return to home page

BRAIN UPTAKE AND UTILIZATION OF FATTY ACIDS: 

Applications to Peroxisomal Biogenesis Diseases

(An International Workshop)

March 2-4, 2000, Holiday Inn Bethesda, Bethesda, Maryland

Session 3 (Morning, March 4, 2000)

Return to Table of Contents

Return to Beginning of Session

The Regulation and Function of DHA in Neurons and Neuronal Membranes

(Saturday, March 4, 2000 Morning Session Summary)

Co-Chairs: Stanley I. Rapoport and Norman Salem, Jr.

Discussion Moderator: Robert E. Anderson

P repared by: Robert Katz and Robert E. Anderson

PUFA in brain phospholipids and plasmalogens.

After uptake into the brain through the cerebrovascular capillary endothelium, omega-6 and omega-3 polyunsaturated fatty acids (PUFA) are incorporated into phospholipids (PLs). PLs are major components of neuronal and glial membranes, where they participate in structural remodeling and synthesis of membranes as well as signal transduction processes. Receptor-initiated activation by phospholipase A2 (PLA2) releases arachidonic acid (AA) and DHA from PLs.

Dr. Stanley I. Rapoport reviewed an in vivo method and model developed in his laboratory that allows localization and quantification of fatty acid (FA) kinetics including long-chain PUFA (AA and DHA) incorporation rates from plasma turnover and half-lives of FAs in brain phospholipids.

The analysis of plasma and brain radioactivity in rat or monkey brains following injected or intravenously infused, radio-labeled, albumin-bound AA and DHA, indicated that about 15-40% of these two PUFA are metabolized daily and are replaced from plasma sources.

The rat model assumes three compartments: a) a plasma pool of albumin-bound FAs at steady state (containing more than 99% of FAs with carbon chains < 22) which yield 5% of free, dissociated FAs during a single passage through the brain lasting 1 or 2 seconds and which can cross the blood-brain barrier (BBB) very rapidly; b) the acyl-CoA pool, where the free FAs that crossed the BBB or the free FAs from PLs by the way of phospholipase A1 (PLA1) or PLA2 are activated to FA-CoA’s by Ac-CoA synthetases (the transient compartment); and c) the steady state phospholipids (the stable compartment). By developing a series of equations and performing pulse chase experiments following administration of [1-1 4 C]- AA and [1-14C]-DHA (in an awake rat) a ratio of FA-CoA-specific activity to plasma FA-specific activity can be calculated. The net turnover rate (% per unit time) of a FA in brain phospholipids can also be estimated and half-lives for phospholipids in the brain can be obtained.

Dr. Rapoport showed that within the brain of awake rats recycling of FAs in phospholipids is very rapid with half-lives of minutes to hours. The small contribution of plasma FAs to this recycling is on the order of 2-4%. This small contribution explains why many weeks may be necessary to recover normal brain PUFA concentration following prolonged dietary deprivation.

Changes in recycling of specific fatty acids in response to centrally acting drugs can help to identify enzyme targets for drug action. For example, recycling of   AA is specifically reduced (by 80%) in rats treated chronically with lithium, a drug effective in bipolar disorder.   This effect reflects down-regulation of gene expression of an AA-specific PLA2. 

Dr. Lloyd A. Horrocks discussed the structure, biochemistry and physiologic functions of plasmalogens (Pls) in the brain. Ethanolamine plasmalogens (EAPs) and choline plasmalogens (CPs) are PLs of neural membranes (concentrated in the white matter of brain tissue), which contain an enol-ether bond in the sn-1 position and often an AA or DHA moiety at the sn-2 position. The CPs are enzymatic methylation products of EAPs and a major portion of the DHA in Pls appears incorporated in EAP.

The first two enzymes required for EAP synthesis reside in peroxisomes and the level of Pls in the brain appears to depend on the degree of myelination. A DHA-Pls deficiency has been demonstrated in Zellweger’s syndrome, an inherited disorder of peroxisomal assembly and function. Dietary DHA supplementation in Zellweger patients results in partial remyelination (see last Session). These observations raise the possibility that DHA-Pls are involved in myelination processes in the brain. A deficiency of Pls may also be responsible for abnormal signal transduction associated with learning, cognitive and visual dysfunctions some of which could be at least partially correctable by dietary supplementation of DHA. Receptor-mediated hydrolysis by Pls-selective PLA 2 (different from the extracellular PLA 2 that hydrolyses the sn-2 position of PLs) provides free AA and DHA and their corresponding metabolites. Pls appear to also contribute to membrane disorder. Since under pathological conditions such as ischemia, spinal cord trauma and Alzheimer’s disease PLs-PLA 2 is stimulated, a decrease in Pls content can destabilize neural membranes by altering membrane order and permeability and lead to neurodegeneration. Due to the sn-1 alpha, beta-unsaturated ether linkage, Pls are susceptible to oxidation and appear to inhibit lipid oxidation as scavengers of peroxy-radicals (two hydroxyl radicals per enol-ether double bond). Thus, plasmalogens might be involved in alleviating oxidative stress.

