From the Lab to the Clinic: Activation of an Alpha-2 agonist Pathway is Neuroprotective in Models of Retinal and Optic Nerve Injury

PURPOSE

The selective alpha-2 agonist brimonidine was used as a pharmacological probe to activate alpha 2 receptor-mediated neuroprotective signaling pathways and quantitate the enhancement of retinal ganglion cell survival and function in animals with ischemic retinal and optic nerve injury.

METHODS

Two animal models were used to achieve different methods of neuronal insult. The first model involved mechanical injury of the rat optic nerve after treatment with a single intraperitoneal (i.p.) dose of brimonidine or a control vehicle. The second model involving acute retinal ischemic/reperfusion injury was used in a variety of experiments in which rats were treated with either intraperitoneal brimonidine or single dose topical brimonidine at various strengths. In all cases retinal ischemia was induced and maintained followed by reperfusion. In some cases, TUNEL staining was performed on histologic sections of the retinas of rats that had been sacrificed after 24 hours. To examine the activation of neuronal survival pathways at the molecular level, rats were injected with i.p. brimonidine followed by the isolation of mRNAs from whole retinas 24 hours after ischemic injury.

RESULTS

Intraperitoneal brimonidine enhanced rat RGC survival and function in the partial crush injury model, and neuroprotection was dose-dependent. Topical application of brimonidine 1 hour before injury was effective in decreasing ischemic retinal injury. Ischemic retinas treated with brimonidine resulted with a large decrease in TUNEL staining.

CONCLUSIONS

Treatment with the alpha 2 adrenoreceptor agonist brimonidine was found to confer neuroprotection to retinal ganglion cells in two distinct models of neuronal injury resulting from acute retinal ischemia/reperfusion and calibrated optic nerve compression.

Electrophysiological and Histological Measures of Retinal Injury in Chronic Ocular Hypertensive Monkeys

PURPOSE

This study was performed to determine whether components of the standard ERG (electroretinogram), multifocal ERG, and flash VECP (visually-evoked cortical potential) response might provide a sensitive measure of retinal ganglion cell injury in a monkey model for chronic ocular hypertension.

METHODS

Argon laser treatment of the aqueous outflow tissue was used to induce chronic elevation of intraocular pressure (IOP) in the right eye of 18 young adult cynomolgous monkeys. At 15 months post- IOP elevation, standard methods were used to record ERG and VECP responses. Multifocal ERG responses were also recorded at this time. Loss of retinal ganglion cells due to ocular hypertensive injury was determined by histological analysis of all retinas.

RESULTS

Ocular hypertensive retinal injury was associated with a loss of retinal ganglion cells. There was no histological or electrophysiological evidence for injury to any other retinal cell type. Correlation of electrophysiological response amplitudes with histological measures of retinal ganglion cell loss/survival yielded results which suggest that activity in retinal ganglion cells makes a substantial contribution to components of the 30 Hz flicker ERG, the flash VECP, and both first and second order multifocal ERG responses. Of the electrophysiological measures used in this study, multifocal ERG response amplitude had the greatest sensitivity to retinal ganglion cell loss.

CONCLUSIONS

Components of the multifocal ERG provide a sensitive measure of ganglion cell injury in a monkey model of chronic ocular hypertension. These same measures may have utility in the clinical diagnosis and management of glaucoma.

[Consequences of glaucoma on circadian and central visual systems]

Glaucoma is a chronic optic neuropathy leading to a degeneration of retinal ganglion cells. There is accumulating evidence that glaucomatous damage extends from retinal ganglion cells to vision centers in the brain. Degenerative changes are observed in magnocellular, parvocellular, and koniocellular pathways in the lateral geniculate nucleus, and these changes are related to intraocular pressure and the severity of optic nerve damage. In addition, recent studies show that there are also changes in the visual cortex in relation to varying degrees of retinal ganglion cell loss. In a rat model of glaucoma, we have recently demonstrated a reduction of retinal projections of retinal ganglion cells, not only on the visual system but also on the suprachiasmatic nucleus. Human studies suggest that the ganglion cell degeneration caused by glaucoma could lead to a lesion of the retinohypothalamic tract, which permits the synchronization of circadian rhythms.

