Promoters maintain their relative activity levels under different growth conditions

Leeat Keren, Ora Zackay, Maya Lotan‐Pompan, Uri Barenholz, E. Dekel, Vered Sasson, Guy Aidelberg, Anat Bren, Danny Zeevi, Adina Weinberger, Uri Alon, Ron Milo, Eran SegalMolecular 2013 · DOI

TL;DR

This study reveals that most gene promoters in yeast and E. coli maintain their relative activity levels across different growth conditions, with changes primarily driven by a global scaling factor dependent on the condition, not the promoter. This finding simplifies our understanding of gene regulation, suggesting that many expression changes are due to global cellular resource allocation rather than specific regulatory events.

What Problem Does This Paper Address?

Understanding how gene expression changes in response to different environmental conditions is fundamental to biology. While it's known that many genes alter their expression levels, a key question has been how much of this change is due to specific regulatory mechanisms targeting individual genes versus broader, global cellular responses. Previous studies often focused on identifying up- or down-regulated genes, but lacked a quantitative framework to disentangle these global and specific effects. This ambiguity makes it difficult to pinpoint true regulatory events and understand the underlying principles governing genome-wide expression profiles.

Methods

The researchers employed a high-throughput experimental system using fluorescent reporters (YFP) fused to native yeast and E. coli promoters. This allowed for accurate, live-cell measurements of promoter activity across multiple growth conditions. They developed a robust statistical approach to analyze the data, focusing on identifying a "global scaling factor" that describes how most promoters' activities change proportionally between conditions. Promoters that deviated significantly from this global trend were identified as "condition-specific." They also utilized k-means clustering to group promoters with similar scaling behaviors across all conditions. Finally, they proposed and tested a passive resource allocation model to explain the observed global scaling factors.

Datasets

The study utilized two main datasets:

  • Saccharomyces cerevisiae (yeast): Activities of ~900 native promoters measured across 10 different environmental growth conditions.
  • Escherichia coli (E. coli): Activities of ~1800 native promoters measured across 9 different environmental growth conditions.

These datasets represent newly generated, comprehensive measurements of promoter activity under various conditions, building upon existing knowledge of gene regulation.

Key Insights and Ideas

  • Proportional Scaling is Dominant: The most striking finding is that the vast majority of promoters (60-90%) change their expression between conditions by a constant factor, meaning their relative activities are preserved. This suggests a fundamental principle of gene regulation.
  • Global Factors Drive Most Changes: This proportional scaling implies that most expression changes are driven by global cellular parameters (like resource availability or growth rate) rather than specific gene regulation.
  • Specific Regulation is a "Fine-Tuning" Layer: True specific regulation is reserved for a smaller subset of promoters that deviate from the global trend. These specific responses often involve functionally related groups of genes.
  • A Handful of Scaling Factors Describe the Genome: The entire genome's expression profile across conditions can be surprisingly well-described by a small number of "scaling lines" or clusters, indicating a much simpler underlying structure than previously assumed.
  • Resource Allocation Model: A simple, parameter-free model based on passive resource allocation can quantitatively explain a large portion of the observed global scaling factors, linking them to growth rate and the magnitude of specific responses.

Key Results

  • In both yeast and E. coli, 60-90% of promoters showed proportional scaling of activity between any two tested conditions.
  • This global scaling factor was condition-dependent but promoter-independent.
  • K-means clustering revealed that six clusters could explain 97% of the variance in the yeast promoter activity data across all conditions.
  • A passive resource allocation model, considering growth rate and specific gene responses, accounted for approximately 90% of the variability in global scaling factors.
  • The study demonstrated that a small number of representative promoters (10 in yeast, 5 in E. coli) could accurately predict the activity of hundreds of other promoters in a new condition.

Limitations and Future Work

The authors acknowledge that their library, while extensive, does not include all yeast promoters, which might introduce some inaccuracies in estimating total cellular activity. The resource allocation model, while effective, doesn't fully account for all observed proportional responses, potentially due to other global factors like cell size or macromolecular composition.

Future work could explore:

  • Whether this proportional scaling holds true for mRNA and protein levels, not just promoter activity, and at the single-cell level.
  • The precise molecular mechanisms responsible for this proportionality and the coordination with growth rate.
  • The role of specific cellular components (like cAMP, ppGpp, sigma factors, or nutrient sensing pathways) in mediating these global responses.
  • The extent to which this phenomenon applies to other conditions beyond growth rate, such as dynamic environmental changes, cell cycle stages, or mutations.

Summary updated

Abstract

Most genes change expression levels across conditions, but it is unclear which of these changes represents specific regulation and what determines their quantitative degree. Here, we accurately measured activities of ~900 S. cerevisiae and ~1800 E. coli promoters using fluorescent reporters. We show that in both organisms 60-90% of promoters change their expression between conditions by a constant global scaling factor that depends only on the conditions and not on the promoter's identity. Quantifying such global effects allows precise characterization of specific regulation-promoters deviating from the global scale line. These are organized into few functionally related groups that also adhere to scale lines and preserve their relative activities across conditions. Thus, only several scaling factors suffice to accurately describe genome-wide expression profiles across conditions. We present a parameter-free passive resource allocation model that quantitatively accounts for the global scaling factors. It suggests that many changes in expression across conditions result from global effects and not specific regulation, and provides means for quantitative interpretation of expression profiles.
BibTeX
@inproceedings{keren2013promoters,
  author = {Leeat Keren and Ora Zackay and Maya Lotan‐Pompan and Uri Barenholz and E. Dekel and Vered Sasson and Guy Aidelberg and Anat Bren and Danny Zeevi and Adina Weinberger and Uri Alon and Ron Milo and Eran Segal},
  title = {{Promoters maintain their relative activity levels under different growth conditions}},
  booktitle = {Molecular},
  year = {2013},
  doi = {10.1038/msb.2013.59},
}
Figure