The nervous system has an absolute molecular species requirement for proper function.

A considerable body of biological evidence has accumulated that suggests that docosahexaenoic acid (22:6n3) is an essential component in the nervous system. Moreover, it appears from these studies that long chain polyunsaturates of the n-6 family such as arachidonate (20:4n6) and docosapentaenoate (22:5n6) cannot substitute for 22:6n3. This evidence is briefly reviewed and two hypotheses centering upon either biochemical or biophysical aspects of polyunsaturate function are presented and discussed. It is concluded that a bioactive metabolite of 22:6n3 is not responsible for its function in brain and that the best hypothesis asserts that a membrane function of a 22:6n3-containing species of phospholipid, such as phosphatidylserine, is critical for optimal neural function. Moreover, data are presented indicating that the biophysical properties of various highly unsaturated species of phospholipid are distinguishable. It is further contended that these species are not randomly distributed in membranes and thus the differences in physical properties may be amplified. It is concluded that a conceptual framework is needed in which the distinct membrane roles of phospholipid species may be understood as a function of the positions and numbers of double bonds. Only then may the critical role of the highly unsaturated n-3 polyunsaturates in the brain and retina be understood.