55primary metabolism, like sugars (monosaccharides and oligosaccharides), organic
acids, amino acids, phosphate compounds, and amines, providing thus an extensive
perspective of the primary (central carbon) metabolism. The primary metabolism is
indicative of the energy, redox homeostasis, and growth demands of the plant cells,
while it produces all the precursors of the cellular macromolecules and secondary
metabolites. The changes in the primary metabolism reflect core perturbations in
the metabolic physiology of the plants. Conserved among species metabolic re-
sponses to environmental stress acclimation should be observable within the prima-
ry metabolism. On the other hand, the secondary metabolism is more diverse among
species and presumably reflects the successful adaptation of a species to particular
environmental stresses through the acquisition of novel biosynthetic capacities of
its primary metabolism (Sanchez et al. 2008a). Therefore, even if there are differ-
ences in the secondary metabolism, they can be inferred from the changes in the
concentration profile of precursor molecules in the primary metabolism.
A major advantage of the high-throughput biomolecular analyses is that by ob-
serving a large number of molecular quantities at the same time, correlations be-
tween the activity of various molecular pathways can be determined, new knowl-
edge can be extracted, and the biomolecular networks at different levels of cellular
function (e.g., gene regulation, protein interaction, or metabolic networks) can be
reconstructed. To this end, we are in search of multi-compound biomarker profiles
and patterns of expression, rather than single molecules that can be sensitive sensors
of changes in the physiology of the plants. Thus, the acquired datasets have to be
analyzed with multivariate statistical methods, attempting to identify either clusters
of genes or gene products that have similar expression or concentration, respec-
tively, profiles among various physiological conditions, or physiological states that
are of similar omic profiles. For this purpose, clustering, e.g., hierarchical clustering
(HCL), analysis, and dataset dimension reduction and visualization, e.g., principal
component analysis (PCA), methods are used. Customized multivariate significance
analysis methods for omic data, like significance analysis for microarrays (SAM)
(Tusher et al. 2001 ), have been developed enabling the identification of the genes or
gene products, the change in the expression or concentration, respectively, of which
is characteristic of the difference between two sets of physiological conditions. In
the case of time-series experiments, particular modifications of PCA (Scholz et al.
2005 ) and SAM analyses (Dutta et al. 2007 ) have been proposed to take into con-
sideration that the physiological states of the plants at the different time points are
not independent, but rather part of the same physiological history.
3.4 Omic Analyses of Salinity Stress on Plants
3.4.1 Metabolomic Analyses
The effect of salt stress on plant metabolic physiology using metabolomic analyti-
cal platforms has been investigated in the context of maize (Gavaghan et al. 2011 ),
3 Investigating the Effect of Elevated CO 2 in the Growth Environment ...