Alpha-2 adrenergic receptor agonists are neuroprotective in experimental models of glaucoma

PURPOSE

The glaucomas are characterized by chronic progressive ganglion cell loss over many years. A drug with neuroprotective activity should increase the resistance of retinal ganglion cells (RGC) to chronic stress or injury and therefore enhance survival. Brimonidine is a highly selective and potent alpha-2 adrenergic receptor agonist, which lowers intra-ocular pressure (IOP) and is neuroprotective. Immunohistochemistry data have shown that the specific receptor targets, the alpha-2 receptors, are located in the inner retina.

METHODS

Brimonidine 0.1 mg/kg given intraperitoneally promoted RGC survival compared with vehicle using the optic nerve crush model even when administered up to 24 hours before injury. Using the chronic ocular hypertensive rat model, brimonidine 1 mg/kg/day (with osmotic pump) significantly prevented the loss of RGCs when compared with vehicle or timolol. This ability was due to the neuroprotective action of brimonidine, since it did not affect IOP. In addition, brimonidine 0.1 mg/day reached concentrations in the retina of Sprague-Dawley rats within 30 minutes of injection, which was sufficient to activate the alpha-2 receptor (> or = 2 nM) and maintained these concentrations for 6 hours.

CONCLUSIONS

Having demonstrated that: a. the specific receptor target of brimonidine is located in the retina, which is important for optic neuroprotection, b. the agent shows neuroprotective ability in animal models, c. pharmacological concentrations of the drug can be reached in the retina, clinical trial has been initiated to determine whether brimonidine is neuroprotective in patients with nonarteritic ischaemic optic neuropathy.

Differential effects of carbachol on calcium entry and release in CHO cells expressing the m3 muscarinic receptor

Calcium signalling was examined in CHO-k1 cells that stably express the m3 subtype of the muscarinic receptor. The calcium indicator Fura-2 was retained in these cells only in the presence of probenecid (1 mM), suggesting that Fura-2 efflux was mediated by an organic anion transporter. The addition of carbachol (CCh) to Fura-2 loaded cells in suspension caused a rapid transient increase in intracellular calcium [Ca]i followed by a smaller sustained plateau phase. The transient rise in [Ca]i was dose-dependent with a threshold response of 89 +/- 18 nM above baseline with 10 nM CCh and a maximum stimulation of 734 +/- 46 nM with 10 microM CCh. This phase was accompanied by a similar dose-dependent stimulation of total inositol phosphate production and was assumed to be generated by release from intracellular stores of the endoplasmic reticulum (ER). The sustained increase in [Ca]i was generated by entry from the extracellular bath since it was blocked by pretreatment with La3+ (1 microM) and was absent when bath calcium was chelated with EGTA. This phase was not dependent on CCh dose, and a stimulation of [Ca]i of approximately 90 nM above baseline was observed with CCh concentrations between 50 nM and 10 microM. With this dose range, the rate of Mn2+ quenching of Fura-2 at the Ca-insensitive excitation wavelength of 360 nm was likewise maximally stimulated. At lower CCh concentrations (10-50 nM), it was clear that the activation of Ca entry could not be dissociated from a threshold release of Ca from intracellular stores. The phorbol ester PMA, which uncouples the muscarinic receptor from phospholipase C, reduced the transient rise in [Ca]i by approximately 50% with little or no effect on Ca entry at higher CCh levels (> or = 1 microM). At lower CCh concentrations (< or = 100 nM) however, pretreatment with PMA completely blocked all Ca mobilization and supports the contention that Ca entry is coupled to Ca release from stores or to store depletion. The emptying of inositol trisphosphate-sensitive stores with thapsigargin (10 nM) stimulated Ca entry and also the rate of Mn2+ quenching. Store depletion by incubation in Ca-free media likewise stimulated Mn2+ uptake without a rise in [Ca]i. Our data are therefore consistent with a 'capacitative' coupling model, whereby the activation of the plasma membrane receptor leads to an InsP3-induced change in the degree of filling of the ER Ca pool.(ABSTRACT TRUNCATED AT 400 WORDS)

