A simple and reliable method for delivering small fluid volumes to the brain of a freely moving rat.

A stepper motor-powered micropump for making intracerebral microinjections in freely moving animals is described. The micropump is capable of delivering fluid volumes of up to 20 microl in steps of 17 nl The system does not require a fluid swivel, delivers injections rapidly, maintains long-term calibration accuracy and is not susceptible to blockage of the cannula.

N-docosahexaenoyl, 3 hydroxytyramine: a dopaminergic compound that penetrates the blood-brain barrier and suppresses appetite.

Fatty acids with varying chain lengths (2-22 carbon atoms long) and degrees of unsaturation (0-6 double bonds) were used to synthesize dopaminergic compounds for a study of the carrier mediated transport of dopamine (DA) to the brain. The most active carrier was the all cis C22:6 fatty acid [docosahexaenoic acid, (DHA)]which increased DA uptake through the blood-brain barrier by greater than 7.5 fold. The DHA-DA compound, NMI 8739, depressed the general locomotor activity of mice in a dose dependent manner. It also suppressed the appetite of Balb c mice and Charles River rats by 50% and 95% respectively at a dose of 10 mg/kg. Daily administration of NMI-8739 for a three week period did not induce tolerance. These results demonstrate DHA's potential for the carrier mediated transport of small molecules to the brain.

Evidence for the in vivo polymerization of ependymin: a brain extracellular glycoprotein.

Ependymin, a glycoprotein of the brain extracellular fluid, has been implicated in synaptic changes associated with the consolidation process of long-term memory formation and the activity-dependent sharpening of connections of regenerating optic nerve. In vitro experiments have demonstrated that ependymin has the capacity to form fibrous insoluble polymers (FIP) when the solvent Ca2+ concentration is reduced by the addition of EGTA. Such products, once formed, do not dissolve in 2% sodium dodecyl sulfate (SDS) in 5 M urea. This property was used to develop a method for isolating brain FIP. A reproducible quantity of FIP was found in goldfish and mouse brain. This was highly concentrated in the synaptosomal fraction and had identical immunoreactivity properties to FIP obtained by the polymerization of pure ependymin in vitro as well as a cross-reactivity to other protein components of the extracellular matrix such as fibronectin and laminin. Labeling studies with [35S]methionine showed that labeled FIP aggregates are synthesized in vivo and become associated with the synaptosomal fraction. A comparison of the amino acid sequence of ependymin with those for proteins of the extracellular matrix indicated that common sequences 5-6 amino acids long exist in the molecules. These homologies may explain why antibodies to fibronectin, laminin and tubulin can recognize the FIP prepared from pure ependymin. These results suggest that ependymin can polymerize in vivo to form FIP aggregates which have similar immunoreactivity properties to major components of the brain extracellular matrix.

Synthesis, brain uptake, and pharmacological properties of a glyceryl lipid containing GABA and the GABA-T inhibitor gamma-vinyl-GABA.

1-O-Linolenoyl-2-O-(4-aminobutyryl)-3-O-(4-vinyl-4-aminobutyryl)glycerol (LGV) was synthesized as an example of a prodrug which readily penetrates the blood-brain barrier (brain penetration index 97% +/- 15%) and releases two active substances in the central nervous system (CNS): GABA (4-aminobutanoic acid) and the GABA transaminase inhibitor (GABA-T) of GABA breakdown. In vitro studies showed that the compound can inhibit GABA-T after hydrolysis by CNS esterases and that it enhanced GABAergic inhibition when applied to rat hippocampus slices. In vivo studies indicate that LGV depresses the spontaneous locomotor activity of mice. Its activity on a molar basis was some 300 times greater than that of gamma-vinyl-GABA.

Protein synthesis as a function of depth in slices of rat hippocampus.

Physiologically viable slices of rat hippocampus were incubated in radiolabeled valine, then cut into 20 microns serial sections to evaluate the profile of protein synthesis through the depth of the slice. Maximum radiolabel incorporation was observed near the center of the slice, while at the upper (gas interface) and lower (liquid interface) surfaces radiolabel incorporation per section was reduced by about 30% and 90%, respectively. The results suggest that in properly slices damage due to slicing may be less important to cell viability than are limits on oxygen diffusion into the tissue.

Classical conditioning leads to changes in extracellular concentrations of ependymin in goldfish brain.

