Biology Reference
In-Depth Information
accounts for 10-30% of the total carbon source provided in unstressed conditions, this
may set a limit to the accuracy of the growth yield predictions.
To enhance the usefulness and predictiveness of the model, several avenues could
be followed in the future. First, additional constraints can be overlaid on the network
to reduce the space of possibilities and increase the accuracy of predictions. In addi-
tion to specifi c knowledge of particular enzymatic or transport processes, such con-
straints are best based on high-throughput experimental evidence such as transcrip-
tomic and proteomic data, which are instrumental in expanding genotype-phenotype
relationships in the context of genome-scale metabolic models [67]. Microarray ex-
periments have guided the discovery of metabolic regulons, and usage of microarray
and proteomic data to constrain metabolic models has improved model accuracy for
other systems [23]. Second, P. putida provides a good opportunity for incorporating
kinetic information into a genome-scale model as there are various kinetic models
available and under development for small circuits in P. putida [68-71]. Incorporating
data from these models into the genome-scale reconstruction would provide insights
into the relationships of isolated metabolic subsystems within the global metabolism.
This synthesis would also improve the fl ux predictions of the global model, particu-
larly in areas where current FBA-based predictions methods fail due to their inherent
limitations.
Experimental validation of a genome-scale model is an iterative process that is
performed continuously as a model is refi ned and improved through novel informa-
tion and validation rounds. In this work, we have globally validated iJP815 as well as
specifi c parts thereof by using both up-to-date publicly available data and data gener-
ated in our lab, but there will be always parts of the model that include blocked reac-
tions and pathways that will require further, specifi c validation. As more knowledge
becomes available from the joint efforts of the large P. putida community (e.g., [http://
www.psysmo.org]), focus will be put on these low-knowledge areas for future experi-
mental endeavors. We anticipate that this model will be of valuable assistance to those
efforts.
The metabolic reconstruction, the subsequent mathematical computation and the
experimental validation reported here provide a sound framework to explore the meta-
bolic capabilities of this versatile bacterium, thereby yielding valuable insights into
the genotype-phenotype relationships governing its metabolism and contributing to
our ability to exploit the biotechnological potential of pseudomonads. By providing
the means to examine all aspects of metabolism, an iterative modeling process can
generate logical hypotheses and identify conditions (such as regulatory events or con-
ditional expression of cellular functions) that would reconcile disagreements between
experimental observations and simulation results. Through a detailed in silico analysis
of PHA production, we show how central metabolic precursors of a compound of
interest not directly coupled to the organism's growth function might be increased via
modifi cation of global fl ux patterns. Furthermore, as the species Pseudomonas putida
encompasses strains with a wide range of metabolic features and numerous isolates
with unique phenotypes, the reconstruction presented provides a basic scaffold upon
which future models of other P. putida strains can be built with the addition or subtrac-
 
Search WWH ::




Custom Search