Characterization of muscarinic receptors in cultured human iris sphincter and ciliary smooth muscle cells

Muscarinic receptors present in cultured human iris sphincter and ciliary smooth muscle cells were characterized by both ligand ([3H]QNB binding) and functional (phosphoinositide hydrolysis) studies. Ligand binding studies showed that [3H]QNB represented a single population of binding sites with KD values of 4.02 x 10(-11) M in the ciliary and 5.6 x 10(-11) M in the iris sphincter cells. In competition studies, the selective antagonist, 4-diphenylacetoxy-N-methylpiperidine-methobromide (4-DAMP) was the most potent in displacing [3H]QNB with selectivity of 150-350-fold over pirenzepine (M1) and 450-1700-fold over AF-DX 116 (M2). 4-DAMP recognized one site in the iris sphincter cells (Ki = 0.34 nM) but two sites in the ciliary cells (KH = 0.9 nM and KL = 49 nM). 4-DAMP was also the most potent in inhibiting carbachol-induced hydrolysis of inositol phospholipids (PI) in both cell types. However, the IC50 values for PI hydrolysis were several fold lower than those for [3H]QNB binding. Using these selective antagonists, our data supports the presence of functional muscarinic receptors of M3 subtype in human iris sphincter and ciliary cells. It also shows the presence of a second low affinity site in the ciliary smooth muscle cells that is recognized by 4-DAMP.

Prostaglandin F2 alpha effects on intraocular pressure negatively correlate with FP-receptor stimulation

According to the current working classification for prostanoid receptors, the prostaglandin F2 alpha-sensitive receptor (FP-receptor) may be identified by comparing the rank order of activity of prostaglandin F2 alpha (PGF2 alpha) and its analogues. In order to further understand the pharmacology of PGF2 alpha-induced ocular hypotension, the intraocular pressure response to PGF2 alpha and selected analogues was compared with their rank order of activity in typical FP-receptor preparations such as contraction of the cat iris sphincter and affinity for corporal luteal membrane binding sites. The rank order of potency for decreasing intraocular pressure was as follows: PGF2 alpha greater than PGF1 alpha greater than 16-phenoxytetranor PGF2 alpha greater than 17-phenyltrinor PGF2 alpha = fluprostenol (inactive). For cat iris sphincter contraction, the rank order of potency appears to be fluprostenol = 17-phenyltrinor PGF2 alpha greater than 16-phenoxytetranor PGF2 alpha = PGF2 alpha greater than PGF1 alpha. The rank order of potency for PGF2 alpha analogues in decreasing intraocular pressure appears to negatively correlate with the rank order for cat iris sphincter contraction and literature values for corporal luteal membrane binding. It is concluded that the ocular hypotensive effect of PGF2 alpha is not mediated by the FP-receptor.

Identification of variants of the basophilic leukemia (RBL-2H3) cells that have defective phosphoinositide responses to antigen and stimulants of guanosine 5'-triphosphate-regulatory proteins

