How does pituitary androgen signalling support lifelong health and wellbeing? An integrated transgenic and systems biology approach

Lead Research Organisation: University of Edinburgh
Department Name: MRC Centre for Reproductive Health

Abstract

The pituitary gland holds such influence over the body it is commonly dubbed the 'master gland'. Located at the base of the brain, it responds to local and circulating factors by secreting circulating hormones that influence the function of many organs and tissues to support the body's health and wellbeing. Each hormone made by the pituitary is produced by a unique cell-type, which link together to form their own networks inside the gland. Currently we don't completely understand how the pituitary's cells respond to signals from the rest of the body to control the production of these hormones.

Testosterone is produced by the testes and is required for healthy development and function of men's bodies, but is also made in small amounts by women too. Receptors for testosterone (called androgen receptors) are found inside all of the cell types in the pituitary gland. It was previously thought that testosterone was required for the production of a pituitary hormone called luteinising hormone (that itself controls the production of testosterone by the testes in a negative feedback loop), however in a recent ground-breaking study using transgenic mice we have shown that blocking of the testosterone signal in the pituitary by knocking out androgen receptors, changes production of two of the pituitary hormones: prolactin and growth hormone, but not luteinising hormone.

Taken together, these exciting findings demonstrate that our understanding of the role of androgens in the pituitary is far from complete and that an urgent re-evaluation of this system is required. Therefore in this project, we aim to unequivocally establish the fundamental role(s) for androgen signalling in the pituitary, and how this supports lifelong health and wellbeing.

We will first produce a computer simulation of pituitary hormone production using software established in our previous BBSRC-funded projects. This will coalesce all of the published information on pituitary function into a single location and will be used throughout the project to support experimental design, with results fed back into the model to refine our understanding.
In our laboratory experiments, we will use mice as a model as their pituitaries contain androgen receptor and produce the same hormones as humans. We can also manipulate the genetics of the mouse by the deletion of genes (known as knockouts or by using mouse genes to control the production of fluorescent reporter genes that label specific cell types.
Firstly, we will knockout testosterone signalling just within the prolactin-producing cells (lactotrophs) of the male pituitary to determine whether circulating prolactin concentration is controlled by testosterone in the lactotroph or in another pituitary cell-type. We will determine how testosterone normally suppresses prolactin production, and whether the organisation of the network of lactotroph cells within the pituitary is shaped by testosterone.
We will define the role of testosterone in controlling the growth hormone-producing cells of the pituitary (the somatotrophs). We will determine whether testosterone controls the frequency of growth hormone secretion, and whether the organisation of somatotroph cells within the pituitary is shaped by testosterone.
We will also define the role of testosterone in the luteinising hormone producing cells of the pituitary (the gonadotrophs). We will dissect the interplay between testosterone, prolactin and luteinising hormone, establishing how these hormones work together in the pituitary.
Finally we will determine the localisation of androgen receptor in the female pituitary, and whether blocking the testosterone signal in females impact the estrous cycle or pregnancy.
Completion of this project will result in a comprehensive redefinition of our understanding of the roles for pituitary testosterone signalling; significantly improving our understanding of the maintenance of lifelong health and wellbeing.

Technical Summary

The anterior pituitary is composed of a heterogeneous population of endocrine cell types, each of which expresses androgen receptor (AR). We have recently shown that male mice with a whole pituitary AR ablation (PARKO) have altered lactotroph and somatotroph function, with a possible role in gonadotrophs. This project will define the roles of AR in each of these cell types and determine the localisation and role of AR in the female pituitary.
1. We will produce a dynamic molecular model of pituitary hormone production using Signalling Petri-Net (SPN) modelling, which will be updated throughout the project.
2. We will determine whether prolactin and growth hormone (GH) production and secretion are controlled by AR signalling in the lactotroph, somatotroph or in another pituitary cell type by interrogating the endocrine phenotypes of lactotroph and somatotroph-specific AR knockouts using ELISAs, immunohistohemistry (IHC) and qPCR.
3. We will determine whether the lactotroph and somatotroph cell networks form normally by using iDISCO whole pituitary fixation, IHC and 2-photon microscopy to generate 3D reconstructions of these networks.
4. We will determine the mechanism of AR impact on prolactin and GH production with a combination of electron microscopy, chromatin immunoprecipitation, in silico ARE interrogation andex vivo pituitary imaging of AR and Prl reporter gene expression in cell-specific AR ablation models.
5. We will determine whether high serum prolactin can suppress increased LH production in the PARKO model by pharmacologically lowering prolactin levels using the dopamine receptor agonist cabergoline and performing ELISAs and qPCR for pituitary endpoints.
6. We will determine the localisation of AR in the female pituitary using IHC and define its role in pregnant and non-pregnant females by characterising phenotypic changes in female PARKO mice.
Completion of this project will comprehensively redefine the roles for pituitary androgen signalling.

