Developmental role of fatty acid-binding proteins in mouse brain.

While the functions of the cytoplasmic fatty acid-binding proteins (FABPs) are not well defined, one possibility in neural tissue is in establishing and maintaining the high levels of polyunsaturated fatty acids in membrane lipids characteristic of this tissue and thought essential for normal function. We investigated the reactivity of a protein in developing mouse brain to antiserum prepared against rat heart (H)-FABP. By immunoblot analysis, levels of H-FABP in brain were nearly undetectable until fetal day 17-19, after which levels increased until at least postnatal day 14. Levels of H-FABP were lower in the adult mouse brain, suggesting a function for the protein during differentiation of neural tissue. In immunohistochemical studies with postnatal day 14 mouse brain, the most intensely stained area was the choroid plexus. H-FABP also localized to regions of the somatosensory cortex and to the spinal trigeminal nucleus. In addition, H-FABP was present in the thalamus, entorhinal and piriform cortex, and throughout the pontine and medullary nuclei. Tracts related to the auditory system, including ventral cochlear nucleus and lateral lemniscus, also were H-FABP-positive. In cerebellum, the molecular layer was heavily labeled in cells and processes; in the granule cell layer, there was punctate staining suggestive of mossy fiber terminals. Small cells adjacent to Purkinje cells were intensely stained, while the Purkinje cells were negative. We conclude that H-FABP in brain participates in neurite formation and synapse maturation, and may be related to the similar pattern of expression of GABA related markers.

Retinal FABP principally localizes to neurons and not to glial cells.

The presence of fatty acid-binding protein (FABP) in the embryonic chick retina may be linked to the demand for polyunsaturated fatty acids in this developing neural tissue. There is a decline in the overall level of FABP as the retina matures, suggesting a role for FABP in cellular differentiation. However, this pattern is not present in the chick brain, indicating a unique function for FABP in the retina. Immunohistochemical staining of paraffin sections of chick retina from embryonic day 21 revealed immunopositive photoreceptor inner segments, outer nuclear layer, 'radial processes' in the inner nuclear layer, a subpopulation of cells in the ganglion cell layer, and inner limiting membrane. This pattern suggested that FABP positive cells were photoreceptors, Müller (glial) cells, and possibly ganglion cells. Staining of sections for glutamine synthetase, an enzyme specific for Müller cells, was similar but not identical to the pattern observed with FABP; thus identification of these cells as FABP-positive was not conclusive. However, in retinal cells dissociated from day E14 embryos and cultured for one week, staining with FABP was more intense in the neurons than in the 'flat' cells (presumed to be derived from the Müller cells). Retinal FABP thus appears to be localized predominantly in neurons, and may serve to sequester fatty acids in preparation for neurite outgrowth as the retinal cells differentiate.

Acylation of 1-palmitoyl-sn-glycerophosphocholine by chick brain microsomes is unaffected by fatty acid binding protein.

Rates of incorporation of exogenously supplied fatty acids into 1-palmitoyl-sn-glycerophosphocholine were measured using the microsomal fraction from brains of 14-15 day old chick embryos. The substrate preferences for reacylation were: 18: 2(n-6) = 20: 4(n-6) > or = 20: 5(n-3) = 18: 3(n-3) > or = 18: 1(n-9) > or = 22: 6(n-3) > or = 18: 0. The normalized rate with 18: 0 was significantly lower than all other rates except for 22: 6(n-3), and the acylation rate with 22: 6(n-3) was significantly lower than with 18: 2(n-6) and 20: 5(n-3). With the addition of fatty acid binding protein partially purified from brain cytosol, a decrease (not significant) in the rate of incorporation was observed; the substrate preference was unchanged. In the presence of FABP, normalized rates with 18: 2(n-6) were significantly higher than with 18: 0, 18: 1(n-9), or 22: 6(n-3); rates with 20: 4(n-6) were significantly higher than those with 22: 6(n-3). Preliminary data on the acylation of 1-palmitoyl-sn-glycerophosphoethanolamine showed lower rates of incorporation than for the choline analogue and no clear substrate preference, but a similar lack of effect of fatty acid binding protein. These results do not support the proposed function of fatty acid binding protein in the establishment of a phospholipid composition rich in polyunsaturated fatty acids. The results are consistent, however, with the role of the reacylation reaction in the continual turnover of particular substrates [18: 2(n-6) and 20: 4(n-6)] used to generate second messengers.

