Comparable effects of flickering and steady patterns of light adaptation on photomechanical responses of cones in amphibian (Xenopus laevis) retina

Neuroscience Letters 272 (1999) 163±166 www.elsevier.com/locate/neulet

Comparable effects of ¯ickering and steady patterns of light adaptation on photomechanical responses of cones in amphibian (Xenopus laevis) retina A.R. Angotzi a, b, J. Hirano b, c, S. Haamedi a, R. Murgia b, S. Vallerga b, c, M.B.A. Djamgoz a,* a

Neurobiology Group, Department of Biology, Imperial College of Science, Technology and Medicine, London SW7 2AZ, UK b IMC-International Marine Centre, Torregrande, 09072 Oristano, Italy c Istituto di Cibernetica e Bio®sica (CNR), Sezione di Ecologia Sensoriale, Genova e Oristano, Italy Received 18 May 1999; received in revised form 29 June 1999; accepted 13 July 1999

Abstract The effects of two distinct patterns of light stimulus, steady and ¯icker, on cone photomechanical movements (PMMs) in the Xenopus laevis retina were investigated. For both patterns studied, the effects on PMMs were assessed by quantitative analysis of the cone positions in the outer retina. Steady light adaptation was found to be equally effective as ¯icker in causing cone contractions. This was unlike the situation previously found in the cyprinid ®sh retina, in which ¯ickering light was signi®cantly more effective than steady. This difference could be related to the light-evoked response characteristics and circuitry of dopaminergic retinal neurones in the two vertebrate classes. The role of dopamine and other possible neuromodulator(s) in light adaptive control of vertebrate retinae is discussed. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Light adaptation; Flicker; Steady; Cone photomechanical movements; Dopamine; Nitric oxide; Retina; Teleost; Xenopus

A considerable body of evidence suggests that photoreceptors of lower vertebrates (teleost and amphibian), which lack a pupillary response, undergo structural changes in response to dark/light adaptation [3]. The most visible morphological changes occurring in the outer retina are the photomechanical movements (PMMs). During light adaptation, cones and melanin granules of pigment epithelial cells move towards the outer limiting membrane (OLM) while rods move away from it [4,16]. These responses are reversed during dark adaptation. The PMMs are thought to contribute to photoreceptors' optimal light sensitivity. Previous studies showed that the effects of light adaptation can be mimicked by exogenous dopamine (DA). Application of DA to retinae or isolated cones in darkness elicited light adaptive cone contractions [4,16]. The pattern of light adaptation has been shown to be an important determinant of PMMs in the carp retina. Flickering light was found to be signi®cantly more effective in causing light adaptive cone contractions than steady light of the same intensity, * Corresponding author. Tel.: 144-0207-594-5370/5385; fax: 144-0207-584-2056. E-mail address: [email protected] (M.B.A. Djamgoz)

although the former delivered 50% less photons to the retina [11]. This difference was thought to be due to ¯ickering light being more effective in releasing endogenous DA in the teleost retina [13]. In turn, this could be related to the light-evoked activity of some dopaminergic interplexiform cells (IPCs) having prominent transient depolarizing components [6]. In contrast, in the dark adapted Xenopus retina, exposure to steady light caused DA release but ¯ickering light had little effect [2]; steady and ¯ickering light were equally ef®cient in releasing DA in another amphibian (mudpuppy) retina assessed by endogenous DA-mediated uncoupling effect on horizontal cells (HCs) [8]. The aim of the present study was to determine whether amphibian photoreceptors would correspondingly respond to ¯ickering vs. steady light adaptation differently compared with cyprinid ®sh. All experiments were carried out on retinae of Xenopus laevis (body size, 8±10 cm) obtained from Blade Biological (UK). The animals were maintained in an aerated aquarium for 1 month after purchase before experiments were performed. The animals were kept on a 12:12 h light/dark cycle, with light coming on at 07:00 h. Two different

