Computational Systems Biology

Computational Systems Biology

Lecture 8: Signal Transduction

Tue 21 Feb 2006 with the collaboration of Luna De Ferrari

Images from: D. L. Nelson, Lehninger Principles of Biochemistry, IV Edition Chapter 12 E. Klipp, Systems Biology in Practice, Wiley-VCH, 2005 Chapter 6

Computational Systems Biology

Summary:
Introduction on cell signalling
Metabolism vs. Signal transduction

The signalling paradigm and some typical components:

The receptor G protein MAP kinase cascade

Computational Systems Biology

Cell Signalling

Computational Systems Biology

Signal Transduction
The cell senses extracellular signals:
Hormones, pheromones, heat, cold, light, osmotic pressure, concentration change of glucose, K+, Ca2+ or cAMP.

and commutes them in intracellular signals:

Signaling involves the same type of molecular modification as metabolism: production and degradation of substances, phosphorylation, activation of inhibition of reactions

Whats the difference then?

4

Computational Systems Biology

Metabolism vs. Signal Transduction

Metabolism
Provides mass transfer Quantity of converted material: M or mM A metabolic network is determined by the present set of enzymes The catalyst to substrate ratio is low (quasi-steadystate assumption in Michaelis-Menten kinetics)

Signal Transduction
Provides information transfer Quantities: 10 to 104 molecules per cell A signal pathway may assemble dynamically

Amount of catalyst and substrate in the same order of magnitude

Computational Systems Biology

Characteristics of Signal Transduction

From: D. L. Nelson, Lehninger Principles of Biochemistry, IV Edition Chapter 12 Biosignaling

Computational Systems Biology

Signalling Paradigm: the Receptor

Computational Systems Biology

Receptor
From: E. Klipp, Systems Biology in Practice, 2005 Chapter 6

Computational Systems Biology

Simple receptor activation

The simplest mechanism is the reversible binding of the ligand L to the receptor R to form the complex RL

R + L = RL

From: E. Klipp, Systems Biology in Practice, 2005 Chapter 6

Computational Systems Biology

Scatchard Analysis
The receptor-ligand binding, like the enzyme-substrate binding, depends on the concentration of the interacting components The resulting equilibrium can be described in term of Ka (the association constant) or Kd (the dissociation constant)

R+L

+ k 1 k- 1

RL

[RL] k + 1 1 = = Ka = [R][L] k 1 Kd
10

Computational Systems Biology

A more realistic Receptor activation

Cells can regulate the activity of their receptors, e.g. to weaken the signal transmission during long term stimulation A classic mechanism is the phosphorylation of the receptor (on a serine, threonine or tyrosine aminoacid) on the cytosolic domain Hence, a more realistic depiction should include not only the active and bound receptor, but:
Ri = the inactivated receptor (cannot be activated) Rs = the susceptible receptor (can be activated) Ra = the activated receptor (bound to the ligand)
11

Computational Systems Biology

A more realistic Receptor activation (1):

the model
From: E. Klipp, Systems Biology in Practice, 2005 Chapter 6

12

Computational Systems Biology

A more realistic Receptor activation (2):

kinetic equations
The receptor is present either in the inactive form (Ri) or susceptible form (Rs) Rs can interact with the ligand L to give the active form Ra The inactive or susceptible form is produced from precursors with rates vpi or vps All three forms can be degraded with rates vdi, vds, vda

13

d Ri = vpi vdi vis + vsi + vai dt d Rs = vps vdi vds + vis vsi vsa + vas dt d Ra = vda + vsa vas vai dt

