Ординатура / Офтальмология / Английские материалы / The Retina and its Disorders_Besharse, Bok_2011

Figure 8 Section showing part of a vasoproliferative tumor (reactive retinal glioangiosis) in an eye removed for complications of retinal detachment. The lesion contains scattered RPE cells (arrowheads), revealed (red) by immunohistochemical staining for cytokeratin 7 (hematoxylin counterstain). A few of these cells contain melanin pigment (arrows). Staining of the lesion for glial cells (inset: immunohistochemical staining red, hematoxylin counterstain) confirms that the retinal tissue is gliotic and disorganized. Scale bar: 200mm

retina. Methods employed during the primary retinal detachment surgery may also increase the risk of PVR. Thus, there is evidence that cryopexy, choice of tamponade agent, and vitrectomy impact on the risk of developing PVR.

Clinical Classification of Proliferative

Vitreoretinopathy

There are several attempts to classify PVR according to the clinical features of the retinal detachment. These classifications are descriptive and help surgeons communicate, especially with regard to surgical approaches in the management of the condition. Thus, they are useful for surgical planning and are not based on pathobiology. PVR classifications are also not prognostic. They do not correlate well with visual prognosis or anatomical success with treatment. The classification is also not related to the stages of the disease. It is just the clinical picture at one snapshot in time. Despite their many shortcomings, clinical PVR classifications are widely used in clinical trials and for clinicopathological correlates.

Several classifications have been suggested, based on the clinical manifestations of the disease. For the most part, there is commonality between these systems with regard to the earlier or milder stages of the disease, whereas the schemes tend to diverge in their classification of later or more advanced stages of PVR.

Within these systems, stage A (minimal) is usually regarded as the presence of vitreous haze and pigment

clumps, whereas stage B (moderate) is typically recognized as wrinkling of the retinal surface, decreased vitreous mobility, and increased retinal stiffness.

With respect to the more severe or later stages of PVR, some schemes separate the advanced stages by the number of quadrants of the retina involved. Thus, for example, the Retina Society Terminology Committee classification of 1983 associates stage C (marked) disease with fixed retinal folding and adds a number to reflect the number of quadrants involved (e.g., C-3 is fixed folds involving three quadrants of retina). In this system, stage D (massive) reflects the involvement of all four retinal quadrants. In addition, stage D is graded 1–3 depending on how extensive the folding of the retina is (D-1 being an open funnel of totally detached retina, D-3 being a closed funnel so that the optic nerve head cannot be seen: Figure 3).

Other schemes classify the more severe stages of PVR into anterior and posterior groups, according to the location of the disease with reference to the retinal equator. In these systems, the number of retinal quadrants involved by the disease is again used so that PVR involving fixed folds in, say, two quadrants of retina posterior to the equator would be classed P2 or CP2 (stage D is generally discarded in these schemes). The schemes employing anterior and posterior also add a contraction type, depending on the extent of epiretinal membrane (focal or diffuse), the presence of subretinal membrane, or posterior hyaloid/vitreous base proliferation.

Management

There is much interest in preventing proliferation of membranes after retinal detachment surgery by treating high-risk patients with combinations of agents. One such combination is the antiproliferative drug 5-fluorouracil and low-molecular-weight heparin (which binds growth factors). This combination has been shown to reduce the incidence of PVR in high-risk retinal detachment patients undergoing vitrectomy.

Once established, PVR membranes are removed by microsurgery so that the retina can be reattached. Again, there is much interest in the use of pharmacological measures to stop membrane recurrence after PVR surgery. For example, there is evidence that daunomycin used preoperatively can reduce the requirement for repeat surgery in these patients.

Tamponade agents to maintain retinal attachment are frequently employed during and after surgery for retinal detachment and PVR. These agents incorporate gases (e.g., air) and liquids (e.g., silicone oil). Liquids have a tendency to emulsify in the eye, particularly if they are of low viscosity. The result is the formation of droplets of various sizes in the vitreous cavity or even elsewhere in the eye (such as in the aqueous if tamponade gains access

to the anterior chamber: Figure 9). Thus, emulsification of tamponade agent may impact on the pathology of PVR membranes. It appears that the droplets can stimulate a foreign body-type reaction and attract macrophages to the retinal surface (Figure 9). New membranes may develop and these tissues characteristically have the microscopic appearances of PVR membranes containing granulomata to emulsified oil.

Outcomes

Successful treatment in PVR has often been measured in terms of final retinal reattachment rates. The assumption is that effective anti-PVR therapy would lead to an increased rate of successful reattachment. This assumption may or may not hold true. It is often missed or untreated holes that lead to retinal redetachment and not necessarily PVR, which may or may not be controlled by the anti-PVR drugs. Indeed, a totally ineffective antiPVR treatment may be compatible with anatomical success so long as the epiretinal membranes do not act on the retina to produce another retinal break or cause tractional retinal detachment. Hence, anatomical success rate is a poor proxy for PVR control. Another important point is that anatomical success does not guarantee visual recovery. The advancement of surgery in the last few

years has greatly increased the final anatomical success rate. Disappointingly, this success rate has not been translated into visual improvement. In retinal detachment patients, the fellow eye is also likely to be involved in sight-threatening pathology so that many PVR patients end up with visual impairment in both eyes.

Conclusions

Despite intense research over the last 25 years that has improved our understanding of the condition, PVR remains the major cause of failure after retinal detachment surgery. Nevertheless, continuing advances in both surgical and pharmacological manipulation of the disease, based on an expanding knowledge of PVR pathobiology, can be expected to reduce the impact of the disease in the future.

See also: Rhegmatogenous Retinal Detachment.

Further Reading

Asaria, R. H., Kon, C. H., Bunce, C., et al. (2001). Adjuvant 5-fluorouracil and heparin prevents proliferative vitreoretinopathy: Results from a randomized, double-blind, controlled clinical trial. Ophthalmology 108: 1179–1183.

Charteris, D. G. (1995). Proliferative vitreoretinopathy: Pathobiology, surgical management, and adjunctive treatment. British Journal of Ophthalmology 79: 953–960.

Colthurst, M., Williams, R. L., Hiscott, P., and Grierson, I. (2000). Biomaterials used in the posterior segment of the eye. Biomaterials 21: 649–665.

Fisher, S. K., Lewis, G. P., Linberg, K. A., and Verardo, M. R. (2005). Cellular remodeling in mammalian retina: Results from studies of experimental retinal detachment. Progress in Retinal and Eye Research 24: 395–431.

Heimann, K. and Wiedemann, P. (1989). Proliferative Vitreoretinopathy. Heidelberg: Kaden.

Hiscott, P. and Mudhar, H. (2008). Is vasoproliferative tumour (reactive retinal glioangiosis) part of the spectrum of proliferative vitreoretinopathy? Eye 23: 1851–1858.

Hiscott, P. and Sheridan, C. (1998). The retinal pigment epithelium, epiretinal membranes and proliferative vitreoretinopathy. In: Marmor, M. F. and Wolfensberger, T. J. (eds.) Retinal Pigment Epithelium – Function and Disease, pp. 478–491. New York: Oxford University Press.

Hiscott, P., Morino, I., Alexander, R., Grierson, I., and Gregor, Z. (1989). Cellular components of subretinal membranes in proliferative vitreoretinopathy. Eye 3: 606–610.

Hiscott, P., Sheridan, C., Magee, R., and Grierson, I. (1999). Matrix and the retinal pigment epithelium in proliferative retinal disease. Progress in Retinal and Eye Research 18: 167–190.

Kampik, A., Kenyon, K. R., Michels, R. G., Green, W. R., and de la Cruz, Z. C. (1981). Epiretinal and vitreous membranes. Comparative study of 56 cases. Archives of Ophthalmology 99: 1445–1454.

Kirchhof, B. and Wong, D. (eds.) (2005) Vitreo-Retinal Surgery. Essentials in Ophthalmology. Berlin: Springer.

Machemer, R. and Laqua, H. (1975). Pigment epithelium proliferation in retinal detachment (massive periretinal proliferation). American Journal of Ophthalmology 80: 1–23.

Pastor, J. C., de la Ru´a, E. R., and Martı´n, F. (2002). Proliferative vitreoretinopathy: Risk factors and pathobiology. Progress in Retinal and Eye Research 21: 127–144.

The Retina Society Terminology Committee (1983). The classification of retinal detachment with proliferative vitreoretinopathy. Ophthalmology 90: 121–125.

Wiedemann, P., Hilgers, R. D., Bauer, P., and Heimann, K. (1998). Adjunctive daunorubicin in the treatment of proliferative vitreoretinopathy: Results of a multicenter clinical trial. Daunomycin study group. American Journal of Ophthalmology 126: 550–559.

Relevant Website

http://www.youtube.com – Number of videos concerning PVR and its management.

Glossary

Age-related macula degeneration (AMD) – It comprises a variety of diseases, mainly of the elderly, that involves loss of vision in the central region of the retina.

Endocannabinoids – Natural chemicals in the body that are mimicked by the active component of marijuana.

Monoamine oxidase (MAO) – An enzyme that degrades dopamine.

Snellen acuity – A test of visual acuity that uses a standard sized ‘‘E’’ in four orientations.

Transient receptor potential type vanilloid

1 receptor (TRPV1) – An ionotropric receptor that increases intracellular calcium either by entry through the plasma membrane or from intracellular stores. It is activated by noxious heat, capsaicin, and the endocannabinoid, anandamide.

Vernier acuity – A test of visual acuity that tests the ability to detect displacement of two lines end-to-end.

Marijuana and the Endocannabinoids

The active component of the marijuana plant Cannabis sativa, D9-tetrahydrocannabinol (THC), mimics endogenous chemicals, endocannabinoids (eCBs) that activate membrane receptors. eCBs include a variety of amide, ester, and ether derivatives of arachidonic acid. The most widely studied of these are arachidonoyl ethanolamide (anandamide, AEA) and sn-2 arachidonoyl glycerol (2-AG) (Figure 1). Other eCBs have been identified with varying degrees, or no affinity for cannabinoid receptors, and also compete with AEA and 2-AG for metabolizing enzymes. In this way, they modulate activity by competition at the receptors or by affecting substrate availability for metabolism.

Synthesis and Release

Unlike water-soluble transmitters, AEA and 2-AG are lipophilic and not stored in synaptic vesicles. Rather, membrane phospholipids are metabolized on demand to liberate AEA and 2-AG by calcium-dependent phospholipases. The precursor of AEA is N-arachidonyolphosphatidyl ethanolamine (NAPE), formed by calcium-dependent transfer of

arachidonic acid (AA) from arachidonoylphosphatidylcholine to phosphatidylethanolamine (PE). There are multiple pathways for AEA liberation from the membrane. First, NAPE is hydrolyzed by phospholipase D (PLD) to release AEA and phosphatidic acid. Second, NAPE is hydrolyzed to N-acyl-lyso-PE by phospholipase A1/A2; then, AEA is released by lysophospholipase D. Third, phospholipase C (PLC) cleaves NAPE to generate phosphoanandamide, which is dephosphorylated to liberate AEA. The PLC pathway may be involved in the on-demand synthesis of AEA rather than in maintaining basal tissue levels of AEA. The primary pathway for 2-AG synthesis involves hydrolysis of diacylglycerols (DAG) by DAG lipase isozymes, DAGLa and DAGLb. DAGs may be produced by the PLC b-catalyzed hydrolysis of phophotidylinositol or hydrolysis of phosphatidic acid by a phosphohydrolase. AEA and 2-AG freely diffuse within the membrane where they interact with the active sites of degradative enzymes and receptors. AEA binds reversibly to serum albumin, and it is likely that such binding is critical for the movement of AEA and 2-AG in blood, the extracellular matrix, and the cytoplasm. The presence and localization of AEA and 2-AG are inferred from the distribution of receptors, synthesizing and inactivating enzymes as well as physiological effects on identified cells.

