.; van der Meer, P. G. J.; Liskamp, R. M. J.
Tetrahedron Lett. 1997, 38, 631. (c) Brouwer, A. J.; Mulders,
S. J. E.; Liskamp, R. M. J. Eur. J. Org. Chem. 2001, 1903.
8. For a review on the potential benefits of multivalency, see:
Mammen, M.; Chio, S.-K.; Whitesides, G. M. Angew. Chem.,
Int. Ed. 1998, 37, 2755.
9. (a) Vrasidas, I.; de Mol, N. J.; Liskamp, R. M. J.; Pieters,
R. J. Eur. J. Org. Chem. 2001, 4685. (b) Andre, S.; Pieters,
R. J.; Vrasidas, I.; Kaltner, H.; Kuwabara, I.; Liu, F.-T.; Lis-
kamp, R. M. J.; Gabius, H.-J. ChemBioChem. 2001, 2, 822.
10. BOP=1H-benzotriazol-1-yloxy)tris(dimethylamino) phos-
phonium hexafluorophosphate).
11. 13C NMR data, see ref 9a for numbering of the dendrimer
backbone, Gal=galloyl moiety. 5: 13C NMR (CD3OD,
75.4 MHz) d 171.7, 170.8 (C¼OI, II), 161.1 (C3/5), 146.6 (Gal-
C3/5), 138.1, 136.8 (Gal-C4, C1), 126.0 (Gal-C1), 109.2 (C1),
107.9 (Gal-C2/6), 106.3 (C4), 67.7 (CH2a), 40.5 (CHb2); ESI-MS:
m/z=545.1 [M + H]+ (100%), 567.0 [M + Na]+; 7: 13C
NMR (CD3OD, 75.4 MHz) d 170.8, 170.2 (C¼OI, II, III), 161.3,
0
0
161.1 (C3/5, C3 /5 ), 146.7 (Gal-C3/5), 138.1, 137.5 (Gal-C40, C1,
As shown in this paper, the synthetic analogues 5, 7,
and 9, retain a satisfactory, albeit lower, biological
activity in comparison to tannic acid, but have the
advantage of a much enhanced stability. Therefore,
polygalloyl-dendrimers may serve as potential leads for
the development of new topical drugs to be used in burn
wound treatment. Considering the importance of both
the number of galloyl moieties and possibly molecular
size for the biological activity, it might finally be of
interest to search for other synthetic analogues of tannic
acid in which the backbone is reduced in dimension but
still can accomodate a maximal load of gallic acid residues.
0
0
C1 ), 126.0 (Gal-C1), 109.4, 107.2, 106.1, 105.7 (C2/6, C2 /6 , C4,
a0
0
C4 ), 107.9 0 (Gal-C2/6), 67.8 (CH2 ), 67.5 (CH2a), 40.8 (CHb2),
+
40.5 (CHb2 ); MALDI-TOF-MS: m/z=1293 [M H]+
(100%); 9: 13C NMR (DMSO-d6, 75.4 MHz)0 d 167.0, 166.2
00 00
0
(C¼OI,
IV), 159.8 (C3 /5 ), 159.7 (C3 /5 ), 159.4 (C3/5),
II, III,
10
100
145.8 (C3/5), 136.7, 136.6, 136.5 (Gal-C4, C01, 0 C ,00C ), 124.8
00
0
(Gal-C1), 107.8, 106.2, 104.4, 104.1 (C2/6, C2 /6 , C2 /6 , C4, C4 ,
a00
a0
00
C4 ), 107.1 (Gal-C2/6), 66.6 (CH2 ), 66.5 (CH2 ), 66.3 (CHa2),
0
00
(CH2b,b ,b signals obscured by DMSO-d6 signals).
12. Armitage, R.; Bayliss, G. S.; Gramshaw, J. W.; Haslam,
E.; Haworth, R. D.; Jones, K.; Rogers, H. J.; Searle, T. J.
Chem. Soc. 1961, 1842.
13. Beasley, T. H.; Ziegler, H. W.; Bell, A. D. Anal. Chem.
1977, 49, 238. HPLC analysis was performed on a Gilson
chromatographic system (Gilson 360 pumps combined with a
Gilson 805 manometric module, a Gilson 811C dynamic
mixer, a Gilson 234 autoinjector, and a Gilson 170 diode-
array-detector) equipped with an Alltima Silica 5U column,
250 Â 4.6 mm (Alltech, Deerfield, IL, USA; Cat. No. 88127).
Solvent system: mobile phase A: hexane with 0.5%
Acknowledgements
S.B.A.H. is financially supported by ‘Stichting ACH-
MEA Slachtoffer en Samenleving’, Zeist and the Dutch
Burn Foundation, Beverwijk. I.V. is grateful to the