ELISA measurements showed that brain extracellular fluid (ECF) levels of ependymin decreased for animals that learned to associate a paired presentation of a light stimulus (CS) with the onset of an electric shock (US), whereas no changes were obtained for control goldfish that received the same number of stimuli delivered in a random unpaired order. Studies of the time course of the changes showed an immediate decrease (19%) after training followed by an increase (20%) above baseline by 5 h and a final return to baseline by 25 h. These data extend the findings of previous experiments, which demonstrated a role for ependymin in two training procedures that involved motor learning, to classical conditioning where no motor learning occurs. Thus it appears that ependymin may have a functional role in molecular mechanisms of learning and memory in general.

Novel GABA agonists depress the reward effect of lateral hypothalamic stimulation in rats.

Rats were given systemic injections of one of a series of novel GABA compounds which can penetrate the blood-brain barrier to release GABA into the brain. They were then tested on lateral hypothalamic self-stimulation behavior using a rate-frequency paradigm to discriminate effects on reward from those on motor/performance. Both reward and, to a lesser extent, motor/performance impairments were found with all GABA compounds. In more extensive testing with one compound, LG2, no differences in the effects of three salts (acetate, ascorbate, and tartarate) were found except that the tartarate salt effects decayed more rapidly.

Uptake in brain and neurophysiological activity of two lipid esters of gamma-aminobutyric acid.

Two lipid esters of gamma-aminobutyric acid (GABA), 1-linolenoyl-2,3-bis(4-aminobutyryl)propane-1,2,3-triol and 1,2-dilinolenoyl-3-(4-aminobutyryl)propane-1,2,3-triol, were found to have brain uptake indices of greater than 30% using the single-pass carotid artery injection technique. Both compounds produced dose-dependent inhibition of the evoked population spike in slices of rat hippocampus maintained in vitro. This effect was blocked reversibly by picrotoxin. The magnitude of the inhibition produced by the lipid esters of GABA was comparable to that of similar doses of GABA, but for both compounds the duration of the effect was at least 10 times longer than that produced by GABA. These data are consistent with the idea that the lipid esters of GABA can effectively penetrate the blood-brain barrier and act as prodrugs for the delivery of GABA to the central nervous system.

gamma-Aminobutyric acid esters. 3. Synthesis, brain uptake, and pharmacological properties of C-18 glyceryl lipid esters of GABA with varying degree of unsaturation.

A series of 14C-labeled and unlabeled di-gamma-aminobutyric acid esters of glyceryl lipids having zero to three double bonds (stearoyl, oleoyl, linoleoyl, and linolenoyl) were synthesized. Measurements of the octanol/water partition coefficients of the compounds showed an increase with decreasing number of double bonds (i.e., from linolenoyl to stearoyl). The brain-uptake index went up from 31.5 (linolenoyl) to 45.1 (stearoyl) and similarly the brain-penetration index went up from 15 (linolenoyl) to 28 (stearoyl). Intraperitoneal injections of these di-GABA lipid esters produced a substantial inhibition of the general motor activity in mice at a dose of 30 mg/kg; the most active molecules were those containing two and three double bonds, i.e., the linolenoyl and linolenoyl derivatives. This is in reverse order to that predicted by brain-uptake and lipid-solubility properties, suggesting that the structure of the fatty acid side chain may be an additional factor in influencing biological activity.

Inhibitory effect of cholesteryl gamma-aminobutyrate on evoked activity in rat hippocampal slices.

Cholesteryl gamma-aminobutyrate (C-G) readily crosses the blood-brain barrier and has properties that suggest that it may be a potential gamma-aminobutyric acid (GABA)-mimetic compound. The effect of this compound on the orthodromically-evoked discharge of hippocampal pyramidal cells was investigated using slices of rat hippocampus maintained in vitro. The compound produced dose-dependent inhibition of the discharge of pyramidal cells. The magnitude of the inhibitory effect was somewhat less than that produced by a similar dose of GABA, but the duration of the inhibition was prolonged by about 10-fold over that produced by GABA. The inhibition produced by cholesteryl gamma-aminobutyrate was blocked by the addition of picrotoxin to the incubation medium, and by replacement of chloride with isethionate. In addition, pretreatment of slices with the irreversible esterase inhibitor, phenylmethylsulfonylfluoride, attenuated the effects of cholesteryl gamma-aminobutyrate, but not that of GABA. These results suggest that cholesteryl gamma-aminobutyrate has GABA-like actions in the CNS, and that its activity is largely dependent upon enzymatic release of GABA from the compound by esterases present in the tissue.

