Genomescale Metabolic Network Reconstruction

1
Genome-scale Metabolic Network Reconstruction
2
Elementary flux modessystem boundaries /
external metabolites to be defined
Exchange flux
A
B
C
E
SystemBoundary
D
Internal flux
Flux The production or consumption of mass per
unit area per unit time.
3
Elementary (flux) modes

• Schuster Hilgetag (1994)

4
An elementary mode is a minimal set of enzymes
that can operate at steady state with all
irreversible reactions used in the appropriate
direction
All flux distributions in the living cell are
non-negative linear combinations of elementary
modes
Related concept Extreme pathway (C.H. Schilling,
D. Letscher and B.O. Palsson, J. theor. Biol.
203 (2000) 229) - distinction between internal
and exchange reactions, all internal reversible
reactions are split up into forward and reverse
steps
5
The Stoichiometric Matrix An Example
6
Spanning the Null Space
A set of basis vectors spanning the null space
may be found. This set is not unique. Each vector
represents a pathway in the metabolic network by
specifying the set of fluxes in the system. Are
these pathways biologically feasible ?
7
Spanning the Convex Cone
Basis vectors must be constrained by flux
directionality These constraints define a
convex flux cone emanating from the origin, which
defines the complete range of biochemically
feasible flux distributions. Any vector can be
represented asa non-negative combination of
thegenerating vectors of the cone, fk
8
Extreme Pathways
The generating vectors of the flux cone are a
biochemically feasible basis to the distribution
of pathways in the network.
9
Extreme Pathways
10
Elementary Flux Modes
When the network includes reversible reactions,
the extreme pathway set may not include all
elementary pathways
Elementary flux modes are the set of irreducible
pathways spanning the solution space.
11
Elementary Flux Modesvs Extreme Pathways
Extreme pathways Elementary flux
modes
12
Properties of Elementary Flux Modes Extreme
Pathways

• If e is a EFM or EP then it maintains the
  following
• Pseudo steady state (metabolite balancing
  equations)
• Feasibility flux directionality constraints are
  satisfied
• Non-decomposability There is no vector v
  satisfying 12 s.t. p(v) is a proper subset of
  p(e).
• In addition
• The set of EFMs and set of EPs are uniquely
  defined
• The set of EPs is a subset of the set of EFMs
• The set of EPs are systematically independent.

13
P
S
4
3
non-elementary flux mode
1
1
P
1
3
S
S
1
2
2
2
S
4
P
P
2
1
P
S
4
3
1
1
P
3
1
S
S
S
S
1
2
1
2
1
1
1
1
S
S
4
4
P
P
P
P
2
1
2
1
elementary flux modes
S. Schuster et al. J. Biol. Syst. 2 (1994)
165-182 Trends Biotechnol. 17 (1999) 53-60
Nature Biotechnol. 18 (2000) 326-332
14
Biochemical ApplicationsCan sugars be produced
from lipids?

• Known in biochemistry for a long time that many
  bacteria and plants can produce sugars from
  lipids (via C2 units) while animals cannot

AcCoA is linked with glucose by a chain of
reactions. However, no elementary mode realizes
this conversion in the absence of the glyoxylate
shunt.
15
Elementary mode representing conversion of AcCoA
into glucose. It requires the glyoxylate shunt.
The glyoxylate shunt is present in green plants
and many bacteria (e.g. E. coli). This example
shows that a description by usual graphs in the
sense of graph theory is insufficien.
16
Biological Network Example
Reaction scheme representing part of
monosaccharide metabolism
17
Elementary Flux Modes of Monosaccharide Metabolism
Basic glycolitic pathway
Degradation of G6P to pyruvate and CO2 producing
ATP, NADPH and NADH
18
Elementary Flux Modes of Monosaccharide Metabolism
Conversion of G6P to ribose-5-phosphate and CO2
Conversion of 5 hexoses to 6 pentoses (when need
for R5P is high)
pentose phosphate cycle carbons are cycled
several times before ending in CO2. Produces
NADPH but not NADH and ATP
19
EFM theoretical predictions
Glucose
Red elementary mode Usual TCA cycle
CO2
AcCoA
Pyr
PEP
Cit
Oxac
IsoCit
Gly
Mal
CO2
OG
Fum
Succ
CO2
SucCoA
20
Hungry between optimal growth and
starvation, can be studied in glucose-limited
continuous (chemostat) cultures with very low
glucose concentrations at a rate of growth that
is controlled by the experimenter. Catabolite
repression is absent
21
Physiological relevance Production of NADH
instead of NADPH
22
Basis of 13C metabolic flux analysis
determination of intracellular fluxes
Cell
Intracellular fluxes
Carbon Source 13C1-C2-C3-C4-C5-C6 (labeled)
Glycolysis
v1
v3
v2
TCA Cycle
Measurable extracellular fluxes
Amino Acids
Measurable carbon labeling pattern of isotopomers
23
Determine the isotopomer distribution in key
metabolites
Isotopic labeling of proteinogenic amino acids is
reflective of their precursors in central
metabolism.
24
(No Transcript)
25
Example labeling with pyruvate
Succinylase pathway
Dehydrogenasepathway
26
Ion fragments of derivatized glutamate from cells
incubated with 1, 2 13C2-glucose
C1
C1
C2
80000
C2
198
75000
C3
C3
70000
152
C4
65000
C4
60000
C5
C5
55000
50000
Abundance
P3 P1
45000
40000
M1
P3
P2 or P1
35000
M2
M1
30000
25000
M2
20000
15000
P2
10000
5000
0
145
150
155
160
165
170
175
180
185
190
195
200
205
210
M/Z
27
Journal publications on 13C flux analysis
13C metabolic flux analysis
metabolic flux analysis
Key words Total
Papers Review Earliest
publication 13C Metabolic flux analysis 164
14 1983
Metabolic flux analysis 1664
155 1960s
DNA microarray 23474
2718 1995
Pubmed query
28
Challenge 1 Achieving steady state
Continuous fermentation best for flux analysis,
but very expensive. Shaking flask approximate
approach growth condition is not stable.
Isotopomer distribution curves in amino acids
(shaking flask culture) Shewanella ( gly ?
Ser ? Ala) E.coli growth (? gly ? Ser ?
Ala)
Mini-bioreactor high throughput low cost for
labeled medium (10mL) controlled growth
conditions.
29
Challenge 2 Minimal medium limits the 13C flux
analysis
Addition of nutrients (amino acids) complicates
the isotopomer analysis!!!
Only Ala, Asp, and Glu (?) Labeling was not
affected by addition of amino acids

This makes flux analysis for mammalian cells
difficult !
Effect of addition of 17 non-labeled amino acid
mix (25 µM each) on labeling pattern in
Shewanella proteinogenic amino acids
30
Challenge 3 measurement of metabolites
Measure isotopomer distribution
Measure metabolites Concentrations

• NMR
• Lower sensitivity mM mM

• GC/MS
• Need to derivatize compounds
• Provide total mass and a-carbon
• labeling information

CE-MS(MS) /LC-MS(MS)
CE-FT-ICR
Carbon balance for yeast metabolism