Simultaneous determination of leu-enkephalin localization and [3H]γ-aminobutyric acid uptake in rat striatal cell cultures

The purpose of this study was to investigate the coexistence of gamma-aminobutyric acid (GABA) and leu-enkephalin in single neurons from the corpus striatum. Monolayer cell cultures, started from newborn rat corpus striatum, were grown in serum-free medium and examined using GABA autoradiography and leu-enkephalin immunocytochemistry in a double-label protocol. Examples of cells were found which were positive for one or the other neurotransmitter or for neither transmitter, but not for both. Furthermore, cells which appear similar by morphological criteria alone differed in transmitter specificity. We conclude that the two transmitters tested are not localized within single cells and that morphology alone is inadequate to identify functional cell classes in this area.     

Autoradiographic studies of the γ-aminobutyric acid (GABA) system in the rat pancreas

Summary The -cells of the pancreatic islets have been shown to contain -aminobutyric acid (GABA) together with insulin. Autoradiographic analysis indicated that high affinity GABA binding sites (GABA receptors) are not present in the pancreas. High affinity GABA uptake sites are present, not in -cells, but in a few cells on the periphery of the islets. These observations cast doubt on the suggestion that GABA has a paracrine role in the pancreas.     

High-affinity uptake of γ-aminobutyric acid in cultured glial and neuronal cells

Both glial and neuronal cells maintained in primary culture were found to accumulate [³H]GABA by an efficient high-affinity uptake system (apparentK _m=9 M,V _max=0.018 and 0.584 nmol/mg/min, respectively) which required sodium ions and was inhibited by 1 mM ouabain. Strychnine and parachloromercuriphenylsulfonate (pCS) (both at 1 mM) also strongly inhibited uptake of [³H]GABA, but metabolic inhibitors (2,4-dinitrophenol, potassium cyanide, and malonate) were without effect. Only three structural analogs of GABA (nipecotate, -alanine, and 2,4-diaminobutyrate) inhibited uptake of [³H]GABA, while several other compounds with structural similarities to GABA (e.g. glycine,l-proline, and taurine) did not interact with the system. The kinetic studies indicated presence of a second uptake (K _m=92 M,V _max=0.124 nmol/mg/min) in the primary cultures containing predominantly glioblasts. On the other hand, only one of the neuronal cell lines transformed by simian virus SV40 appeared to accumulate [³H]GABA against a concentration gradient. ApparentK _m of this uptake was relatively high (819 M), and it was only weakly inhibited by 1 mM ouabain and 1 mM pCS. The structural specificity also differed from that of the uptake observed in the primary cultures. Significantly, none of the nontransformed continuous cell lines of either tumoral (glioma, C₆; neuroblastoma, Ml; MINN) or normal (NN; I₆) origin actively accumulated [³H]GABA. It is suggested that for the neurochemical studies related to GABA and requiring homogeneous cell populations, the primary cultures offer a better experimental model than the continuous cell lines.     

Involvement of γ-aminobutyric acid transporter 2 in the hepatic uptake of taurine in rats

Taurine is essential for the hepatic synthesis of bile salts and, although taurine is synthesized mainly in pericentral hepatocytes, taurine and taurine-conjugated bile acids are abundant in periportal hepatocytes. One possible explanation for this discrepancy is that the active supply of taurine to hepatocytes from the blood stream is a key regulatory factor. The purpose of the present study is to investigate and identify the transporter responsible for taurine uptake by periportal hepatocytes. An in vivo bolus injection of [(3)H]taurine into the rat portal vein demonstrated that 25% of the injected [(3)H]taurine was taken up by the liver on a single pass. The in vivo uptake was significantly inhibited by GABA, taurine, β-alanine, and nipecotic acid, a GABA transporter (GAT) inhibitor, each at a concentration of 10 mM. The characteristics of Na(+)- and Cl(-)-dependent [(3)H]taurine uptake by freshly isolated rat hepatocytes were consistent with those of GAT2 (solute carrier SLC6A13). Indeed, the K(m) value of the saturable uptake (594 μM) was close to that of mouse SLC6A13-mediated taurine transport. Although GABA, taurine, and β-alanine inhibited the [(3)H]taurine uptake by > 50%, each at a concentration of 10 mM, GABA caused a marked inhibition with an IC(50) value of 95 μM. The [(3)H]taurine uptake exhibited a significant reduction when the GAT2 gene was silenced. Immunohistochemical analysis showed that GAT2 was localized on the sinusoidal membrane of the hepatocytes predominantly in the periportal region. These results suggest that GAT2 is responsible for taurine transport from the circulating blood to hepatocytes predominantly in the periportal region.     

