Regulation of the Serotonergic System by Kainate in the Avian Retina

Abstract

Serotonin (5-HT) has been recognized as a neurotransmitter in the vertebrate retina, restricted mainly to amacrine and bipolar cells. It is involved with synaptic processing and possibly as a mitogenic factor. We confirm that chick retina amacrine and bipolar cells are, respectively, heavily and faintly immunolabeled for 5-HT. Amacrine serotonergic cells also co-express tyrosine hydroxylase (TH), a marker of dopaminergic cells in the retina. Previous reports demonstrated that serotonin transport can be modulated by neurotransmitter receptor activation. As 5-HT is diffusely released as a neuromodulator and co-localized with other transmitters, we evaluated if 5-HT uptake or release is modulated by several mediators in the avian retina. The role of different glutamate receptors on serotonin transport and release in vitro and in vivo was also studied. We show that L-glutamate induces an inhibitory effect on [³H]5-HT uptake and this effect was specific to kainate receptor activation. Kainate-induced decrease in [³H]5-HT uptake was blocked by CNQX, an AMPA/kainate receptor antagonist, but not by MK-801, a NMDA receptor antagonist. [³H]5-HT uptake was not observed in the presence of AMPA, thus suggesting that the decrease in serotonin uptake is mediated by kainate. 5-HT (10-50 μM) had no intrinsic activity in raising intracellular Ca²⁺, but addition of 10 μM 5-HT decreased Ca²⁺ shifts induced by KCl in retinal neurons. Moreover, kainate decreased the number of bipolar and amacrine cells labeled to serotonin in chick retina. In conclusion, our data suggest a highly selective effect of kainate receptors in the regulation of serotonin functions in the retinal cells.

Keywords: Glutamate receptors; Kainate; Retina; Serotonin.

Fig. 1

Immunodetection of serotonin in cells in both retinal cell cultures and embryonic retina. a In culture at E8C6, staining was concentrated primarily in somata (arrows), although some labeling was also noticed in the cellular processes (arrowheads, see insert). Bar = 10 μm. b In the embryonic avian retina (E14), amacrine and bipolar cells are, respectively, heavily and weakly labeled by anti-serotonin antibody (TH). Amacrine tyrosine hydroxylase positive cells (TH) also show serotonin immunolabeling. The labeling with DAPI reveals all nuclear layers of the retina. Bar = 20 μm

Fig. 2

[³H]5-HT uptake in retinal cell cultures is decreased by l-glutamate through kainate receptors. a The effect of several neurotransmitter agonists in the 5-HT uptake was evaluated. Cells were pre-treated with histamine (50 μM and 100 μM), 2-chloroadenosine (50 μM and 100 μM), methacholine (50 μM and 100 μM), GABA (50 μM and 100 μM), and l-glutamate (50 μM) for 10 min prior to the [³H]5-HT uptake assay. Among all agonists only l-glutamate showed an effect. n = 5. b Effect of glutamate receptor agonists on the 5-HT uptake. Cells were pre-incubated with l-glutamate (50 μM), NMDA (50 μM), AMPA (100 μM), kainate (1 μM), and Trans-ACPD (100 μM) by 10 min prior to the [³H]-HT uptake assay. n = 5. c The inhibition of [³H]-5-HT uptake by l-glutamate in retinal cell cultures is only prevented by antagonist of kainate/AMPA receptor (CNQX), but not by NMDA receptor antagonist (MK-801). MK-801 (10 μM) and CNQX (50 μM) were pre-incubated before l-glutamate treatment. n = 5. d The effect of iGluRs agonists, kainate or NMDA, depends on selective receptor activation. NMDA receptor antagonist, MK-801 (10 μM), and kainate/AMPA receptor antagonist, CNQX (50 μM), were pre-incubated before 10 μM agonist treatment and blocked the respective agonist action. n = 5. Fluoxetine (10 μM) was used to evaluate the specific [³H]5-HT uptake through serotonin transporter. Data are expressed as percent control (mean ± S.E.M.). ANOVA t test *p < 0.01 vs. control, #p < 0.01 vs. fluoxetine treatment

Fig. 3

Effects of glutamatergic receptor agonists and antagonists on the retinal culture 5-HT release. a glutamate (100 μM) and kainate (50 μM) induces 5-HT release, while AMPA (100 μM) and NMDA (100 μM) had no effect. The glutamate and kainate effects were blocked by pretreatment with kainate/AMPA receptor antagonist CNQX (100 μM). Data are expressed as [³H]5-HT released as percent basal (mean ± S.E.M.). ANOVA test-t. *p < 0.01 vs. Basal. b Graph shows the effect of Ca⁺⁺ on [³H]5-HT release evoked by Kainate (50 μM) or Basal, and of fluoxetine (10 μM) treatment on [³H]5-HT release evoked by Kainate or Basal. Data are expressed as [³H]5-HT released as percent basal (mean ± S.E.M.). ANOVA t-test *p < 0.01 vs. Basal

Fig. 4

Intracellular calcium ([Ca²⁺]i) transients in the presence of 5-HT, potassium chloride (KCl), or adenosine triphosphate (ATP) in mixed retina neuroglial cells. a Bright field and b Fura-2 fluorescence images reveal neurons and glia from embryonic retinal cell cultures. c There was no [Ca²⁺]i increase induced by 10 µM 5-HT or d 50 µM 5-HT in neuroglia cells in culture. e Addition of 50 µM 5-HT in the presence of 50 mM KCl decreases [Ca²⁺]i shifts in neurons compared to activation promoted by 50 mM KCl (red) as quantified in (f). This difference is significant when the respective areas under the curve (AUC) are compared (p < 0.05). g However, addition of 1 mM ATP + 10 µM 5-HT-induced Ca²⁺ shifts similar to 1 mM ATP (red), with no significant changes between [Ca²⁺]i curves, quantified in H. as comparison of AUC (p > 0.05). N = 4 for all conditions, at least 1000 cells analyzed

Fig. 5

Antagonism of the glutamatergic receptor prevents [³H]-5-HT uptake in retinal cell cultures. a Photomicrographs of chick retina radial sections immunolabeled for serotonin after different treatments. Control retinas kept in Locke’s solution for 50 min (Control, left panel); retinas kept in Locke’s solution for 20 min plus 10 μM kainate for 30 min (KA, middle panel); retinas treated with DNQX for 20 min followed by the exposure to 10 μM kainate plus DNQX for 30 min (KA + DNQX, right panel). (b–d) Histograms show optical densitometry of bipolar cells (b), amacrine cells (c), and IPL (d). The values are expressed as mean ± SEM, n = 3. (*p < 0.05; *** p < 0.001, with respect to controls; #p < 0.05; ###p < 0.001, with respect to kainate). Scale bar: 20 μm. ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer