ESPOD Proposal Vidal‐Puig/Dougan (WTSI) and Steinbeck (EBI)
Identification of new players in human brown and beige adipocyte recruitment/activation
as an anti-obesity strategy
Antonio Vidal-Puig (TVP, WTSI), Christoph Steinbeck (CS, EBI), Gordon Dougan (GD,
WTSI),
Background – An important Biomedical Problem and a new original approach
Obesity and associated disorders are a major and increasing problem for the western world
for which there is an urgent need to develop novel therapies. Brown adipose tissue (BAT) is an energy
dissipating “machine” with great therapeutic potential. However, BAT in humans is limited and
poorly characterised. BAT is the main site of non-shivering thermogenesis in mammals and is
activated and recruited in response to cold and a hypercaloric diet, a process referred to as adaptive
thermogenesis [1, 2]. BAT activation decreases body fat content, reduces plasma triglycerides [3]
and improves glucose homeostasis in mice [4-6].BAT is also present and functional in young lean
human adults, and its tissue mass is inversely related to human adiposity suggesting a potential
causal relationship [7-8]. This evidence supports the concept that increasing BAT mass/activity
or browning of WAT could be a safe and efficient strategy to prevent or treat obesity [4] and
its associated complications. Of note activation of BAT has recently been linked to the
functional state of adipose tissue macrophages (ATMs). Thus, we propose to use human
pluripotent stem cells as a tool to give us a unique insight into the biology of human brown
adipocytes and its interactions with human macrophages.
Preliminary results: Towards this goal, Vidal-Puig and Dougan’s labs have already
developed step by step protocols for directed differentiation of human iPS cells into WAT and
BAT and macrophages respectivel [9,10]. Both groups are partnering in the study of molecular
interactions between adipocytes and ATMs using advanced co-culture systems. In addition to
hypothesis driven mechanistic studies, we are generating large scale RNASeq, proteomics,
phosphoproteomics and metabolomics datasets after co-culturing iPS derived BAT, WAT and
Macs under different polarising conditions [13]. Preliminary analysis of the RNA-seq data
generated on the developmental states of adipocytes has already uncovered significant changes in
gene expression during the course of this development.
Specific Objective: In this project, we propose to combine adipocyte and macrophage
expertise of TVP lab and Dougan lab (both laboratories are currently co-localised within Sanger
Institute) with bioinformatics expertise of Steinbeck group to integrate these multi-omics dataset
to build a quantitative model. This model will be then mined to identify the key molecular
interactions between adipocyte and ATMs that promote BAT activation and WAT browning. We
will use this model to understand key changes during adipocyte differentiation, the mechanisms of
cell commitment to these changes, possibly hidden cell states, and to identify candidate targets to
develop novel therapeutics to induce BAT recruitment/WAT browning and thereby treat metabolic
disorders. The experience gained on this will be transferable and applicable to other metabolic
disorders.
Research plan
In a first phase, the fellow will develop an integrated network model of signal transduction,
gene regulation and metabolism in differentiating and mature brown/beige adipocytes; and
establish role of polarised macrophages in these processes This model will be built using the
transcriptomics, metabolomics/lipidomics, and proteomics data generated in TVP and Dougan labs
from human PSC-derived brown/beige adipocytes and macrophages along with prior knowledge
of the underlying molecular pathways obtained from repositories at EBI (e.g. Reactome,
1
ESPOD Proposal Vidal‐Puig/Dougan (WTSI) and Steinbeck (EBI)
Biomodels) and elsewhere (KEGG, BioCyc, WikiPathways). Data coming from BAT isolated
from mouse embyos and cold vs thermoneutrality exposed mice will be used to validate the model.
While well-established mathematical models for individual levels exist, integrating these different
layers is an important challenge [11]. During this project the fellow will receive training in
dynamic (ODE) and stationary models (FBA, FVA, etc) of metabolism, and discrete logical
networks for gene regulation and signalling. A differential equation model of relevant areas of
metabolism will be integrated with signalling and regulation as described in [12]; Larger Flux
Balance Analysis stationary models of metabolism will be integrated with expression data using
different approaches [13]. Both strategies will provide complementary views, one more focused
and the other more general. The data integration will require RNA-Seq data analysis, to be done
with the existing IRAP pipeline [14] developed within the EBI, and metabolomics data analysis,
which will be done with existing metabolomics workflow environments [15, 16], scaled within the
PhenoMeNal project of CS lab. Senior members at the CS lab are well acquainted with the IRAP
RNA-Seq pipeline, metabolomics data analysis [17], and dynamic/stationary modelling of
metabolism [18, 19]. CS lab has close collaborators as well with a track record in modelling, such
as Kieran Patil at EMBL.
The human PSC-derived brown/beige adipocytes used to build the model will be
differentiated using chemically define protocols mimicking the physiological/natural
differentiation processes of these cells, from the mesodermal to the mature active state. There is
limited knowledge about the development of human BAT and its potential recruitment in adults
and the role that macrophages may play on this process. Therefore until now we relied on insights
from knockout and Cre-recombinase-based lineage tracing mouse models. It is crucial to validate
these insights in a human developmental model system since understanding the origins of the
BAT/brown like lineages and mapping the transcriptomes, signalling pathways and metabolism of
BAT/beige cells precursors is key in order to understand their function in the adult and in metabolic
disease.