Involvement of sialic acid in the regulation of γ--aminobutyric acid uptake activity of γ-aminobutyric acid transporter 1

The γ-aminobutyric acid (GABA) transporters (GATs) have long been recognized for their key role in the uptake of neurotransmitters. The GAT1 belongs to the family of Na(+)- and Cl(-)-coupled transport proteins, which possess 12 putative transmembrane (TM) domains and three N-glycosylation sites on the extracellular loop between TM domains 3 and 4. Previously, we demonstrated that terminal trimming of N-glycans is important for the GABA uptake activity of GAT1. In this work, we examined the effect of deficiency, removal or oxidation of surface sialic acid residues on GABA uptake activity to investigate their role in the GABA uptake of GAT1. We found that the reduced concentration of sialic acid on N-glycans was paralleled by a decreased GABA uptake activity of GAT1 in Chinese hamster ovary (CHO) Lec3 cells (mutant defective in sialic acid biosynthesis) in comparison to CHO cells. Likewise, either enzymatic removal or chemical oxidation of terminal sialic acids using sialidase or sodium periodate, respectively, resulted in a strong reduction in GAT1 activity. Kinetic analysis revealed that deficiency, removal or oxidation of terminal sialic acids did not affect the K(m) GABA values. However, deficiency and removal of terminal sialic acids of GAT1 reduced the V(max) GABA values with a reduced apparent affinity for extracellular Na(+). Oxidation of cell surface sialic acids also strongly reduced V(max) without affecting both affinities of GAT1 for GABA and Na(+), respectively. These results demonstrated for the first time that the terminal sialic acid of N-linked oligosaccharides of GAT1 plays a crucial role in the GABA transport process.     

γ-aminobutyric acid in the central nervous system of metamorphosing and mature Manduca sexta

We have begun to examine the factors controlling the accumulation of the neurotransmitter γ-aminobutyric acid (GABA) in the central nervous system (CNS) of the sphinx moth Manduca sexta. Analysis of soluble amino acids in CNS structures from mature moths outlines the regional distribution of GABA. Analysis of amino acids in the antennal lobes (the primary olfactory centres) of Manduca during metamorphosis reveals that GABA accumulates gradually and continuously through most of adult development until eclosion; within 18 hr after eclosion, levels of GABA abruptly increase 27–50%. The activity of the biosynthetic enzyme glutamic acid decarboxylase (EC 4.1.1.15), assayed in extracts of antennal lobes from developing moths, does not change after eclosion. Extracts of hemolymph from mature moths contain low levels of glutamate ( <0.2 mM) and higher levels of certain other amino acids such as serine, glutamine and proline. The concentration of proline in hemolymph increases up to 2-fold after eclosion. Glutamate, glutamine and proline are interconvertible in the CNS, and each can serve as precursor for GABA synthesis both in vivo and in vitro. The efficiency of the precursor role in vitro is similar for each amino acid, as estimated from the ratio of the specific radioactivities of GABA and glutamic acid in the ganglion derived from each precursor. Exogenous proline and glutamine can equilibrate rapidly with the ganglionic pools of the same amino acids while glutamic acid is relatively excluded. Taken together, the findings of this study show that proline and glutamine may contribute substantially to synthesis of GABA in the CNS of M. sexta.     

Autoradiographic localization of α-bungarotoxin-binding sites in the carotid body of the rat

Summary Radioiodinated -bungarotoxin (-Bgt) was used to localize -Bgt-acetylcholine receptors in the carotid body of the rat. The gamma spectrometer analyses indicated a high uptake of [¹²⁵I] -Bgt in carotid bodies incubated in vitro (1.51 fmole per organ). Incorporation of the isotope was effectively blocked by pretreatment of carotid bodies with d-tubocurarine and unlabeled -Bgt, but not by atropine. Light microscopic autoradiography showed a heavy labeling of some parenchymal cells. Electron-microscopic autoradiography revealed that labeling was localized along the interface between parenchymal cells, especially where their cytoplasmic processes engage in complex interdigitations. The silver grain counts on electron-microscopic autoradiographs suggest that labelings are preferentially associated with the plasma membrane of certain Type I cells. It is suggested that these Type I cells in the rat's carotid body probably are provided with nicotinic acetylcholine receptors on their plasma membranes.This research was partly supported by a grant from the Edward G. Schleider Foundation of New Orleans.The authors extend appreciation to Drs. Akira Arimura, Dept. of Medicine, for supplying a part of the radioisotope and to James Fisher, Dept. of Pharmacology, Tulane Medical School, for use of the gamma spectrometer. Appreciation also is extended to Mrs. Lia Pedroza for technical assistance     

