The Influence of Promoter Architecture on Targeted Gene ... · The Influence of Promoter...
1
The Influence of Promoter Architecture on Targeted Gene Activation by Complex Transcription Factor Signaling Jeffrey B. Wang, Anders S. Hansen, and Erin K. O’Shea FAS Center for Systems Biology, Harvard University, Cambridge MA 02138 Abstract Our studies focused on Msn2, a transcription factor in budding yeast responsible for mediating the response to a number of stresses, such as oxidative stress and starvation. Based on the identity of the stressor, Msn2 enters the nucleus with different patterns, such as entering and leaving the nucleus rapidly or periodically. However, how the promoter is able to understand these different signals is not well understood. Our project’s goal was to elucidate how the positioning of histones within promoter sequences allowed genes to be differentially activated by these various signals. We selected the DDR2 gene promoter, which possesses Msn2 binding sites both in and out of its well-defined nucleosome binding positions. Then, we systematically modified the position of its binding sites relative to the nucleosomes and utilized a fluorescent reporter that allowed us to quantify relative expression levels. We focused on how the promoter processes three aspects of the incoming Msn2 signal: the duration (how long Msn2 is present in the nucleus), the intensity (the maximum amount of Msn2 within the nucleus), and the frequency of oscillations (how quickly we move Msn2 in and out of the nucleus). We then fit these results to a set of differential equations describing how transcription factor binding can lead to protein production. With these results, we hope to identify how all of these binding sites cooperate to analyze complex Msn2 signaling. Background ? Experimental Approach Analyzing Gene Expression Stress TF TF Response Upon stress, the cell will often activate transcription factors, which subsequently translocate to the nucleus to induce expression of appropriate stress response genes Msn2 is a general stress transcription factor found in Saccharomyces cerevisiae [1]. It has been shown to translocate into the nucleus with different patterns based on different types of stresses [2] Purpose: How is the same transcription factor able to activate a presumably different gene expression profile using these different signals? Quantitatively characterizing how the promoter region is able to process these complex signals would be illuminating not only for the general stress response, but also gene regulation in higher organisms Chromatin might be part of the answer, since it has been shown to influence the threshold and dynamic range of promoters [3] Specific Goal: To elucidate how the positioning of Msn2 binding sites relative to nucleosomes influences the promoter’s ability to process these complicated signals Generation of Promoter Variants We selected the DDR2 promoter for its well defined nucleosomes and multiple Msn2 binding sites both in and out of the nucleus We then formed a dozen mutants manipulating the positioning of Msn2 Four of the mutants consisted of a single binding site knockout Two of the mutants were a “slow” and “fast” variant consisting of three binding sites within and outside of a nucleosome, respectively The remaining represented a spectrum of promoter speeds from the slow and fast promoters Promoter Architecture for DDR2: Nucleosome positioning was determined with MNase-seq and depicted as gray peaks. Red triangles depict locations of Msn2 Binding Motifs Slow variant of the DDR2 Promoter Msn2- mCherry Nhp6a CFP YFP pDDR2* pDDR2* Site-Directed Mutagenesis Plasmid Yeast Transformation Promoter variants were first formed using site-directed mutagenesis Mutants were then transformed upstream of a YFP and CFP dual reporter in a preexisting yeast strain with Msn2 and Nhp6a (a nuclear protein) fluorescently tagged to track Msn2 and identify the nucleus Controlling Msn2 Signals Bcy1p Bcy1p Stress: low cAMP levels TPK 1/2/3 No stress: high cAMP levels Bcy1p Bcy1p cAMP TPK 1/2/3 Msn2 P P P Cytoplasm Nucleus Cytoplasm Nucleus Msn2 Stress Gene Transcription OFF No stress + 1-NM-PP1 Bcy1p Bcy1p TPK 1/2/3 Cytoplasm Nucleus Msn2 1-NM-PP1 YFP CFP YFP CFP Msn2 CFP YFP TPK 1/2/3 Msn2 is normally prevented from entering the nucleus by phosphorylation by PKA By adding an inhibitor for PKA, it is possible to force Msn2 in and out of the nucleus on the order of a few minutes cAMP 0 20 40 60 80 100 0 500 1000 1500 Average Msn2 Localization Time (min.) Amt. of