Dr. Eduardo Mendoza

1
Lecture 8
Yeast Glycolysis Canonical Models

• Dr. Eduardo Mendoza
• Oct 22, 2003 Physics Department
• Mathematics Department Center for
  NanoScience
• University of the Philippines
  Ludwig-Maximilians-University
• Diliman Munich, Germany
• eduardom_at_math.upd.edu.ph
  Eduardo.Mendoza_at_physik.uni-muenchen.de

2
Topics to be covered

• 8.1 Features of the Fermentation Pathway
• 8.2 Definition of Variables
• 8.3 Setting Up the GMA Equations
• 8.4 Deriving the S-System Model
• 8.5 Heat Shock Model

3
8.1 Features of the fermentation pathway

• Yeast Glycolysis Background
• Yeast are unicellular fungi that are versatile
  laboratory microorganisms. They grow rapidly and
  have simple nutritional requirements. When yeast
  degrade nutrients in the absence of oxygen, they
  use the process of glycolysis to produce energy
  in the form of ATP.
• For millennia, humans have used the alcoholic
  fermentation capability of yeast of the genus
  Saccharomyces to produce breads, crackers and a
  variety of fermented beverages including beer and
  wine. The general equation for the fermentation
  reaction is
• Substrate Glycolytic Enzymes ? Ethyl Alcohol
  (C2H5OH) CO2 ATP

4
References

• VOTO00 E. Voit, N. Torres Darias Canonical
  Modeling of Complex Pathways in Biotechnology,

5
7.1 Evolution of the model VOTO00
6
Evolution of the model (2)
7
Key processes in Yeast Glycolysis (1)
8
Key processes in Yeast Glycolysis (2a)
9
Key processes in Yeast Glycolysis (2b)
10
Key processes in Yeast Glycolysis (3)
11
Key processes in Yeast Glycolysis (4)
12
Key processes in Yeast Glycolysis (5)
13
Key processes in Yeast Glycolysis (6)

• In addition to the above reactions of the main
  pathway, several exchanges occur between ADP and
  ATP levels

14
7.2 Definition of variables (1)

• What about ADP, ATP, NAD ?
• There is no generally applicable, perfect
  solutionit will depend on focus of analysis.

15
Definition of variables (2)
16
Definition of variables (3)
17
Overview of model variables
Observed steady-state concentrations
18
7.3 Setting up the GMA model (1)
19
Setting up the GMA model (2)
20
Setting up the GMA model (3)
21
7.4 The S-System Model
22
Yeast Model in PLAS

• Additional syntax
• THE DECLARATION
• In sensitivity analysis, the concept of
  "external" or "independent" variable is often
  used, in which case the sensitivities relative to
  these parameters are called "logarithmic gains".
• In PLAS external variables are declared as
  constants but if any of them appears in a list
  following a double "", then they will be
  considered external variables for the purpose of
  sensitivity analysis.

23
Further references

• E. Voit, T. Radivoyevitch Biochemical systems
  analysis of genome-wide expression data,
  Bioinformatics 11 (2000)
• E. Voit Metabolic modeling a tool of drug
  discovery in the post-genomic era, Drug Discovery
  Today, 7 (2002)

24
Yeast Fermentation (Glycolysis) Model

• Based on work by Galazzo Bailey (1991) and
  Cascante, Curto Sorribas (1995)
• Used by Torres et al (1997) to illustrate methods
  of flux optimization in a biotechnological
  setting
• Used to analyze and explain gene expression data
  for heat schock (same data used by Eisen et al
  1998, Toh Horimoto 2002)

25
Data used

• Schena et al, 1995 expression levels for 2000
  genes at 0,10,20,40,80 160 minutes after
  transition from 25 to 37oC
• http//rana.Stanford.EDU/clustering/Figure2.txt
• Wodicka et al 1997, Holstege et al 1998 baseline
  mRNA expression levels and transcription rates
• http//gaiberg.wi.mit.edu/cgi-bin/young_public/li
  sts.cgi?typeH

26
Model Extension

• Inclusion of G6PDH branch
• G6PDH oxidizes G6P to 6-phosphogluconate
• G6PDH reduces to NADP to NADPH, the latter being
  important in defense against oxidative stress
• NADPH is further needed in sphingolipid
  metabolism, which is important in heat shock
  response

27
Results of analysis

• Observed heat schock profile is not intuitive
  (Voit)
• Observed profile satisfies primary goals of
• Increased ATP, trehalose NADPH production
• Maintaining intermediate metabolites at
  reasonable levels
• Systematic exploration of alternative,
  hypothetical expression profiles ? observed
  profile outperforms other profiles

28
Methods (1)

• Confirmation of steady-state concentrations and
  fluxes published
• Enzyme activities specified in accordance with
  the observed gene expression profile (VORA00,
  Table 1)
• Hypothetical profiles implemented
• single enzyme catalyzes a reaction ? observed
  change in gene expression corresponds to change
  in enzyme activity
• Two or enzymes catalyze ? importance weighted by
  corresponding numbers of mRNA copies/cell
• Factor in effects at transcription level

29
Methods (2)

• Performance metric for profiles based on the
  following criteria
• (primary) sufficient production of ATP, trehalose
  and NADPH
• (secondary) unneeded accumulation of
  intermediates, which would strain the cells
  solubility capacity
• (secondary) cost of overexpression

30
Methods (3)

• Primary summands
• ln ATP fluxhypo/ATP fluxhs
• similar expressions for trehalose, NADPH
• Secondary summands
• Cost of overexpressing gene or inducing enzyme
• - ln expr-levelhypo/expr-levelbasel
• Deviations of intermediate metabolites
• - abs(ln Concentrhypo/Concentrhs
• Examples of profiles explored (s. paper)

31
Thanks for your attention !

• Questions?