Data driven modelling of cellular phospholipid homeostasis

Altered cell membrane phospholipid compositions are associated with obesity, heart disease, liver disease, cancer, Alzheimer’s disease and many other twenty first century public health concerns. Whilst many of the mechanisms through which cells synthesise different phospholipids are understood, there is a significant gap in scientific knowledge as to what drives cells to alter their phospholipid composition. This phenomenon is known as phospholipid homeostasis.

There are several current theories of phospholipid homeostasis, which each state that cells control their phospholipid composition to ensure that critical cell membrane properties are regulated, such that the function of biomolecules (proteins) are stable under many different environmental conditions.

In this project we use lipidomic data, which provides detailed quantitative information on the phospholipid composition of cells under different environmental conditions and in different disease states, to develop a molecular understanding of phospholipid homeostasis that will impact our understanding of the role of phospholipids in health.

DDM-lipid-homeostasis

Figure 1 the data driven modelling process for lipidomic data

Project timeframe

This research project began in 2011 and is ongoing.

Project aims

The project aims to:

develop new computational tools to be used in conjunction with lipidomic data sets accelerating the rate of data analysis in ‘big data’ sets.
use data driven modelling to evaluate two prevalent rival theories of phospholipid homeostasis (‘homeoviscous adaptation’ and the ‘intrinsic curvature hypothesis’) and to assess which theory can best explain the phospholipid compositional changes observed in the extensive sets of lipidomic data we have obtained.
build new fundamental molecular scale models of how individual phospholipid molecules contribute to maintaining phospholipid homeostasis in cells.
To identify the connection between cellular disease states and phospholipid homeostasis, with a view to develop system-wide biomarkers.

Project finding and impact

To date we have:

demonstrated that the data driven modelling approach can be used to find complex correlations in and rapidly analyse large lipidomic datasets.
found evidence for the intrinsic curvature hypothesis of phospholipid homeostasis and developed a proxy measure of this membrane property.
found evidence for the homeoviscous adaptation theory of phospholipid homeostasis and developed a proxy measure of this membrane property.

Our results suggest that the two theories of phospholipid homeostasis might be correlated in cells; we are currently exploring this correlation.

Impact

The development of a systematic molecular model of phospholipid homeostasis is a fundamental advancement that has the potential to impact our understanding of the role of lipids in many public healthcare concerns.

The methodology we have developed can be extended to develop lipidomic proxies as biomarkers of disease or adapted to find lipidomic proxies of environmental markers.

Research team

Dr Marcus Dymond

Output

Dymond MK. 2016 Mammalian phospholipid homeostasis: evidence that membrane curvature elastic stress drives homeoviscous adaptation in vivo. J. R. Soc. Interface 13: 20160228. http://dx.doi.org/10.1098/rsif.2016.0228

Dymond, M. K., Hague, C. V., Postle, A. D., & Attard, G. S. (2013). An in vivo ratio control mechanism for phospholipid homeostasis: evidence from lipidomic studies. Journal of The Royal Society Interface, 10(80), 20120854.

Hague, C. V., Postle, A. D., Attard, G. S., & Dymond, M. K. (2013). Cell cycle dependent changes in membrane stored curvature elastic energy: evidence from lipidomic studies. Faraday discussions, 161, 481-497.

Dymond, M. K. (2015). Mammalian phospholipid homeostasis: Homeoviscous adaptation deconstructed by lipidomic data driven modelling. Chemistry and Physics of Lipids, 191, 136-146.

Partners

George Attard (University of Southampton).