Transcript flhDC

Chemotaxis and Motility in E. coli
Examples of Biochemical and Genetic Networks
• Background
• Chemotaxis- signal transduction network
• Flagella gene expression – genetic network
Dictyostelium- adventures in multicellularity
Julie Andreotti – Oscillations in a Biochemical Network
Bacterial Chemotaxis
Flagellated bacteria “swim” using a reversible rotary motor linked by
a flexible coupling (the hook) to a thin helical propeller (the flagellar
filament). The motor derives its energy from protons driven into the
cell by chemical gradients. The direction of the motor rotation
depends in part on signals generated by sensory systems, of which
the best studied analyzes chemical stimuli.
Chemotaxis - is the directed movement of cells towards an
“attractant” or away from a “repellent”.
• For a series of QuickTime movies showing swimming bacteria with fluorescently
stained flagella see: http://www.rowland.org/bacteria/movies.html
• For a review of bacterial motility see Berg, H.C. "Motile behavior of bacteria".
Physics Today, 53(1), 24-29 (2000). (http://www.aip.org/pt/jan00/berg.htm)
A photomicrograph of three cells
showing the flagella filaments.
Each filament forms an extend helix
several cell lengths long.
The filament is attached to the cell
surface through a flexible ‘universal
joint’ called the hook.
Each filament is rotated by a reversible rotary motor, the direction of the motor
is regulated in response to changing environmental conditions.
The E. coli Flagellar Motor- a true rotary motor
Rotationally averaged reconstruction of electron micrographs of purified hook-basal
bodies. The rings seen in the image and labeled in the schematic diagram (right)
are the L ring, P ring, MS ring, and C ring. (Digital print courtesy of David DeRosier,
Brandeis University.)
Tumble
(CW)
Smooth Swimming
or Run
(CCW)
Chemotactic Behavior of Free Swimming Bacteria
No Gradient
Increasing attractant
Increasing repellent
A ‘Soft Agar’ Chemotaxis Plate
A mixture of growth media and a low concentration of agar are mixed in
a Petri plate. The agar concentration is not high enough to solidify the
media but sufficient to prevent mixing by convection.
The agar forms a mesh like network making
water filled channels that the bacteria can
swim through.
A ‘Soft Agar’ Chemotaxis Plate
Bacteria are added to the center of the plate and allowed to grow.
A ‘Soft Agar’ Chemotaxis Plate
As the bacteria grow to higher densities, they generate a gradient
of attractant as they consume it in the media.
Attractant
Concentration
cells
cells
A ‘Soft Agar’ Chemotaxis Plate
The bacteria swim up the gradients of attractants to form
‘chemotactic rings’ .
This is a ‘macroscopic’ behavior. The chemotactic ring is the
result of the ‘averaged” behavior of a population of cells. Each
cell within the population behaves independently and they
exhibit significant cell to cell variability (individuality).
A ‘Soft Agar’ Chemotaxis Plate
‘Serine’ ring
‘Aspartate’ ring
Each ‘ring’ consists of tens of millions of cells. The cells outside the rings are
still chemotactic but are just not ‘experiencing’ a chemical gradient.
Serine and aspartate are equally effective “attractants”, but in this assay the
attractant gradient is generated by growth of the bacteria and serine is
preferentially consumed before aspartate.
Videos of motile bacteria:
1)
2)
3)
4)
5)
6)
7)
Free swimming bacteria
Swimming in soft agar
Tethered cells
Latex bead tethered to flagellum
Surface swarming behavior
Swarm cells mixed with swim cells
Aggregation / patterns formation
Videos of motile bacteria:
1) Free swimming bacteria
2)
3)
4)
5)
6)
7)
Swimming in soft agar
Tethered cells
Latex bead tethered to flagellum
Surface swarming behavior
Swarm cells mixed with swim cells
Aggregation / patterns formation
Watch for sudden
changes of direction
= tumbles
Videos of motile bacteria:
1) Free swimming bacteria
2) Swimming in soft agar
3)
4)
5)
6)
7)
Tethered cells
Latex bead tethered to flagellum
Surface swarming behavior
Swarm cells mixed with swim cells
Aggregation / patterns formation
GFP labeled cells
Cells are stuck most
of the time but when
the video is run at
5X they look almost
like cells in aqueous
environments.