High affinity uptake of L-glutamate and γ-aminobutyric acid inDrosophila melanogaster

Preparations having properties resembling those of synaptosomes have been isolated from whole fly homogenates ofDrosophila melanogaster using ficoll gradient floatation technique. These have been characterized by marker enzymes and electron microscopy and binding of muscarinic antagenist³H Quinuclidinyl benzilate. An uptake system for neurotransmitter, ã-Aminobutyric acid has been demonstrated in these preparations. A high affinity uptake system for L-glutamate has also been studied in these subcellular fractions. This uptake of glutamate is transport into an osmotically sensitive compartment and not due to binding of glutamate to membrane components. The transport of glutamate has an obligatory requirements for either sodium or potassium ions. Kinetic experiments show that two transport systems, withK _m values 0.33 X 10^-6M and 2.0 X 10^-6M, respectively, function in the accumulation of glutamate. ATP stimulates lower affinity transport of glutamate. Inhibition of glutamate uptake by L-aspartate but not by phenylalanine and tyrosine indicates that a common carrier mediates the transport of both glutamate and aspartate. β-N-oxalyl-L-β β-diamino propionic acid and kainic acid, both inhibitors of glutamate transport in mammalian brain preparations, strongly inhibited transport of glutamate inDrosophila preparations Comparison with uptake of ã-aminobutyric acid and glutamate in isolated larval brain is presented to show that the synaptosome-like preparations we have isolated are rich in central nervous system derived structures, and presynaptic endings from neuromuscular junctions.     

Role of uptake inγ-aminobutyric acid (GABA)-mediated responses in guinea pig hippocampal neurons

Intracellular recordings were obtained from hippocampal pyramidal neurons maintained in vitro. Measurements were made of the conductance change induced by iontophoretically applied gamma-aminobutyric acid (GABA) and, using voltage-clamp techniques, of inhibitory postsynaptic currents resulting from activation of inhibitory pathways. Analysis of GABA iontophoretic charge-response curves indicated that there was considerable variation among neurons with respect to the slope of this relation. The placement of the GABA-containing pipette did not appear to be responsible for the observed variation, since vertical repositioning of the pipette did not alter the slope of the charge-response relationship. Steady iontophoresis of GABA from one barrel of a double-barreled pipette markedly affected the charge-response relation obtained when short pulses were applied to the other barrel. The curve was shifted to the left, and the slope was decreased. Concomitantly, the enhanced GABA-induced responses were prolonged. Similar alterations in GABA responsiveness were observed when the uptake blocker, nipecotic acid, was iontophoretically applied. Furthermore, bath application of saline containing a reduced sodium concentration (25% of control) also produced a prolongation of GABA-mediated responses. Under voltage clamp, inhibitory postsynaptic currents were observed to have biphasic decays. The initial, fast decay was prolonged by an average of 18% by nipecotic acid, whereas the later, slow phase was prolonged by 23%. The results of these studies support the hypothesis that a saturable GABA uptake system is responsible for the observed variation in the charge-response curves and, in turn, underlies the apparent sensitizing effect of excess GABA application. The results also suggest that a reduction of transmitter uptake affects the time course of inhibitory postsynaptic currents in the hippocampus.     

Entropy as a factor in the binding of γ-aminobutyric acid and nipecotic acid to the γ-aminobutyric acid transport system

Nipecotic acid is one of the most potent competitive inhibitors and alternative substrates for the high-affinity -aminobutyric acid transport system in neurons, but the structural basis of this potency is unclear. Because -aminobutyrate is a highly flexible molecule in solution, it would be expected to lose rotational entropy upon binding to the transport system, a change which does not favor binding. Nipecotic acid, in contrast, is a much less flexible molecule, and one would expect the loss of conformational entropy upon binding to be smaller thus favoring the binding of nipecotic acid over -aminobutyric acid. To investigate this possibility, the thermodynamic parameters, G°, H°, and S°, were determined for the binding of -aminobutyrate and nipecotic acid to the high affinity GABA transport system in synaptosomes. In keeping with expectations, the apparent entropy change for nipecotic acid binding (112±13 J·K^–1) was more favorable than the apparent entropy change for -aminobutyric acid binding (61.3±6.6 J·K^–1). The results suggest that restricted conformation per se is an important contributory factor to the affinity of nipecotic acid for the high-affinity transport system for -aminobutyric acid.This work was conducted when both authors were at the Department of Chemistry, University of Maryland, College Park.Special issue dedicated to Dr. Elling Kvamme.     