0304-3940/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 9 9) 00 60 6- 0

164

A.R. Angotzi et al. / Neuroscience Letters 272 (1999) 163±166

patterns of light adaptation were used: (i) `steady', i.e. continuous; (ii) square, ¯ickering at 3 Hz with zero basal level, as in the case of the previous study [11]. All experiments were carried out using animals that had been dark adapted for 3±5 h during their normal dark phase. For the experiment, the frogs were placed in a black bucket and exposed to a uniform ®eld of white light at an intensity of 15 or 30 mW/cm 2 (at the level of the animals). After the treatments, the animals were killed, retina-eye cups were dissected and ®xed in 2.5% paraformaldehyde and 2% glutaraldehyde in 0.067 M sodium cacodylate buffer (pH 7.2) for 2 h at 48C. After ®xation, the tissue was washed in 3% sucrose in the buffer, dehydrated in a graded series of ethanol and then kept for 15 min in propylene oxide. For the embedding, epoxy resin (Agar Scienti®c Ltd.) was used. Thin (2 mm) sections were cut and stained by Richardson's stain [17]. It has been reported that mainly cones undergo photomechanical movement in the Xenopus retina [3]. In the

present study, the cone index (CI) was de®ned as x/y where x was length of cone (the distance between the distal border of the myoid region of the cone and the OLM) and y was retinal thickness (from the OLM to the inner limiting membrane) to correct for possible artefacts introduced by embedding and sectioning [10]. For each experiment, a total of 50 cones (12±13 cones from each of four different regions) were measured in a given retina, one from a different animal. Data from a given retina were pooled; averages were obtained from three separate animals for each experiment and expressed as means ^ standard errors. Statistical analysis were performed using non-parametric tests, Mann± Whitney and Kruskal±Wallis tests. All measurements were performed using a light microscope (DMRB, Leica) and image analysis software package (Quantimet 500, Leica). The dark-adapted control state was assessed in retinaeeyecups taken from Xenopus in the dark phase and kept in the dark for a further period of 30 min in the same position

Fig. 1. Typical light micrographs showing mainly the outer retinae of Xenopus under various light/dark adaptive conditions. (a) Darkadapted control retina taken from a fully dark-adapted animal. (b) Light-adapted control retina taken from a fully light-adapted animal. (c) Retina following test light adaptation with 30 mW/cm 2 steady light. (d) Retina following test light adaptation with 30 mW/cm 2, 3 Hz ¯icker light. In each case, the arrowhead denotes the external limiting membrane. The scale bar (20 mm) applies to all the panels.

A.R. Angotzi et al. / Neuroscience Letters 272 (1999) 163±166

Fig. 2. Summary of data from light adaptation experiments involving different patterns and intensities of background light adaptation. CI, cone index; Dark, fully dark-adapted control; S1 and F1: steady and ¯icker (3 Hz) light adaptation of 15 mW/cm 2. S2 and F2: steady and ¯icker (3 Hz) light adaptation of 30 mW/ cm 2. Light, fully light-adapted control.

as for light adaptation; the CI value was 0:25 ^ 0:012. Fully light-adapted control was obtained from animals taken directly from the aquarium during the middle of the normal light phase. In this case, the value of CI was 0:16 ^ 0:012. Application of either 3 Hz ¯ickering (F) or steady (S) background light of the same intensity to living Xenopus in the normal dark phase produced light adaptive cone contractions. These results are illustrated qualitatively in Fig. 1 and quantitatively in Fig. 2. After treatment with steady and ¯ickering lights of intensity 15 mW/cm 2, the CI values became 0:20 ^ 0:01 (S1) and 0:22 ^ 0:006 (F1), respectively. Increasing the illumination level to 30 mW/cm 2 caused a further reduction in CI values for both steady and ¯ickering light: 0:17 ^ 0:01 (S2) and 0:19 ^ 0:008 (F2), respectively. Importantly, the effect of the steady light was not signi®cantly different from that of the ¯icker at either intensity (Kruskal±Wallis test, P , 0:05). A previous study on the carp retina showed that the pattern of light is an important determinant of the light adaptive process by demonstrating that ¯ickering light induced greater cone contraction than steady, although the latter delivered twice the number of photons in the given time [11]. In contrast, the present study on the Xenopus retina has shown that ¯ickering and steady lights are equally effective in inducing light adaptive cone PMMs. This complements an earlier electrophysiological study on amphibian (mudpuppy) HCs showing transient uncoupling induced equally effectively by ¯ickering and steady background adaptation suggesting that both ¯ickering and steady stimuli could induce DA release [8]. An earlier DA release study in the dark-adapted Xenopus retina showed that exposure to steady light caused DA release but ¯ickering light had little effect [2]. However, in that study, the frequency of the ¯ickering light used (0.05 Hz) was much lower than in