Antigen immunoglobulin E-mediated secretion of histamine from RBL-2H3 cells is associated with substantial hydrolysis of membrane inositol phospholipids and a rise in the concentration of cytosol Ca2+ (calcium signal). Such responses differed among cloned variant lines of the RBL-2H3 cell line from undetectable (1A3 bromodeoxyuridine-resistant (BUDRR), 2B1 BUDRR, and 1B3 BUDRR lines) to about 80% of those in the parent RBL-2H3 cells. In all but one line (1B3 thioguanine-resistant (TgR)), the intensities of the phosphoinositide response and of the calcium signal were correlated with the secretory response. The 1B3 TgR line had no detectable calcium signal (as measured by quin 2 fluorescence or uptake of 45Ca2+) but paradoxically showed modest rates of hydrolysis of inositol phospholipids and of secretion. The responses of the 1B3 TgR line were, however, dependent on the presence of external Ca2+ ions. The induction of secretion with antigen, therefore, was invariably associated with the hydrolysis of inositol phospholipids, but it was not necessarily associated with a change in concentration of cytosol Ca2+. All antigen unresponsive clones could secrete when synergistic signals were induced by exposure to the Ca2+ -ionophore, A23187 and the phorbol ester, 12-O-tetradecanoylphorbol 13-acetate. These lines, otherwise, had immunoglobulin E receptors and had no obvious defect in their capacity to synthesize the inositol phospholipids or in their phenotypic expression of phospholipase C as measured in cell extracts. One finding of possible relevance to the role of guanosine 5'-triphosphate-regulatory proteins in the activation of phospholipase C was the inability of one antigen-nonresponsive line to respond to NaF (in intact cells) or to guanosine 5'-(3-O-thio)triphosphate (in electrically permeabilized cells).

Identification of variants of the basophilic leukemia (RBL-2H3) cells that have defective phosphoinositide responses to antigen and stimulants of guanosine 5'-triphosphate-regulatory proteins

Antigen immunoglobulin E-mediated secretion of histamine from RBL-2H3 cells is associated with substantial hydrolysis of membrane inositol phospholipids and a rise in the concentration of cytosol Ca2+ (calcium signal). Such responses differed among cloned variant lines of the RBL-2H3 cell line from undetectable (1A3 bromodeoxyuridine-resistant (BUDRR), 2B1 BUDRR, and 1B3 BUDRR lines) to about 80% of those in the parent RBL-2H3 cells. In all but one line (1B3 thioguanine-resistant (TgR)), the intensities of the phosphoinositide response and of the calcium signal were correlated with the secretory response. The 1B3 TgR line had no detectable calcium signal (as measured by quin 2 fluorescence or uptake of 45Ca2+) but paradoxically showed modest rates of hydrolysis of inositol phospholipids and of secretion. The responses of the 1B3 TgR line were, however, dependent on the presence of external Ca2+ ions. The induction of secretion with antigen, therefore, was invariably associated with the hydrolysis of inositol phospholipids, but it was not necessarily associated with a change in concentration of cytosol Ca2+. All antigen unresponsive clones could secrete when synergistic signals were induced by exposure to the Ca2+ -ionophore, A23187 and the phorbol ester, 12-O-tetradecanoylphorbol 13-acetate. These lines, otherwise, had immunoglobulin E receptors and had no obvious defect in their capacity to synthesize the inositol phospholipids or in their phenotypic expression of phospholipase C as measured in cell extracts. One finding of possible relevance to the role of guanosine 5'-triphosphate-regulatory proteins in the activation of phospholipase C was the inability of one antigen-nonresponsive line to respond to NaF (in intact cells) or to guanosine 5'-(3-O-thio)triphosphate (in electrically permeabilized cells).

