Searchable abstracts of presentations at key conferences in endocrinology
Endocrine Abstracts (2015) 37 EP163 | DOI: 10.1530/endoabs.37.EP163

1Laboratories for Integrative Neuroscience and Endocrinology, School of Clinical Sciences, University of Bristol, Bristol, UK; 2School of Mathematics, University of Bristol, Bristol, UK; 3Texas A&M University, College Station, Texas, USA; 4Department of Mathematics, College of Engineering, Mathematics and Physical Sciences, University of Exeter, Exeter, UK.


One way in which cells receive information about their environment is through hormones binding to appropriate receptors on their cell surface. Single cell measurements of signalling proteins typically reveal high cell–cell variability raising questions about how reliably individual cells sense their environment in order to make decisions. Information theoretic approaches can be used to explore such sensing, treating cell signalling pathways as ‘noisy’ communication channels. Mutual information (MI) can be calculated between system inputs and outputs as a statistical measure of the reliability of sensing. GnRH acts via Gq/11-coupled seven-transmembrane receptors to stimulate ERK and nuclear factor of activated T-cells (NFAT), but information transfer has not previously been quantified for these (or other) hormone receptors. Here, we do so using automated fluorescence microscopy to quantify dual-phosphorylated (pp)ERK and the nuclear fraction (NF) of NFAT1c–EFP as activation readouts. In gonadotroph-derived LβT2 cells, GnRH increased ppERK and NFAT-NF levels with maximal effects at 5–60 min. MI between GnRH and ppERK or between GnRH and NFAT-NF followed similar time courses but did not exceed 0.8 bits, implying that information transfer through these pathways is insufficient for an individual cell to unambiguously distinguish even two states of the environment. We also used imaging reporters for transcription activation (Egr1-driven or NFAT-RE-driven asRED) and again MI between GnRH and these outputs was always <0.8 bits. As a possible explanation for such unreliable sensing we consider GnRH receptor (R) number, using recombinant adenovirus to express GnRHR at sub-physiological to super-physiological levels (<20 000–>170 000 sites/cell) in HeLa cells. This revealed that information transfer is indeed dependent on GnRHR number but MI values were <0.8, even at the highest GnRHR levels. Accordingly, this novel approach to quantification of information transfer suggests that individual pathways and cells are very inefficient sensors of GnRH concentration.

Disclosure: This work was supported by the Biochemical and Biophysical Science Research Council (BBSRC grant number BB/J014699/1).

Article tools

My recent searches

No recent searches.