165

the present study (3 Hz). Our results may indicate also that measuring light adaptive cone PMMs could be more sensitive than direct measurement of endogenous DA release (see also Ref. [8]). The differences in the effects of ¯ickering and steady light on light adaptive cone PMMs between the two vertebrates (teleost and amphibian) could be due to a number of reasons, as follows: First, there may be difference(s) in the light-induced DA release mechanisms between teleosts and amphibians. For example, in both teleost and amphibian retinae, lightinduced DA release was inhibited by g -aminobutyric acid (GABA), suggesting tonic inhibitory control of DA release by the OFF bipolar cells (BCs) and GABAergic amacrine cells (ACs) [8,13]. In the teleost retina, however, lightinduced DA release was also inhibited by l-glutamate and kainate [13] and activation of N-methyl-d-aspartate receptors has been suggested to mediate the ¯ickering lightinduced DA release [12]. These results, taken together, would suggest that the OFF BC pathway is more dominant in controlling DA release in the teleost retina. In contrast, in the mudpuppy retina, 2-amino-4-phosphonobutyrate (APB) blocked the DA-mediated HC uncoupling effect of light [2] and it has also been reported that APB inhibited the lightinduced DA release in the Xenopus retina [1]. These results would strongly indicate ON-BC control of DA release in the amphibian retina. The light-evoked response characteristics of dopaminergic cells in the two species might also be different. In teleost ®sh (gold®sh), dopaminergic IPCs appear to respond to light with prominent depolarizing transients [6]. However, in amphibian (tiger salamander) retina, dopaminergic ACs have been shown to produce much slower responses to light [22]. Second, the synaptic output contacts of the DA cells in teleost and amphibian retinae appear to be different. Dopaminergic IPCs in the teleost retina have well-developed plexuses in the OPL. Previous studies have shown that gold®sh DA-IPCs make junctional apposition with photoreceptors thereby suggesting that DA could act directly on these cells [19]. This is unlike the Xenopus DA-ACs, which lack a plexus in the OPL and only some 20% of the ACs have ®ne processes extending towards the OPL [21]. Also, the densities of the DA cells are different: ~100 cells/mm 2 in carp vs. ~20 cells/mm 2 in Xenopus [7,21]. The released DA can diffuse throughout the retina to reach pharmacologically active levels several tens of microns away from its local production site [20]. The paracrine spread of the released DA in the Xenopus retina, therefore, could result in a uniform concentration being achieved in the outer retina, irrespective of the original release pattern. Third, the neurochemical bases of the light-evoked cone PMMS may be different in teleost and amphibian retinae. In fact, evidence from several independent lines has shown that light adaptation of the vertebrate retina may involve multiple modulator(s) in addition to DA [5,7]. In particular, a light adaptive role for nitric oxide (NO) has been

166

A.R. Angotzi et al. / Neuroscience Letters 272 (1999) 163±166

suggested from its effects on carp cones and HCs [9,10,15], and direct evidence has been presented for NO release during light adaptation of retinae of carp [18] and rabbit [14]. However, it is not yet known if NO has a similar role in the Xenopus retina and if endogenous NO is involved in light adaptation-induced cone contraction in either species. This work has partly been supported by the STRIDE programme provided both by the European Commission and Regione Autonoma della Sardegna. [1] Boatright, J.H., Gordon, J.R. and Iuvone, P.M., Inhibition of endogenous dopamine release in amphibian retina by L-2amino-4-phosphonobutyric acid (L-AP4) and trans-2aminocyclopentane-1,3-dicarboxylate (ACPD). Brain Res., 649 (1994) 339±342. [2] Boatright, J.H., Hoel, M.J. and Iuvone, P.M., Stimulation of endogenous dopamine release and metabolism in amphibian retina by light and K 1-evoked depolarization. Brain Res., 482 (1989) 164±168. [3] Burnside, B. and Dearry, A., Cell motility in the retina. In R. Adler and D. Faber (Eds.), A Model for cell Biology Studies, Part I, Academic Press, Orlando, FL, 1986, pp. 151±206. [4] Dearry, A. and Burnside, B., Dopaminergic regulation of cone retinomotor movement in isolated teleost retinas. Induction of cone contraction is mediated by D2 receptors. J. Neurochem., 46 (1986) 1006±1021. [5] Djamgoz, M.B.A., Vallerga, S. and Wagner, H-J., Functional organization of the outer retina in aquatic and terrestrial vertebrates: comparative aspects and possible signi®cance to the ecology of vision. In S.N. Archer, M.B.A. Djamgoz, E.R. Loew, J.C. Partridge and S. Vallerga (Eds.), Adaptive Mechanisms in the Ecology of Vision, Dodrecht, Kluwer, 1999, pp. 329±382. [6] Djamgoz, M.B.A., Usai, C. and Vallerga, S., An interplexiform cell in the gold®sh retina: light-evoked response pattern and intracellular staining with horseradish peroxidase. Cell Tissue Res., 264 (1991) 111±116. [7] Djamgoz, M.B.A. and Wagner, H.-J., Localization and function of dopamine in the adult vertebrate retina. Neurochem. Int., 20 (1992) 139±191. [8] Dong, C.J. and McReynolds, J.S., Comparison of the effects of ¯ickering and steady light on dopamine release and horizontal cell coupling in the mudpuppy retina. J. Neurophysiol., 67 (1992) 364±372. [9] Furukawa, T., Yamada, M., Petruv, R., Djamgoz, M.B.A. and Yasui, S., Nitric oxide, 2-amino-4-phosphonobutyric acid and dark/light adaptation modulate short-wavelength-