Inactivation

AEA and 2-AG are inactivated following intracellular accumulation by fatty acid amide hydrolase (FAAH), monoacylglycerol lipase (MGL), cyclooxygenase-2 (COX-2), and lipoxygenase (LOX). AEA and 2-AG are hydrolyzed by FAAH into AA and ethanolamine or glycerol, respectively. 2-AG, but not AEA, is hydrolyzed by MGL. Following hydrolysis of AEA or 2-AG, AA is incorporated into membrane phospholipids. COX-2 oxidizes arachidonic acid, AEA, and 2-AG to prostamides or prostaglandin glyceryl esters, leading to prostaglandins. In addition, oxidation of AA by LOX produces 12-(S)-hydroperoxyei- cosatetraenoic acid (15-(S)-HPETE), 5-(S)-HETE, and leukotriene B4, all of which are agonists of TRPV1 receptors (Figure 2).The effects of AEA and 2-AG are modulated by the balance of metabolic enzymes that is specific to each cell type.

Receptors

Effects of cannabinoids are mediated by metabotropic (G-protein-coupled receptors (GPCRs)) and ionotropic

FAAH

AA

Membrane phospholipids

Syn

HPETE

HETE

TRPV1

Isomerizes to

PGH2

PGE2 + PGD2

(ion channel) receptors (Figure 2). In general, activation of cannabinoid 1 receptors (CB1Rs), via heterotrimeric guanosine-5’-triphosphate (GTP)-binding proteins Gi/o (Gi/o), modulates voltage-gated Kþ and Ca2þ conductances, resulting in a reduction of neurotransmitter release, particularly g-aminobutyric acid (GABA) and glutamate. CB2 receptors, which also signal through Gi/o, are expressed in cells of the immune system and the central nervous system (CNS), particularly in astrocytes. There is evidence for additional cannabinoid receptors, perhaps GPCR 55. AEA, but not 2-AG, activates the ionotropic transient receptor potential type vanilloid 1 receptor (TRPV1) that increases intracellular calcium

either by entry through the plasma membrane or from intracellular stores. Prostamides and prostaglandin glycerol esters, produced by eCB oxidation by COX-2, bind to a variety of prostaglandin receptors. eCBs are ligands for peroxisome proliferator-activated receptors (PPARs), members of the nuclear receptor superfamily that are involved in lipid metabolism, insulin sensitivity, regulation of inflammation, and cell proliferation.

Distribution and Function

CB1Rs are the most numerous GPCRs in the brain. eCBs, their receptors, and metabolizing enzymes are enriched in

brain regions associated with the physiological and psychomotor effects of cannabis. AEA and 2-AG have shortand long-term effects on synaptic plasticity and neuroprotection. The effects depend largely on retrograde transmission in which postsynaptic dendrites release an eCB that binds to presynaptic CB1Rs to reduce transmitter release. Retrograde release of eCBs is evoked by two mechanisms. In a voltage-dependent mechanism, depolarization of postsynaptic dendrites by L-glutamate opens voltage-gated calcium channels. The increase in intracellular Ca2þ activates Ca-dependent PLD to release an eCB. A second mechanism involves activation of heterotrimeric GTP-binding protein Gq/11 (Gq/11) coupled metabotropic receptors, usually group I metabotropic glutamate receptors (mGluRs), mGluR1 and mGluR5, and muscarinic receptors (M1 and M3). By enzymatic cascades that may or may not release calcium from intracellular stores, eCBs are released from the plasma membrane. The eCB-induced reduction of presynaptic glutamate and GABA release contributes to synaptic plasticity, while the reduction of glutamate release inhibits excitotoxicity following ischemia. Evidence implicates 2-AG more so than AEA in plasticity, while both AEA and 2-AG are involved in neuroprotection.

