In the previous chapter, numerical data obtained by mass spectrometric analyses were translated into heat maps to easily visualise regions or metabolites that show differential behaviour when cells were cultured on each substrate (plain, 2 kPa F2/S, 6 kPa F2/S and 38 kPa F2/S). The heat maps and volcano plots were investigated in conjunction with one another in order to choose metabolites that may be of further interest, that is, metabolites that showed unique behaviour to a particular substrate. Although the heat maps give an insight into the abundance of metabolites the volcano plots serve as an added confidence, singling out molecules that are also of statistical significance (Figure 4-1). To reduce the search size, selected metabolites were taken from those identified as being involved with lipid metabolism.
While it is acknowledged that metabolomics carried out for assessment of the lipid population using HILIC is far from ideal, indeed a lipidomics approach optimising methods toward lipid detection is advisable were it the only facet of interest. The presently used HILIC technique however, is not necessarily a method that is considered entirely devoid of information on cellular lipids. In this vein, metabolite masses putatively identified as being involved in lipid metabolism were chosen for further investigation for a number of reasons. Samples within the lipid metabolism pathway showed the most amount of change on differentiation compared to the other investigated pathways. It also gave the most visible differential change between the F2/S substrates when compared to the other pathways for which the most contrasting change was noted only for the 2 kPa F2/S hydrogel. Lastly, within cell biology generally, functional changes are widely researched with regards to the genomic and proteomic changes. Relative to these, limited information is available about the role of lipids as regulators of cell behaviour and function as is highlighted in the previous chapter.
It therefore stands to reason that this facet of cell biology be explored further as the metabolomics study allows the means for doing so.
Due to the limitations associated with performing a global metabolomics study as discussed in the previous chapter, a number of precautions were taken when selecting metabolites for further study which also contributed to downsizing the search field:
a) The confidence value: the measure of confidence makes use of an arbitrary value between 1 and 10 (10 indicating the highest confidence) based on the retention time drift of the measured mass from the calculated/predicted retention when performing metabolite identification in Ideom. Selection was restricted to metabolites with a confidence factor of 6 and above.
b) The number of putative metabolites identified for a chosen mass from database cross-referencing is kept as low as possible. This reduces the risk of choosing false positives from metabolites with similar masses that could not be distinguished through chromatographic separation. It is acknowledged however, that this condition applies to the confines of the analytical tools used in this thesis. That is, metabolite identification of noted isomers is restricted to the scope of the databases that Ideom uses for identification.
c) Metabolite(s) that showed a significant decrease in abundance that was unique to a particular substrate. It also had to be detectable on the plain surfaces (control sample) indicating its presence within the cells when it is in a ‘resting’ state and showing depletion when the substrate properties are changed, i.e, cultured on hydrogel surfaces.
d) Other factors out with metabolite mass detection and behaviour such as commercial availability and current research done on each compound, which may give an inclination on the likelihood to affect cell behaviour (supporting the hypothesis), were also taken into consideration.
Using the aforementioned criteria, the metabolites cholesterol sulphate (CS), 1- octadecanoyl-sn-glycero-3-phosphate (GP18:0) and sphinganine were singled out for further investigation (Figure 4-2).
Figure 4-1 Simplified schematic illustrating metabolite selection process. Metabolite chosen for further investigation were LC-MS metabolites that had been putatively identified as being involved in lipid metabolism. Selections were whittled down based on a number of defined criterion inclusive of measured abundances singled out by cluster analysis and statistical significance as determined from the volcano plots.
Figure 4-2 Average peak intensities of metabolites isolated for further investigation.
Metabolites were chosen from those identified as being involved with lipid metabolism. An initial condition for selection was that each metabolite showed considerable depletion on a singular substrate compared to the remaining three biomaterials suggesting it may play a functional role that occurs uniquely in cells cultured on that particular substrate. Error bars denote standard deviations from the mean; n = 3 replicates; * notes statistical significance to the total population where p < 0.05, ** where p < 0.01 and *** where p < 0.001 calculated using one way ANOVA.