Bilateral NSF/BIO-BBSRC: Quantifying cellular signalling by dynamically pulsing transcription factors

Lead Research Organisation: University of Edinburgh
Department Name: Sch of Biological Sciences

Abstract

When the environment surronding yeast cells change, the cells usually experience stress. This stress causes a protein to change its location within the cell moving from the cytoplasm to the cell nucleus. Surprisingly, the protein does not change location once but moves in and out of the nucleus multiple times in a pulsatile manner. In our own cells, p53, a tumour supressor protein, behaves in a similar fashion when DNA is damaged.

It is not well understood why cells should response to change with proteins that have pulsatile dynamics. Using budding yeast and time-lapse microscopy to follow the behaviour of single cells, we will determine if pulsatile dynamics allow the cell to encode environmental stress into the characteristics of the pulses, with each type of environmental change generating a particular dynamics. To respond to stress, cells produce new proteins, but to match the types of new proteins they produce to the stress they are experiencing, they need to be able to determine the type of stress from the dynamics of the proteins that pulse in response to the stress. Therefore, we will determine the extent to which the dynamics of pulsatile factors can be interpreted by downstream genes to allow protein production to be tailored to the environment. Further, we will ask if mild stress is used by yeast cells as a warning for the imminent occurrence of new stress and if cells can better survive if they use such anticipatory behaviour.

Technical Summary

Recent single-cell studies have revealed a temporal complexity of eukaryotic signalling previously obscured by population averaging. In particular, diverse systems appear to activate transcription factors in a pulsatile fashion in response to inputs. In these systems, each pulse involves many molecules of a transcription factor being simultaneously activated and then deactivated. Such pulses occur repetitively. For example, in mammalian cells, the tumour suppressor p53 can be activated in pulses when subjected to DNA damage stress. Similarly, in yeast, the general stress response regulator Msn2 also exhibits pulses in response to various stresses.

Pulsatile dynamics may allow the cell to encode environmental signals into characteristics of pulses, which in turn can be decoded into downstream target output. The extent of information encoding and decoding as well as the long-term physiological impacts of pulsing remain, however, largely unknown.

Here we will address these questions in yeast by collaboration between the Swain and the Elowitz laboratories. Using time-lapse microscopy, microfluidics, and techniques from information theory, we will map and quantify the relationship between the dynamics of multiple stress-responsive transcription factors and the nature of extracellular stress. We will thus quantify how well the dynamics encode the type of extracellular stress. Further, we will determine and quantify the relationship between the transcription factors dynamics and the expression levels of their downstream target genes, thus measuring the ability of cells to transfer information from one level of control (transcriptional regulators) to another (downstream gene expression). Finally, we will test the hypothesis that variation in the response of isogenic cells to stress enables history-dependent and anticipatory behaviours and determine if such behaviours increase fitness by measuring single-cell growth rates under sequentially applied stresses.

Planned Impact

We see two main groups of beneficiaries:

(i) industry producing anti-microbial drugs, particularly fungicides;

(ii) the general public through increasing their awareness and understanding of the interdisciplinary nature of today's bioscience.

First, it has long been recognized that microbes respond heterogeneously to anti-microbial drugs, but it is only recently that technology has developed sufficiently to quantitatively monitor those responses in real time. This heterogeneity can generate persister cells: those that can survive drug treatment and potentially re-grow and re-initiate infection. Drugs are a form of stress and it is believed that persister cells are in a stress-resistant state.

Importantly, our focus is on yeast: there are much fewer classes of antifungal agents compared to antibacterial drugs limiting therapeutic options. Surprisingly, medical advances have actually increased the number of life-threatening fungal infections through, for example, organ transplants and new treatments for cancers. Invasive fungal infections were estimated to cost $2.6 billion in the United States in 2008.