Cannabinoids and Ocular Tissues

Marijuana induces conjunctival vasodilation and reduces intraocular pressure (IOP), but is not a mydriatic. These effects are mediated locally by eCBs as demonstrated in the ciliary body, iris, choroid, and trabecular meshwork in mammalian tissues. THC, as low as 10–12 M, increases monoamine oxidase (MAO) activity in the bovine trabecular meshwork, choroid, and ciliary processes but not in the iris. Hydrolysis of anandamide has been measured in the porcine iris, choroid, lacrimal gland, and optic nerve. CB1 mRNA and CB1R-immunoreactivity (IR) have been detected in the ciliary body, trabecular meshwork, and conjunctival epithelium of rat, mouse, bovine, and human. AEA and 2-AG have been measured by gas chromatography in human ocular tissues. The content of eCBs varies in certain disease states, suggesting the importance of eCBs in maintaining ocular homeostasis. For example, 2-AG levels are lower in the ciliary body of patients with glaucoma. However, in diabetic retinopathy there are higher levels of 2-AG only in the iris, and increased levels of AEA in the retina, ciliary body, and cornea. Eyes of patients with age-related macula degeneration (AMD) also show increases of AEA in the retina, choroid, ciliary body, and cornea. Topically applied AEA reduces IOP by activation of CB1R and activation of the prostaglandin E 2 receptor (EP2R) after conversion of AEA to prostamides (see Figure 2). Administration of either AEA or THC to human nonpigmented epithelium (NPE) cells

induces COX-2 expression, indicating a relationship among prostaglandins, COX-2, and eCBs in lowering IOP. In addition, EP2 receptors have been localized in the NPE of mouse, porcine, and human ciliary body.

Cannabinoids – Retinal Anatomy

Early studies of the effects of cannabis on vision were performed in concert with the effects of alcohol in order to examine the influence on visual motor behaviors as they related to driving. Anecdotal reports also came from studies citing side effects of cannabis when used as an analgesic. Effects on vision are subtle and include blurred and double vision, a reduction in vernier and Snellen acuity, alterations in color discrimination, an increase in photosensitivity and an increase in recovery from foveal glare. It is unlikely that all of these effects of marijuana are due to cortical or preretinal sites because processes of light–dark adaptation take place in the retina. Knockout mice that lack CB1Rs or FAAH are not blind, but the effects on vision have not been studied.

Biochemical Assay

The first evidence for cannabinoids in the retina was the demonstration that THC induced an increase in MAO activity, indicating a role in dopaminergic transmission. Later, FAAH-mediated hydrolysis of 3H-AEA was shown in homogenates of porcine, bovine, and goldfish retinas. AEA and 2-AG were detected in mammalian retina by gas chromatography. Release of AEA from bovine retinal extracts in a physiological buffer demonstrated that the extracts contained the metabolic machinery necessary for eCB release, the precursor NAPE, and PLD.

Localization – Cannabinoid Receptors

CB1Rs have been localized by immunohistochemistry in the retinas of numerous species, including human, monkey, mouse, rat, chick, salamander, and goldfish. Despite differences in detail, there is a common theme. In general, the most prominent label is in cells of the through pathway: photoreceptors, bipolar cells, and ganglion cells. Cone pedicles in all species contain CB1Rs. Rod spherules appear to be labeled in all species except goldfish. Ultrastructural analysis has been performed exclusively on goldfish cones. CB1R-IR is on plasma membrane at the perimeter of the pedicle as well as within the invagination. CB1R-IR is not immediately apposed to the synaptic ribbon, but is at some distance from it. Regarding bipolar cells in mammals, CB1R-IR is restricted to rod bipolar cells as confirmed by double labeling with antisera against PKC. In goldfish, there is a higher proportion ( 3:1) of CB1R-IR in ON bipolar cells compared to OFF bipolar

cells. This difference holds for mixed rod–cone bipolar cells as well as for cone bipolar cells. CB1R-IR, on the bipolar cell synaptic terminal membrane, is not adjacent to the synaptic ribbons. Rather, the CB1R-IR is always some distance removed from the ribbon, the same as observed for the cone pedicles.

Regarding rat horizontal cells, CB1R-IR is confined to the cell bodies and is not present on the dendrites, unlike bipolar cells. CB1R-IR is also found on a population of large amacrine cells, identified in rat as a rare type that is immunoreactive for PKC and GABA. In goldfish, CB1RIR is on presynaptic membrane of amacrine cell boutons. These boutons appear throughout the depth of the inner plexiform layer and are presynaptic to bipolar cell terminals and small processes derived from ganglion cells. It is likely that these CB1R-immunoreactive processes are from a single type of diffuse amacrine cell.

CB1R-IR is on Mu¨ller’s cells in goldfish but not in any other preparation. There are inconsistent reports of CB1R-IR in mammalian astrocytes, microglia, and oligodendrocytes. Activation of CB1Rs inhibits excitatory amino acid transport and induces glutamate release from astrocytes in the mammalian brain. CB1R and CB2R are involved in gliotic responses to injury. The interaction of eCBs and glia has not been investigated in the retina. CB2 mRNA was described in all cellular layers of the rat retina; this could include glial labeling, particularly Mu¨ller’s cells.

Localization – Metabolizing Enzymes

There is relatively little information regarding the distribution of eCB metabolizing enzymes in the retina. The distribution of FAAH-IR in the rat and mouse is quite different from that in the fish. FAAH-IR, in rat and mouse, is most prominent in medium size and large ganglion cells, while weaker FAAH-IR is observed in the soma of horizontal cells, large dopaminergic amacrine cells, dendrites of starburst amacrine cells, and Mu¨ller’s cells. FAAH-immunoreactive bipolar cells in rat and mouse are exclusively cone bipolar cells, in contrast to CB1RIR that is exclusively in rod bipolar cells. In goldfish, FAAH-IR is present over cone photoreceptors, Mu¨ller’s cells, and some amacrine cells, not ganglion cells as in mouse and rat. The distribution of FAAH-IR as it relates to FAAH activity was studied in goldfish retina. 3H-AEA is hydrolyzed by FAAH with 3H-AA rapidly incorporated into membrane phospholipids. Silver-grain deposition represents the trapping of 3H-arachidonic acid in the plasma membrane. FAAH-IR and specific 3H-AEA uptake showed the same pattern over cone photoreceptors, Mu¨ller’s cells, and some amacrine cells. The codistribution of FAAH-IR and 3H-AEA uptake indicates that the bulk clearance of AEA from the extracellular

space in the retina occurs as a consequence of a concentration gradient across the plasma membrane created by FAAH activity.

AEA is a ligand for the TRPV1 receptor whose binding site is on an intracellular domain. As FAAH and TRPV1 are integral membrane proteins of the endoplasmic reticulum and plasma membrane, respectively, FAAH activity may regulate the levels of AEA for TRPV1 activation. Also, following the hydrolysis of AEA by FAAH, LOX metabolites of AA could activate TRPV1. AEA then could act as an intracellular mediator by being produced from and/or degraded by the same neurons that express TRPV1 receptors. Supporting anatomical evidence for this scheme was provided first in goldfish in which co-localization of TRPV1-IR with FAAH-IR occurs in three types of amacrine cells, two of which are GABAergic. These cells ramify in interplexiform layer sublaminae a and b, indicating a general function in the OFF, ON, and ON/OFF pathways. The role would depend on the downstream cascade following the increase in calcium concentration.

MGL-IR co-localizes with FAAH-IR over medium size and large ganglion cells and with CB1R-IR in all rod bipolar cells in rat and mouse. In rat retina, COX- 2-IR is constitutive in horizontal, amacrine, and ganglion cells. Following transient ischemia, COX-2 is upregulated in these cell types and induced in Mu¨ller’s cells. This pattern in rat differs from mouse in which COX-2 is restricted to bipolar cell bodies and their axons. COX-2 bipolar cells are a mixed group, with 65% rod bipolar cells and 35% cone bipolar cells. Also rod bipolar cells are of two types: 68% contain COX-2-IR and 32% do not.