Videos of motile bacteria:
1) Free swimming bacteria
2) Swimming in soft agar
3) Tethered cells
4)
5)
6)
7)
Latex bead tethered to flagellum
Surface swarming behavior
Swarm cells mixed with swim cells
Aggregation / patterns formation
wt - motor switches regularly
 cheY – motor does not switch
 cheZ – motor switched more frequently
A cell is stuck to the
coverslip by a sheared
flagella. The motor still
turns but since the
flagella can’t the cell
body rotates.
Videos of motile bacteria:
1) Free swimming bacteria
2) Swimming in soft agar
3)
Tethered cells
4) Latex bead tethered to flagellum
5) Surface swarming behavior
6) Swarm cells mixed with swim cells
7) Aggregation / patterns formation
A cell is stuck to the
coverslip and a latex
bead is attached to a
single flagella. The
flagella rotation can be
visualized by the bead.
Videos of motile bacteria:
1) Free swimming bacteria
2) Swimming in soft agar
3)
Tethered cells
4) Latex bead tethered to flagellum
5) Surface swarming behavior
6) Swarm cells mixed with swim cells
7) Aggregation / patterns formation
Bacteria can move over
a solid surface in a
process call swarming.
The movement is
relatively slow
compared to swimming
and is coordinated.
Videos of motile bacteria:
1) Free swimming bacteria
2) Swimming in soft agar
3)
Tethered cells
4) Latex bead tethered to flagellum
5) Surface swarming behavior
6) Swarm cells mixed with swim cells
7) Aggregation / patterns formation
Swarms cells are
elongated relative to
normal swimming
cells.
Videos of motile bacteria:
1) Free swimming bacteria
2) Swimming in soft agar
3)
Tethered cells
4) Latex bead tethered to flagellum
5) Surface swarming behavior
6) Swarm cells mixed with swim cells
7) Aggregation / patterns formation
Dilute cells placed under conditions where they release attractants will
aggregate into large masses of cells (~30’ video  ~2’).
The Molecular Machinery of Chemotaxis
INPUT
Attractant concentration
Signal
Transduction
OUTPUT
Direction
of
rotation
The Molecular Machinery of Chemotaxis
INPUT
Tsr
Tar
Tap
Trg
Attractants bind receptors at the cell
surface changing their “state”.
(methylated chemoreceptors MCPS).
Signal
Transduction
OUTPUT
Direction
of
rotation
The Molecular Machinery of Chemotaxis
INPUT
Tsr
Tar
Tap
Trg
CheA
(CheW)
The MCPs regulate the activity of a
histidine kinase - autophosphorylates
on a histidine residue.
P~
OUTPUT
Direction
of
rotation
The Molecular Machinery of Chemotaxis
CheA transfers its phosphate to a
signaling protein CheY to form
CheY~P.
INPUT
Tsr
Tar
Tap
Trg
CheA
(CheW)
CheY
P~
P~
OUTPUT
Direction
of
rotation
The Molecular Machinery of Chemotaxis
CheY~P binds to the “switch” and
causes the motor to reverse direction.
The signal is turned off by CheZ
which dephosphorylates CheY.
INPUT
Tsr
Tar
Tap
Trg
CheA
(CheW)
CheY
CheZ
P~
P~
OUTPUT
Direction
of
rotation
Excitatory Pathway
At ‘steady state’, CheY~P levels in the cell are constant and there is some
probability of the cell tumbling. Binding of attractant of the receptorkinase complex, results in decreased CheY~P levels and reduces the
probability of tumbling and the bacteria will tend to continue in the same
direction.
MCP
CheA
(CheW)
CheY~P
CheZ
Motor
+ attractant
CheY
inactive
The Molecular Machinery of Chemotaxis
Adaptation involves two proteins, CheR
and CheB, that modify the receptor to
counteract the effects of the attractant.
INPUT
Tsr
Tar
Tap
Trg
CheA
(CheW)
CheY
CheZ
CheR
CheB
P~
P~
OUTPUT
Direction
of
rotation
Adaptation Pathway
CheR
MCP
CheA
(CheW)
Less active
CheB~P
MCP~CH3
CheA
(CheW)
More active
Adaptation Pathway
CheR
MCP-(CH3)0
MCP-(CH3)1
MCP-(CH3)2
MCP-(CH3)3
MCP-(CH3)4
MCP-(CH3)0
MCP-(CH3)1
MCP-(CH3)2
MCP-(CH3)3
MCP-(CH3)4
+Attractant
+Attractant
+Attractant
+Attractant
+Attractant
CheB~P
In a receptor dimer there will 65 possible states (5 methylation states and two
occupancy states per monomer). If receptors function in receptor clusters,
essentially a continuum of states may exist.