Fatty acid composition of phospholipids from chick neural retina during development.

The fatty acid composition of retina phospholipids from developing chicks was investigated to determine what changes, if any, occur in the relative levels of polyunsaturated fatty acids. Embryonic chicks were killed at 3-day intervals from day 6 through hatching (day 21), and at 1 week post-hatch. Fatty acids were prepared from retina phospholipids and were analysed by capillary gas-liquid chromatography. A comparison of the composition of yolk taken on day 6 with retinas isolated on that day revealed a much greater proportion of polyunsaturated fatty acids in the latter, suggesting an ability of the embryo to metabolize selectively unsaturated fatty acids at this early stage of development. Throughout the time course studied, saturated fatty acids constituted 50% of all fatty acids, most of which was due to palmitic acid (16:0; 33-41%). Among other saturated fatty acids, myristic acid (14:0) increased to maximal levels by day 18, then declined, while stearic acid (18:0) was minimal on day 12 and then increased. Polyunsaturated fatty acids varied between 14 and 23% of total fatty acids, depending on the developmental stage. One of the most remarkable changes in polyunsaturated fatty acids occurred in the levels of 22:4 (n-6). The proportion of this single fatty acid decreased from 9.4 to 2.4% between days 15 and 18. Relative levels of 22:5 (n-6) increased significantly between day 21 and 1 week post-hatch, from 1.1 to 3.2%. In this same time period, the proportion of 22:6 (n-3), the fatty acid known to be prominent in the outer segments of rod-dominant retinas, did not change.(ABSTRACT TRUNCATED AT 250 WORDS)

Phospholipid synthesis by chick retinal microsomes: fatty acid preference and effect of fatty acid binding protein.

The acylation of 1-palmitoyl-sn-glycerophosphocholine (1-16:0-GPC) or 1-palmitoyl-sn-glycerophosphoethanolamine (1-16:0-GPE) was measured using the microsomal fraction prepared from retinas of 14-15-day-old chick embryos. Rates of incorporation of exogenously supplied fatty acids into diacyl-GPC were generally 5-7 times greater than into diacyl-GPE. Substrate preferences for incorporation into diacyl-GPC and diacyl-GPE were, respectively, 18:2 greater than 18:3 = 20:5 greater than 20:4 greater than 18:1 greater than 22:6 = 18:0 and 18:2 greater than 22:6 greater than or equal to 18:3 = 18:0 greater than or equal to 20:4 = 18:1 greater than 20:5. The apparent selectivities were not consistent with the reported fatty acid compositions of these lipid classes. The addition of partially purified fatty acid binding protein (FABP) to the reaction had no effect either on overall rates of incorporation or on the substrate preference. When fatty acyl-CoA substrates were used, rates of incorporation of the 18:0 derivative were much higher than with the fatty acid, while rates with other fatty acyl-CoA were similar to those with the respective fatty acid. Substrate preferences for CoA derivatives incorporated into diacyl-GPC were: 18:0 greater than 20:4 greater than 18:2 greater than or equal to 22:6, and into diacyl-GPE: 20:4 = 22:6 greater than 18:0 greater than 18:2. Polyunsaturated fatty acyl CoA (PUFA-CoA) were thus favored for incorporation into diacyl-GPE, and to a lesser extent into diacyl-GPC, a result that is consistent with composition data.(ABSTRACT TRUNCATED AT 250 WORDS)

The blood-retinal barrier: leucine transport by the retinal pigment epithelium.