Manoalide, a natural sesterterpenoid that inhibits calcium channels

Manoalide is a marine natural product that has anti-inflammatory and anti-proliferative activities and is an irreversible inhibitor of phospholipase A2 and phospholipase C. It is now shown that the compound is a potent inhibitor of Ca2+ mobilization in several cell types. In A431 cells the increase in epidermal growth factor receptor-mediated Ca2+ entry and release from intracellular Ca2+ stores were blocked by manoalide in a time-dependent manner with an IC50 of 0.4 microM. The effect of manoalide on phosphoinositide metabolism, namely the production of inositol monophosphate, did not coincide with its effect on the epidermal growth factor response. In GH# cells, manoalide blocked the thyrotropin-releasing hormone-dependent release of Ca2+ from intracellular stores without inhibition of the formation of inositol phosphates from phosphatidylinositol 4,5-bisphosphate. Manoalide also blocked the K+ depolarization-activated Ca2+ channel in these cells as well as the activation of the channel by Bay K8644 with an IC50 of 1 microM. In addition, manoalide also inhibited the Ca2+ influx induced by concanavalin A in mouse spleen cells in a time- and temperature-sensitive manner with an IC50 of 0.07 microM. However, neither forskolin-activated adenylate cyclase in A431 cells nor the distribution of the potential sensitive dye, 3,3'-dipropylthiodicarbocyanide iodide in GH3 cells was affected by manoalide. Thus, manoalide acts as a Ca2+ channel inhibitor in all cells examined. This action may account for its effects on inflammation and proliferation and may be independent of its effect on phospholipases.

Changes in histidine uptake and histamine synthesis during the growth cycle of rat basophilic leukemia (2H3) cells

Rat basophil leukemia (2H3) cells, like normal rat peritoneal mast cells, were shown to take up and decarboxylate histidine. Uptake was mediated by a temperature-dependent system with high affinity (apparent Km 24 +/- 4 microM) for histidine. Newly formed histamine was incorporated into the intracellular pool of histamine. In confluent cultures, substantial amounts of histamine were lost to the medium while intracellular histamine levels remained constant. As calculated from the rate of appearance of histamine in the medium, the maximum turnover time for the intracellular histamine pool (2-7 nmol/10(6) cells) was about 12 hr. Variation in histamine content, histidine uptake and histidine decarboxylation was noted with different passages of 2H3 cells, but with all passages there were characteristic changes in these parameters during growth and division of the cells. Separation of cells into fractions of different size by elutriation indicated low rates of histidine uptake and decarboxylation in the smallest 2H3 cells, a progressive increase in ability to take up and decarboxylate histidine as the cells approached the S phase of growth and marked decline in this ability in fractions containing the larger cells. The changes in kinetic constants suggested that fluctuation in histidine uptake during the life cycle of the 2H3 cell was due to changes in the number of active carriers or sites of histidine transport and that during cell division all components associated with histamine synthesis (i.e., histidine uptake and decarboxylation) were diminished.

Changes in histamine synthetic activity, histamine content and responsiveness to compound 48/80 with maturation of rat peritoneal mast cells

Intact rat peritoneal mast cells were separated by elutriation into fractions differing in cell size, the smallest cells (less than 12 microns) exhibited high histamine synthetic activity (200-300 pmol of histidine decarboxylated per hr/10(6) cells) and contained little histamine (less than 2 pg/cell). With increasing cell size, histamine synthetic activity diminished to less than 20 pmol/hr/10(6) cells and histamine content increased to greater than 12 pg/mast cell. A gradation in histamine release in response to Compound 48/80 was also observed; small mast cells were resistant, those of intermediate size (12-13 microns diameter) were partially responsive and large mast cells (14-17 microns) were fully responsive to Compound 48/80. These changes were related to the state of maturity of the mast cells as indicated by their histochemical reactions to Alcian blue and safranin, which stain nonsulfated heparin precursors and heparin, respectively. Fractions of small mast cells were predominantly immature cells (70-85% of the cells were Stage I) with few granules, whereas fractions of large cells contained mostly mature cells (greater than 80% Stages III and IV) with numerous heparin-containing granules. Fractions in between contained mast cells at intermediate stages (Stages II and III) of development. Degenerative changes, namely decreased viability and diminished responsiveness to Compound 48/80, were evident, however, in fractions of the largest cells (greater than 17 microns). This and other work indicate that, even in adult rats, peritoneal mast cells are heterogeneous and can be separated on the basis of size into subpopulations that differ markedly in histamine content, histamine synthetic activity and ability to release histamine in response to Compound 48/80.