The conformational transition
between T and R states of the MCP-CheACheW ternary complex probably involves an
alteration in the positioning of methylated
helices within a coiled coil structure. This
transition is modulated by changes in the
electrostatic potential between helices effected
by the conversion of anionic glutamyl side
chains to neutral methyl glutamyl groups and
vice versa. Ligand binding between the
sensory
domain would act to perturb the T/R
equilibrium by altering the relative positioning
of monomers within the cytoplasm (see Fig. 6).
This interplay between methylation and
stimulation could operate to control the relative
positioning of signaling domains and their
associated CheA subunits so as to regulate the
transphosphorylation activity of CheA, which
through CheY controls the swimming behavior
of the bacterial cell.
Some Issues in Chemotaxis:
• Sensing of Change in Concentration not absolute concentration
i.e. temporal sensing
• Exact Adaptation
• Sensitivity and Amplification
• Signal Integration from different Attractants/Repellents
The range of concentration of attractants that will cause a chemotactic
response is about 5 orders of magnitude (nM  mM)
References on Modeling Chemotaxis
Barkai, N. & Leibler, S. (1997) Nature (London) 387: 913–917.
Spiro, P. A., Parkinson, J. S. & Othmer, H. G. (1997) Proc. Natl. Acad. Sci. US
94: 7263–7268.
Tau-Mu Yi, Yun Huang , Melvin I. Simon, and John Doyle (2000)
Proc. Natl. Acad. Sci. USA 97: 4649–4653.*
Bray, D., Levin, M. D. & Morton-Firth, C. J. (1998) Nature (London)
393: 85–88. *
* - these models have incorporated the Barkai model.
Robustness in simple biochemical networks
N. Barkai & S. Leibler
Departments of Physics and Molecular Biology, Princeton University,
Princeton, New Jersey 08544, USA
Simplified model
of the chemotaxis
system.
Mechanism for robust adaptation
E is transformed to a modified form, Em, by the
enzyme R; enzyme B catalyses the reverse
modification reaction. Em is active with a probability
of am(l), which depends on the input level l. Robust
adaptation is achieved when R works at saturation
and B acts only on the active form of Em. Note that
the rate of reverse modification is determined by
the system’s output and does not depend directly
on the concentration of Em (vertical bar at the end
of the arrow).
Some parameters used to characterize the network.
Tumble frequency
Steady-State Tumble Frequency
Adaptation Time
Adaptation precision
Chemotactic response and adaptation in the Model.
The system activity, A, of a model system which was subject to a series of
step-like changes in the attractant concentration, is plotted as a function of
time. Attractant was repeatedly added to the system and removed after 20
min, with successive concentration steps of l of 1, 3, 5 and 7 mM. Note the
asymmetry to addition compared with removal of ligand, both in the
response magnitude and the adaptation time.
How robust is the model with respect to variation in parameters?
Adaptation precision
Adaptation Time
Adaptation precision (i.e. exact adaptation) is Robust
Adaptation time is very sensitive to parameters
Testing the predictions of the Barkai model
Robustness in bacterial chemotaxis.
U. Alon, M. G. Surette, N. Barkai & S. Leibler
• The concentration of che proteins were altered as a simple method to
vary network parameters.
• The behavior of the cells were measured (adaptation precision,
adaptation time and steady-state tumble frequency).
• In each case the predictions of the model we observed.
Data for CheR
As predicted by the model the
adaptation precision was robust
while adaptation time and
steady-state tumble frequency
were very sensitive to
conditions.
Regulation of flagella gene expression:
A three tiered transcriptional hierarchy
Positive transcriptional regulators
Alternative sigma factors
Ant-sigma factors
Temporal regulation
The Flagellar Transcription Hierarchy
1. The Master Regulon
2. The FlhCD Regulon
CRP,H-NS,OmpR
other?
Chemotaxis
proteins
Motor
proteins
FlhCD
inside
outside
FlgM
FliA
Basal Body
and Hook
other?
3. The FliA Regulon
Filament
The flhDC promoter integrates inputs from
multiple environmental signals
flhDC
?
CRP - catabolite repression, carbohydrate metabolism
OmpR - osmolarity
IHF - growth state of cell?
HdfR - ?
FliA Regulation by FlgM
FlhDC expression leads to activation of Level 2 genes including the
alternative sigma factor FliA and an anti sigma factor FlgM
FlgM accumulates in the cell
and binds to FliA blocking
its activity (i.e. interaction
with RNA polymerase)
preventing Level 3 gene
expression.
inside
outside
Level 3 Genes
FliA Regulation by FlgM
Other level 2 genes required for Basal body and hook assembly are
made and begin to assemble in the membrane.