Our research through its focus on individual cells and its characterization of responses to time-varying stress, the adaption to stress, and single-cell fitness provides novel, quantitative means to characterize the efficacy of anti-microbial drugs. For example, we can flow time-varying concentrations of a fungicide over cells using microfluidics, quantify the resulting stress response of the cells with fluorescent markers, and determine the fitness of each cell by counting numbers of divisions. We can evaluate the effectiveness both of different regimens of applying the drug and of different combinations of drugs, including, for example, those that might inhibit the stress response. Further, we can monitor the recovery of persister-like cells (our experiments can run for days) and determine if there are periods when these cells are vulnerable to a re-application of the drug or to a different fungicide.

Second, our research has the capacity to engage the public with its focus on fluorescence and time-lapse microscopy, which provide powerful visual tools for education and increasing scientific awareness.

One goal is to demonstrate the interdisciplinary approaches necessary to understand biological phenomena. Our research uses techniques from communication engineering and informatics, as well as cell and molecular biology. Most high school and many undergraduates are not aware of interdisciplinary research with curricula still typically following traditional silos. Yet breakthroughs often come from those working at the edge of disciplines, and we aim to encourage both those interested in the physical sciences to consider research in biology and those already interested in biology to realize the value of training in the physical sciences and mathematics.

A second goal is to illustrate the importance of "blue skies" research, particularly that studies on yeast can inform on our own physiology. For example, yeast undergo an adaptive stress response, where mild stress allows cells to survive more severe stress, and an analogous phenomena also occurs in humans. Similarly, low nutrients, a form of stress, can prolong life in yeast cells, and such dietary restriction reduces aging in many organisms, including primates. Another example, key to our research, is the study of cell-to-cell heterogeneity, which appears at first sight to be of only academic interest, but is now recognized to be fundamental for understanding anti-microbial resistance. Finally, taken together, these illustrations will underpin the importance of having an evolutionary perspective for comprehending and manipulating our world.

Our third goal is to show how anti-microbials work and the importance of following prescribed drug regimens, which we will illustrate in real time through aperiodic applications of a drug using microfluidics.

Publications

10 25 50

publication icon
Granados AA (2018) Distributed and dynamic intracellular organization of extracellular information. in Proceedings of the National Academy of Sciences of the United States of America

 
Description We have quantitatively analyzed a library of fluorescently tagged transcription factors (22 transcription factors are being studied) in four different types of stress: carbon stress, nitrogen stress, osmotic stress, and oxidative stress. For each stress and each transcription factor we have time-lapse data for approximately 200 cells followed for 10 hours showing whether the transcription factor enters or leaves the nucleus. Further, we have developed, in collaboration with Gasper Tkacik and Sarah Cepeda at the IST Austria, novel methods to measure the degree of correlation between the nuclear translocation of the transcription factor and the type of stress the cells experience. Our results shown that transcription factors can encode enough information in the dynamics of their nuclear translocations to unambiguously report an environmental change if that change is sufficiently large, that the nature of the change can also be encoded although with some degree of error, that how the information is encoded alters for changes of different magnitudes, and that no single transcription factor can accurately encode both the nature and magnitude of environmental change.
Exploitation Route We believe the algorithm developed in collaboration with Tkacik and Cepeda of the IST Austria will be of interest to others analyzing time-series data.
Sectors Manufacturing, including Industrial Biotechology

 
Description Development of information theory methodologies to characterize single-cell data 
Organisation Institute of Science and Technology Austria
Country Austria 
Sector Academic/University 
PI Contribution We have provided single-cell data and contributed to the development of the information theoretic methods.
Collaborator Contribution The team headed by Gasper Tkacik have led the development of the algorithm.
Impact The collaboration is multi-disciplinary involving molecular biology and statistics and computer science.
Start Year 2016
 
Title Software for estimating mutual information from stochastic time series data by decoding 
Description Software to estimate the mutual information - a general measure of statistical dependency - between a discrete set of inputs and outputs that are distributions of time series. 
Type Of Technology Software 
Year Produced 2018 
Impact N/A 
URL https://github.com/swainlab/mi-by-decoding