Msn2 in Nucleus (AU) Duration Amplitude 0 20 40 60 80 100 0 500 1000 1500 Average Msn2 Localization Time (min.) Amt. of Msn2 in Nucleus (AU) Number of Pulses 0 20 40 60 80 100 0 500 1000 1500 Average Msn2 Localization Time (min.) Amt. of Msn2 in Nucleus (AU) Frequency of Pulses 3 different microscopy experiments were conducted measuring YFP/CFP intensity, while manipulating key features of the incoming Msn2 signal Fast variant of the DDR2 Promoter For each cell over a time period of 2.5 hours, Msn2 localization and CFP/YFP expression were measured The results were fit to a set of ordinary differential equations (ODEs) describing activation of the promoter region followed by transcription of mRNA and translation of fluorescent protein [3] 0 20 40 60 80 100 0 500 1000 1500 Average Msn2 Localization Time (min.) Amt. of Msn2 in Nucleus (AU) YFP Expression Time (min) 0 10 30 50 70 90 110 130 150 2000 4000 6000 8000 10000 0 50 100 150 0 2000 4000 6000 8000 10000 Time (min) Fluorescence 50 Minute Pulse, 3 μM: YFP Expression matured YFP measured YFP Typical Results of Microscopy Experiment Gene Expression Data from all 30 Conditions 0 1 2 3 4 5 x 10 4 0 2000 4000 6000 8000 10000 12000 Msn2 Area Under Curve (AU) Max YFP expression DDR2: Amplitude modulation 10 Min. Pulse 20 Min. Pulse 30 Min. Pulse 40 Min. Pulse 50 Min. Pulse ODE Prediction Response of Wild-Type DDR2 Promoter to Changes in the Total Integrated Amount of Nuclear Msn2 over time. Different colored points indicate different durations of signals. Note how all of the points follow the same general pattern regardless of duration. Msn2 Nuclear Localization Time (min) 0 10 30 50 70 90 110 130 150 DM 10 Min.: 140 nM DM 10 Min.: 330 nM DM 10 Min.: 760 nM DM 10 Min.: 3 μM DM 20 Min.: 140 nM DM 20 Min.: 330 nM DM 20 Min.: 760 nM DM 20 Min.: 3 μM DM 30 Min.: 140 nM DM 30 Min.: 330 nM DM 30 Min.: 760 nM DM 30 Min.: 3 μM DM 40 Min.: 140 nM DM 40 Min.: 330 nM DM 40 Min.: 760 nM DM 40 Min.: 3 μM DM 50 Min.: 140 nM DM 50 Min.: 330 nM DM 50 Min.: 760 nM DM 50 Min.: 3 μM FM: 2 pulses: 760 nM FM: 3 pulses: 760 nM FM: 4 pulses: 760 nM FM: 5 pulses: 760 nM FM: 6 pulses: 760 nM FM: 8 pulses: 760 nM FM4: 7.5 min Intervals: 760 nM FM4: 10 min Intervals: 760 nM FM4: 15 min Intervals: 760 nM FM4: 20 min Intervals: 760 nM 0 200 400 600 800 1000 0 1 2 3 4 5 6 x10 4 0 2000 4000 6000 8000 10000 12000 Msn2 Area Under Curve (AU) Max YFP expression DDR2: Single Pulse vs. Multiple Pulses Single Pulse Multiple Pulses Response of Wild-Type DDR2 Promoter to single vs. multiple pulses. As can be seen, DDR2 responds similarly with both the single pulse and multiple pulses Conclusions DDR2 appears to respond strongly to Amplitude Modulation, but not Duration or Frequency Modulation. It effectively integrates the total amount of Msn2 in the nucleus over time Other properties of the DDR2 promoter, such as the threshold and dynamic range, can also be inferred from the data The next immediate step is to characterize all of the mutants and compare the behavior with the wild- type Futures studies could: Characterize using ChIP-Seq how Msn2 binding changes over time after induction with the inhibitor Try to create a completely synthetic promoter and predict its behavior given predicted promoter architecture and positioning of Msn2 binding sites Acknowledgments This project was supported by the generous contributions of the Howard Hughes Medical Institute (HHMI). Part of the work was conducted at the Center for Nanoscale Systems at Harvard University. Many thanks go to the rest of the O’Shea Lab for their patient and constant help and support for the duration of the summer. Last but not least, I am grateful for the wonderful and inspiring community within the Harvard College Program for Research in Science and Engineering (PRISE) program and the FAS Center for Systems Biology summer internship, who truly made this summer not only enriching, but also enjoyable. References 1. Hao, N. and O'Shea, E.K., 'Signal-dependent dynamics of transcription factor translocation controls gene expression". Nature Structural & Molecular Biology, 2011. 19(1): p. 31-9. 2. Lam, F.H., Steger, D.J., and O'Shea, E.K., "Chromatin decouples promoter threshold from dynamic range". Nature, 2008. 453(7192): p. 246-50. 3. Kim, H.D. and O'Shea, E.K., "A quantitative model of transcription factor-activated gene expression". Nature Structural & Molecular Biology, 2008. 15(11): p. 1192-8. YFP CFP
Transcript of The Influence of Promoter Architecture on Targeted Gene ... · The Influence of Promoter...