Distribution and uptake of glycine,glutamate and γ-aminobutyric acid in the vagal nuclei and eight other regions of the rat medulla oblongata

In order to study the central neurochemical control of the vagus nerve, the contents of glycine, GABA, glutamate and five other amino acids have been measured in ten anatomically distinct regions of the rat medulla oblongata. Additionally, the high affinity uptake of glycine, GABA, glutamate, and leucine were measured in the same ten medullary regions. The data support published evidence for glutamatergic and GABAergic transmission in the nucleus of the tractus solitarius (NTS), and glycinergic inhibition in the hypoglossal nucleus. The data also lead to the suggestion that GABA and glutamate may be taken up into glial cells which exist along fiber tracts.     

Ab-initio SCF investigation ofγ-aminobutyric acid

Summary The potential energy surface of the neutral form of-aminobutyric acid was investigated by means of ab-initio 4-31 G SCF calculations. Geometries, energies, and selected wave numbers of all 62 symmetry unique local minima are reported. Intramolecular interactions and all reactions, which involve the intramolecular hydrogen bond, are discussed and compared with those present in the homologues-alanine and glycine.Abbreviations GABA -aminobutyric acid - PES potential energy surface - RHF Roothaan-Hartree-Fock - SCF self consistent field     

Autoradiographic study on the localization of estradiol-17β in neonatal mouse uterus and cervix

Summary The uptake of ³H-estradiol-17 in the neonatal mouse uterus and cervix has been studied by an autoradiographic method. When the radio-active hormone is administered in vivo and in vitro, grains are found to be concentrated above the nuclei both in the uterine and cervical epithelium and stroma. Grain counts revealed that the nuclear concentration of grains is higher at 4 h than at 2 h after isotope injection. The cervical epithelium has a higher nuclear concentration than the uterine epithelium both in vivo and in vitro. In the stroma, this situation is reversed except after in vitro treatment of the tissues.In the cervix, more of the hormone seems to be located within the nucleus while in the uterus a higher proportion of the grains are found in the vicinity of the nuclear periphery.Although the nuclear concentration of grains is higher at 4 h than at 2 h, the number of grains above the sections is lower at 4 h. Both in vivo and in vitro, the number of grains is higher above the stromal than above the epithelial compartments of the uterus and cervix.Five days old animals showed the same labeling pattern. The differences in uptake and distribution of ³H-estradiol are discussed in relation to other known differences in the hormone responsiveness in these tissues.We are greatly indebted to Professor W.E. Stumpf and the Laboratories for Reproductive Biology, University of North Carolina Medical School for the opportunity to study the method of dry mount autoradiography. The work has been supported by the Norwegian Research Council for Science and the Humanities and by the Norwegian Cancer Society (Landsforeningen mot Kreft)     

Light- and electron-microscopic visualization of γ-aminobutyric acid and GABA-transaminase in the oviduct of rats

Summary The distribution in the rat oviduct of -aminobutyric acid and its catabolic enzyme GABA-transaminase was studied by the use of immunocytochemical and enzymehistochemical techniques. At the light-microscopic level, both GABA immunoreactivity and GABA-transaminase enzyme reactivity were found primarily in the tubal epithelium while in the muscle layers of the organ only a faint GABA and GABA-transaminase positive staining could be detected. Electron-microscopic evaluation of the GABA immunoreactivity revealed a heavy labelling of the basal bodies (kinetosomes) and a moderate staining of the cilia. These findings indicate that the role of GABA in the oviduct is not related to neurotransmission but may be related to ciliary functions.     

Electrophysiological characterization of ivermectin triple actions on Musca chloride channels gated by l-glutamic acid and γ-aminobutyric acid

Ivermectin (IVM) is a macrocyclic lactone that exerts antifilarial, antiparasitic, and insecticidal effects on nematodes and insects by acting on l-glutamic acid-gated chloride channels (GluCls). IVM also acts as an allosteric modulator of various other ion channels. Although the IVM binding site in the Caenorhabditis elegans GluCl was identified by X-ray crystallographic analysis, the mechanism of action of IVM in insects is not well defined. We therefore examined the action of IVM on the housefly (Musca domestica) GluCl and γ-aminobutyric acid (GABA)-gated ion channel (GABACl). For both channels, IVM induced currents by itself, potentiated currents induced by low concentrations of agonists, and inhibited currents induced by high concentrations of agonists. Despite exerting common actions on both types of channels, GluCls were more susceptible to IVM actions than GABACls, indicating that GluCls are the primary target of IVM. Substitution of an amino acid residue in the third transmembrane segment (G312M in GluCls, and G333A and G333M in GABACls) resulted in significantly reduced levels or loss of activation, potentiation, and antagonism of the channels, indicating that these three actions result from the interaction of IVM with amino acid residues in the transmembrane intersubunit crevice.     

Identification and characterization of γ-aminobutyric acid uptake system GabPCg (NCgl0464) in Corynebacterium glutamicum

Corynebacterium glutamicum is widely used for industrial production of various amino acids and vitamins, and there is growing interest in engineering this bacterium for more commercial bioproducts such as γ-aminobutyric acid (GABA). In this study, a C. glutamicum GABA-specific transporter (GabP(Cg)) encoded by ncgl0464 was identified and characterized. GabP(Cg) plays a major role in GABA uptake and is essential to C. glutamicum growing on GABA. GABA uptake by GabP(Cg) was weakly competed by l-Asn and l-Gln and stimulated by sodium ion (Na(+)). The K(m) and V(max) values were determined to be 41.1 ± 4.5 μM and 36.8 ± 2.6 nmol min(-1) (mg dry weight [DW])(-1), respectively, at pH 6.5 and 34.2 ± 1.1 μM and 67.3 ± 1.0 nmol min(-1) (mg DW)(-1), respectively, at pH 7.5. GabP(Cg) has 29% amino acid sequence identity to a previously and functionally identified aromatic amino acid transporter (TyrP) of Escherichia coli but low identities to the currently known GABA transporters (17% and 15% to E. coli GabP and Bacillus subtilis GabP, respectively). The mutant RES167 Δncgl0464/pGXKZ9 with the GabP(Cg) deletion showed 12.5% higher productivity of GABA than RES167/pGXKZ9. It is concluded that GabP(Cg) represents a new type of GABA transporter and is potentially important for engineering GABA-producing C. glutamicum strains.     

Nitrate uptake and utilization is modulated by exogenous γ-aminobutyric acid in Arabidopsis thaliana seedlings

Exogenously applied GABA modulates root growth by inhibition of root elongation when seedlings were grown in vitro on full-strength Murashige and Skoog (MS) salts, but root elongation was stimulated when seedlings were grown on 1/8 strength MS salts. When the concentration of single ions in MS salts was individually varied, the control of growth between inhibition and stimulation was found to be related to the level of nitrate (NO₃^?) in the growth medium. At NO₃^? concentrations below 40 mM (full-strength MS salts level), root growth was stimulated by the addition of GABA to the growth medium; whereas at concentrations above 40 mM NO₃^?, the addition of GABA to the growth medium inhibited root elongation. GABA promoted NO₃^? uptake at low NO₃^?, while GABA inhibited NO₃^? uptake at high NO₃^?. Activities of several enzymes involved in nitrogen and carbon metabolism including nitrate reductase (NR), glutamine synthetase (GS), glutamate synthase (NADH-GOGAT), NADP-dependent isocitrate dehydrogenase (NADP-ICDH), and phosphoenol pyruvate carboxylase (PEPCase) were regulated by GABA in the growth medium. Supplementing 1/8 strength MS medium with 50 mM GABA enhanced the activities of all of the above enzymes except ICDH activities in root tissues. However, at full-strength MS, GABA showed no inhibitory effect on the activities of these enzymes, except on GS in both root and shoot tissues, and PEPCase activity in shoot tissues. Exogenous GABA increased the amount of NR protein rather than its activation status in the tissues. This study shows that GABA affects the growth of Arabidopsis, possibly by acting as a signaling molecule, modulating the activity of enzymes involved in primary nitrogen metabolism and nitrate uptake.     