DHA functions in neuronal membranes

Dr. Ephraim Yavin discussed the effects of oxidative stress during in utero accumulation of brain DHA in rats. Given the fact that DHA is a major PUFA in the adult rat brain it is important to understand the mechanism of its enhanced accumulation (together with AA) during the last trimester of gestation, the so-called “DHA accretion spurt.” The elevated level of DHA is located mainly in phosphatidylethanolamine (PE)- and phosphatidylserine (PS) phospholipids around synapses with exceptionally high levels in the retina. Susceptibility to oxidative stress during this period can be considerable due to immaturity of the developing brain with respect to antioxidant activity, cerebral blood flow auto regulation, and simultaneous subjection to active, on-going, programmed cell death. The DHA is supplied to the fetus from maternal stores. If maternal dietary n-3 PUFA deficiency occurs, DHA accretion by the fetus is reduced, however normal levels can be restored through direct, intra-amniotic injections of millimolar concentrations of ethyl-DHA (Et-DHA) and the danger of stress-caused intrauterine growth retardation can be avoided. Fetuses treated with Et-DHA also formed less oxidation products after in utero ischemic stress (as detected by thiobarbituric acid) than controls. Furthermore, Et-DHA treatment prevented the formation of adducts of a spin trap, DMPO, with hydroxyl radicals generated by Fe2+/ H 2 O 2- induced oxidation, as measured by electron spin resonance technique. Fetuses treated with ethyl oleate did not prevent DMPO-OH aduct formation. Thus, it appears that DHA accretion in the fetal brain increases oxidative risk, while simultaneously increasing free radical trapping capacity.

These paradoxical, antagonistic effects seem to be occurring due to metabolic alterations in the neuronal lipid bilayer. Yavin considers PL species highly enriched in DHA and DHA-Pls moieties, both capable of serving as targets for hydroxyl radicals, as a possible explanation for the dual, antagonistic pro-oxidant and antioxidant roles of DHA.

Dr. Burton J. Litman addressed the issue of whether DHA-containing phospholipids impart unique structural and functional properties to biological membranes.   In particular, the mechanism of how the variation of membrane lipid and cholesterol composition modulates signal transduction in G protein-coupled receptor systems was addressed.   In these studies the visual transduction pathway was used as a model G protein-coupled receptor system.   The site of visual transduction is the retinal rod outer segment disk membrane, whose phospholipids are very high in DHA acyl-chains, as is the case for other neuronal tissue.   In order to study the role of the variation of membrane lipid composition of G protein-coupled signaling, the components of the visual pathway were isolated and reconstituted into bilayer vesicles of defined lipid composition. 

The results obtained in these studies show that DHA-containing lipid bilayers: 1. allow the highest levels of formation of MII, the G protein activating form of rhodopsin; 2.   are highly dynamic and support the fastest rates of MII-G complex formation, a diffusion dependent process; 3. yield the highest levels of MII-G complex formation; 4.   buffer the inhibitory effects of   the presence of cholesterol relative to MII formation; 5.   are involved in the formation of lateral clusters or microdomains rich in rhodopsin.   These studies demonstrate that DHA-containing bilayers optimize the signaling sensitivity of the visual pathway.   The findings reported here highlight a critical role of DHA-containing phospholipids in optimizing membrane signaling processes through the modulation of receptor activation, the creation of receptor-containing lateral microdomains, and the formation of highly dynamic membranes.   If signaling in other G protein-coupled receptors systems, such as neurotransmitter receptors, is similarly effected by DHA-containing phospholipids, then a deficiency in DHA would be expected to result in visual and cognitive deficiencies, which is in good agreement with experimental observations.  

Dr. Robert E. Anderson, discussant for this session, raised the issue of a possible dual function for DHA in retinal membranes. In vivo and in vitro studies showed the importance of DHA in the visual process and a large number of studies showed the vital importance of DHA in optimal development of the nervous system.  However, Dr. Anderson raised the possibility that normal high levels of DHA in the retina may be reduced under conditions of oxidative stress.  A specific example cited was the rearing of albino rats in bright cyclic light. Under these conditions, the normal high levels of DHA in ROS membranes (45-50%) were reduced to low levels (15-20%), at the same time the glutathione antioxidant enzymes, vitamin C, and vitamin E, were elevated in the retina. This result is consistent with the bright cyclic rearing producing an oxidative stress that resulted in up-regulation of protection mechanisms and down- regulation of a substrate for lipid peroxidation. 

Dr. Hee-Yong Kim reported on the potential involvement of DHA in apoptotic neuronal cell death. Direct exposure of Neuro 2A or PC-12 cells to DHA at 1-25 µM concentrations or preincubation with this fatty acid for 5 h did not prevent DNA fragmentation caused by serum starvation.  However, when these cells were enriched with DHA for 24 hours prior to the serum deprivation, DNA fragmentation decreased considerably, and this protective effect was further potentiated after 48 h of enrichment. Under the same condition, oleic acid did not exert any effect.  DNA fragmentation as well as caspase-3 activity consistently indicated the protective effect of DHA on apoptotic cell death after a prolonged enrichment period. Examination of the lipid composition altered by DHA enrichment was examined in relation to the membrane translocation of Raf-1 kinase which has been implicated in anti-apoptotic signaling revealed an increase in the PS content and Raf-1 translocation, a down-regulation of caspase-3 activity and prevention of apoptotic cell death.  The correlation observed between the degree of protective effect and the extent of PS accumulation as well as the Raf-1 translocation enhanced by DHA enrichment strongly suggests that the protective effect of DHA may be mediated at least in part through the promoted accumulation of PS in neuronal membranes.  

PUFA and regulation of gene expression.

Dr. James M. Ntambi has focused on identification of mechanisms by which PUFA regulate lipogenic gene expression. Previous research indicates that PUFA regulate both the gene expression and proteolytic maturation of a group of transcription factors termed sterol regulatory element binding proteins (SREBPs) and this effect in turn accounts for hepatic lipogenic gene expression. PUFA reduce lipogenic gene expression in the liver and adipose tissue of lean and obese Zucker rats. They also reduce the expression of stearoyl-CoA desaturase, S14 and glucose transporter-4 genes in cultured 3T3-L1 adipocytes, in cells of the immune system, small intestine, pancreas and brains of neonatal mice. The PUFA-mediated induction of genes of beta-oxidation is regulated by a common transcription factor named peroxisome proliferator activated receptor alpha (PPAR-alpha). However, work on the PPAR-alpha null mouse indicated that while it mediated PUFA-induction of fatty acid beta-oxidation gene transcription, it did not appear to mediate PUFA-mediated repression of lipogenic genes. Therefore, the search for PUFA-specific transcription factor is now an important focus in PUFA research. Dr. Ntambi’s recent studies on the transcriptional regulation of the stearyl-CoA desaturase (SCD1 and SCD2) genes indicate that PUFA can suppress gene transcription by an as yet unknown mechanism independent of SREBP maturation.

Other speakers already mentioned the need for de novo synthesis of saturated and unsaturated fatty acids by brain cells. This high level of lipid biosynthesis by the brain is correlated with high expression of SREBP mRNA, nevertheless, despite high levels of SREBP feeding, long chain omega-3 PUFA (DHA) and safflower oil appear to have no effect on expression of SCD2 in adult mouse brain. In nursing pup mice SCD2 gene expression in the brain was reduced by maternal feeding of an essential fatty acid deficient diet compared with controls. Due to scarcity of information on lipogenesis in the brain there is a need for further studies in this area.

Return to Table of Contents

  Return to Beginning of Session

The Regulation and Function of DHA in Neurons and Neuronal Membranes

(Saturday, March 4, 2000 Morning Session Recommendations for Future Research)

Uptake rates of nutritionally supplied fatty acids by the brain

The method, model and “operational equations” presented at the workshop, for quantifying in vivo brain turnover rates and half-lives of fatty acids during uptake from plasma to the brain, can be applied to the elucidation of uptake rates of nutritionally -provided omega-6 PUFA (e.g. linoleic acid, LA; gamma-linolenic acid, GLA, arachidonic acid, AA), omega-3 PUFA (e.g. alpha-linolenic acid, ALA; eicosapentaenoic acid, EPA and docosahexaenoic acid, DHA) and of  dietary saturated and monounsaturated fatty acids.   In the case of saturated and monounsaturated fatty acids, studies could help clarify the possible differences between their uptake versus de novo synthesis in the developing and in the adult brains.

Effects of centrally acting drugs on the steady state of fatty acids

The above approach can also be applied to the assessment of changes in steady states in response to centrally acting drugs and pathological changes in bipolar and other disorders. When combined with neuroimaging intravenously injected radiolabeled PUFA can also be utilized to examine neuroplastic remodeling of brain fatty acids and lipid membranes.

Role of plasmalogens in glial and neuronal tissue

-It was noted that AA- and DHA-containing plasmalogens (the glycerophospholipids containing an enol-ether group at the sn-1 position) play important roles as potential protectors of neurons from oxidative stress and suppliers of free AA and DHA to the cell. They can function also as modulators of neuronal membrane physical state and permeability both in the developing and in the mature brain. Plasmalogens are formed in peroxisomes, therefore they could be involved in a wide range of pathological states such as peroxisomal biogenesis disorders (Zellweger's syndrome and neonatal leukodystrophy) as well as ischemia or neurodegerative diseases, such as Alzheimer’s disease, and even spinal cord trauma. All of these issues warrant rapid further elucidation.

