Mental illnesses are some of the most difficult to define and diagnose. There are at least two scientific and medical/clinical disciplines devoted to it, a number of specialised hospitals, a wealth of mainstream prejudice and a long history of investigation and experimentation—and yet it is an area that is still regarded as one of the most mysterious regarding human sickness.

One treatment that relies upon lipids, and in particular membranes, has been around for some time. Lithium carbonate or citrate can be taken, leading to (positively charged) lithium ions entering the bloodstream. This is useful because they cannot pass over the membranes of neurones, increasing the net positive charge outside the cell. This changes the electronic potential across the membrane, and so changes the threshold at which neurotransmitters such as glutamic acid and serotonin, are released. This has been shown to have a direct impact on the mood of patients with serious mental illnesses.

The question I am curious to answer, in the view of that evidence, is why should those membranes need it, where others do not? Is there something different about the membranes of someone with mild (type II) bipolar disorder, someone with serious and incurable paranoid schizophrenia and someone with an anxiety condition?

Several recent papers in the scientific literature suggest that there may be, and so tantalisingly for sufferers, some of the mystery of mental illness may be in the process of being dispelled. The most recent work was focussed on shifts in the lipid profile of cells in different parts of the human brain with respect to serious mental illness.

Faria et al. [1] have shown that stress-induced depression in mice actually changes the lipid profile of membranes across the whole brain. The levels of phosphatidyl inositol (PI) fall by about 50%. This is made up for by slight increases in bulk lipids phosphatidylethanolamine (PE) and phosphatidylcholine (PC). Although there is evidence that a form of PC has a crucial role in brain function (Nguyen et al. [2], Ben-Zvi et al. [3]), it cannot replace PI in its functions—PI has been found to be significant in cell division but it is also the precursor to several important signalling lipids. However, when the pool of starting material is diminished by half, we might expect that the scope for normal signalling in the brain is compromised.

Although this points towards a biological effect of changing the lipid profile, the fact that it has changed at all raises several questions. What are the physical properties of the membrane after the change in lipid profile? Is the adaptation to the stress stimulus effective? The question I am interested in is the more precise effects. What this study shows is that the result of an emotional experience of a mammal has a direct effect on the molecular profile and thus the function of its brain, but it does not show what the shifts are within the tissues of that organ, or within the organelles of those cells.

This work is therefore an excellent basis for further research. However, it would be disingenuous to consider it only in those terms. This work can be used to inform the understanding we have about the mode of action of certain anti-depressants. Aboukatwa and Undieh [4] showed that the fraction size of a lipid called CDP-diacylglycerol, and PI were both increased on treatment with imipramine, paroxetine and maprotiline (also in mice). This is significant because CDP-diacylglycerol is the precursor to PI, and thus the whole production of inositides is increased by these drugs.

This exciting connection is marvellous grist to the mill of those interested in the behaviour of inositides in biological systems (viz. this author, [5, 6]), but it is also a good connection to another, well-defined set of mental conditions, anxiety disorders. Gormanns et al. [7] used a systems biology approach to explore the changes to PI signalling pathways that were observed in certain psychiatric disorders. They found that dysregulation of the PI signalling pathway was the result of changes in the transcriptome in anxiety disorders.

This fits rather well with the observations about stress-induced changes in the size of the PI fraction described above [1] and thus contributes to an exciting set of studies that link inositides with mental illnesses. It begins to provide a context for older observations about changes in inositides in schizophrenia patients [8, 9]. What is less clear from a clinical point of view is the therapeutic target. Should a particular part of the brain be targeted? How can this be achieved? I predict that analyses of individual parts of the brain will give further insight, along with individual tissue types within those. The data that emerges can then be used to inform our understanding of the underlying molecular mechanism, and thus provide the probable explanation for, psychiatric disorders. Prejudice may yet be beaten by phosphatidylinositol.


[1] R. Faria, M. M. Santana, C. A. Aveleira, C. Simones, E. Maciel, T. Melo, D. Santinha, M. M. Oliveira, F. Peixoto, P. Domingues, C. Cavadas, M. R. M. Domingues, Neuroscience, 2014, 273, 1-11.

[2] L. N. Nguyen, D. Ma, G. Shui, P. Wong, A. Cazenave-Gassiot, X. Zhang, M. R. Wenk, E. L. K. Goh, D. L. Silver, Nature, 2014, 509, 503-506.

[3] A. Ben-Zvi, B. Lacoste, E. Kur, B. J. Andreone, Y. Mayshar, H. Yan and C. Gu, Nature, 2014, 509, 507-511.

[4] M. A. Aboukhatwa, A. S. Undieh, BMC Neuroscience, 2010, 11, 10. DOI:

[5] S. Furse, N. J. Brooks, A. M. Seddon, R. Woscholski, R. H. Templer, E. W. Tate, P. R. J. Gaffney, O. Ces. Soft Matter, 2012, 8, 3090-3093. DOI:

[6] S. Furse, L. H. Mak, E. W. Tate, R. H. Templer, O. Ces, R. Woscholski, P. R. J. Gaffney, Organic and Biomolecular Chemistry, 2015, Advance article. DOI:

[7] P. Gormanns, N. S. Mueller, C. Ditzen, S. Wolf, F. Holsboer, C. W. Turck, Journal of Psychiatric Research, 2011, 45, 973-979. DOI:

[8] I. Das, M. A. Essali, J. de Belleroche, S. R. Hirsch, Schizophrenia Research, 1994, 12, 265-268.

[9] O. Fedorenko N. Strutz-Seebohm, U. Henrion, O. N. Ureche F. Lang, G. Seebohm, U. E. Lang, Psychopharmacology, 2008, 199, 47–54. DOI: