Lucifer yellow stains displaced amacrine cells of the chicken retina during embryonic development

We have used Lucifer Yellow for histological tracing of displaced amacrine cells within the ganglion cell layer (GCL) during the embryonic development of the chicken retina. Incubating whole eyes in the dye leads to bright staining of all displaced amacrine cells, whereas ganglion cells and glial cells are not stained. A subpopulation of cells of the inner part of the inner nuclear layer (INL) are also stained (for further details see ref. 13). Kainic acid, which is known to interfere with and kill amacrine cell systems, blocks the staining of these cells fully. This in addition to histological evidence confirms that the LY-stained cells in the GCL are displaced amacrine cells. Of the cells in the GCL, 23% (+/- 3%) are of the displaced amacrine type. Further, we find that the cytoarchitectural arrangement of these cells changes significantly during development.     

The ganglion cell response to optic nerve injury in the cat: differential responses revealed by neurofibrillar staining

Summary The early responses of cat retinal ganglion cells to axotomy have been examined using neurofibrillar and Nissl-stained wholemounts. We were interested to learn whether the enhanced neurofilament expression, seen in a number of neuronal systems, was also present in different neuronal populations of the cat retina and could be used to study the distribution of these cells. We found that beta ganglion cells degenerate very rapidly after axotomy with the nuclei becoming pyknotic within a few days. Few beta cells showed increased neurofibrillar staining of the dendrites. The cell body degenerated prior to any visible degenerative changes in the axon. A proportion of the alpha and gamma ganglion cells degenerated in the first two to three weeks after axotomy. The alpha cells underwent markedly enhanced neurofibrillar staining of their dendrites prior to degeneration. The Nissl material of the cell bodies diminished as the cells degenerated but we have not observed pyknotic nuclei. The dendritic trees of some axotomised gamma cells were also revealed by the neurofibrillar stain three weeks after axotomy. These results show that retinal ganglion cells do not degenerate by a dying back process. We suggest that the rapid degeneration of the beta ganglion cell population comes about by excitotoxic cell death, a consequence of their large glutamatergic input from bipolar cells. The degenerating beta ganglion cells have the morphological appearance of cells undergoing apoptosis.     

GABA_A和GABA_C受体各亚单位基因在鲫鱼视网膜内的分布──原位杂交组织化学法研究

应用原位杂交组织化学技术，利用同位素［￣（３５）Ｓ］－ｄＡＴＰ标记的寡核苷酸探针，在鲫鱼视网膜观察了含ＧＡＢＡＡ受体α１、α３、α４、α６，β１－３，γ１－２及ＧＡＢＡＣ受体ρ１亚单位ｍＲＮＡ的神经元分布。在外核层，所有测试的亚单位均无表达；而在内核层和神经节细胞层，除α４和γ２亚单位外，均有不同程度的表达。在不同区域标记神经元的数量和标记强度各不相同，α１亚单位广泛分布在内核层的远端、中部及神经节细胞层，呈强阳性；α３亚单位相对稀少，主要分布在内核层近端和神经节细胞层，呈中等阳性；α４和α６亚单位几乎无阳性表达，仅α６亚单位在神经节细胞层呈弱阳性。β１和β２亚单位在内核层及神经节细胞层呈中等阳性；β３亚单位主要分布在内核层，在神经节细胞层标记细胞较少，呈弱阳性。γ１亚单位分布在整个内核层，在神经节细胞层有零星阳性表达。ＧＡＢＡＣ受体主要分布在内核层，ρ１亚单位主要分布在内核层的远端及中间部分，呈强阳性，而在神经节细胞层表达相对较弱。这种独特的表达型式与其功能密切相关。     

Substance P-immunoreactive neurons in the retina of two lizards.

The dendritic morphology and retinal distribution of substance P(SP)-immunoreactive neurons was determined in two Australian lizard species Pogona vitticeps and Varanus gouldii, by using immunohistochemistry on retinal wholemounts and sectioned materials. In both species, two classes of SP-immunoreactive neurons were described in the inner nuclear layer (INL) and classified as amacrine cells (types A and B). Type A amacrine cells had large somata and wide-field, bistratified dendrites branching in sublaminas 1 and 5 of the inner plexiform layer (IPL). Their morphology and retinal distribution differed between the two species. Type B amacrine cells in both species had small somata and small-field dendritic branching. A population of SP-immunoreactive neurons with classical ganglion cell morphology were identified in the ganglion cell layer (GCL). Immunostained ganglion cells occurred in larger numbers of Varanus gouldii than in Pogona vitticeps. In both species type B SP cells were the most numerous and were estimated to be about 60,000-70,000. They were distributed non-uniformly with a high density band across the horizontal meridian of the retina, from where the density decreased towards the dorsal and ventral retinal margins. In both species type A amacrine cells occurred in small numbers distributed sparsely in the peripheral retina. The faint immunostaining of SP-immunoreactive neurons in the GCL, did not allow us to reliably determine their numbers and retinal distribution. The functional significance of SP-immunoreactive amacrine and ganglion cells in the lizard retina remains to be determined.     

Characterization of green fluorescent protein-expressing retinal cells in CD 44-transgenic mice

Sensory information in the retina is transferred from rod and cone photoreceptors to higher visual centers via numerous parallel circuits that sample the photoreceptor mosaic independently. Each circuit consists of a unique combination of ganglion cell, bipolar and amacrine cell types. The morphology and physiological responses of many amacrine cells have been characterized. However, the synaptic connections and retinal circuits in which they participate are only rarely understood. A major problem that has prevented fuller characterization of retinal circuitry is the need for specific cellular markers for the more than 50 inner retinal cell types. One potential strategy for labeling cells is to use transgenic expression of a reporter gene in a specific cell type. In a recent study of cluster of differentiation 44 (CD44)-enhanced green fluorescent protein (EGFP) transgenic mice, we observed that the green fluorescent protein (GFP) was expressed in a population of amacrine and ganglion cells in the inner nuclear layer (INL) and the GCL. To characterize the morphology of the GFP-labeled cells, whole mount preparations of the retina were used for targeted iontophoretic injections of Lucifer Yellow and Neurobiotin. Furthermore, immunocytochemistry was used to characterize the antigenic properties of the cells. We found that many GFP-expressing cells were GABAergic and also expressed calretinin. In addition to the somatic staining, there was a strong GFP(+)-band located about 50-60% depth in the inner plexiform layer (IPL). Double labeling with an antibody to choline acetyltransferase (ChAT) revealed that the GFP-band was located at strata 3 inner retina. The best-labeled GFP-expressing cell type in the INL was a wide-field amacrine cell that ramified in stratum 3. The GFP-expressing cells in the GCL resemble the type B1, or possibly A2 ganglion cells. The CD44-EGFP mice should provide a valuable resource for electrophysiological and connectivity studies of amacrine cells in the mouse retina.     

Co-localization of serotonin and GABA in neurons of the Xenopus laevis retina

Serotonin-synthesizing neurons in the retina of Xenopus laevis have been identified using anti-phenylalanine hydroxylase (PH) antibody which recognizes tryptophan 5-hydroxylase, the rate-limiting enzyme for serotonin synthesis. Double-labelling experiments, using anti-PH antibody and anti-serotonin antibody/5,7-dihydroxytryptamine (5,7-DHT) uptake, have shown that some serotonin-like immunoreactive/5,7-DHT-labelled neurons exhibit PH-like immunoreactivity (PH-LI) (serotonin-synthesizing neurons), but the others do not (serotonin-accumulating neurons). In the present study, triple-labelling experiments were performed using 5,7-DHT uptake and antibodies raised against GABA and PH, to determine the possible co-localization of γ-aminobutyric acid (GABA) in serotonin-synthesizing and/or -accumulating neurons in the Xenopus retina. All 5,7-DHT-labelled bipolar cells lacked PH-LI; all of them were immunoreactive to GABA. In contrast, all 5,7-DHT-labelled large amacrine cells exhibited PH-LI, but none of them expressed GABA-LI. Small amacrine cells labelled with 5,7-DHT but not PH-LI exhibited GABA-LI, whilst the small amacrine cells with PH-LI lacked GABA-LI. These observations indicate that GABA is co-localized in serotonin-accumulating amacrine and bipolar cells, whereas serotonin-synthesizing large and small amacrine cells do not contain GABA-LI.     

Increased expression of ciliary neurotrophic factor receptor alpha mRNA in the ischemic rat retina

Using in situ hybridization, we investigated the expression of ciliary neurotrophic factor receptor ((CNTFRalpha) mRNA in the rat retina rendered ischemic by elevation of the intraocular pressure (IOP). The IOP was increased to 120 mmHg and maintained for 60 min. The rats were sacrificed on the day of reperfusion (DRP) 1, 3, 7, 14, and 28. In the normal retina, the signal for CNTFRalpha mRNA was present in retinal cells in the inner nuclear layer (INL) and in the ganglion cell layer (GCL). On DRP 1, numerous cells in the INL and GCL showed a CNTFRalpha mRNA signal. From DRP 3 onwards, CNTFRalpha mRNA appeared in photoreceptor cells located in the outer part of the outer nuclear layer. The signal in these cells increased up to DRP 14 and then decreased at DRP 28. Our findings suggest that cells expressing CNTFRalpha mRNA may resist the degenerative processes induced by ischemic insult in the rat retina.     

Co-localization of serotonin and GABA in neurons of the Xenopus laevis retina

Serotonin-synthesizing neurons in the retina of Xenopus laevis have been identified using anti-phenylalanine hydroxylase (PH) antibody which recognizes tryptophan 5-hydroxylase, the rate-limiting enzyme for serotonin synthesis. Double-labelling experiments, using anti-PH antibody and anti-serotonin antibody/5,7-dihydroxytryptamine (5,7-DHT) uptake, have shown that some serotonin-like immunoreactive/5,7-DHT-labelled neurons exhibit PH-like immunoreactivity (PH-LI) (serotonin-synthesizing neurons), but the others do not (serotonin-accumulating neurons). In the present study, triple-labelling experiments were performed using 5,7-DHT uptake and antibodies raised against GABA and PH, to determine the possible co-localization of -aminobutyric acid (GABA) in serotonin-synthesizing and/or -accumulating neurons in the Xenopus retina. All 5,7-DHT-labelled bipolar cells lacked PH-LI; all of them were immunoreactive to GABA. In contrast, all 5,7-DHT-labelled large amacrine cells exhibited PH-LI, but none of them expressed GABA-LI. Small amacrine cells labelled with 5,7-DHT but not PH-LI exhibited GABA-LI, whilst the small amacrine cells with PH-LI lacked GABA-LI. These observations indicate that GABA is co-localized in serotonin-accumulating amacrine and bipolar cells, whereas serotonin-synthesizing large and small amacrine cells do not contain GABA-LI.     

Expression of Nav1.1 in rat retinal AII amacrine cells

In retinal ganglion cells (RGCs), the expression of various types of voltage-gated sodium channel (Nav) alpha-subunits (Nav1.1, Nav1.2, Nav1.3, and Nav1.6) has been reported. Like RGCs, certain subsets of retinal amacrine cells, including AII amacrine cells, generate tetrodotoxin (TTX)-sensitive action potentials in response to light; however, the Nav subtypes expressed in these cells have not been identified. We examined the Nav subtypes expressed in rat retinal amacrine cells by in situ hybridization (ISH) using RNA probes specific for TTX-sensitive Na(v)s (Nav1.1, Nav1.2, Nav1.3, Nav1.6, and Nav1.7). Our results confirmed that Nav1.1, Nav1.2, Nav1.3, and Nav1.6 are localized in the ganglion cell layer (GCL). Interestingly, Nav1.1 was expressed not only in the GCL, but also in the inner nuclear layer (INL). The cell bodies of the Nav1.1-positive cells in the INL were located at the INL/inner plexiform layer (IPL) border. The cell bodies of AII amacrine cells are located close to the INL/IPL border, and these cells can be labeled with antibodies against parvalbumin (PV). Therefore, we combined ISH with immunohistochemistry and discovered that most of the PV-immunoreactive cells located at the INL/IPL border express Nav1.1. Our results show that AII amacrine cells express Nav1.1.     

Is Thy-1 expressed only by ganglion cells and their axons in the retina and optic nerve?

Summary The distribution of Thy-1 in the retina and optic nerve has been examined immunohistochemically, and compared to that of the astrocytic marker glial fibrillary acidic protein. The axons and cell bodies of ganglion cells were found to be Thy-1 positive as were processes within the inner plexiform layer. Transection of the optic nerve in the neonatal rat results in the rapid degeneration of the ganglion cells but some Thy-1 staining remains in the inner plexiform layer. We have estimated using an immunoassay of normal and optic nerve transected retinae that about 70% of the Thy-1 in the retina is on ganglion cells and their axons and the remainder is on cells which contribute processes to the inner plexiform layer, presumably amacrine, bipolar or Müller cells.In the optic nerve the Thy-1 was found to be limited to the fascicles of optic nerve fibres and the intrafascicular spaces, containing astrocytes and their processes, were not stained. Axotomy of the adult nerve, which produced axonal degeneration and astrocytic proliferation, led to a loss of over 95% of the Thy-1 from the nerve. We found no evidence that the astrocytes of the retina or optic nerve were Thy-1 positive in normal animals or during degeneration.     

The birthdates of GABA-immunoreactive amacrine cells in the rat retina

The birthdates of GABAergic amacrine cells in the rat retina were investigated by immunocytochemistry using anti-GABA and anti-bromodeoxyuridine (BrdU) antisera. The ratio of co-localization of GABA to BrdU increased gradually from embryonic-day 13 (E13) and showed a peak value on E18 in the central retina and on E20 in the periphery. After birth, until postnatal-day 3 (P3), a few co-localized cells were observed in the inner nuclear layer (INL). However, in the peripheral retina, co-localized cells were observed in the INL and ganglion cell layer until P5. Our results suggest that the birthdates of GABA-immunoreactive cells vary, depending on cell-type and that there is a temporal lag in the GABA-immunoreactive cell production in the peripheral retina relative to the central retina. Received: 11 January 1999 / Accepted: 20 April 1999     

Distribution of GABA immunoreactivity in kainic acid-treated rabbit retina

Ischaemic retinal cell degeneration seems to involve both NMDA and non-NMDA receptor over stimulation. However, different retinal cell types differ largely in their susceptibility to excitatory amino acid induced neurotoxicity. We have investigated the vulnerability of GABAergic cells in the rabbit retina to the non-NMDA receptor agonist kainic acid (KA). The distribution of GABA immunoreactivity (GABA-IR) was examined in the central inferior retina at different survival times (5 h–6 days) following an intra-ocular injection of 140 nmol KA and compared to that of control and untreated retinas. In the normal retina, the majority of GABA-positive cells (79%) were located in the inner nuclear layer (INL), in one to four cell rows next to the inner plexiform layer (IPL), and in one cell row next to the outer plexiform layer (OPL). The remainder (21%) were found in the ganglion cell layer (GCL). Dense immunoreactivity was seen throughout the IPL. In the OPL, stained dots and occasional immunoreactive large processes could be seen. KA-exposed retinas processed for GABA immunocytochemistry 5 and 24 h after the injection showed an 85% reduction in the number of GABA immunoreactive cells. About the same degree of depletion was seen among GABA-IR cells located at different retinal levels. However, at these survival times, immunostaining was observed in three distinct bands in the IPL, indicating that the vulnerability to KA is not uniformly distributed among all GABAergic cells. At 48 h, an additional decrease in the number of labelled cells was noted, but immunoreactive cells were still found both in the INL and GCL. Even 6 days after KA treatment, a few stained cell bodies were seen in the INL next to the IPL, as well as a few processes in the IPL. The study shows that KA receptor overstimulation induces a marked depletion of the endogenous cellular GABA pools of the central rabbit retina, most likely as a result of GABAergic cell loss. However, a small population of GABAergic cells located in the INL appears to be less vulnerable to the toxic effects of 140 nmol KA.     

Chondroitin sulfate proteoglycans and microglia prevent migration and integration of grafted Müller stem cells into degenerating retina

At present, there are severe limitations to the successful migration and integration of stem cells transplanted into the degenerated retina to restore visual function. This study investigated the potential role of chondroitin sulfate proteoglycans (CSPGs) and microglia in the migration of human Müller glia with neural stem cell characteristics following subretinal injection into the Lister hooded (LH) and Royal College of Surgeons (RCS) rat retinae. Neonate LH rat retina showed minimal baseline microglial accumulation (CD68-positive cells) that increased significantly 2 weeks after transplantation (p < .001), particularly in the ganglion cell layer (GCL) and inner plexiform layer. In contrast, nontransplanted 5-week-old RCS rat retina showed considerable baseline microglial accumulation in the outer nuclear layer (ONL) and photoreceptor outer segment debris zone (DZ) that further increased (p < .05) throughout the retina 2 weeks after transplantation. Marked deposition of the N-terminal fragment of CSPGs, as well as neurocan and versican, was observed in the DZ of 5-week-old RCS rat retinae, which contrasted with the limited expression of these proteins in the GCL of the adult and neonate LH rat retinae. Staining for CSPGs and CD68 revealed colocalization of these two molecules in cells infiltrating the ONL and DZ of the degenerating RCS rat retina. Enhanced immune suppression with oral prednisolone and intraperitoneal injections of indomethacin caused a reduction in the number of microglia but did not facilitate Müller stem cell migration. However, injection of cells with chondroitinase ABC combined with enhanced immune suppression caused a dramatic increase in the migration of Müller stem cells into all the retinal cell layers. These observations suggest that both microglia and CSPGs constitute a barrier for stem cell migration following transplantation into experimental models of retinal degeneration and that control of matrix deposition and the innate microglial response to neural retina degeneration may need to be addressed when translating cell-based therapies to treat human retinal disease.     

大鼠视网膜神经元NOS与Parvalbumin的共存研究

用NADPH脱氢酶组化及Parvalbumin免疫组化双标记技术观察了正常大鼠视网膜一氧化氮合酶(NOS)与Parvalbumin(PV)的分布,结果显示NOS阳性神经元主要位于内核层内缘带第二列,少数位于节细胞层,胞体圆形/卵圆形,直径8～12μm,细胞一侧发出突起伸向内网层1、3、5亚层,以第3亚层最为明显,PV免疫反应(PV—Ⅰ)神经元位于内核层最内缘第一列,少数位于第二列、中间部及节细胞层,胞体卵圆形,直径6～10μm,由胞体一端发出突起伸向内网层第1、5亚层.神经纤维层可见PV～Ⅰ纤维.内核层内缘第二列可见少数双标阳性细胞,在它们的PV免疫反应胞质内散布有NOS颗粒.实验结果表明 NOS阳性神经元与PV～Ⅰ神经元均为无长突细胞,分属不同的亚型,少效PV～Ⅰ神经元属节细胞,个别双标细胞可能为另一种亚型的无长突细胞,提示NOS与PV在视觉信息传递中可能存在某些联系.     

Three-Dimensional Electron Microscopy Reconstruction of Degenerative Dopaminergic Neurons Surrounded by Activated Microglia in Substantia Nigra

There is an urgent need to investigate the reason for the pathogenic mechanism of intractable central neurological diseases such as Parkinson’s disease. It has been reported that the activation of microglial cells is involved in the pathology of these diseases. However, due to technical difficulties, the relationship between degenerative neurons and activated microglial cells remains unclear. Therefore, we tried the improved analysis technique to clarify the spatial relationship between these cell types. We were able to establish an analysis technique that consists of a three-dimensional reconstruction method using serial immunoelectron micrographs after having identified both degenerative neurons and activated microglial cells under optical microscope. Using this technique, we have relatively easily been able to clarify the spatial relationship between degenerative neurons and activated microglial cells. Furthermore, using this technique it is possible to determine the neuronal degeneration process in detail, because it is able to identify structures implicated in degeneration, such as accumulation of lipofuscin in degenerated neuronal somata and phagocytotic structures of microglial cells. In future, this technical approach may be applied to elucidate the relationship between degenerative neurons and activated glial cells in human diseases.     

Ischemia-reperfusion alters the immunolocalization of glial aquaporins in rat retina

Glial cells control the retinal osmohomeostasis, in part via mediation of water fluxes through aquaporin (AQP) water channels. By using immunohistochemical staining, we investigated whether ischemia-reperfusion of the rat retina causes alterations in the distribution of AQP1 and AQP4 proteins. Transient ischemia was induced in retinas of Long–Evans rats by elevation of the intraocular pressure for 60 min. In control retinas, immunoreactive AQP1 was expressed in the outer retina and by distinct amacrine cells, and AQP4 was expressed by glial cells (Müller cells and astrocytes) predominantly in the inner retina. After ischemia, retinal glial cells in the nerve fiber/ganglion cell layers strongly expressed AQP1. The perivascular staining around the superficial vessels altered from AQP4 in control retinas to AQP1 in postischemic retinas. The data suggest that the glial cell-mediated water transport in the retina is altered after ischemia especially at the superficial vessel plexus.     