[10] [11]

[12]

[13]

[14] [15]

[16]

[17] [18]

[19]

[20]

[21]

[22]

sensitive synaptic transmission to retinal horizontal cells. Neurosci. Res., 27 (1997) 65±74. Greenstreet, E.H. and Djamgoz, M.B.A., Nitric oxide induces light-adaptive morphological changes in retinal neurones. NeuroReport, 6 (1994) 109±112. Haamedi, S.N. and Djamgoz, M.B.A., Effects of different patterns of light adaptation on cellular and synaptic plasticity in teleost retina: comparison of ¯ickering and steady lights. Neurosci. Lett., 206 (1996) 93±96. Harsanyi, K., Wang, Y. and Mangel, S.C., Activation of NMDA receptors produce dopamine-mediated changes in ®sh retinal horizontal cell light responses. J. Neurophysiol., 75 (1996) 629±647. Kirsch, M. and Wagner, H-J., Release pattern of endogenous dopamine in teleost retinae during light adaptation and pharmacological stimulation. Vision Res., 29 (1989) 147±154. Neal, M.J., Cunningham, J.R. and Matthews, K., Selective release of nitric oxide from retinal amacrine and bipolar cells. Invest. Ophthalmol. Vis. Sci., 39 (1998) 850±853. Petruv, R., Furukawa, T., Yasui, S. and Djamgoz, M.B.A., Sodium nitroprusside, a nitric oxide donor, generates chromatic difference in the receptive ®eld size of H1 horizontal cells in isolated retina of carp. J. Physiol., 473 (1993) 163. Pierce, M.E. and Besharse, J.C., Circadian regulation of retinomotor movements. Interaction of melatonin and dopamine in the control of cone length. J. Gen. Physiol., 86 (1985) 671±689. Richardson, K.C., Jarett, L. and Finke, E.H., Embedding in epoxy resin for ultrathin sectioning in electron microscopy. Stain Technol., 35 (1960) 313±323. Sekaran, S., Matthews, K.L., Cunningham, J.R., Neal, M.J. and Djamgoz, M.B.A., Nitric oxide release during light adaptation of the carp retina. J. Physiol., 515P (1999) 102±103. Van Haesendonck, E., Marc, R.E. and Missotten, L., New aspects of dopaminergic interplexiform cell organization in the gold®sh retina. J. Comp. Neurol., 333 (1993) 503± 518. Witkovsky, P., Nicholson, C., Rice, M., Bohmaker, K. and Meller, E., Extracellular dopamine concentration in the retina of the clawed frog Xenopus laevis. Proc. Nat. Acad. Sci. USA, 90 (1993) 5667±5671. Witkovsky, P., Zhang, J. and Blam, O., Dopaminergic neurons in the retina of Xenopus laevis: amacrine vs. interplexiform subtypes and relation to bipolar cells. Cell Tissue Res., 278 (1994) 45±56. Yang, C.Y., Lukasiewicz, P., Maguire, G., Werblin, F.S. and Yazulla, S., Amacrine cells in the tiger salamander retina: morphology, physiology, and neurotransmitter identi®cation. J. Comp. Neurol., 312 (1991) 19±32.