In a second phase, the fellow will use the model to identify novel pharmaceutical strategies
to induce BAT generation/WAT browning. The model will be used to evaluate in silico the
potential effect of drugs on adipocytes. In combination with this model and the existing data, we
will use available data sets describing gene expression upon drug treatment. It has been shown that
gene expression data of this sort provides insight into a drug’s mode of action and can suggest
novel actions [20]. Methods developed in the CS group and elsewhere to leverage this data will be
used. The best candidate molecules will be applied to the human pluripotent stem cell models to
confirm their capacity to induce brown/beige adipogenesis in vitro. Experiments will be performed
with the support of Stefania Carobbio at WTSI.
In summary, the fellow will (i) build a mechanistic models of human brown fat
development/ activation by macrophages and WAT browning; ii) use it to propose novel
therapeutic interventions (e.g. involving macrophages) to induce and activate brown and brownlike adipocytes, and (iii) test these hypothesis on human stem cell models. This project will provide
the fellow state of the art interdisciplinary training in the analysis and integration of different
‘omics’ data, network modelling and systems pharmacology, as well as in pluripotent stem cells,
adipocyte’s biology and brown/beige fat physiology.
2
ESPOD Proposal Vidal‐Puig/Dougan (WTSI) and Steinbeck (EBI)
References
1.
Bachman, E.S., et al., betaAR signaling required for diet-induced thermogenesis and obesity
resistance. Science, 2002. 297(5582): p. 843-5.
2.
Cannon, B. and J. Nedergaard, Brown adipose tissue: function and physiological significance.
Physiol Rev, 2004. 84(1): p. 277-359.
3.
Feldmann, H.M., et al., UCP1 ablation induces obesity and abolishes diet-induced thermogenesis
in mice exempt from thermal stress by living at thermoneutrality. Cell Metab, 2009. 9(2): p. 203-9.
4.
Kopecky, J., et al., Expression of the mitochondrial uncoupling protein gene from the aP2 gene
promoter prevents genetic obesity. J Clin Invest, 1995. 96(6): p. 2914-23.
5.
Leonardsson, G., et al., Nuclear receptor corepressor RIP140 regulates fat accumulation. Proc
Natl Acad Sci U S A, 2004. 101(22): p. 8437-42.
6.
Lowell, B.B., et al., Development of obesity in transgenic mice after genetic ablation of brown
adipose tissue. Nature, 1993. 366(6457): p. 740-2.
7.
Cypess, A.M., et al., Cold but not sympathomimetics activates human brown adipose tissue in
vivo. Proc Natl Acad Sci U S A, 2012. 109(25): p. 10001-5.
8.
Orava, J., et al., Different metabolic responses of human brown adipose tissue to activation by
cold and insulin. Cell Metab, 2011. 14(2): p. 272-9.
9.
Hale C1, et al, Dougan G1, Mukhopadhyay S2. Induced pluripotent stem cell derived macrophages
as a cellular system to study salmonella and other pathogens.PLoS One. 2015 May 6;10(5):e0124307.
doi: 10.1371/journal.pone.0124307. eCollection 2015.
10.
Alasoo K Transcriptional profiling of macrophages derived from monocytes and iPS cells
identifies a conserved response to LPS and novel alternative transcription
Sci Rep. 2015. PMID: 26224331PMCID: PMC4519778
11.
Goncalves, E., et al., Bridging the layers: towards integration of signal transduction, regulation
and metabolism into mathematical models. Mol Biosyst, 2013. 9(7): p. 1576-83.
12.
Ryll, A., et al. A model integration approach linking signalling and gene-regulatory logic with
kinetic metabolic models. Biosystems. 2014. 124: p 26-38
13.
Machado, D., et al Systematic Evaluation of Methods for Integration of Transcriptomic Data into
Constraint-Based Models of Metabolism. PLoS Comput Biol. 2014 Apr; 10(4): e1003580.
14.
Fonseca NA, et al. RNA-Seq Gene Profiling - A Systematic Empirical Comparison. PLoS ONE
9(9): e107026
15.
Giacomoni, F., et al. Workflow4Metabolomics: a collaborative research infrastructure for
computational metabolomics. Bioinformatics. 2015 May 1;31(9):1493-5
16.
Davidson RL., et al. Galaxy-M: a Galaxy workflow for processing and analyzing direct infusion
and liquid chromatography mass spectrometry-based metabolomics data. Gigascience. 2016 Feb 23;5:10.
17.
Beisken, S., et al. Metabolic differences in ripening of Solanum lycopersicum 'Ailsa Craig' and
three monogenic mutants. Sci Data 2014 16;1:140029
18.
Moreno, P., et al, Gene regulation of metabolic pathways: A continuous model for Escherichia
coli carbon sources uptake. 8th International Conference of Systems Biology, California 2007
19.
Method to increase the production of extracellular polymeric substances (eps) in a
acidithiobacillus ferrooxidans culture by the inhibition of enzymes of tricarboxilic acid cycle US Patent
8,927,246, 2015
20.
Iorio, F., J. Saez-Rodriguez, and D. di Bernardo, Network based elucidation of drug response:
from modulators to targets. BMC Syst Biol, 2013. 7: p. 139.
3