and non-reducing conditions as un-treated VEGF (Supporting
Information Fig. 1). This strongly suggests that covalently
linked VEGF-polyphenol adduct(s) were not formed and that
binding between VEGF and the two polyphenols was the result
of (strong) non-covalent binding. We also attempted to
determine if there were differences in the pI of polyphenoltreated
VEGF and un-treated VEGF but the high intrinsic
pI of VEGF (pI = 8.0–8.5) [25] precluded us from visualising
VEGF or putative VEGF adducts on the gels, presumably
because the VEGF migrated off the top (cathodic) edge
of the gels (data not shown). We also analysed VEGF and
polyphenol-treated VEGF using MALDI-TOF MS to try and
detect changes in the mass of VEGF after polyphenol treatment.
However, the mass spectra obtained for VEGF and
polyphenol-treated VEGF were very similar and no modifications
were observed (data not shown). The lack of observed increases
in the mass of VEGF post-treatment with the polyphenols
is not consistent with the formation of covalent bonds
between the VEGF and the polyphenols, and it likely indicates
that the non-covalent binding involved in complex formation
is disrupted during ionisation such that only masses
corresponding with free VEGF are observed in the MALDITOF
analysis. The combination of results obtained from
SDS-PAGE and MALDI-TOF MS do not support the notion
that the binding between VEGF and polyphenol is due to a covalent
interaction. Bearing inmind that VEGF activity was not
recovered from polyphenol-treated VEGF samples following
dialysis, it ismost likely that the polyphenol-mediated inhibition
of VEGF is the result of strong non-covalent interactions
between VEGF and the polyphenols.