The cannabinergic system in vertebrate retinas, as indicated by CB1R-IR, FAAH-IR, COX-2-IR, and MGL-IR, is concentrated in the through pathway of photoreceptors, bipolar cells, and ganglion cells (Table 1).These cells for the most part use L-glutamate as their neurotransmitter. Cells of the inhibitory lateral pathways, horizontal cells and amacrine cells, do not feature as prominently. Exceptions are some horizontal cells, dopaminergic amacrine cells, and cholinergic starburst amacrine cells that label weakly for FAAH-IR. Bipolar cells tend to be of the on type and differentiated by rod and cone input. CB1R-IR and MGL-IR are restricted to rod bipolar cells, FAAH-IR to cone bipolar cells, and COX-2-IR to subtypes of rod and cone bipolar cells.

Cannabinoids – Retinal Physiology

Effects on Transmitter Release

Stimulation of CB1Rs via Gi/o reduces voltageand Ca2þ-evoked release of [3H]-noradrenaline and [3H]- dopamine in guinea pig retina. Agonists of CB1Rs, but not CB2Rs, inhibit Kþ- and ischemia-evoked [3H]

D-aspartate release from isolated bovine retina. Uptake of [3H] D-aspartate identifies high-affinity uptake sites for L-glutamate and L-aspartate in photoreceptors, a small percentage of ganglion cells and Mu¨ller’s cells. The rank order of potency for the CB1 agonists differs for Kþ- and ischemia-evoked release. As photoreceptors are more resistant to ischemia than ganglion cells, the difference in the rank order could reflect the relative potencies of the agonists on CB1Rs on these cell types.

Effects on Ganglion cells

CB1R-mediated activity was demonstrated in rat retinal ganglion cells by [35S]GTPgS autoradiography and reverse transcription polymerase chain reaction (RT-PCR). Voltageactivated Ca2þ currents in cultured rat ganglion cells are suppressed by cannabinoid agonist, WIN 55,212-2, an effect that is blocked by CB1 antagonists, SR141716A and AM281. The presence of CB1R function on rat retinal ganglion cells appears unusual in that CB1Rs tend to be at presynaptic boutons. One possibility is that CB1Rs are present on associational ganglion cells, whose axons and axon collaterals do not leave the retina. Rat and mouse ganglion cells also contain FAAH and MGL, putting them in position to regulate AEA and 2-AG as potential retrograde transmitters for suppression of bipolar cell and amacrine cell activity.

Effects on Bipolar Cells

CB1-mediated inhibition of L-type calcium (Ica) and delayed rectifier (IK(V)) currents has been reported for ON-bipolar cells of salamander and goldfish. As yet there are no data on OFF-bipolar cells. The voltageactivation range of the currents is not altered, but simply scaled down over the entire activation range. Goldfish mixed rod–cone (Mb) bipolar cells also have D1 dopamine

receptors that enhance ICa and IK(V) via G protein Gs. CB1R agonists and dopamine oppose each other to modulate IK(V) of Mb bipolar cells. Co-application of WIN 55,212-2 (0.1–0.25 mM) reversibly blocks the enhancement induced by 10 mM dopamine even though low concentrations of WIN 55,212-2 have no effect when applied alone. The effects of dopamine and cannabinoid agonists on IK(V) occur within the physiological range of Mb bipolar cell function ( –25 to 0 mV). IK(V) would be activated during the on portion of the response and, as a counter current, would modulate the peak:plateau ratio of the response. CB1R activation should make the Mb bipolar cell on response more tonic by suppressing the hyperpolarizing effect of IK(V), whereas D1 receptor activation should make the on response more phasic by enhancing IK(V). The effect on ganglion cells should be relatively tonic responses in scotopic (dark-adapted) conditions and relatively phasic responses in photopic (light-adapted) conditions. CB1-induced suppression of calcium currents should reduce transmitter release and reset sensitivity to further increments.