The Influence of Promoter Architecture on Targeted Gene Activation by Complex Transcription Factor Signaling
Jeffrey B. Wang, Anders S. Hansen, and Erin K. O’Shea FAS Center for Systems Biology, Harvard University, Cambridge MA 02138
Abstract Our studies focused on Msn2, a transcription factor in budding yeast responsible for mediating the response to a number of stresses, such as oxidative stress and starvation. Based on the identity of the stressor, Msn2 enters the nucleus with different patterns, such as entering and leaving the nucleus rapidly or periodically. However, how the promoter is able to understand these different signals is not well understood. Our project’s goal was to elucidate how the positioning of histones within promoter sequences allowed genes to be differentially activated by these various signals. We selected the DDR2 gene promoter, which possesses Msn2 binding sites both in and out of its well-defined nucleosome binding positions. Then, we systematically modified the position of its binding sites relative to the nucleosomes and utilized a fluorescent reporter that allowed us to quantify relative expression levels. We focused on how the promoter processes three aspects of the incoming Msn2 signal: the duration (how long Msn2 is present in the nucleus), the intensity (the maximum amount of Msn2 within the nucleus), and the frequency of oscillations (how quickly we move Msn2 in and out of the nucleus). We then fit these results to a set of differential equations describing how transcription factor binding can lead to protein production. With these results, we hope to identify how all of these binding sites cooperate to analyze complex Msn2 signaling.
Background
?
Experimental Approach Analyzing Gene Expression
Stress TF
TF
Response
Upon stress, the cell will often activate transcription factors, which subsequently translocate to the nucleus to induce expression of appropriate stress response genes
Msn2 is a general stress transcription factor found in Saccharomyces cerevisiae [1]. It has been shown to translocate into the nucleus with different patterns based on different types of stresses [2]
Purpose: How is the same transcription factor able to activate a presumably different gene expression profile using these different signals? Quantitatively characterizing how the promoter region is able to process these complex signals would be illuminating not only for the general stress response, but also gene regulation in higher organisms
Chromatin might be part of the answer, since it has been shown to influence the threshold and dynamic range of promoters [3]
Specific Goal: To elucidate how the positioning of Msn2 binding sites relative to nucleosomes influences the promoter’s ability to process these complicated signals
Generation of Promoter Variants We selected the DDR2 promoter for its well defined nucleosomes
and multiple Msn2 binding sites both in and out of the nucleus We then formed a dozen mutants manipulating the positioning of
Msn2 Four of the mutants consisted of a single binding site knockout Two of the mutants were a “slow” and “fast” variant consisting
of three binding sites within and outside of a nucleosome, respectively
The remaining represented a spectrum of promoter speeds from the slow and fast promoters
Promoter Architecture for DDR2: Nucleosome positioning was
determined with MNase-seq and depicted as gray peaks. Red
triangles depict locations of Msn2 Binding Motifs
Slow variant of the DDR2 Promoter
Msn2-mCherry
Nhp6a
CFP
YFP pDDR2*
pDDR2*
Site-Directed Mutagenesis
Plasmid
Yeast Transformation
Promoter variants were first formed using site-directed mutagenesis
Mutants were then transformed upstream of a YFP and CFP dual reporter in a preexisting yeast strain with Msn2 and Nhp6a (a nuclear protein) fluorescently tagged to track Msn2 and identify the nucleus
Controlling Msn2 Signals
Bcy1p
Bcy1p
Stress: low cAMP levels
TPK 1/2/3
No stress: high cAMP levels
Bcy1p
Bcy1p cAMP
TPK 1/2/3
Msn2 P P P
Cytoplasm Nucleus
Cytoplasm
Nucleus
Msn2 Stress Gene
Transcription OFF
No stress + 1-NM-PP1
Bcy1p
Bcy1p
TPK 1/2/3
Cytoplasm
Nucleus
Msn2
1-NM-PP1
YFP
CFP
YFP
CFP
Msn2
CFP
YFP
TPK 1/2/3
Msn2 is normally prevented from entering the nucleus by phosphorylation by PKA By adding an inhibitor for PKA, it is possible to force Msn2 in and out of the nucleus on the order of a
few minutes
cAMP
0 20 40 60 80 1000
500
1000
1500
Average Msn2 Localization
Time (min.)
Am
t. of
Msn
2 in
Nuc
leus
(AU
)
Duration
Amplitude
0 20 40 60 80 1000
500
1000
1500Average Msn2 Localization
Time (min.)
Am
t. of
Msn
2 in
Nuc
leus
(AU
) Number of Pulses
0 20 40 60 80 1000
500
1000
1500Average Msn2 Localization
Time (min.)