Dual localization of β-glucuronidase and acid phosphatase in lysosomes and in microsomes

Summary The dual localization of certain hydrolases in lysosomes and in endoplasmic reticulum as studied in enzyme staining reactions is now supported by cytobiochemical studies on mouse liver and kidney -glucuronidase and acid phosphatase. Use was made of the renal -glucuronidase response to endogenous androgen for both studies. Accordingly, sucrose homogenates were prepared of liver and kidney of male BALB/C mice previously injected with gonadotrophin along with control animals receiving saline instead. The homogenates were subjected to differential ultracentrifugation yielding six fractions. These were characterized as to their organelle composition by measurements of marker enzymes and by observations with the electron microscope. In all subcellular fractions, -glucuronidase was uniformly increased 5 to 8 times over the corresponding control value and, in fractions rich in lysosomes, this enzyme was easily released by alternate freezing and thawing. On the other hand, the microsomal -glucuronidase and acid phosphatase enzymes were not liberated by freezing and thawing nor were they after treatment with 0.1 % Triton X-100 and by employing other reagents and conditions which are known to release lysosomal enzymes. In contrast to microsomal acid phosphatase, microsomal -glucuronidase activity could be liberated by treatment with hyaluronidase. This soluble -glucuronidase showed the same optimum pH, Michaelis Constant and heat inactivation behavior as the lysosomal -glucuronidase prepared by freezing and thawing treatment. These observations define two populations of microsomal vesicles each identifiable by an individual membrane-associated acid hydrolase. One of these -glucuronidase, increases in specific activity in the animal on androgens and is released by hyaluronidase and the other, acid phosphatase, does not respond to androgen and is not released by hyaluronidase. There would appear to be a variety of mechanisms by which hydrolases enter into association with the membranes of the endoplasmic reticulum and from there, a variety of routes to the lysosomes. A comment is made concerning the question of acid phosphatases and -glucuronidase as enzyme markers for lysosomes in mouse kidney.Aided in part by Research Grant, P-106, of the American Cancer Society, Inc., New York, and by U.S.P.H.S. Grant CA-07538 and by a Research Career Award, CA-K6-18453 to William H. Fishman.     

Regulation of γ-aminobutyric acid synthesis in the vertebrate nervous system

The regulation of glutamic decarboxylase (GAD) activity is undoubtedly the key to the control of the steady-state concentrations of 4-aminobutyric acid (GABA) in the central nervous system. Those factors that might influence GAD activity are reviewed. They include repression and induction of GAD synthesis; the interconversion of the holo- and apo-form of GAD; the availability of substrate and cofactor; the competitive inhibition of GAD by endogenous substances, including GABA; and the involvement of calcium ions in whole-cell preparations. Where possible mechanisms of action are described, and the likelihood that each is of physiological importance is discussed. Experiments are suggested that would help clarify (1) the role of GABA in GAD repression; (2) the possible phosphorylation of GAD; and (3) the existence of multiple forms of the enzyme. In addition, a kinetic mechanism is proposed to explain the possible regulation of GAD by the interconversion of the holo- and apo-forms of the enzyme. It is concluded that the overriding factors responsible for GAD regulation are not yet understood. However, a possible mechanism relying on the direct feedback action of GABA on GAD activity has many attractive features.     

The Organization of the lamina ganglionaris of the prawn,Pandalus borealis (Kröyer)

The Lamina ganglionaris (first optic neuropile) of the decapod crustacean Pandalus borealis has its optic cartridges (synaptic compartments) arranged in horizontal rows. Each optic cartridge contains seven receptor axon terminals and the branching axis fibres of five monopolar second order neurons. Four types of monopolar neurons are classified. Their cell bodies are arranged in two layers. The inner layer contains the cell bodies of exclusively one of these types, and each cartridge is invaded by two neurons of this neuron type (type M 1:a and M 1:b). The outer layer contains the cell bodies of the remaining three types (M 2, M3 and M4). One gives rise to a large radially branched axis fibre in the centre of the cartridge. The other two have wide branches which may make inter-cartridge contacts, one proximally and the other distally in the plexiform layer, which is clearly bistratified. The receptor axons terminate in two levels corresponding to these strata. Two sets of tangenital fibres form networks in the proximal and the mid-portion of the lamina. Both networks have fibres with primary branches in the vertical plane and secondary branches in the horizontal plane. The fibres of the networks are derived from axons that pass from the second optic neuropile, the medulla externa.