Transcript Noise

Distribution of fluorescence values of
cells w/ LacY::YFP
Cells begin to show “all or nothing” behavior at about 30-40 uM TMG.
Fluorescence intensity is on the X axis, number of cells is on the Y axis
Images of “all or nothing” behavior at
30 um TMG in LacY::YFP cells
Note that the “nothing” cells have a few
LacY::YFP molecules
Most “nothing” cells have 0-4
LacY::YFP molecules
~400 molecules of LacY::Yfp are needed
for induction in 40 uM TMG
Measurement of the threshold of permease molecules for induction. (A) Single-cell time traces of fluorescence intensity,
normalized by cell size, starting from different initial permease numbers. The initial LacY-YFP numbers were prepared
through dilution by cell division of fully induced cells after removal of the inducer. Upon adding 40mM TMG at time 0,
those cells with low initial permease numbers lost fluorescence with time as a result of dilution by cell division and photo
bleaching,whereas those cells with high initial permease numbers exhibited an increase in fluorescence as a result of
reinduction. Permease molecule numbers were estimated from cell fluorescence(28). The dashed red line indicates the
determined threshold. (B) The probability of induction of a cell within 3 hours as a function of the initial permease number
was determined using traces from 90 cells.The probability of induction (p) was fit with a Hill equation p = y4.5/(y4.5 +
3754.5) for the initial permease number y. The threshold of permease numbers for induction was thus determined
tobe375molecules. Error bars are the inverse square root oft he sample size at each point.
Novick has something to contribute to the
analysis of data in Choi et al.
Cells are rapidly induced in high level of
TMG
The slope is a measurement of speed of induction. It is
proportional to the amount of b-gal activity made per
generation.
Induction in low levels of TMG is slower
Slope = .005
After one generation cells should be ~30%
induced and induction should be quick
Should begin to see an increase in
expression right away at 40 uM TMG
(at least 30% induced in one
generation). Go back to slide 1 and
see that MUCH less than 30% of the
cells were induced after 24 hours in
the 40 uM TMG experiment.
If lac can’t loop, then induction
occurs more readily
C: With O3 removed the lac region
can ’ t loop. Such cells are able to
induce lacY even when they start out
with very low numbers of LacY initially.
Compare to Fig 2A where ~400 LacY
molecules were needed for induction
D: DNA looping is needed for
bistability. Without looping population
are not bimodal—compare with first
slide.
LacY is made in bursts-small and
large
The Experiment: done in lacY- cells with 200 uM TMG; the lac promoter is driving anoth
membrane protein (Tsr) fused to YFP.
The Results:
Cells rarely make new LacYs and when
they do, they only make a few (top panel).
Every once and a while they make many,
over an extended period (bottom panel).
When lac region can’t loop large
bursts are common
In panel D, note that the x-axis has MUCH higher values—cells w/o lac looping make
have many bursts that make a lot of product relative to cells that still loop the lac
region (panel C)
Model for expression bursts
At high TMG concentrations, TMG binds to LacI and causes it to fall off of the DNA.
Once off of the DNA, it tends to stay off, leading to large bursts of transcription.
Note that the Km for TMG binding to DNA-bound LacI is ~1mM (that is when internal
TMG concentrations are ~1mM, LacI has a 50:50 chance of binding TMG and falling
off of the DNA
Model for expression bursts (cont)
At low TMG concentrations one side of the LacI tetramer will fall off of O1 and a small
burst of transcription will take place before rebinding. If the whole tetramer should fall
off (this is rare), then it will bind to TMG and stay off, giving a large burst.
Note that the Km for TMG binding to DNA-bound LacI is ~1mM BUT its Km for LacI
not bound to DNA is ~ 10 uM. This means that 40 uM TMG is not enough to bump
LacI off of the operators, but it is enough to keep it off once it falls off on its own.
In strains w/o lac looping, every time LacI falls off one operator it finds itself in
solution and
quickly bound by TMG. This tends to make every burst a large burst.
Experimental set up used in Robert et al.
This allows them to follow induction in single cells
by microscopy as they grow.
Typical (?) results
Figure 2 Establishment of bimodality in
microcolonies.
(A) Fluorescent images (YFP channel) of
a microcolony of LCY1 just before TMG
introduction (inset) and 2 h.
(B) Fluorescence intensity of all
individual cells (black symbols) and mean
fluorescence of the population (yellow) as
a function of time. The green and red lines
are representative examples of single-cell
trajectories. The dashed gray line
indicates the time of
TMG introduction.
(C) Distribution of YFP fluorescence of
single cells 2 h after TMG introduction
Induction levels of single
cells and their
descendents were tracked
over time. Cells appeared
to be correlated—
descendents of one cell
often behaved the same
way all inducing or not
inducing. The circled cell
shows one cell whose
daughters all induced,
even though the two
daughters were separated
from each other before
TMG was added (at the
dashed line.