Protein release from hippocampus in vitro.

Physiologically viable slices of rat hippocampus in vitro continuously release protein into the superfusion medium at a rate of about 2 micrograms/mg tissue/h. Assays of a cytoplasmic marker enzyme (lactate dehydrogenase) indicate that this material is not the result of cell lysis. Pulse-chase experiments using [3H]valine indicate that a substantial fraction of the newly synthesized proteins eventually appear in the incubation medium (18.7% +/- 3% of the total TCA precipitable radioactivity during a 6-h superfusion) and that the releasable protein pool has an apparent half-life of about 4 h. Simultaneous labeling of newly synthetized proteins with [3H]fucose and [14C]valine showed a 3-fold higher ratio of [3H]fucose to [14C]valine in the released protein fraction compared to the soluble cytoplasmic protein and to the crude membrane protein fraction, suggesting that the soluble released proteins are more highly glycosylated than the proteins retained in the tissue. Electrophoretic migration patterns on SDS-polyacrylamide gels with both labeled and unlabeled proteins show differences between the released proteins and the soluble cytoplasmic proteins of the tissue. Several molecular weights between 14 kdalton and 86 kdalton appear to be characteristic of the released protein fraction. These results suggest that a distinct group of proteins and glycoproteins exists in hippocampal tissue which is destined to be selectively released into the extracellular space.

Proteins of the brain extracellular fluid: evidence for release of S-100 protein.

Extracellular protein fractions were obtained (1) by mild, isotonic irrigation of freshly perfused brain tissue; (2) by collection of proteins released into superfusing medium by physiologically viable slices of rat hippocampus; and (3) by sampling the CSF of anesthetized rats. Analysis of the S-100 protein content of these fractions gave values of 2.8, 4.2, and 1.8 micrograms S-100/mg protein, respectively. These values were three- to sixfold higher than the S-100 content of the soluble cytoplasmic protein fractions from the same tissue. This several-fold higher S-100 content of the extracellular protein fractions relative to the intracellular cytoplasmic protein fractions indicates that S-100 is selectively released into the extracellular spaces of the brain. We suggest that the biological function of this CNS protein may involve intercellular transfer.

Chronic zinc deficiency alters neuronal function of hippocampal mossy fibers.

Low-frequency stimulation of hippocampal mossy fiber axons in zinc-deficient adult rats produced synaptic responses that declined in amplitude with successive stimuli. This response decrement is abnormal and suggests that the heavy deposits of zinc in mossy fiber boutons are important for synaptic transmission.

ERP response decrement and recovery as a function of stimulus type and scalp location.

Response decrement and recovery of visual event-related potentials (ERPs) at different electrode sites to different types of repetitive stimuli were studied. Nine blocks of stimuli were presented in the form AAAAAABB. Stimuli were of four types: diffuse flashes of light, geometric shapes, words of similar meanings, and words of dissimilar meanings. The relative amounts of ERP response decrement and recovery differed as a function of ERP response components measured (early vs. late), electrode site (left frontal, left temporoparietal, and left occipital), and stimulus type. Early components were sensitive to sensory characteristics of the stimuli. Late components from the left temporoparietal site were sensitive to the meanings of the linguistic stimuli. The research suggests a useful paradigm for the study of habituation-like responses and of cerebral specificity in an intact normal population.

Psychotic children's free-choice selection of natural and distorted speech.

The relative preferences of psychotic, language-handicapped, and linguistically normal children for natural intelligible speech or various forms of distorted speech were assessed by a two-choice operant procedure. The stimulus preferences of the psychotic group were found to be qualitatively different from those of either nonpsychotic group. The nonpsychotic children preferred natural intelligible speech in every case, although the language-handicapped children were less sensitive than the normals to the distinctions between communicative speech and unintelligible speech. In contrast, the psychotic group showed an overall preference for unnatural, distorted speech. They apparently attended primarily to the acoustic novelty of the stimuli and had little regard for linguistic intelligibility. These results suggest that speech and language may have distinctly different functional significance for these psychotic children than for normals.