Planned Impact

We anticipate that the impact of our research will be wide-ranging and will result in beneficiaries from the following sectors:

Academic trainees: Investment in this project will result in capacity building in the UK science base through the delivering and training of highly skilled researchers, including supporting the career development and training of a young female Researcher Co-Investigator and training MSc and BSc students as part of this project. The timescale of this impact is within the three years of the grant but also beyond, as the trainees develop their careers on the strong foundation of training received during this project.

Academic peers: This project will enhance the knowledge economy by providing a platform of understanding to support downstream fundamental and translational research in this area both by ourselves, our collaborators, and the wider international scientific community. More immediately, our development and utilisation of new and innovative methodologies with significant potential for wider applications will benefit the research community, as this information will be widely disseminated, with data and resources made freely available to others for non-commercial use. We will also promote the use of in silico modelling as a cross-disciplinary tool for biologists to use for pathway mapping, hypothesis generation and reduction of use of animals in research. The timescale of this impact will begin as soon as results are disseminated, which will be within the first two years of the project.

Clinical researchers: The fundamental understanding of pituitary function gained from this project will provide a basic understanding for downstream research into clinical pituitary pathologies such as pituitary cell type hyperplasia and tumours. It is challenging for clinicians to understand these pathologies and for appropriate therapies to be developed when the most fundamental understanding of the underlying physiology is still incomplete. Whilst not a clinical project, with no specific focus on pathology, this research may also identify biomarkers predisposing to poor endocrine profiles, or reflecting perturbed HPG axis function and identifying potential pathways and targets for therapeutic development. We anticipate that the timescale for these advances will be within ten years of project completion.

Veterinary researchers: Results generated from this project benefit veterinarian and livestock researchers working on the role of androgens in reproductive and general health in animals with additional relevance for productivity in farm animals due to the influence of prolactin on seasonal breeding and milk production. Inhibition of androgen receptor is also a target for novel contraceptives and sterilants, (see Smith's recent Michelson grant award http://www.michelsonprizeandgrants.org/michelson-grants/current-grantee-profiles). We anticipate that the timescale for these advances will be within five years after completion of this project.

The UK public: Our research will benefit the public by providing a platform of fundamental understanding of the processes underpinning pituitary hormone production, which will add to public health advice for maintaining a long and healthy life and if necessary, therapeutic approaches to pituitary disease. Through our results and communication plan we will raise public awareness of men's health as a significant societal issue. We anticipate that the timescale for these results will begin within as soon as results are disseminated and our public engagement plan is actioned.

The UK economy: Spin-out company development will be facilitated by the University of Edinburgh's strong commitment to translation and commercialisation, and their investments in this area. We have extensive experience of such commercialisation with Freeman founding and acting as Director of two spin-out companies: Fios Genomics (www.fiosgenomics.com/) and Kajeka Ltd (www.kajeka.com).
 
Description As of March 2020, the award has resulted in the generation of a large amount of data that has already resulted in the publication of one paper (PMID: 29690918), another that is currently in revision after review at the Journal of Neuroendocrinology, and two manuscripts that are currently being prepared for submission.

Some of the key findings from the award include

1. Generation of in silico models of testicular androgen production, and the control of pituitary prolactin production.
2. Refinement of the understanding of the control mechanism of pituitary prolactin production by androgen receptor signalling in the cells of the anterior pituitary
3. Refinement of the understanding of the control mechanism of pituitary gonadotrophin production by androgen receptor signalling in the hypothalamic kisspeptin neurons
4. Characterisation of the pituitary folliculostellate cell type in mice


1. We had previously developed a novel in silico modelling system that integrates clear and unambiguous diagrams of biological pathways with dynamic model production (PMID: and )
We aimed to use this to make a model of androgen receptor signalling in the pituitary to assist with subsequent studies (Objective 1: Generate a dynamic in silico model of the pituitary: toward a digital organism). This resulted in the production of a model of steroidogenesis (PMID: 29690918) and a large model of the control of pituitary prolactin production which we aim to integrate into a publication.