Computational Systems Biology

A more realistic Receptor activation (3):

rate approximation
The degradation terms are assumed to be linearly dependent on the concentration of substrates (in this case the receptor R) (here the asterisks means any species, either i, s or a):

vd* = kd * R *
Also the state changes can be approximated linearly as first guess: is is i

v = k R

The receptor activation is dependent on ligand concentration. A linear approximation is:

vsa = ksa Rs L
14

Computational Systems Biology

A practical example

15

Computational Systems Biology

and its time course

16

Computational Systems Biology

Signaling Paradigm: the structural components G protein cycle

MAP kinase cascade

17

Computational Systems Biology

18
From: E. Klipp, Systems Biology in Practice, 2005 Chapter 6

G protein

Computational Systems Biology

Guanosine nucleotide-binding protein

(G protein)
The G protein is an heterotrimer, consisting of 3 different subunits (alpha, beta and gamma) The alpha subunit bound to GDP (guanosine di-[two] phosphate) has high affinity for the beta and gamma subunits When it is stimulated by the activated receptor, the alpha subunit exchanges bound GDP for GTP (guanosine tri-[three] phosphate) The GTP-bound alpha subunit dissociates from the beta/gamma components, and it binds to a nearby enzyme, altering its activity GTP is then hydrolysed to GDP (looses a phosphate group) by the intrinsic GTPase activity of the alpha subunit, that regains its affinity for the beta/gamma subunits

19

Computational Systems Biology

20

The G protein cycle

From: D. L. Nelson, Lehninger Principles of Biochemistry, IV Edition

Computational Systems Biology

G protein Coupled Receptors

The human genome encodes more than 1000 Gprotein Coupled Receptors (GPCR), that transduce messages as diverse as light, smells, taste, and hormones An example is the beta-adrenergic receptor, that mediates the effects of epinephrine on many tissues:

21

Computational Systems Biology

An example of G protein Coupled Receptor

22

From: D. L. Nelson, Lehninger Principles of Biochemistry, IV Edition

Computational Systems Biology

and the following signalling cascade

An example of signal transduction amplification

23

From: D. L. Nelson, Lehninger Principles of Biochemistry, IV Edition

Computational Systems Biology

A more realistic G protein cycle

Note: in this model in addition to the intrinsic GTPase activity of the G protein (slow rate vh0), there is a faster hydrolysis (fast rate vh1) supported by RGS (Regulator of G protein Signaling)

From: E. Klipp, Systems Biology in Practice, 2005 Chapter 6

24

Computational Systems Biology

The G protein model:

25

ODE = ordinary differential equation

kinetic equations

Computational Systems Biology

The G protein model:

time course
From: E. Klipp, Systems Biology in Practice, 2005 Chapter 6

26

Computational Systems Biology

Signaling Paradigm: the structural components

G protein cycle

MAP kinase cascade

27

Computational Systems Biology

28
From: E. Klipp, Systems Biology in Practice, 2005 Chapter 6

MAPK

Computational Systems Biology

Mitogen-activated protein kinases

(MAPKs)
MAPKs belong to a family of serine/threonine kinases (kinases add
phosphate groups to/phosphorylate other proteins)

They participate in cell growth, differentiation, transformation, apoptosis (controlled cell death) and others They are conserved from yeast to mammal, but their names differ in different species:

29

Computational Systems Biology

The MAPKs cascade

From the first activation step: MAPKKKK stands for MAPKKK Kinase because it phosphorylates MAPKKK MAPKKK (MAPKK Kinase) phosphorylates MAPKK MAPKK (MAPK Kinase) phosphorylates MAPK, which is the last enzyme of the cascade

30

MAPKs are phosphorylated on two separate sites (aminoacids)

Computational Systems Biology

Reading
D. L. Nelson, Lehninger Principles of Biochemistry, IV Edition:

Parts of Chapter 12 on Biosignalling

E. Klipp, Systems Biology in Practice, Wiley-VCH, 2005:

Parts of Chapter 6 on Signal Transduction

31

Computational Systems Biology

additional example of G protein coupled receptor

Appendix (1):

32

From: D. L. Nelson, Lehninger Principles of Biochemistry, IV Edition

Computational Systems Biology

Appendix (2):

additional example of G protein coupled receptor

33

From: D. L. Nelson, Lehninger Principles of Biochemistry, IV Edition