Level 3 Genes
inside
outside
Basal Body
and Hook
Assembly
FliA Regulation by FlgM
The Basal body and hook assembly are completed.
Level 3 Genes
inside
outside
Completed Basal Body
and Hook
FliA Regulation by FlgM
The Basal body and hook assembly are completed.
FlgM is exported through
the Basal Body and Hook
Assembly
Level 3 Genes
inside
outside
Completed Basal Body
and Hook
FliA Regulation by FlgM
Level 3 gene expression is initiated.
FlgM is exported through
the Basal Body and Hook
Assembly.
FliA can interact with RNA
polymerase and activate
Level 3 gene expression.
Level 3 Genes
inside
outside
Completed Basal Body
and Hook
FliA Regulation by FlgM
Level 3 gene products are added to the motility machinery including the
flagella filament, motor proteins and chemotaxis signal transduction system.
inside
outside
Filament
Using reporter genes to measure gene expression
RNA polymerase
Regulator
Organization of operon on chromosome.
flhD
flhDC promoter
flhC
Using reporter genes to measure gene expression
RNA polymerase
Regulator
Organization of operon on chromosome.
flhD
flhC
flhDC promoter
Clone a copy of the promoter into a reporter plasmid.
Reporter gene
Using reporter genes to measure gene expression
RNA polymerase
Regulator
flhD
flhC
Both the flhDC genes and the reporter
plasmid are regulated in the same way
and thus the level of the reporter
indicates the activity of the promoter.
Reporter gene
Note that the strain still has
a normal copy of the genes.
Gene Expression
in Populations
Gene Expression
in Single Cells
Multi-well plate reader
Video microscopy
- sensitive, fast reading
- high-throughput screening
- liquid cultures
- colonies
- mixed cultures
- “individuality”
- cell cycle regulation
- epigenetic phenomenon
Automation: Both approaches are amenable to high throughput robotics
Fluorescence of flagella reporter strains as a function of time
Fluorescence
relative to max
0.6
0.1
0.01
0
600
Time [min]
The order of flagellar gene expression is the order of assembly
Early
Cluster 1 Class 1 flhDC
Cluster 2
Late
Class 2 fliL
Class 2 fliE
Class 2 fliF
Class 2 flgA
Class 2 flgB
Class 2 flhB
Class 2 fliA
Class 3 fliD
Class 3 flgK
Class 3 fliC
Class 3 meche
Cluster 3Class 3 mocha
Class 3 flgM
Master regulator
Activator of class 3
Simple Mechanism for Temporal Expression Within an Regulon
[protein]
Induction of
positive
regulator
Time
Promoters with
decreasing
affinity for
regulator
Simple Mechanism for Temporal Expression Within an Regulon
[protein]
Using Expression Data to Define and Describe Regulatory Networks
With the flagella regulon, current algorithms can distinguish Level 2 and
Level 3 genes based on subtleties in expression patterns not readily
distinguished by visual inspection.
Using our methods for expression profiling (sensitive, good time resolution)
we have been able to demonstrate more subtle regulation than previously
described.
The Challenge:
Can this type of experiment and analysis be used to describe the details of the
flagella regulon? (our ‘model’ network)
Can this be applied on a genomic scale?
Synchronization of the population occurs only
under some growth conditions
0
600
0
600
Time [min]
Condition A
Time [min]
Condition B
(No pre-existing flagella)
(Pre-existing flagella)
1:600 dilution
1:60 dilution
flhDC
activation
Level 2 genes
Level 2 & 3 genes
Level 3 genes
Variability in 22 E. coli flhDC Promoters
500000
400000
Relative Promoter Activity (max)
300000
200000
100000
0
***
1000000
100000
10000
1000
100
***
The Promoter for flhDC varies significantly between E. coli Isolates
• In several randomly cloned E. coli flhDC promoters, there is a large
distribution in promoter strength
• Quantitative differences in promoter strength can not be inferred from
promoter sequence nor from swim rates on soft agar plates.
• The same promoter behaves differently in different strain backgrounds
which implies variability in regulators acting on the promoter
(CRP,OmpR etc.)
• Correct temporal patterning of gene expression and assembly of
flagella occurs despite significant variation in the level of gene
expression between strains. Where is the source of the ‘robustness’ in
this genetic network?