2. We had previously shown that circulating prolactin in a whole-pituitary androgen receptor knockout male mouse model is 10x greater than controls (PMID: 25799562
), determining androgen signalling as a novel repressor of pituitary prolactin production. We aimed to specify the location of this repression ('Objective 2: Define the role of androgen signalling in the lactotrophs of the male pituitary).
To address this objective, we made both lactotroph specific, and Pit1 lineage-specific (lactotroph, somatotroph and thyrotroph) androgen receptor knockout male mice, and compared them to the whole pituitary androgen receptor knockout mice (Objective 3: define the role of androgen signalling in the somatotrophs of the male pituitary). The lactotroph and Pit-1 lineage specific androgen receptor knockout mice did not show the 10x greater increase, suggesting that the location of androgen receptor negative feedback on prolactin release at the pituitary was not the lactotroph, thyrotroph or somatotroph.
We further investigated whether the large increase in circulating prolactin was due to a non-specific ablation of androgen receptor in the hypothalamus. The hypothalamus controls pituitary prolactin production through the production of inhibitory dopamine by TIDA neurons. An inhibition of dopamine production would cause an increase in pituitary prolactin release. We found that androgen receptor was expressed by these neurons but not non-specifically ablated in the whole pituitary androgen receptor knockout model. Circulating prolactin did also not increase in a neuron-specific androgen receptor knockout model.
Although the generation of these three new knockout models did not reveal the location of the androgen inhibition on prolactin production, we have ruled out several mechanisms. Further investigation would centre on the role of androgen signalling in the gonadotrophs, or on the non-endocrine folliculostellate cells on pituitary prolactin production.

3. In our previous publication (PMID: 25799562) we had also shown that circulating gonadotrophins luteinising hormone (LH) and follicle stimulating hormone (FSH) do not change in a model of male mouse pituitary androgen receptor ablation. To investigate whether this unexpected result was due to repression of the HPG axis by the high circulating prolactin, we treated these mice with a dopamine receptor agonist to reduce pituitary prolactin production (Objective 4: Define the role of androgen signalling in the gonadotrophs of the male pituitary). However the repression of prolactin production in these mice did not reveal an underlying increase in pituitary gonadotrophin production in the absence of androgen receptor signalling, so we concluded that pituitary androgen signalling is not required to control gonadotrophin production.

4. These studies were complemented by the investigation of the pituitary folliculostellate network in mice. This non-endocrine cell type in the anterior pituitary is important in the coordination of physiological feedback and release of hormones by the endocrine cell networks. Through characterisation and lineage tracing studies we determined that the folliculostellate cell network in mice expresses mural cell markers, and may behave like an extension of the mural cell network. This study is currently in revision after review at the Journal of Neuroendocrinology.
Exploitation Route Outcomes of the award redefine our understanding of the role of androgens in the pituitary, this information will be of value to future studies of endocrine disorders
Sectors Healthcare

 
Description The results described in key findings point 2 have been presented at two large international conferences: European Congress of Endocrinology, Lisbon, 2017 International Congress of Neuroendocrinology, Toronto, July 2018 Both posters obtained active engagement from scientists, clinicians and industrial partners. The researcher co-investigator was also invited to give a research seminar on the data generated by the project at the University of Newcastle, Australia in August 2018. The researcher co-investigator is actively involved in teaching reproductive biology and endocrinology at two institutions (University of Edinburgh, UK and Edinburgh-Zhejiang Institute, China) and incorporates selected results from her research into her teaching. Two students have completed projects relating to the grant: one MSc student from the University of Edinburgh MSc Reproductive Sciences course and one honours student from the University of Edinburgh BSc Medical Sciences course.
First Year Of Impact 2017
 
Description Project support grant: Characterisation of PDGFR beta cells in the anterior pituitary
Amount £5,000 (GBP)
Organisation British Society for Neuroendocrinology 
Sector Charity/Non Profit
Country United Kingdom
Start 06/2017 
End 05/2018
 
Title Development and roll out of an in silico modelling system, 
Description There is a need for formalised diagrams that both summarise current biological pathway knowledge and support modelling approaches that explain and predict their behaviour. Here, we present a new, freely available modelling framework that includes a biologist-friendly pathway modelling language (mEPN), a simple but sophisticated method to support model parameterisation using available biological information; a stochastic flow algorithm that simulates the dynamics of pathway activity; and a 3-D visualisation engine that aids understanding of the complexities of a system's dynamics. We present example pathway models that illustrate of the power of approach to depict a diverse range of systems. 
Type Of Material Technology assay or reagent 
Year Produced 2016 
Provided To Others? Yes  
Impact Only just published. No impacts as yet 
URL https://www.ncbi.nlm.nih.gov/pubmed/27509052