Cannabinoids and Photoreceptors

Voltage-Gated Currents

CB1-mediated modulation of photoreceptor membrane currents has only been reported for tiger salamander and goldfish. The voltage-activation ranges of these currents are not affected. Salamander rods and cones responded differently to WIN 55,212-2. IK is suppressed in single cones and rods, whereas ICa is suppressed in cones but enhanced in rods. The differential effect on ICa and IK in rods would increase transmitter release, resulting in the reduction of sensitivity, which is an apparent counteradaptive effect. Goldfish cones show a biphasic response

to WIN 55,212-2: an enhancement of IK, ICl, and ICa via Gs at concentrations <1 mM, and suppression via Gi/o at concentrations >1 mM. The data obtained with retrograde suppression, to be described below, suggest that the enhancement produced by WIN 55,212-2 may be due to agonist-specific trafficking in which binding of agonists to CB1Rs favors coupling to different G proteins. For example, WIN 55,212-2 increases intracellular calcium by Gq/11 coupling in human trabecular meshwork cells, while other CB1 agonists, including THC, 2-AG, CP55940, and methanandamide couple to Gi/o but not to Gq/11.

The caution is that the data obtained with WIN 55,212-2 may or may not apply to other agonists or the eCBs.

An effect of evoked released of eCBs in the retina was demonstrated in goldfish. Retrograde suppression of membrane currents in goldfish cones in a retinal slice was achieved by applying a puff of saline with 70 mM KCl or an mGluR1 agonist, RS-3,5-dihydrophyenylglycine (DHPG), through a pipette at an Mb bipolar cell body while recording IK(V) from cone inner segments under whole-cell voltage clamp (Figure 3(a)). Retrograde inhibition of IK(V) was reversible and stable over several hours

IK(V)

OPL

K+puff (b)

50 nA

50 ms

Control #1

Control #2

K+ #2

K+ #1

100 nA

10 ms

(Figures 3(b) and 3(c)). It had a latency of about 200 ms after a Kþ puff, was reduced on average by 25%, and had a halftime of 3.4 min to recover (Figure 3(d)). Retrograde suppression of IK(V) was unaffected by a combination of the GABA receptor antagonist, picrotoxin, and a-amino-3-hydroxyl-5-methyl- 4-isoxazole-propionate (AMPA) glutamate receptor antagonist, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), but blocked completely by the CB1 antagonist, SR141716A, indicating mediation by CB1 receptors. Experiments with the FAAH inhibitor (URB597), a COX-2 inhibitor (nimesulide), and a blocker of 2-AG synthesis (Orlistat) indicated that 2-AG, rather than AEA, is the retrograde eCB.

Two conditions evoke 2-AG release from Mb bipolar cells, strong depolarization and activation of mGluR1, corresponding to voltage-dependent and voltageindependent mechanisms. Rods and cones release glutamate at a steady rate under any ambient illumination; this rate is increased by decrements of light intensity and

decreased by increments in light intensity. Voltagedependent release of 2-AG would occur following depolarization of the Mb bipolar cell in response to a light flash. The retrograde suppression of glutamate release from cones would be a positive feedback that would amplify the reduction in cone transmitter release initially caused by increasing light intensity. This may not be physiologically relevant because the halftime to recover from a 50-ms stimulus is several minutes. The long halflife of the suppressive effect should make this mechanism insensitive to rapid changes in intensity.

The voltage-independent mechanism that provides negative feedback on glutamate release may be more functional. Hypothetically, glutamate, released during ambient illumination, stimulates mGluR1a on Mb bipolar cells tonically to maintain a steady release of 2-AG via a Gq/11 mechanism. The degree of feedback inhibition of glutamate release from cones varies inversely with ambient illumination; the

100 pA