Molecular genetics

1
Molecular genetics networks

• Bioinformatics at Caltech
• C. elegans genetic programming at its best
• Stochastic to deterministic development
• Stereotyped behavior
• Signal transduction
• Genes, pathways, networks
• Modeling

2
Bioinformatics
Classical bioinformatics functional
genomics Genome sequencing and analysis DNA
microarrays informatics Model Organism
Databases sea urchin genome database
WormBase biological annotation of C. elegans
genome Modeling Simulation
Plans abstractions from the data Integrate
analysis of biological networks with biological
databases Automated extraction of information
from the literature Intermediate throughput
biology (automated data collection) Data
analysis clustering, pattern recognition
3
Alliance for Cell Signaling

• Multi-institution consortium
• Al Gilman (UTSW Med) Mel Simon (Caltech)
• Caltech Molecular Biology Core John Doyle
• Dissect signaling networks systematically in two
  cell types (B lymphocytes cardiomyoctes)
• Single cell might respond to 100s of inputs
• Examine effects of single multiple inputs on
  many outputs in standardized assays
• Integrate modeling simulation

4
Biological Databases

• Genome sequence
• Protein-protein interactions
• Gene expression
• Images
• Regulatory interactions
• System behavior

5
WormBase
6
Deletion of single gene
Deletion of single neuron
7
Example of genetic interactions
eat-16 RGS
egl-30 Gq
PIP2
PLCb
dgk-1/sag-1
DAG
IP3
PA
IP2
lfe-1/itr-1 IP3R
tpa-1 PKC
egl-19 Ca channel
DEATH
8
Multi-organismic network views
2
1
4
Species 1
6
3
5
Species 2
1
2
4
6
3
5
Interactions Protein-protein, protein-DNA,
genetic
9
Bioinformatics
Classical bioinformatics functional
genomics Genome sequencing DNA microarray
informatics (clustering) Model Organism
databases sea urchin genome database
WormBaseBiological annotation of C. elegans
genome Modeling Simulation
Plans pull out abstractions Integrate analysis
of biological networks with biological
databases Automated extraction of information
from literature Intermediate throughput biology
(automation) Data analysis clustering, pattern
recognition
10
Nematode worms

• 106 species on planet
• programmed development and behavior
• Caenorhabditis elegans
• 19,000 genes
• 1000 cells 959 (female), 1031 (male)
• Transparent
• Two generations per week

11
Programming of Nematode Development

• Circuits that implement developmental functions
  with high fidelity
• Making a hole to connect two tubes
• Deterministic development might occur by tuning
  of stochastic development during evolution
• Modular sub-programs

12
Different species have distinct but invariant
development
Species 1
Species 2
A
A
C
F
B
C
B
D
E
Mutant of Species 1
F
13
Vulval development Formation of a Hole
14
AC
Vulva
AC
Vulva
15
1. AC patterns uterus and vulva
utse
uterus
uv1
uv1
AC
vulF
vulF
vulva
vulE
vulE
lumen
2. uv1 bonds to vulF
3. AC fuses with utse
utse
uv1
AC
uv1
uv1
uv1
utse
vulF
vulF
vulF
vulF
vulE
vulE
vulE
vulE
16
Stochastic Cell Fate Specification
Belly-up developing worm gonad
Z4
Z1
0.50
AC
VU
0.50
VU
AC
Shapessubprograms
17
Bipotential AC/VU Cells
1
2
Receptor
Signal
2
Receptor
1
Signal
VU
AC
1
2
Receptor
Signal
2
1
18
Evolution of Anchor Cell Specification
0.5
e.g., C. elegans
0.8
1.0
e.g., Panagrellus
19
Molecular genetics of signal transduction

• Genetics
• indicates necessity sufficiency
• Orders proteins into pathways
• Detects interactions
• Dissect feedbacks and branches
• Complexity
• More components than you would expect
• Evolvability Control Fidelity

20
EGF family growth factor precursor
Anchor Cell
LIN-3
LET-23
EGF-R tyrosine kinase
SH3-SH2-SH3
SEM-5
SLI-1
Sos (RAS GNEF)
LET-341
c-cbl
ras
LET-60
ARK-1
LIN-45
raf-1 ser/thr kinase
Tyr kinase
MEK-2
MEK
GAP-1
MAPK/ERK
MPK1/SUR1
rasGAP
VPC
Vulval Differentiation
Epidermal

Differentiation
21
Signal EGF
OUT
ras
RAF MAPKKK
IN
MAPKK
SOS rasGNEF
MAPK
EGF- RECEPTOR
GRB2 Adaptor
Universal EGF-receptor pathway Humans
proto-oncogenes Worm vulval induction by the
AC Fly developmental inductions
22
Amplification -NOT THE CASE!
J. Ferrell (Stanford University) The MAP kinase
module displays ultra-sensitivity because of the
multiple events needed for enzyme activation
23
Ultrasensitivity Goldbetter Koshland PNAS 1981
Normal (Michaelis-Menten) sensitivity
0.9
Output
Ultra-sensitivity
0.1
Input
Hyperbolic sensitivity input/(1
input) Ultrasensitivity inputn/(1 inputn)
24
MAPK cascade
Upstream signal
MAPKKK
P
PP
MAPKK
P
PP
MAPK
Downstream effect
Andre Levchenko, Shuki Bruck
25
MAPK cascade pathway
SCAFFOLD PROTEINS
covalent activation
activation by localization
Andre Levchenko, Shuki Bruck
26
Potential Roles of Scaffolds for MAP Kinase Module
1. Insulation sequesters each instance of the
module, thereby preventing cross-talk. 2.
Acceleration pre-assembled complex is activated
in one step 3. Tuning of response threshold
to graded
27
Mathematical model
of3 C3t
of1 C2t
P
P
- on1 MEKt C1t
P
P
C1
C2
C3
C4
C5
of2 C4t - on2 MAPKt C1t
of4 C5t
d/dt C1t -C1t (on1 MEKt on2 MAPKt)
of1C2t of2 C4t of3 C3t of4 C5t
P
P
MAPKK
MAPK
MAPK
MAPKK
P
P
28
Kinetics of MAPK activation
Scaffold concentrations
0.2 ?M
0.1 ?M
output
0.05 ?M
0.01 ?M
None
1 ?M
Time (seconds)
29
Existence of an optimal Scaffold for signaling
MAPK, C8
Signal output
1/Scaffold
C6
scaffold protein concentration
30
What do Adaptor proteins do?
Protein-protein interaction domains
adaptors
T. Pawson
SH3
SH2
SH3
SEM-5
Clark, Stern Horvitz (Nature 1992)
31
Adaptor proteins might recruit both positive
and negative regulators
LET-23
Kinase
pTyr
SEM-5
SH3
SH2
SH3
Pro-rich
Pro-rich
SOS-1
ARK-1
Ras GNEF
32
Neil Hopper
LET-23 EGF-R
SEM-5 Grb2
IP3
ARK-1
LET-341 SOS
IP3 Receptor
LET-60 RAS
Fertility
Vulva
33
Molecular genetics of signal transduction

• Genetics
• indicates necessity sufficiency
• Orders proteins into pathways
• Detects interactions
• Dissect feedbacks and branches
• Complexity
• More components than you would expect
• Evolvability Control Fidelity

34
Molecular genetics networks

• Bioinformatics at Caltech
• C. elegans programming
• Signal transduction

35
Amplification -NOT THE CASE!
J. Ferrell (Stanford University) The MAP kinase
module displays ultra-sensitivity because of the
multiple events needed for enzyme activation