Am
t. of
Msn
2 in
Nuc
leus
(AU
)
Frequency of Pulses
3 different microscopy experiments were conducted measuring YFP/CFP intensity, while manipulating key features of the incoming Msn2 signal
Fast variant of the DDR2 Promoter
For each cell over a time period of 2.5 hours, Msn2 localization and CFP/YFP expression were measured
The results were fit to a set of ordinary differential equations (ODEs) describing activation of the promoter region followed by transcription of mRNA and translation of fluorescent protein [3]
0 20 40 60 80 1000
500
1000
1500Average Msn2 Localization
Time (min.)
Amt.
of M
sn2
in N
ucle
us (A
U)
YFP Expression
Time (min)
0 10 30 50 70 90 110 130 150
2000 4000 6000 8000 10000
0 50 100 1500
2000
4000
6000
8000
10000
Time (min)
Fluo
resc
ence
50 Minute Pulse, 3 µM: YFP Expressionmatured YFPmeasured YFP
Typical Results of Microscopy Experiment
Gene Expression Data from all 30 Conditions
0 1 2 3 4 5x 104
0
2000
4000
6000
8000
10000
12000
Msn2 Area Under Curve (AU)
Max
YFP
exp
ress
ion
DDR2: Amplitude modulation10 Min. Pulse20 Min. Pulse30 Min. Pulse40 Min. Pulse50 Min. PulseODE Prediction
Response of Wild-Type DDR2 Promoter to Changes in the Total Integrated Amount of Nuclear Msn2 over
time. Different colored points indicate different durations of signals. Note how all of the points follow
the same general pattern regardless of duration.
Msn2 Nuclear Localization
Time (min)
0 10 30 50 70 90 110 130 150
DM 10 Min.: 140 nMDM 10 Min.: 330 nMDM 10 Min.: 760 nMDM 10 Min.: 3 µMDM 20 Min.: 140 nMDM 20 Min.: 330 nMDM 20 Min.: 760 nMDM 20 Min.: 3 µMDM 30 Min.: 140 nMDM 30 Min.: 330 nMDM 30 Min.: 760 nMDM 30 Min.: 3 µMDM 40 Min.: 140 nMDM 40 Min.: 330 nMDM 40 Min.: 760 nMDM 40 Min.: 3 µMDM 50 Min.: 140 nMDM 50 Min.: 330 nMDM 50 Min.: 760 nMDM 50 Min.: 3 µM
FM: 2 pulses: 760 nMFM: 3 pulses: 760 nMFM: 4 pulses: 760 nMFM: 5 pulses: 760 nMFM: 6 pulses: 760 nMFM: 8 pulses: 760 nM
FM4: 7.5 min Intervals: 760 nMFM4: 10 min Intervals: 760 nMFM4: 15 min Intervals: 760 nMFM4: 20 min Intervals: 760 nM
0 200 400 600 800 1000
0 1 2 3 4 5 6 x104
0
2000
4000
6000
8000
10000
12000
Msn2 Area Under Curve (AU)
Max
YFP
exp
ress
ion
DDR2: Single Pulse vs. Multiple PulsesSingle PulseMultiple Pulses
Response of Wild-Type DDR2 Promoter to single vs. multiple pulses. As can be seen, DDR2 responds
similarly with both the single pulse and multiple pulses
Conclusions DDR2 appears to respond strongly to Amplitude Modulation, but not Duration or Frequency
Modulation. It effectively integrates the total amount of Msn2 in the nucleus over time Other properties of the DDR2 promoter, such as the threshold and dynamic range, can also be
inferred from the data The next immediate step is to characterize all of the mutants and compare the behavior with the wild-
type Futures studies could: Characterize using ChIP-Seq how Msn2 binding changes over time after induction with the
inhibitor Try to create a completely synthetic promoter and predict its behavior given predicted promoter
architecture and positioning of Msn2 binding sites
Acknowledgments This project was supported by the generous contributions of the Howard Hughes Medical Institute (HHMI). Part of the work was conducted at the Center for Nanoscale Systems at Harvard University. Many thanks go to the rest of the O’Shea Lab for their patient and constant help and support for the duration of the summer. Last but not least, I am grateful for the wonderful and inspiring community within the Harvard College Program for Research in Science and Engineering (PRISE) program and the FAS Center for Systems Biology summer internship, who truly made this summer not only enriching, but also enjoyable.
References 1. Hao, N. and O'Shea, E.K., 'Signal-dependent dynamics of transcription factor translocation controls gene expression". Nature
Structural & Molecular Biology, 2011. 19(1): p. 31-9. 2. Lam, F.H., Steger, D.J., and O'Shea, E.K., "Chromatin decouples promoter threshold from dynamic range". Nature, 2008.
453(7192): p. 246-50. 3. Kim, H.D. and O'Shea, E.K., "A quantitative model of transcription factor-activated gene expression". Nature Structural &
Molecular Biology, 2008. 15(11): p. 1192-8.
YFP CFP