The transport of leucine across the isolated bullfrog pigment epithelium-choroid was studied in a modified Ussing chamber. With the same concentration of leucine on both sides of the tissue, there was a net flux in the apical-to-basal (retina-to-choroid) direction. The concentration dependence of the apical-to-basal flux comprised at least 2 components: a carrier-mediated flux, with an apparent Km of 0.76 mM and a Vmax of 298 nmol/cm2/hr, and a linear component with a slope of 33.4 cm/hr. The basal-to-apical flux was linear to 10 mM leucine, with a slope of 60 cm/hr. In the absence of Na+, the apical-to-basal flux was inhibited 20%, while the basal-to-apical flux was unaffected. The apical-to-basal flux could be inhibited to the level of the basal-to-apical flux by removing glucose from the solution and adding KCN. Otherwise both fluxes were relatively unaltered by metabolic poisons. There appear to be 3 contributors to the apical-to-basal flux: a weakly Na+-dependent carrier, a linear component (diffusion), and a countertransport mechanism. The basal-to-apical flux must have 2 linear mechanisms, diffusion to the same extent as the opposite flux, and possibly pinocytosis.

The movement of organic solutes between the retina and pigment epithelium.

The transport of several membrane precursors across the bullfrog retinal pigment epithelium (RPE)-choroid was studied in a modified Ussing chamber. The substrates choline, ethanolamine, and glucose did not exhibit a net flux across the tissue, while all amino acids tested showed a net flux in the retina to choroid direction. The 'sink' for a particular amino acid, leucine, was measured in bullfrog retinas in vitro. Decarboxylation of leucine was the major fate of the incorporated amino acid, while much less was directed toward protein synthesis. The subretinal fluid concentration of leucine was estimated to be in the order of 10 microM, approximately eight-fold less than in plasma, and consistent with levels in cerebrospinal fluid. The sink for leucine in the retina at 10 microM leucine is smaller than the fluxes across the RPE-choroid under conditions of an eight-fold gradient. This suggests that the RPE, not the retina, is responsible for controlling the concentrations of solutes in the subretinal fluid.

Incorporation of polyunsaturated fatty acids into lipids of rainbow trout hepatocytes.

Hepatocytes from 5- or 20 degrees C-acclimated rainbow trout (Salmo gairdneri) were incubated with [1-14C]oleic, -linoleic, or -linolenic acid. Both acclimation groups demonstrated greater incorporation of derivatives from linolenic and linoleic acids into phospholipids when assayed at 5 and 20 degrees C; few derivatives of oleic acid were formed. Cells from cold-acclimated trout, when assayed at 5 degrees C with linolenic acid, incorporated a large proportion of radioactivity into free fatty acids. Analysis of each lipid fraction revealed a relatively specific incorporation of certain fatty acids. For example, "dead end" elongation products of the three substrates were preferentially incorporated into neutral lipids, while delta 6 desaturation products of the three acids were retained in the free fatty acid fraction. Twenty-carbon acid derivatives of linoleic and linolenic acids were directed into the phospholipid fraction. Incorporation of the delta 5 desaturation products was temperature sensitive in cells from cold-acclimated but not warm-acclimated trout. The results suggest that selectivity of incorporation of specific fatty acids into phospholipids may be of importance in restructuring membranes of poikilotherms during thermal adaptation.

Fatty acid and sterol synthesis by hepatocytes of thermally acclimated rainbow trout (Salmo gairdneri).

Incorporation of tritium from tritiated water into lipid fractions was measured in isolated hepatocytes from rainbow trout (Salmo gairdneri) acclimated to 5 degrees C and 20 degrees C. Hepatocytes from cold-acclimated trout exhibited significantly higher rates of tritium incorporation into both fatty acid and sterol fractions at assay temperatures of 15 degrees C and 20 degrees C than did hepatocytes from warm-acclimated trout. Tritium incorporation into the fatty acid fraction was nearly temperature independent in hepatocytes from warm-acclimated trout (Q10 = 1.39) but markedly temperature dependent (Q10 = 2.63) in hepatocytes from cold-acclimated trout; in contrast, rates of sterol synthesis were more temperature dependent in warm-acclimated trout. At 5 degrees C, fatty acid lipogenesis comprised a significantly greater percentage of the total tritium incorporation in hepatocytes from warm-acclimated trout and the percentage of total lipogenesis attributable to fatty acids decreased significantly in warm-acclimated trout as the assay temperature increased; the opposite trends were observed in cold-acclimated trout.