Biology Reference
In-Depth Information
Hevea brasiliensis
latex [
69
] and the cytoplasmic proteome from
maize [
69
], and more recently spinach leaves [
71
]. At the time of
writing, there were no publications describing the use of
ProteoMiner for analysis of seed proteins. Our attempts to use
ProteoMiner kit, under the conditions developed for spinach
leaves, to either deplete samples from developing soybeans of the
abundant SSP or to prepare samples that yielded an increased
number of protein ID's have been without success to date (JAM,
unpublished). The method is known to be sensitive to pH and
solvent characteristics, and will likely have to be customized for
individual applications. Furthermore, we speculate that the often
unusual solubility of SSP might be problematic in terms of nonspe-
cifi c adsorption.
4
The Future of Seed Proteomics
The short-term future of seed proteomics will feature more and
better protein IDs. This goal will be achieved through combined
advances in instrumentation [
72
], bioinformatic analyses and
methods for protein identifi cation [
73
], improved databases (e.g.,
phytozome.net/), and, of course, improvements in the methods
used to deplete the supra-abundant SSP from input samples. In the
latter case, the strategy of using combinatorial-ligand random-
peptide beads appears to have substantial potential. One obvious
improvement would be to remove the “random” component. In
part because of their abundance, genes and cDNAs for SSP were
among the fi rst sequenced at the beginning of the genomics era
[
3
]. Because the sequences of many SSP are extant [
74
-
76
], it
should be relatively simple to prepare “designer sequence”—beads
that would have both the capacity and specifi city to effi ciently
remove these supra-abundant components of the seed proteome.
This strategy should additionally be amenable to incorporation
into high-throughput experimental designs.
Another obvious need is for more comparative analyses, espe-
cially in terms of gymnosperm seeds, although in many cases this
will require parallel development of better genomic/EST resources
to facilitate protein identifi cation.
There can be no debate about the need to move from qualita-
tive to quantitative proteomic analyses, and there are proponents
of both protein labelling and of label-free methods to achieve this
end [
14
]. While the debate will likely continue into the future, an
increasing number of researchers are using a simple, inexpensive
label-free method termed spectral counting [
77
-
79
]. Regardless of
the method ultimately adopted, it is clear that future proteomics-
based studies of seeds can only benefi t from employing quantita-
tive approaches.
4.1 More, Better,
and Quantitative