人视网膜星形胶质细胞发育及与血管前体细胞的关系

目的：研究人视网星菜胶质细胞的发育及与血管前体细胞的关系。材料和方法：收集１３４例发育各期胎儿视网膜和４例成人视网膜、石蜡包埋切片整装铺片，四种抗体免疫组分染色，光镜观察，结果：星形胶质细胞分为三种：（１）Ｓ－１００（＋）／胶质纤维酸性蛋白（ＧＦＡＰ）（＋）的双极形星形胶质细胞，视盘进入视网膜，与跟随其后的纤连蛋白（Ｆｎ）（＋）的血管前体细胞接触并相伴向锯齿缘迁移，足月后和成人此类星形胶质细胞主要     

N-methyl-D-aspartate-induced excitotoxicity in adult rat retina is antagonized by single systemic injection of MK-801

Intravitreal injection of N-methyl-D-aspartate (NMDA) produced a substantial damage to the adult rat retina that was largely restricted to inner retinal layers, including the ganglion cell layer (GCL), inner nuclear layer (INL), inner, and outer plexiform layers. This retinal damage was significantly reduced by a systemic injection of a low dose of MK-801 (0.5 mg/kg), a potent NMDA-receptor antagonist. This neuroprotection was dose dependent and was most effective when the antagonist was given 1 h before NMDA insult. An intraperitoneal injection of 0.5 mg/kg MK-801 provided a virtually complete protection to the retina to the NMDA-induced toxicity, as indicated quantitatively by the number of DiI-filled retinal ganglion cells, the number of cells in the GCL and INL that undergo DNA fragmentation, and the edematous changes in retinal thickness. A post-lesion administration of MK-801 was still able to provide an effective neuroprotective effect to the retina, but this protection was lost when MK-801 was given 4 h after NMDA exposure. The current results indicate a therapeutic potential of systemic application of MK-801 in protecting the adult rat retina from neurologic disorders related to excessive activation of NMDA receptors.     

Dopamine and serotonin in cat retina: electroretinography and histology

Summary Anatomical structures of the cat retina were related to functional changes induced by the application of dopaminergic and serotonergic substances. We report on the contribution of dopaminergic and serotonin accumulating retinal neurones to retinal activity as reflected by the electroretinogram. The effect of dopaminergic neurones was investigated by the application of the neurotoxin 6-hydroxydopamine (6-OHDA) which is known to destroy dopaminergic neurones, and injections of either serotonin (5-HT) or the analogue 5,6-dihydroxytryptamine (5,6-DHT) were used to monitor the effects of indoleamines. In control experiments aminophosphonobutyric acid (APB), an agonist of glutamate transmission, was injected. Conventional immunohistochemical methods identified dopaminergic and serotonin accumulating neurones, and the electrophysiological data obtained from the same animals were related to the anatomical structures influenced by the respective substances. Destruction of dopaminergic amacrine cells by 6-OHDA increased the ERG b-wave amplitude. Accumulation of indoleamines by certain amacrine cells also caused an increase of the ERG b-wave. However, intravitreal injection of APB completely blocked the bwave. The data show that ERG mass responses can be used to monitor transmitter-specific effects on retinal circuitry.     

The generation and changing retinal distribution of displaced amacrine cells in Bufo marinus from metamorphosis to adult.

Summary The generation and retinal distribution of displaced amacrine cells (DAs) were studied from metamorphosis to adult in the cane toad Bufo marinus. Displaced amacrine cells were identified by inducing chromatolytic changes in ganglion cells (GCs) following optic nerve section. Cells that did not chromatolyse in the ganglion cell layer (GCL) of the retina were regarded as DAs. The number of DAs increased considerably from an estimated 10000 at metamorphosis to 211000 in the adult toad, and was accompanied by a substantial decrease of average cell density. In contrast to the reported 6:1 cell density gradient of all cells of the GCL in adult toad (Nguyen and Straznicky 1989) only a shallow 1.6:1 density gradient of DAs from the visual streak to the dorsal and ventral retinal margins was detected. Consequently, the incidence of DAs increased from 15% of all cells of the GCL in the visual streak to 30% in the dorsal and ventral peripheral retina. These results indicate that the ratio of the newly generated DAs and GCs at the ciliary margin must be changing during development. More GCs are generated before and around metamorphosis then DAs, in contrast to the relative increase of the percentage of DAs generated after metamorphosis. The possible control of the numbers of DAs in the GCL is discussed.