Figure 3 Typical ‘Lineage history tree’
established for the microcolony presented in
Figure 2A. Individual cells are plotted as horizontal
lines where the color corresponds to fluorescence
intensity (YFP channel) as a function of time
(horizontal axis). At division time, a vertical line is
drawn to connect the mother cell and its two
daughters. The dashed line corresponds to TMG
introduction. The black circle indicates an
example of division of a cluster’s common
ancestor before TMG introduction.
Parents tend to leave descendants that are
all either “On” or “Off”
For each cell present at time tI (TMG introduction), the proportion of induced cells in its final progeny was calculated. The
distribution of these proportions exhibits two peaks at 0 and 1 (Figure 3B, left), suggesting that most of the initial cells have in
their progeny either almost only induced cells or almost only uninduced cells.
Statistical evidence that induction is
correlated in descedents
Figure 4 (A) Typical genealogical tree with pairs of sister cells present at TMG introduction (red) and pairs of sister cells in the
population of mothers (blue) and grandmothers (green) of cells present at TMG introduction. The vertical black line corresponds to
TMG introduction. (B) Correlation of responses between sister cells and first and second cousin cells (from four different
experiments). The mean fluorescence of the progeny of a cell is plotted against the mean fluorescence of its sister’s progeny, for
cells present at TMG introduction (red) or cells in the population of mothers (blue) and grandmothers (green, inset) of cells present at
TMG introduction
Show movie!!
What caused correlated behavior?
Why might all the descendents of the circled cell be readily induced? The authors show 2
things influence the correlated behavior:
1. LacI levels in the initial (circled) cell. Less LacI in a cell means that its descendants are
more likely to induce in response to low levels of TMG. This could be caused by larger/longer
bursts in cells with little LacI.
2. Growth rate: slow growth of an initial cell also makes it more likely that descendants will
induce.
Paper by Elowitz et al.
Effects of bursting and stochastic
expression
Bursting
mRNA levels
Protein levels
Effect on other genes
Signal noise and gene regulation
noise affect gene expression
Signal noise and gene regulation
noise affect gene expression
Paper by Elowitz et al.
Extrinsic noise: Next. Noise in a population of cells that that arises from global conditions that
can vary from cell to cell. Such global conditions may be levels of RNA polymerase, internal
concentration of inducer, internal concentration of important regulatory proteins.
Intrinsic noise: Nint. This is noise that would come about even if there was no extrinsic noise.
Elowitz et al. point out that Nint can be seen as the variation in expression of 2 identical
promoters in the same cell.
A Nint value of .25 means that expression of 2 such promoters would vary by
25%.
Ntot=(Standard deviation of expression)/(mean of the expression)
Ntot2 =Next2 + Nint2
High expression levels lead to low noise
CFP (colored green) and YFP (colored red)
each driven by plac(O1) in
a lacI+ cell with IPTG
CFP (colored green) and YFP (colored red)
each driven by plac(O1) in
a lacI- cell
Ntot=.117
Next=.098
Nint=.063
Ntot=.077
Next=.054
Nint=.055
Low expression levels lead to high noise
(~25x dimmer than cells w/ IPTG)
CFP (colored green) and YFP (colored red)
each driven by plac(O1) in
a lacI+ cell with no IPTG
CFP (colored green) and YFP (colored red)
each driven by plac(O1) in
a lacI+ cell with no IPTG
Ntot=.41
Next=.33
Nint=.25
Ntot=.37
Next=.32
Nint=.19
Noise in two stains, one noisy, the other
less so
M22: recA+,
D22 recA+
Noise as a function of expression levels
Increasing fluorescence/expression
Cell death by antibiotics- a typical
experiment
Cell death by antibiotics- a typical
experiment
-if you regrow the persisters, they are normally resistant, and give
rise to a small percentage of persisters---so their resistance is no
due to mutation.
Persisters
-slow growing or non-growing cells that are resistant to:
•antibiotics (causes problems clinically)
•heat, dessication
•phage
•other stresses
-usually make up a small percentage of a population
-the actually percentage is a function of how frequently the
environment changes
-mutants can be found that make persisters more often. One
classic persister++ mutation is in the gene hipA
--> if HipA is made
by itself, cells cannot
grow!
The hipA gene was put under control
of the tetracycline repressor
-when tetracycline was added to cells, they made HipA in proportio
to the amount of tet added.
The amount of HipA effects when
colonies appear (ie affects when cells
start to grow)
-Show movie
High HipA causes colonies
to appear late and with
large variability in appearance
time
growth rate is the same for
all levels of HipA
No HipA made
HipA made
see persisters
only when HipA
is made. Only slow
appearing cells
are resistant!
+Amp
The effects of high HipA can be seen
on plates too-it inhibits the start of
growth
-cell with highest amount of HipA does not divide until late
HipA inhibition happens at a threshold
(blue arrow w/o HipB, orange arrow w/
HipB)
Cells above HipA threshold put off
growth, cells below grow right away
Model
red = free HipA
How might this happen? How might
there be enough free HipA to trigger
various levels of persistance?
Low TMG data from Novick
Note that at 10 uM TMG 43% of the
cells are induced after several
generations. Choi et al. had very few
induced cells after 24 hours at 30
uM!!