The emerging phospholipid hypothesis takes as its basis increasing evidence that phospholipid metabolism is abnormal in schizophrenia. Broadly, it suggests that as a result of this altered metabolism, the cell membranes of schizophrenia sufferers are depleted of certain polyunsaturated fatty acids (PUFAs), including arachidonic acid (AA), docosahexaenoic acid (DHA), and eicosapentaenoic acid (EPA) (Horrobin et al., 1994; Peet et al., 1999). Currently, there are thought to be several possible mechanisms responsible for this. All of them offer an explanation for schizophrenia that brings together the genetic and the environmental, the neurotransmitter, and the lipid all bound up in a package that suggests that diet may have a crucial role to play in ameliorating or exacerbating the symptoms of this much misunderstood illness.
Evidence for the phospholipid hypothesis is arriving from various sources and growing rapidly. Studies at several centers have shown depleted PUFAs in red blood cell (RBC) membrane phospholipids (Peet et al., 1995; Glen et al., 1994; Yao et al., 1994), although not all reports are consistent (Doris, et al., 1998). In drug-treated schizophrenic patients, RBC membranes have been shown to have an apparent bimodal distribution of AA and DHA, where the lower mode is related to negative schizophrenic symptoms (Glen et al., 1994). Reduced PUFA levels have also been reported in cultured skin fibroblasts (Mahadik, et al., 1996). RBC PUFA levels have also been related to tardive dyskinesia, where the depletion of AA and DHA in particular were related to the severity of the abnormal movements (Vaddadi, 1989).
There is, however, a need for direct measurement within the brain. This is notoriously difficult to carry out, but in 1991, Pettegrew et al. used 31P magnetic resonance spectros-copy (MRS) to make a direct examination of brain phospholipid metabolism. Using a group of first-episode drug-free schizophrenic patients, they found significantly raised levels of phosphodiesters (breakdown products of phospholipid) and significantly decreased levels of phosphomonoester (involved in phospholipid synthesis) when compared with controls (Pettegrew et al., 1991). For some, this study, along with others, including those of Stanley et al. (1994, 1995) showing similar results, remains a landmark investigation showing abnormal brain phospholipid metabolism in schizophrenic patients. Most recently, Yao et al. (2000) have shown a significant reduction in PUFAs, particularly AA, within RBC membrane phospholipids in the caudate region in the postmortem brain of a group of schizophrenics compared to well-matched, healthy controls.
A method for indirectly assessing prostaglandin production has come from the application of what has become known as the niacin skin flush (Ward et al., 1998). Previously, it has been shown that normal, healthy controls show a generalized skin flush response when given a high dose of oral niacin. This is caused by a release of prostaglandin D2 in skin (Morrow et al., 1992). Prostaglandin D2 is the major cyclo-oxygenase metabolite of AA in skin (Ruzicka & Printz 1982). Any reduction in availability of AA would therefore be expected to affect the subsequent availability of prostaglandin D2, hence a reduction in skin flush. For many years, it has been observed that people with schizophrenia do not flush in response to oral niacin (Horrobin, 1980). This effect has recently been substantially confirmed (Hudson et al., 1997; Rybakowski & Weterle, 1991; Glen et al., 1996) but confounded by a possible medication effect. Recently, however, studies have been conducted in drug-free schizophrenia sufferers. Using a skin-patch test that utilizes topical aqueous methyl nicotinate developed by the Highland Psychiatric Research Group (Ward et al., 1998), Shah et al. (2000) looked at the flush reaction of a number of medicated and unmedicated schizophrenic patients plus a group of well-matched controls. The results confirm that compared to healthy controls, schizophrenics show a highly significant difference in flush response to topical niacin, with the unmedicated patients showing a similar response to the medicated. This is the most telling evidence accrued thus far that the reduced flush response in schizophrenia sufferers is not the result of any medication effect. It also suggests that metabolic abnormalities in schizophrenia affect the whole body, not just the brain.
Another indirect measure of physiological function relating to PUFA metabolism comes from studies of the electroretinogram (ERG). It is known that adequate n-3 PUFA levels are essential for normal retinal function. Our group has shown that schizophrenic patients show ERG changes similar to those that occur in states of experimental dietary deprivation of n-3 in primates (Warner et al., 1999).
One explanation for this evidence of reduced membrane PUFA may come from studies showing that within the plasma, platelets, and brain of schizophrenic patients, there are elevated levels of phospholipase A2 (PLA2) (Gattaz et al., 1990, 1995; Ross et al., 1997, 1999). Existing in several forms, PLA2is an enzyme that is involved in the turnover of PUFA in cell membranes. Thus, any overactivity or underactivity of PLA2 will have a significant effect on phospholipid metabolism. Until recently, studies of PLA2 activity in schizophrenia have delivered mixed results, suggesting either raised (Gattaz et al., 1987) or normal (Albers et al., 1993) PLA2 activity. However, Ross et al. (1997) have looked at the possibility that different investigators had studied different subtypes of PLA2. Subsequently, they have followed up their report showing increased PLA2 activity in blood, by studying two classes of PLA2, calcium stimulated and calcium-independent, in the brain of schizophrenics (Ross et al., 1999). These data show a 45% increase in calcium-independent PLA2in the temporal cortex within the brain of schizophrenics when compared to well-matched controls and a group with bipolar disorder. Furthermore, animal studies by the same group show that this is unlikely to be a medication effect.
As evidence from a host of sources has now firmly posited schizophrenia as being the result of a disease process, the search for genetic markers has grown apace. The evidence associating PLA2 with schizophrenia has indicated that one area to study may be among those genes that regulate PLA2 activity. Indeed, recent findings have demonstrated significant differences in allele distribution between schizophrenic patients and controls on polymorphic sites in the region of the gene for PLA2 (Hudson et al., 1996; Peet et al., 1998; Wei et al., 1998), although there are also negative reports (Strauss et al., 1999; Doris et al., 1998). This raises the possibility that the abnormality of cytosolic PLA2 is genetically determined.
Thus far, we have reviewed evidence demonstrating that a depletion of PUFAs in RBC membranes and an over activity of PLA2 may be responsible for stripping away those PUFA. Other evidence has suggested that sufferers of schizophrenia also show increased levels of oxidative stress. Mahadik et al. (1998) reported elevated plasma lipid peroxides in first-episode nonaffective psychosis. Recently, the Scottish Schizophrenia Research Group (2000) has suggested that such findings could be an artifact caused by smoking, which, in itself, causes oxidative stress. However, work by our group (Zhang, 1999) has demonstrated increased oxidative stress in unmedicated schizophrenic patients who do not smoke. Oxidative stress could be either a cause or a consequence of increased breakdown of membrane PUFA. Yao et al. (1998) showed a reduction in plasma total antioxidant capacity in schizophrenic patients, which was not attributable to antipsychotic drug treatment. Mukerjee et al. (1996) also showed impaired antioxidant defense enzymes at the onset of psychosis. This suggests that there may be a primary abnormality of antioxidant defenses in schizophrenia, possibly of genetic origin (Edgar et al., 2000).
Was this article helpful?