-It would also be important to understand the contribution of plasmalogens to the free fatty acid pool and therefore their contributions to prostaglandin-mediated processes in the developing and mature neuron in health and disease.

The role of DHA in the developing and adult brain

-Results reported in Session 2 and 3 settle neither the question whether saturated and monounsaturated fatty acids are taken up by the brain or whether they are synthesized de novo by the brain, nor the question whether developing brain is similar or identical with adult brain in this regard. Studies on the gene regulatory effects of EPA and DHA in the liver and other organs on lipogenic enzymes do not extend to the brain in a satisfactory fashion. A strong need exists in elucidating the molecular biology and genetics of brain lipogenesis and its regulation by long-chain, omega-3 fatty acids (EPA and DHA) in both the developing and the adult brain. 

-The accelerated in utero uptake of DHA (by the rat embryo) appears characteristic both to animals and humans. Since the mother is the major supplier of DHA to the developing rat brain, a low circulating level of plasma DHA in the mother could lower the embryo’s DHA brain levels. Direct intra-amniotic injection of DHA ethyl ester restores embryonic brain uptake of DHA to normal levels. Additional research into the mechanisms of DHA accretion and depletion in embryonic and developing rat brain and into the potential behavioral or cognitive consequences of below normal DHA levels in the embryonic and postnatal brain should shed light on why DHA and possibly AA supplementation of infant formulas would be important for normal brain development. Elucidating the consequences of DHA deficiency during the developmental period is especially important due to the reported potential role of DHA to protect the developing brain from oxidative stress.

-The extension of these in vivo studies to humans would render the development of appropriate non-invasive techniques for quantitation of neuronal changes due to DHA presence or absence including, but not limited to the development of new highly desirable imaging and tracer approaches.

-The question of the role of DHA in the management of environmental stress and oxidative stress in the adult brain is also worthy of exploration. 

-A major breakthrough in understanding the role of DHA in signal transduction was recently reported in the outer rod segments (ROS) of the retina.  It appears that the role of DHA in ROS membranes is attributable to its six double bonds that provide and ensure a “most favored” microenvironment in which rhodopsin can absorb a photon of light and undergo the rapid conformational changes that result in the activation of a G-protein. This activation initiates the chain of biochemical events in the visual transduction pathway that ultimately leads to hyperpolarization of the plasma membrane. This observation, based on solid methodological determinations indicates that the high concentrations of DHA in the outer rod segments might be of paramount importance to efficient signal transduction.  DHA might have similar functions in the transduction of the neuronal signals through the membrane and synapses. Validating this potential role of DHA in the different neurons of different areas of the brain and should contribute significantly to our understanding of the functioning of the central nervous system.

-It is also important to understand what implications does DHA deficiency, during brain development, have on neuronal signal transduction and transmission.

Intermediary metabolism of PUFA and brain disorders

Answers should be sought to questions involving post-uptake shuttling and metabolism of AA, EPA and DHA.   Such questions are exemplified in the following: How does DHA, taken up by the brain through the BBB or synthesized from its precursors by astrocytic peroxisomes, reach its neuronal locations? Are there special DHA transporters in the brain? Are astrocytes mostly devoid of DHA and function primarily as elongation/ desaturation tools of EPA and its higher homolog docosapentaenoic acid (DPA)? What are the functions of EPA and DPA in astrocytes? Are EPA and DHA naturally segregated between astrocytes (EPA) and neurons (DHA)? What are the eicosanoids produced from EPA and what are their functions in the brain? What are docosanoids and what are their functions in the brain? Are these functions related to the pathophysiology of neuropsychobehavioral disorders such as: bipolar disorder, unipolar depression and schizophrenia? Do EPA, DPA and DHA exert the same therapeutic effect in these disorders? Etc.

DHA, EPA and apoptosis of glial and neuronal cells

-The finding that DHA can potentially inhibit neuronal apoptosis in two cell lines and that this inhibitory effect appears related to an increase in PS concentration raises several questions on yet another potential physiological role of DHA namely that of long-term survival of the neuronal cell. Does depletion of DHA result in neuronal death?  If that effect of DHA is proven, is this effect reversible?  Is there a connection between DHA depletion and the pathophysiology of inherited neurodegenerative diseases or Alzheimer’s disease? Is the anti-apoptotic effect of DHA responsible for its apparent segregation to the neuron?  What would be the consequences of long-term DHA storage in astrocytes, which appear to eliminate free DHA as it is formed from its precursors.  How would the anti-apoptotic effect persist in animal models?  All of these questions deserve to be answered.

Return to Table of Contents

Return to Beginning of Session

return to home page